close

Вход

Забыли?

вход по аккаунту

?

Target-Specific Microwave Antenna Optimization for Pre-Clinical and Clinical Bladder Hyperthermia Devices

код для вставкиСкачать
 Target‐Specific Microwave Antenna Optimization for Pre‐Clinical and Clinical Bladder Hyperthermia Devices by Sara Salahi Department of Biomedical Engineering Duke University Date:_______________________ Approved: ___________________________ Mark W. Dewhirst, Supervisor ___________________________ Paul R. Stauffer ___________________________ Paolo F. Maccarini ___________________________ James R. MacFall ___________________________ Zeljko Vujaskovic ___________________________ Kathryn R. Nightingale Dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Biomedical Engineering in the Graduate School of Duke University 2012
UMI Number: 3519100
All rights reserved
INFORMATION TO ALL USERS
The quality of this reproduction is dependent on the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if material had to be removed,
a note will indicate the deletion.
UMI 3519100
Copyright 2012 by ProQuest LLC.
All rights reserved. This edition of the work is protected against
unauthorized copying under Title 17, United States Code.
ProQuest LLC.
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, MI 48106 - 1346
UMI Number: 3519100
All rights reserved
INFORMATION TO ALL USERS
The quality of this reproduction is dependent on the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if material had to be removed,
a note will indicate the deletion.
UMI 3519100
Copyright 2012 by ProQuest LLC.
All rights reserved. This edition of the work is protected against
unauthorized copying under Title 17, United States Code.
ProQuest LLC.
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, MI 48106 - 1346
ABSTRACT Target‐Specific Microwave Antenna Optimization for Pre‐Clinical and Clinical Bladder Hyperthermia Devices by Sara Salahi Department of Biomedical Engineering Duke University Date:_______________________ Approved: ___________________________ Mark W. Dewhirst, Supervisor ___________________________ Paul R. Stauffer ___________________________ Paolo F. Maccarini ___________________________ James R. MacFall ___________________________ Zeljko Vujaskovic ___________________________ Kathryn R. Nightingale An abstract of a dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Biomedical Engineering in the Graduate School of Duke University 2012 Copyright by Sara Salahi 2012 Abstract
We have yet to establish the optimum combination of hyperthermia with radiation and/or chemotherapy for effective treatment of bladder cancer. Convenient and affordable microwave applicators capable of well‐localized non‐invasive heating of murine, canine and human bladder cancers is essential for logical progression of studies from pre‐clinical to multi‐institution clinical trials, as needed to investigate the effects of hyperthermia as an adjuvant treatment for bladder cancer. The primary objective of this research was to utilize state‐of‐the art segmentation and simulation software to optimize target‐specific microwave antennas for more uniform heating in pre‐clinical and clinical investigations of bladder hyperthermia. The results of this research are: 1. The development of a reliable simulation‐based approach for optimizing microwave applicators; 2. The design, construction and testing of an applicator for heating murine bladder to 40‐43°C while maintaining surface and core temperatures normothermic; 3. The optimization, construction and testing of a fundamentally different type of antenna (metamaterial) for heating pediatric and/or canine bladder; iv
4. A preliminary effort towards the optimization, construction and testing of metamaterial antennas for heating adult bladder. One significant implication of this work is to enable essential pre‐clinical bladder hyperthermia studies with the development of a reliable microwave applicator for heating murine bladder to 40‐43˚C while maintaining surface and core temperatures normothermic. It is clear that hyperthermia enhances the effects of chemo‐ and radio‐ therapies, and this device will allow scientists to investigate the basic principles underlying this phenomenon more systematically. Another significant contribution of this work is the development of metamaterial antennas for deep tissue hyperthermia. These antennas decrease the cost and increase the comfort and portability of bladder hyperthermia devices. These improvements will enable the multi‐institutional clinical trials required to apply for insurance reimbursement of deep‐tissue thermal therapy and the subsequent widespread use of hyperthermia as an adjuvant to current cancer therapies.
v
Contents Abstract ......................................................................................................................................... iv List of Tables ................................................................................................................................. ix List of Figures ................................................................................................................................ x Acknowledgements ...................................................................................................................xiv 1. Introduction ............................................................................................................................... 1 1.1 Urinary Bladder Anatomy and Bladder Cancer .......................................................... 1 1.2 Bladder Cancer Epidemiology and Etiology ................................................................ 3 1.3 Bladder Cancer Therapy .................................................................................................. 6 1.3.1 Surgery and Radiation ................................................................................................ 6 1.3.2 Chemotherapy and Intravesical Therapy ................................................................ 9 1.4 Bladder Hyperthermia and Radiofrequency (RF) Devices ....................................... 10 1.4.1 Metamaterial Antennas ............................................................................................ 13 1.5 Temperature‐Sensitive Liposomes ............................................................................... 15 1.6 Pre‐Clinical Bladder Hyperthermia Studies ............................................................... 16 1.7 Overview ......................................................................................................................... 18 1.7.1 Objectives .................................................................................................................... 18 1.7.2 Significance ................................................................................................................. 19 2. Antenna Optimization for Medical Device Applications .................................................. 23 2.1 Electromagnetic Wave Equations ................................................................................. 23 2.2 Finite Element Method .................................................................................................. 24 vi
2.2.1 Galerkin Method ........................................................................................................ 27 2.2.2 Elemental Approach for Multi‐Dimensions .......................................................... 29 2.3 Optimizing Specific Absorption Rate (SAR) Pattern and S Parameters ................. 29 2.4 Computational Models for Patients and Animals ..................................................... 32 2.5 Thermal Simulations for Hyperthermia Applications .............................................. 34 3. Devices for In Vivo Murine Hyperthermia Studies ............................................................ 37 3.1 Murine Bladder Hyperthermia ..................................................................................... 38 3.1.1 Methods ...................................................................................................................... 39 3.1.1.1 Antenna Design Optimization ......................................................................... 39 3.1.1.2 Thermal Modeling ............................................................................................. 42 3.1.1.3 S Parameter and SAR Validation ..................................................................... 43 3.1.1.4 In Vivo Murine Bladder Heating Studies ........................................................ 44 3.1.2 Results ......................................................................................................................... 45 3.1.3 Discussion ................................................................................................................... 54 3.1.4 Conclusions and Future Work ................................................................................. 59 3.2 Murine Brain Hyperthermia ......................................................................................... 61 3.2.1 Methods ...................................................................................................................... 62 3.2.1.1 Accounting for the Thermal Effects of the Superior Sagittal Sinus ............ 63 3.2.1.2 Phantom Studies for Thermometry ................................................................. 64 3.2.1.3 In Vivo Murine Brain Tumor Heating Studies ............................................... 66 3.2.2 Results ......................................................................................................................... 67 3.2.3 Discussion ................................................................................................................... 74 vii
3.2.4 Conclusions and Future Work ................................................................................. 77 4. Metamaterial Antennas for Pediatric Bladder Hyperthermia .......................................... 78 4.1 Metamaterial Antenna Design ...................................................................................... 79 4.1.1 Benefits of this Design .............................................................................................. 85 4.2 Antenna Optimization for Pediatric Bladder Hyperthermia ................................... 87 4.2.1 Average Patient Model ............................................................................................. 88 4.2.2 EM‐Thermal Simulation‐Based Optimization ....................................................... 89 4.2.2.1 One vs. Two Antennas ...................................................................................... 95 4.2.3 Experimental Validation of Metamaterial Antennas ........................................... 96 4.3 Expected Antenna Performance for Pediatric Bladder Heating ............................ 100 4.4 Conclusions and Future Work .................................................................................... 102 5. Metamaterial Antennas for Bladder Cancer Hyperthermia Treatments ...................... 104 5.1 Single Antenna vs Phased Array Heating ................................................................ 104 5.2 Shielding ........................................................................................................................ 111 5.3 Conclusions and Future Work .................................................................................... 112 Concluding Remarks ................................................................................................................ 114 Appendix B: Electromagnetic Simulations ............................................................................ 131 Appendix C: Coupled EM‐Thermal Simulations ................................................................. 134 Appendix D: Phantom Formulations ..................................................................................... 143 Works Cited ............................................................................................................................... 146 Biography ................................................................................................................................... 158 viii
List of Tables
Table 1.1 Stage Distribution and 5‐year Relative Survival by Stage at Diagnosis for 2001‐
2007, All Races, Both Sexes .......................................................................................................... 4 Table 3.1 Properties of murine tissues at 2.45GHz(57) ............................................................. 40 Table 3.2 Properties of murine tissues at 4.2GHz(57, 86) ........................................................... 63 Table 5.1 Thermal properties of human tissues(101, 102) ......................................................... 108 Table 5.2 Backward radiation levels measured 5cm from the back of the antenna with input power of 10W compared to FCC Requirements ........................................................ 111 ix
List of Figures
Figure 1.1 Anatomical rendering of an adult urinary bladder ............................................... 1 Figure 1.2 Transitional epithelium of urinary bladder; 1 – Free surface, 2 – Transitional epithelial layer, 3 – Nucleus, 4 – Basement membrane ........................................................... 2 Figure 1.3 Synergo® endovesical bladder heating technology ............................................ 11 Figure 1.4 BSD technology for non‐invasive bladder heating .............................................. 12 Figure 2.1 S parameter definition for a two‐port network .................................................... 31 Figure 2.2 Example of S parameter plot ................................................................................... 31 Figure 2.3 Thermal map of mouse without effects of anesthesia ......................................... 36 Figure 3.1 a) Parameters varied to optimize SAR and S11; b) Front‐view picture of 2.45GHz applicator; c) Bottom‐view picture of 2.45GHz applicator; d) HFSS™ model of applicator on mouse; e) Snapshot of experimental setup for murine bladder hyperthermia study with applicator positioned on the skin overlying the pelvis. ........... 46 Figure 3.2 S11: HFSS™ simulation compared to measurements using an Agilent E5071C network analyzer. ....................................................................................................................... 47 Figure 3.3 SAR: HFSS™ simulation compared to measurements with all values normalized to the maximum value at 3mm depth: a) profile across x‐axis, b) profile across y‐axis, c) depth profile, and d) 3D SAR measurements in muscle phantom .......... 49 Figure 3.4 HFSS™ SAR simulation in mouse model a) x‐axis profile and b) y‐axis profile
....................................................................................................................................................... 50 Figure 3.5 2D slice temperature profiles in mouse body, bladder, vagina and rectum with bolus temperature (a) unregulated (b) constant 41oC, (c) constant 38oC and (d) constant 35oC and RF power set at (a) 5W and (b), (c), (d) 12W. .......................................... 51 Figure 3.6 COMSOL simulation of average temperatures as a function of bolus temperature .................................................................................................................................. 52 Figure 3.7 Temperature data taken from five independent murine bladder hyperthermia studies. Temperature increases occur when RF power is turned on (around the 2‐
x
5minute mark) and decreases occur when RF power is turned off (around the 18‐
20minute mark). The dotted line signifies 42˚C. ..................................................................... 53 Figure 3.8 a) Schematic of brain phantom setup with temperature sensor channel placement, b) picture of temperature sensor channels with window cutouts, c) picture of final box with brain phantom, Mylar® window and antenna placed on top .................... 65 Figure 3.9 Cranial Window Model ........................................................................................... 66 Figure 3.10 Prototype 4.2GHz applicator ................................................................................ 68 Figure 3.11 S11: HFSS™ simulation compared to measurements using an Agilent E5071C network analyzer. ....................................................................................................................... 68 Figure 3.12 HFSS™ SAR simulation in mouse model a) x‐axis profile and b) y‐axis profile ............................................................................................................................................ 69 Figure 3.13 COMSOL simulation of 2D temperature profiles in mouse brain and brain tumor undergoing hyperthermia treatment a) with the effects of the sagittal sinus and b) without the effects of the sagittal sinus .................................................................................... 70 Figure 3.14 Change in temperature of the brain phantom as measured in four locations before and after 22W is applied ................................................................................................ 71 Figure 3.15 Zeiss 780 confocal microscope images taken of mouse brain through cranial window model a) after free doxorubicin injection, b) after doxorubicin‐loaded TSL injection, and c) after doxorubicin‐loaded TSL injection with heat treatment. Scale on images indicates 100μm. ............................................................................................................ 74 Figure 4.1 Schematic of Metamaterial Antenna ...................................................................... 79 Figure 4.2 Antenna slot and rectangular patch dimensions in terms of effective wavelength ................................................................................................................................... 82 Figure 4.3 At the magnitude S21 minimum (860MHz), a positive phase slope and negative group delay are observed (highlighted in red), while the antenna is impedance matched across the 200‐500MHz band (highlighted in green). ............................................ 84 Figure 4.4 (a) The metamaterial antenna can have improved power deposition at depth when compared to the equivalent waveguide applicator; (b) the focus point at depth can be manipulated by varying the antenna geometry – in this example, Layer 1 thickness xi
has been varied to move the location of the focus point; (c) the equivalent of (b) except that these images reveal the SAR pattern instead of E‐field. ................................................ 86 Figure 4.5 (a) Fat layer thickness, bladder isocenter depth and volume as determined by 40 patient CT and Ultrasound images; (b) 3D average patient model created by image segmentation in VSG Avizo® and imported in HFSS™. ...................................................... 89 Figure 4.6 Reflected power of the 915MHz antenna on the child model, with and without a water cooling bolus .................................................................................................................. 90 Figure 4.7 SAR pattern in pediatric bladder ........................................................................... 91 Figure 4.8 (a) SAR pattern in the virtual phantom study model; (b) Percent power deposited in the bladder phantom as distance between the phantom and the Mylar window is increased ................................................................................................................... 93 Figure 4.9 Plot of temperatures in the virtual phantom study model. The fat phantom temperature is maintained between the cooling bolus and muscle phantom temperatures while the bladder phantom is heated to 41.8°C. The arrows indicate flow pattern in the circulated muscle phantom that is maintained at 37°C. ........................................................ 94 Figure 4.10 Plot of T90 in the bladder phantom as a function of power for the case of 1 antenna vs. 2 antennas ............................................................................................................... 96 Figure 4.11 Comparison of simulated reflected power and measured reflected power on 3 independent subjects ............................................................................................................... 97 Figure 4.12 (a) Picture of laboratory phantom study setup; and (b) data from a phantom study where power was turned on at the 8 minute mark and turned off at the 45 minute mark (approximately 10 minutes after steady state was reached) ....................................... 99 Figure 4.13 Steady‐state temperature profile in two cross‐sections (xy and xz) across the center of the average child model bladder after two‐antenna heating ............................. 101 Figure 4.14 (a) Avizo® model of a golden retriever; (b) HFSS model of beagle with tumor (red) on bladder wall (yellow) ................................................................................................ 103 Figure 5.1 Picture of 915MHz (4x5cm) and 180MHz (12x20cm) metamaterial antennas with US$0.25 (to indicate scale) and strap to secure antennas around torso ................... 105 xii
Figure 5.2 Comparison of simulated reflected power and measured reflected power on 3 independent subjects ................................................................................................................ 106 Figure 5.3 Simulated vs. measured data for the low frequency metamaterial antenna show good agreement .............................................................................................................. 107 Figure 5.4 (a) Conformal model of 180MHz metamaterial antenna array in HFSS (single antenna model not shown, but placed directly above bladder); (b) temperature profile of single antenna heating in COMSOL; (c) temperature profile of four‐antenna array heating in COMSOL ................................................................................................................. 110 xiii
Acknowledgements
First and foremost, I would like to thank my best friend and soon‐to‐be husband Tyler Louie for his constant support in my personal and professional endeavors. I would also like to thank his family, Cheryl, Wei and Tamara Louie, for all the advice and good times we shared – they helped me stay grounded and were the key ingredient to my success in graduate school. I would like to extend my deepest gratitude to my primary advisor Professor Mark Dewhirst for allowing me the opportunity to “start over” on a more positive and productive team that works closely with medical professionals to design, build and test hyperthermia and radiometric devices. My thesis would not be as impressive and my time here would not have been as meaningful without the wisdom and kindness bestowed upon me by my secondary advisors, Professor Paul Stauffer and Assistant Professor Paolo Maccarini. From the nights we spent grant‐writing to the days spent perfecting our experiments, they have always allowed me to get in there with them – to learn even if it meant getting my hands dirty and to make important, tangible contributions to our team’s work. I will forever be grateful for these experiences shared. Not only do they develop expert engineers, my advisors maintain a wonderful “professional family” here at work. I believe that’s partly due to their caring and xiv
supportive wives, so I would be remiss in not thanking Deborah Stauffer, Irina Maccarini and Nancy Dewhirst. Biomedical Engineering PhD students are required to have 5 people on their committee, but I have 6 – all of whom care about my professional and academic development. I would like to thank my entire committee for their dedication to my advancement: Professors Mark Dewhirst, Paul Stauffer, James MacFall and Kathy Nightingale, Assistant Professor Paolo Maccarini and Dr. Zeljko Vujaskovic. The Hyperthermia Physics Lab is always full of the best people from all over the world. Here’s a big THANK YOU to the group that made working in our lab a remarkable experience: Drs. Kavitha Arunachalam, Cory Wyatt, Yu Yuan, Yngve Birkelund, Oystein Klemetsen, Erdem Topsakal, Satoru Kikuchi, Alessandro Di Carlofelice, and Valeria De Luca, Tiago Oliveira, Fabio D’Isidoro, Dario Rodrigues, Titania Juang, Sneha Rangarao and Alina Boico. My research projects involved a lot of teamwork, and for this reason I have a lot of people to thank. For the murine bladder experiments, I thank Dr. Brant Inman, Wiguins Etienne, Chelsea Landon and Dr. Katherine Hansen. For the murine brain experiments, I thank Drs. Gerald Grant and Christy Wilson. For the VUR detection project, I thank the folks at ThermImage, Inc: Doug Reudink, Doug Turnquist and Dr. Brent Snow. For the metamaterial antenna‐based projects, I thank Dr. Gerard Aknine. xv
Special thanks also to Don Pearce for the hours we spent in the Medical Instrument Shop to make the medical devices I designed a reality – “just do it.” Before arriving to Duke, I had very supportive mentors who inspired and encouraged me to pursue my PhD. I’d like to thank Drs. Allen Taflove, Alan Sahakian and Xu Li at Northwestern University and Lynn Davenport at Medtronic, Inc. Without their guidance and instruction, I would have never thought to pursue a doctorate in engineering. And lastly, I would like to thank my father, Dr. Farouk Salahi, for showing me that success is loving what you do, and my mother, Maral Salahi, whose “stubbornness” may be what I call “determination” when I describe myself. Research was supported by NIH grant CA42745‐21‐23, ThermImage Inc., and an NSF Graduate Research Fellowship. Software was supported by Ansys, COMSOL and VSG.
xvi
1. Introduction
1.1
1 Urinarry Bladde
er Anatom
my and Bladder Ca
ancer
Figure
e 1.1 is an an
natomical reendering of aan adult urin
nary bladderr, which is aa hollow and elastic organ that collectss urine via th
he ureters an
nd passes urrine via the urethra. The b
bladder walll is made of the detruso
or muscle and
d the trigonee, a smooth reegion between the two u
ureteral orifiices. When th
he trigone iss stretched to
o a certain degree, the parasympath
hetic nervouss system is siignaled to co
ontract the d
detrusor mu
uscle an
nd expel thee urine. Both
h the autonom
mically conttrolled intern
nal sphincteer, located in
n the bladder neck,, and the volluntarily con
ntrolled exteernal sphinctter must be opened to ex
xpel he urine. th
Figure 1
1.1 Anatomiical renderin
ng of an adu
ult urinary b
bladder 1
The m
mucosal linin
ng of the blad
dder wall, u
ureters and u
urethra is traansitional ep
pithelium (F
Figure 1.2). T
Transitional epithelium consists of m
multiple layeers of cells th
hat ca
an contract a
and expand to accommo
odate for chaanges in urin
ne volume capacity. 95%
% of bladder canceers arise in the transition
nal epitheliu
um and are k
known as Trransitional C
Cell Carcinoma (T
C
TCC) (1). Wh
hen TCC is o
only found in
n the transittional epitheelial layer, it is cllassified as S
Stage 0 bladd
der cancer.
Fig
gure 1.2 Tran
nsitional epithelium of urinary blaadder; 1 – Free surface, 22 – Tra
ansitional ep
pithelial lay
yer, 3 – Nuclleus, 4 – Bassement mem
mbrane The la
ayer underneeath the tran
nsitional epitthelium is k
known as thee basement membrane. If
m
f TCC has in
nvaded the b
basement meembrane, bu
ut has not pro
oliferated in
nto su
urrounding tissues, it is considered carcinoma i n situ and iss classified as Stage 1 blaadder ca
ancer. When
n TCC invad
des the muscles in the blaadder wall ((Stage 2) and
d the fatty tisssue, prostate gland, vagina orr uterus (Stag
ge 3) but nott to the lymp
ph nodes or other organ
ns, it iss considered
d muscle‐invasive bladdeer cancer. 2
Once cancer has proliferated to the lymph nodes, pelvic or abdominal wall, and/or other organs, it is considered to be metastatic cancer and is classified as Stage 4 bladder cancer. Bladder cancer is classified as recurrent when it appears again in the urinary bladder or another nearby organ after having been treated. 1.2 Bladder Cancer Epidemiology and Etiology
This section is a summary of facts found in the American Cancer Society’s document on Bladder Cancer published in 2011 (1). In 2011, it is estimated that 69,250 new cases of bladder cancer (52,020 men and 17,230 women) will be diagnosed in the US. With over 500,000 bladder cancer survivors in the US, around 14,990 deaths will result from bladder cancer this year. 90% of bladder cancer cases occur in people over the age of 55, with the average age at time of diagnosis at 73 years old. Men have a 1 in 26 change of developing bladder cancer during their life and are 3 times more likely to get bladder cancer than women. Bladder cancer is the 4th most common cancer diagnosed in men. Whites are diagnosed with bladder cancer twice as often as blacks, and Hispanics have an even lower rate than blacks. However, blacks are more likely than whites to have more advanced disease at time of diagnosis. Around 50% of cases are diagnosed while the bladder cancer is confined to the inner layer of the bladder (Stage 0 and Stage 1). 35% of cases are diagnosed after the 3
bladder cancer has invaded into deeper layers but is still contained in the bladder (Stage 2). Most of the remaining cases are diagnosed at Stage 3, where the cancer has spread to nearby tissues. It is rare (~4%) for this cancer to be diagnosed after it has spread to distant sites. 5‐year survival rate is defined as the percentage of patients who live at least 5 years after their cancer is diagnosed. The 5‐year relative survival rate assumes that some people will die of other causes and compares the observed survival with that expected for people without the cancer. The latter is considered to be a more accurate measure of the chances of dying from a particular type and stage of cancer. The following statistics for 5‐year relative survival rates are based on SEER’s data collected over thousands of patients during 2001‐2007. The overall 5‐year survival rate was 78.1%; 80.1% for white men, 73.8% for white women, 70.3% for black men and 54% for black women. Table 1.1 reveals the stage distribution (a measure of disease progression) and 5‐year relative survival rates based on stage at diagnosis. Table 1.1 Stage Distribution and 5‐year Relative Survival by Stage at Diagnosis for 2001‐2007, All Races, Both Sexes Stage at Diagnosis In situ (stage 0 and 1) Localized (stage 2) Regional (stage 3) Distant (stage 4) Unknown Stage Distribution (%) 5‐year Relative Survival (%) 51 35 7 4 3 96.6 70.7 34.6 5.4 49.1 4
The probability that an individual will develop bladder cancer and/or survive after receiving a diagnosis of bladder cancer depends on his/her previous, current and future exposure to many risk factors. The most important risk factor for bladder cancer is smoking. Smokers are more than twice as likely to get bladder cancer as nonsmokers. Smoking also causes about half of the deaths from bladder cancer among men and a third among women. Certain industrial chemicals found in some workplaces are also linked with bladder cancer. The industries that carry the highest risks include makers of rubber, leather, textiles, and paint products. Workers are also at an increased risk if they use products such as paint or hair dyes, or inhale fumes such as diesel exhaust (e.g. truck drivers) in their everyday work. Race, ethnicity, age and gender also contribute to a person’s risk. For example, whites are twice as likely to develop bladder cancer, and men have a much higher risk than women. The risk of developing bladder cancer increases with age; those over the age of 55 have the highest risk. People who experience chronic bladder irritation and infections, or those who have been diagnosed with any kind of urothelial carcinomas in the past also have an increased risk for developing bladder cancer. Patients who have bladder‐related birth 5
defects, such as exstrophy, or those who have family members with bladder cancer are at an increased risk for bladder cancer. Long‐term uses of some chemotherapy drugs and/or radiation treatments to the pelvic region are also linked to increases in risk of developing bladder cancer. And finally, not drinking enough fluids daily or drinking fluids that contain high levels of arsenic have been found to increase the risk of bladder cancer. 1.3 Bladder Cancer Therapy
This section outlines the most common methods for treating bladder cancer. These methods include surgery, radiation, intravesical therapy and chemotherapy. In nearly all cases, a combination of these treatment methods is used. Bladder hyperthermia, increasing the temperature of the bladder region by 4‐5˚C, is being explored as an adjuvant to radiation, intravesical therapy and chemotherapy. In the case of chemotherapy, hyperthermia treatments are being explored in conjunction with drugs that are packaged in temperature‐sensitive liposomes as a means to achieve localized drug delivery. 1.3.1 Surgery and Radiation
Two types of surgery are used in the treatment of bladder cancer. For early stage or superficial bladder cancers, a transurethral resection (TUR) of the bladder tumor is most common. “Transurethral” refers to method in which a bladder tumor is accessed. 6
This procedure is done with an instrument that is passed through the urethra rather than cutting through the abdomen to access the bladder. After a TUR procedure, other steps may be taken to ensure the tumor is completely destroyed. Burning the base of the tumor (fulguration) and/or applying a high‐energy laser using a cystoscope are two examples of common post‐TUR procedures. Side effects of this procedure are considered to be generally mild (short‐term bleeding and pain during urination). However, recurrence is common and the use of fulguration alone may be used in patients with multiple recurrences to avoid extensive damage to the bladder. The second type of surgery is for invasive bladder cancer is cystectomy, in which all or part of the bladder is removed. All cystectomy procedures require accessing the bladder via incisions made into the abdomen. A partial cystectomy, the removal of a piece of the bladder and nearby lymph nodes, is done in a small portion of cases when the cancer has invaded the muscle but is localized and not very large. The main concern with this type of surgery is that bladder cancer can still recur in another part of the bladder wall. When the cancer is larger or is not localized to one part of the bladder, a radial cystectomy is required. This procedure removes the entire bladder and nearby lymph nodes. In men, the prostrate is also removed. In women, the ovaries, fallopian tubes, uterus and small portion of the vagina are often removed with the bladder. The 7
complications and side effects of this procedure can be severe. If the whole bladder is removed, reconstructive surgery is necessary to create a reservoir connected to the front of the abdomen by an opening or inside the abdomen with valve access on the skin. In both cases the patient must manually empty the reservoir once it is full. A new bladder that functions similar to normal bladders can also be created often out of a piece of intestine. Side effects of these procedures include pain, blood clots, infections, incontinence, pouch stones and blockage of urine flow. In all cases of radial cystectomy, the patient is rendered infertile and oftentimes sexual function is also impaired. The physical changes that result from cystectomy can have major emotional and psychological impacts as well. Due to the severe complications and side effects associated with cystectomy, some urologists prefer bladder preservation using TUR of the bladder cancer along with radiation therapy and chemotherapy. External‐beam radiation is the type of radiation that is most often used to treat bladder cancer. The procedure is painless and only lasts a few minutes, with treatments administered 5 days a week for several weeks. Radiation is often used in conjunction with other bladder cancer treatments. It is the main treatment for those who can’t have surgery, and it is also used in palliative care for those with advanced bladder cancers. Side effects include skin changes in areas that receive radiation, nausea and vomiting, 8
burning with urination, blood in the urine, diarrhea, fatigue, and a compromised immune system that increases risk of infection. 1.3.2 Chemotherapy and Intravesical Therapy
Systemic chemotherapy is an anticancer drug treatment that is given either in pill form or by injection. It is used to treat bladder cancer, especially in the case where the bladder cancer has already progressed to invasive cancer and/or metastatic cancer (Stages 2‐4). Chemotherapy is often used before or after surgery to shrink large tumors or kill any remaining cancer cells as a method for preventing recurrence. Sometimes it is used in combination with radiation therapy to make the radiation more effective (although side effects are also increased). In this case, the most common drugs used are Cisplatin and Mitomycin. If chemotherapy is administered before surgery or radiation, it is referred to as neoadjuvant therapy. If administered after surgery or radiation, it is referred to as adjuvant therapy. To treat bladder cancer with only chemotherapy, combining chemotherapy drugs is more effective than using any single drug. The combinations that are most commonly used are Gemcitabine with Cisplatin, Methotrexate with Vinblastine, Doxorubicin and Cisplatin, and Carboplatin with either Paclitaxel or Docetaxel. Side effects associated with chemotherapy for bladder cancer can be especially hard to tolerate, particularly due to the advanced age usual in the bladder cancer population. 9
Intravesical therapy is treatment with a drug that is put into the bladder via catheter. It mainly affects the cells lining the bladder and is therefore only used for non‐
invasive or minimally invasive bladder cancers (Stages 0 and 1). The most effective intravesical immunotherapy is Bacillus Calmette‐Guerin (BCG). Its side effects are minor except if the BCG spreads through the body which is characterized by a sustained high fever. Since BCG is related to the germ that causes tuberculosis (TB), the sustained fever may be treated with the antibiotics used to treat TB. Chemotherapy drugs can also be used in intravesical therapy. One of the main advantages of giving chemotherapy into the bladder rather than systemically is that the drugs don’t reach other parts of the body, allowing patients to avoid many of the unwanted side effects. The most common chemotherapy drugs used in intravesical therapy are Mitomycin‐C (MMC) and Thiotepa. Delivery of MMC into the bladder along with heating the inside of the bladder may work even better than giving intravesical MMC the usual way (2). Heating the bladder is achieved with Hyperthermia devices which are discussed in the next section. 1.4 Bladder Hyperthermia and Radiofrequency (RF) Devices
Bladder hyperthermia, the procedure of raising tissue temperature to 40‐43°C, is being explored as an adjuvant therapy. Several clinical trials have revealed that hyperthermia can enhance the effects of radiation and/or chemotherapy for the 10
trreatment of b
bladder canccer (3‐6). Rad
diofrequenccy (RF) devicces are used to induce bladder hypeerthermia. O
One example of an RF dev
vice for end
dovesical heaating of blad
dder wall is the Sy
w
ynergo® tech
hnology seen
n in Figure 11.3. An RF an
ntenna at the end of the ca
atheter heatss the bladder wall. Temp
perature is m
monitored by
y thermocou
uples placed
d in seeveral locations along th
he bladder w
wall. The cath
heter is also designed to
o circulate co
ooled ch
hemotherapeutic drug in
nto and out of the bladd
der. Figure 1
1.3 Synergo®
® endovesiccal bladder h
heating tech
hnology http
p://www.me
edical‐enterp
prises.com/ttreatment.httml A preliminary clin
nical study d
demonstrateed the feasib
bility of using
g this devicee in an
n out‐patien
nt proceduree, and conclu
uded that thee anti‐tumorral efficacy is significanttly en
nhanced com
mpared to ussing intravesical chemottherapy alon
ne (7). In som
me cases, necrosis or bu
urning of thee bladder wa
all occurred where the aantenna is pllaced againsst the bladder. Thiss suggests th
hat there is ro
oom for imp
provement in
n the techniq
ques used to
o heat bladder, with
h preference to a method
d that is non‐‐invasive an
nd heats the bladder walll uniformly. 11
Non‐iinvasive blad
dder heating
g is achieved
d with the BS
SD technolo
ogy pictured in Figure 1.4. Diipole antenn
nas are arran
nged around
d the cylindrrical (Sigma 60) or elliptiical Sigma Ellipsse and Sigma
a Eye) appliccators. The aantennas aree water‐coup
pled to the (S
patient. Phasee and amplittude steering is used to create a focu
used hot spo
ot in the blad
dder. Figure 1
1.4 BSD tech
hnology for non‐invasiv
ve bladder h
heating http:///www.bsdm
medical.com
m/products_b
bsd2000_3D
D.php A receent 16‐patien
nt pilot study using the B
BSD Sigma Ellipse and Sigma 60 pplicators has demonstrrated the fea
asibility of heeating bladd
der to 40‐43°°C while ap
ad
dministering
g MMC (8). BSD applica
ators are larg
ge (not portaable), expenssive and req
quire a shielded roo
om for opera
ation. Also, some patien
nts are not ab
ble to fit into
o BSD ap
pplicators. N
New and imp
proved technologies mu
ust be explorred to overco
ome these limitations. 12
1.4.1 Metamaterial Antennas
Metamaterials are materials that are engineered to have properties not found in nature. This is done by combining naturally occurring materials in such a way that the overall macroscopic effect of the composite created by the new structure has its own effective material properties that allow for the manipulation of electromagnetic waves in ways that are not found in naturally occurring materials. The primary research in metamaterials is the development of materials with an effective negative index of refraction. Refractive index (n) is a number that describes how electromagnetic radiation propagates through a medium. As radiation propagates from one medium into the next, the index of refraction in both mediums can be described using the angles of incidence and transmission through Snell’s law of refraction: both and . In the case where are positive, the radiation is transmitted on the opposite side of the incident normal. However, if is negative, the transmitted radiation will appear on the same side as the incident normal, creating a focus point inside the medium. The focus point in negative index materials allows for the near‐field manipulation necessary in the creation of superlenses for subwavelength imaging (9‐13). For hyperthermia, a negative index of refraction material can be used to focus energy into near‐field tissue targets. To create an effective negative index of refraction in microwave antennas for hyperthermia, the effective permittivity (
) and effective permeability (
13
) need to be negative. This can be achieved with magnetic and electric resonant structures. One popular example of magnetic resonant structure is the Split‐Ring Resonator. The magnetic resonance induces rotating currents in the split rings which produce their own flux that opposes an incident field. Combined with a negative dielectric constant of another structure (more easily found in natural metals), an effective negative index material can be created. In this dissertation, a microstrip‐fed dual‐rectangular patch loaded slot antenna is designed to have metamaterial‐like lensing to focus energy into a tissue target. The standard antennas in current hyperthermia systems create a focused heating zone only with the use of multiple antennas that operate as phased arrays. Now with the advent of this novel antenna, a focused heating zone can be achieved with an individual antenna. Multiple metamaterial antennas, each creating its own focused heating zone, can still be phased to create larger heating zones. Furthermore, these antennas are designed to be efficient over a broad frequency band, so they provide the additional benefit of optimal operation across a broader patient population that is bound to exhibit anatomical and dielectric load differences. The novel metamaterial antennas have a low profile and are small enough to be integrated into a belt that can be comfortably worn by patients of all sizes. The portability of this belt is achieved with the use of shield cloths. By lining the belt with 14
low‐profile absorbers and thin shield cloths, the device would not require a shield room for operation. This would cut down on infrastructure costs and make this device more accessible (e.g., could potentially be used in a private clinic). And finally, these antennas are MR‐compatible which is especially useful for the cases when Magnetic Resonance Imaging (MRI) is used for thermometry during hyperthermia treatments. 1.5 Temperature-Sensitive Liposomes
The previous sections describe the use of hyperthermia as an adjuvant to intravesical chemotherapy. For cases when bladder cancer is invasive, systemic chemotherapy is used as an adjuvant or neoadjuvant therapy to improve survival. However, normal tissue and organ toxicities limit the amount of therapeutic dose that can be administered. A higher dose can be administered by encapsulating the drug in a non‐toxic vesicle that targets cancer cells, while rendering normal tissues and organs unexposed to the drug. One such vesicle is a liposome. Liposomes are phospholipids that are arranged in a bilayer to form a vesicle. When liposomes were initially used in vivo, they were rapidly taken up by the liver, spleen and bone marrow. This limited circulation time to 30 minutes in humans. Polyethylene glycol‐derivatized (PEG) lipids have since been added to the lipid bilayer to improve circulation time by preventing opsonization and lipid aggregation (14‐16). 15
Liposomes can be loaded with anticancer drugs such as doxorubicin and MMC. They can also specifically target cancer cells when antibody or receptor targets are attached to them. However, drug accumulation in the tumor is inadequate due to the slow release from the liposomes. To address this issue, temperature‐sensitive liposomes (TSLs) have been developed. TSLs release their contents more rapidly when their temperature is increased above a certain threshold. In most cases, the threshold is around 41˚C, defined by the melting point of dipalmitoyl phosphatidylcholine (DPPC). In a study by Hosokawa et al., the temperature‐dependent release of MMC‐loaded DPPC TSLs was characterized in rat plasma. The results show a 99% release of the drug within 10 minutes of heating at 42˚C (17). Using TSLs with hyperthermia as a drug delivery mechanism for treating invasive bladder cancers has yet to be explored in both pre‐clinical and clinical studies. 1.6 Pre-Clinical Bladder Hyperthermia Studies
Pre‐clinical bladder cancer models are well‐developed and have proven to be an effective tool for studying the effects of hyperthermia as an adjuvant to standard cancer therapies (18, 19). Methods for heating localized regions at depth in mice are currently being explored. A few examples include ultrasound transducers (20) and RF heating with parallel‐plate configurations (21) which have been shown to effectively heat superficial murine flank tumors. Researchers are also investigating the use of ferrite 16
powder (22, 23) or iron oxide nanoparticles (24) which heat up with the application of an external magnetic field. Although potentially effective, these methods are expensive and may require invasive methods to deliver the metal powders and particles to the region of interest. The most common method used for inducing hyperthermia in murine tumor is by submersing the tumor in a hot water bath. In previous pre‐clinical bladder cancer hyperthermia studies, the tumor was grown on the flank of a mouse. One reason for this is that using a hot water bath to achieve hyperthermia in murine bladder would require submersing the lower half of the mouse into the hot water bath, which would result in elevating its core body temperature to dangerous levels, likely killing the mouse over the course of a one‐hour treatment. The tumor environment present in murine bladder differs from murine flank; namely, the bladder contains urine which is an ideal absorber of microwave power, beneficial for microwave hyperthermia treatment of bladder cancer. However, up to now, there has been no device that selectively heats small volumes at depth in mice while avoiding systemic stress. To investigate hyperthermia in combination with MMC in murine bladder, a device is developed to locally heat murine bladder while maintaining surface and core body temperatures. 17
1.7 Overview
There is now an argument for the use of hyperthermia as an adjuvant therapy to both intravesical chemotherapy and as part of the TSL‐based drug delivery mechanism for systemic chemotherapy. The development of effective, low‐cost and non‐invasive RF technologies are necessary to conduct the studies necessary to achieve wide‐use of hyperthermia for the treatment of both non‐invasive and invasive bladder cancers. 1.7.1 Objectives
The objectives of this dissertation were three‐fold. Objective 1: Develop a reliable target‐specific antenna design optimization method; then model, build and test a novel microwave antenna for localized heating of murine bladder. Hypothesis: A simulation‐based approach can be used to reliably optimize a miniature microwave device to locally heat murine bladder to 40‐43˚C. Results: Using state‐of‐the‐art tissue segmentation and electromagnetic modeling software, an applicator was designed specifically for localized heating of murine bladder. After fabrication, its performance was validated with quantitative laboratory measurements and in vivo temperature distribution characterization. Chapter 3 describes this work in detail. Objective 2: Model, build and test a two‐element metamaterial antenna array for heating pediatric bladder. Hypothesis: A novel metamaterial antenna can be built 18
for use with non‐invasive MR‐thermometry to heat pediatric bladder to 40‐43˚C with simple clinical coupling bolus. Results: Using the methods developed in Objective 1, a broadband metamaterial antenna was optimized for heating pediatric bladder. After fabrication, its performance was validated with laboratory measurements and phantom studies. Chapter 4 describes this work in detail. Objective 3: Design low‐frequency metamaterial antennas for heating human adult bladder as a potential substitute for the BSD Sigma applicator. Hypothesis: A metamaterial antenna array can be designed for heating human adult bladder to 40‐43˚C that is smaller and more cost effective than previous applicators. Results: Using the methods developed in Objective 1, a metamaterial antenna array has been designed to heat human adult bladder. Chapter 5 describes this work in detail. 1.7.2 Significance
Many studies have demonstrated that hyperthermia, the procedure of raising tumor temperature to 40‐43˚C, enhances the effectiveness of chemotherapy and/or radiation therapy (25‐29). However, insurance reimbursement for hyperthermia procedures in the US is limited to interstitial devices, intracavitary devices, and external devices for heating superficial tissues (depths of 4cm or less). Insurance reimbursement is not currently in place for non‐invasive hyperthermia treatment of deep tissues, such as the bladder. 19
Bladder cancer is currently treated using mitomycin‐C (MMC). However, as with other chemotherapy drugs, tumor cells eventually develop a resistance to MMC and render the treatment less effective or completely ineffective (30). In vitro studies have shown that heating mitomycin‐C (MMC) resistant cells to 40‐43°C increases the cellular uptake of MMC suggesting that MMC resistant tumor cells may be more effectively treated with combination MMC and heat (2). Optimized devices to uniformly heat bladder cancer in vivo would allow us to investigate the basic principles underlying this phenomenon more systematically. Currently, there exists no device to selectively heat small volumes in mice while avoiding systemic stress. One significant contribution of this work is to enable essential pre‐clinical bladder hyperthermia studies with the development of a reliable microwave applicator for heating murine bladder to 40‐43˚C while maintaining surface and core temperatures normothermic. For human hyperthermia bladder treatments, there is one intracavitary heating device for bladder hyperthermia, the Synergo® endovesical technology. This technology does not heat the bladder wall uniformly and, in some cases, necrosis or burning of the bladder wall occurs when the antenna comes in contact with the bladder wall (7). To limit toxicity, a non‐invasive heating device is preferred. A BSD Sigma applicator is a non‐invasive technology that could potentially be used for bladder hyperthermia. BSD Sigma applicators have large (>12cm long) antennas in order for the 20
device to operate at a low enough frequency to induce deep heating. The antennas, which are located around a fixed aperture applicator, couple energy into the body via a large water cooling bolus that encircles the patient’s abdomen. Because the antennas are incorporated into a fixed aperture water bolus, a segment of the patient population cannot receive treatment because they cannot fit into the applicator. Furthermore, the device is large (i.e., not portable) and requires a shielded room for operation. These requirements prevent widespread use of this technology which is especially prohibitive for the multi‐institutional clinical trials necessary to apply for FDA approval and insurance reimbursement. Designing bladder hyperthermia devices with metamaterial antennas can overcome all of these limitations, and furthermore, recent simulation‐
based studies have demonstrated the ability to design metamaterial antennas for an improved focus region of heating (31‐33) suggesting the potential for improved performance. It is now necessary to move the field beyond simulation by constructing and validating the performance of a simulation‐optimized metamaterial antenna array with laboratory measurements and in vivo studies. This contribution is significant because the development of metamaterial antennas for deep tissue hyperthermia enables the multi‐
institutional clinical trials required to apply for insurance reimbursement of deep‐tissue thermal 21
therapy and the subsequent widespread use of hyperthermia as an adjuvant to current cancer therapies. Incorporating metamaterial antenna arrays into a conformal belt that is size‐
adjustable allows patients who could not fit into the fixed aperture BSD Sigma applicator to undergo bladder hyperthermia treatments. The mechanism for shielding that is incorporated into the belt will facilitate use outside a shielded room to avoid the increased cost and infrastructure changes to hospitals and clinics that do not have available shield rooms in place. The development of metamaterial antennas for hyperthermia devices is a complete paradigm shift and will set a new precedent for all future antenna development in hyperthermia applicators.
22
2. Antenna Optimization for Medical Device Applications
Before electromagnetic simulation techniques were available, RF antenna optimization was accomplished through an iterative build‐and‐test method. Some scientists still employ this method, while others use simulation techniques to reduce the overall cost and time required to develop a new device from concept to working product. HFSS™ (Ansys Inc., Canonsburg, PA) is an effective simulation tool for designing RF antennas. It employs the finite‐element method to approximate the solution to the partial differential equations that define electromagnetic wave propagation. 2.1 Electromagnetic Wave Equations
Maxwell’s equations describe the relationship between electromagnetic sources and fields. In antenna design, we are mainly concerned with solving for the electric field in 3‐dimensions (3D). HFSS™ solves for the electric field in 3D using the finite element method. For mathematical simplicity, we will walk through the 1‐dimensional (1D) problem. In a 1D inhomogeneous medium, the partial differential equation (PDE) for the electric field (Ey) is defined as the following, where and are the relative magnetic permeability and relative electric permittivity of the medium, and k0 is the wavenumber: 23 ,
∈
,
Eqn. 1 The source term is defined as the following, where Jy and Mz are the electric and magnetic current densities of the source: . 2.2 Finite Element Method
The finite‐element method is a commonly used numerical technique for finding approximate solutions to PDEs or integral equations. This method is often used to solve thermal, electromagnetic, fluid, and structural problems with high accuracy and relatively short computational times over complex geometries. The following explanation is a simplified version (1D, rather than 3D, PDE) of the mathematics behind using the finite element method to solve for the electric field. We start by multiplying Eqn. 1 with a testing function wm(x) and integrating over the interval [a,b]: Taking Eqn. 2 and integrating‐by‐parts, we obtain: 24 Eqn. 2 ∙
Eqn. 3 In order to solve the problem on a computer, the computational domain has to be finite and therefore it is necessary to define the boundaries of the domain. There are many boundary conditions that can be imposed: perfectly magnetic conductor, perfectly electric conductor, absorbing boundaries, etc. Radiation boundary conditions are used in this project. Radiation boundaries are used when the sources within the computational domain are the only sources that need to be considered and for unbounded problems where waves will propagate outside of the computational domain ∈
If the medium is homogenous for where and with wavenumber are the relative permeability and permittivity for ,
. / , √
and is the speed of light in vacuum, the radiation boundary condition for the left side is: ,
And similarly for the right side, the boundary condition is: ,
If the incident field (i.e., plane wave for the 1D case), is from the left 25 , we have: 2
Eqn. 4a And similarly, if the incident field is from the right 2
, we have: Eqn. 4b Substituting Eqn. 4a and Eqn. 4b into Eqn. 3, we get: ∙
Eqn. 5 2
To solve for the electric field , let’s call the right‐hand side of Eqn. 5 V. Assuming there are N elements inside the domain and expanding the field in terms of the basis functions , we can obtain an impedence matrix Z from the left‐hand side so that the solution for the electric field will be . ∙
Eqn. 6 26 2
Eqn. 7 2.2.1 Galerkin Method
The Galerkin method is used to discretize continuous operator problems like Eqns. 6 and 7. After discretizing these problems, constraints are applied on the function space with a finite set of basis functions. For the case of triangular basis and test , the impedance matrix and right‐hand side become: functions, ∙
,
,
,
,
2
2
and Where are: 27 ,
,
1
1
,
∆
,
1
,
∆
,
,
, ∆
,
∆
0,
,
1
,
1
, ∆
,
1
∙
,
1
, ,
,
1
∆
, 3
1
,
,
∆
6
1
,
,
6
∆
, 1
, 1
0,
Finally, in the special case where a point electric current source is located at , the source term becomes , and the right‐hand side is evaluated as: 2
,
,
Now that the impedance matrix and right‐hand side vector are completely discretized, solving for the electric field is trivial: . This is an overly simplified, 1D example of how HFSS™ uses finite element method to solve for electric field. 28 2.2.2 Elemental Approach for Multi-Dimensions
The above example is in 1D, so looping through indices m and n is straight‐
forward and relatively fast. For higher dimensions, an elemental approach is easier to implement. Essentially the problem is broken up into elements so that the Z matrix and V vector are solved for in pieces and then assembled together at the end to solve for the electric field. For 3D problems, this elemental approach is ideal – each element can be solved in parallel, reducing computational time significantly. 2.3 Optimizing Specific Absorption Rate (SAR) Pattern and S
Parameters
Specific absorption rate (SAR) is the rate at which energy is absorbed by a body exposed to an RF electromagnetic field. In the work presented in this thesis, the SAR is calculated using the solved E‐field and the electrical conductivity (
and density of the tissue model (Eqn. 8). The SAR is used to determine the power deposition pattern of a transmit antenna or, by the principle of reciprocity, the volume over which electromagnetic emissions would be collected by a receive antenna. For example, in bladder hyperthermia devices, the antenna would be optimized for an SAR pattern that exhibits maximum power deposition in the bladder. |
|
29 Eqn. 8 The term impedance is used to describe the resistance of a system to an energy source. In delivering power from our antenna (the energy source) into tissue (the “system”), the antenna must be impedance‐matched to the tissue for maximum power to be transmitted into the tissue. For a simplified model, assume that a certain percent of the input power into the antenna will be transmitted into the tissue and the remaining will be reflected back into the source. An antenna that is not impedance‐matched to the tissue will reflect more power. In a two‐port network (such as the antennas described in this dissertation), Port 1 is the source of power and Port 2 is the interface between the antenna and the load. S parameters are used as a measure of the amount of transmitted and reflected power in linear electric networks (Figure 2.1). S11 represents the amount of input power that is reflected back into the source and therefore not transmitted into the load. S21 represents the amount of power transmitted into the load. Optimizing the S parameters of an antenna for hyperthermia means aiming to achieve less than 10% reflected power and therefore over 90% transmitted power in the operating frequency band. When this is achieved, the antenna is considered to be impedance‐matched to the tissue. 30 a1
b1
b2
Figure
e 2.1 S param
meter definiition for a tw
wo‐port netw
work S para
ameters are rratio measurres of two po
ower levels.. This ratio iss usually reepresented o
on a logarith
hmic scale in
n Decibels by
y evaluating 10 times thee base‐10 lo
ogarithm of tthe ratios. Fiigure 2.2 dep
picts an exam
mple plot off S parameteers. In the plot, ‐
10
0dB represen
nts 10% pow
wer reflected
d (S11) or tran
nsmitted (S211). In a logariithmic scale,, decreasing by
y a decade reepresents a d
decrease by 10‐fold. Theerefore, a sim
mple calculation demonstratess that ‐20dB correspondss to 1% and ‐‐30dB corressponds to 0.1%, etc. Figure 2.2 Example o
of S parameeter plot 311 2.4 Computational Models for Patients and Animals
Using HFSS™ to optimize RF antennas for tissue loads requires developing geometrically and electrically accurate 3D models of animals and humans. Currently, two classes of computational models are used for the evaluation of dose distributions resulting from internal or external radiation sources. These two classes are 1) stylized phantoms, based on mathematical equations that define organ position and geometrical shape, and 2) voxel phantoms, developed using segmentation software to extract 3D geometries from MR or Computed Tomography (CT) images. The first stylized phantom of a human was created at Oak Ridge National Laboratory. It represented the average adult male and was named the Medical Internal Radiation Dose (MIRD) phantom (34). Extended versions of the original MIRD now exit for adult male and female phantoms (35), a series of pediatric phantoms (36, 37), and a pregnant adult female at three stages of gestation (38). Voxel phantoms have a similar series made available for dosimetry studies (39‐45). Some work is being done to combine the strengths of both methods to achieve a more accurate computational model (46) since several investigators have performed comparative studies between stylized and voxel phantoms (47‐56) and have reported clear differences in dosimetry results that are attributed to differences in body shape, age, and organ size, depth and position. 32 Up to now, computational models for the “average” patient have been used for dosimetry assessment in phantom studies. In order to use a computational model for treatment planning, a voxel‐based model is developed on a patient‐specific basis to account for variations in body shape, age and organ size, depth and position. In both dosimetry assessment and treatment planning cases, the power emission and/or deposition of a source is evaluated using computational models. In this dissertation, voxel‐based computational models are developed for mouse and patient models in technology development, specifically to optimize the power deposition pattern of antennas used in hyperthermia treatments. As far as we know, this is the first such case when a geometrically accurate animal and human computation model is used in antenna optimization for pre‐clinical and clinical devices. Because of the variations in human anatomy, a statistical study was done on two samples within the specific pediatric patient population in order to create an “average” model. This is described in further detail in Chapter 4. The geometrically accurate patient model is determined and created using segmentation methods in VSG Avizo (see Appendix A for details). After the patient or animal model is developed, the 3D surfaces are imported to HFSS™ to be utilized as the load in antenna optimizations. The electrical properties of anatomical tissues are well 33 documented in the literature (57‐59) and can be assigned to each tissue once imported into HFSS™. 2.5 Thermal Simulations for Hyperthermia Applications
To plan for effective spatial temperature distributions without damaging healthy tissues, a precise thermal model of the tissue load is required. In 1948, Pennes proposed and experimentally validated an analytical bioheat transfer equation (Eqn. 9) that (60). This included a heat loss term to account for the effects of blood perfusion bioheat equation also accounts for other critically important thermal mechanisms such , and thermal as thermal storage, heat conduction, heat generated by metabolism energy absorbed from external sources. In our case, the effect of an external source is characterized by SAR. Eqn. 9 The parameters density ( ), specific heat capacity ( ), thermal conductivity ( and arterial blood temperature (
) can be considered temperature invariant parameters since these parameters do not change significantly in the narrow range of temperatures used in hyperthermia (61). However, blood perfusion (
generation (
) and metabolic heat ) are temperature dependent, and also vary under anesthesia. Experiments by Song et al. (62) demonstrated that in rodents (i.e., rats) there exists a 34 threshold at 43˚C where blood perfusion temperature dependence changes from linear to nonlinear behavior. Temperatures in hyperthermia studies typically do not exceed 43˚C; therefore, linear behavior can be assumed. Blood perfusion of 0.019
10
2.008
represents the average results for ten anesthetized rats (61), and is the expression used for the pre‐clinical studies in this dissertation. For temperatures below 43˚C, metabolism also has a linear behavior with respect to temperature as defined by Gillooly et al. (63). His theory states that temperature governs metabolism through its effects on biochemical reaction rates which vary with temperature according to Boltzmannʹs Law: Eqn. 10 From experiments by Barclay et al. (64) conducted on murine muscle, the activation energy of 10,760
was determined as 0.67
which corresponds to a metabolic rate in a body at 37˚C. Using these values in the bioheat transfer model and taking into consideration the convection of surrounding air at 25˚C, the resulting simulation in COMSOL gives a steady state core temperature of 37˚C which is consistent with experimental results (Figure 2.3). This validation of our thermal model allows us to confidently use COMSOL thermal simulations to assess the performance of our novel heating devices for pre‐clinical applications. 35 Figure 2.3
3 Thermal m
map of mousse without eeffects of an
nesthesia Finally, Sessler et al. (65) veriffied that witth anesthesiaa there is a g
general drop
p of 30
0% in basal m
metabolism,, from which
h the expecteed metabolissm becomess 7530
. Using an inve
U
erse analysiss of Eqn. 10 we would e xpect a coree temperaturre drop of 4˚C
C affter anesthessia due to th
his lower metabolic heat production. This was allso confirmeed with experim
w
mental measu
urements. Th
hus, for futu
ure calculatio
ons, the initiial core teemperature iin thermal simulations o
of rodent mo
odels will bee 33˚C to tak
ke into accou
unt th
he effect of a
anesthesia. 366 3. Devices for In Vivo Murine Hyperthermia Studies
The most commonly used method for inducing hyperthermia in murine tumor models is to submerse the tumor bearing portion of the mouse into a hot water bath (66‐
70). Since this method cannot provide selective power deposition into tissues at depth, it has been useful mainly for tumors in the hind limb where the region immersed can be limited primarily to the target tumor mass. Water bath heating of deep‐seated tumors and/or organs such as bladder would require circulation of heated water through the region of interest or submersing the mouse’s body. The former is not technically achievable for small volumes such as a 0.15ml bladder and the latter would heat the large regions in the mouse. In most cases, heating larger regions would elevate the mouse’s central body temperature to dangerous levels, which could be toxic. Other methods for locally heating deep‐seated organs like the urinary bladder have been explored. A few examples include ultrasound transducers (20) and RF heating with parallel‐plate configurations (21) which have been shown to effectively heat superficial murine flank tumors. Ferrite powder (22, 23) or iron oxide nanoparticles (24) could be instilled into the bladder, which would then heat up with the application of an external magnetic field. Although potentially effective, these methods are expensive and may require invasive methods to effectively deliver the metal powders and particles to 37 the region of interest. There is an unmet need for a cost‐effective and non‐invasive device that can reliably heat deep‐seated organs and small tumors in mice while maintaining surface and core body temperatures. The work described in this chapter details the design optimization, construction and validation of novel microwave devices to address this unmet need by selectively heating murine bladder and brain tumors. 3.1 Murine Bladder Hyperthermia
There is a strong interest in using established murine bladder cancer models to study the mechanisms behind the increased effectiveness observed in combination hyperthermia and intravesicular MMC treatments in order to better assess the future potential of such combinatory treatment protocols. This work presents a novel simulation‐based approach to RF antenna design for pre‐clinical hyperthermia studies to facilitate the creation of cost‐effective and non‐invasive microwave heat applicators to locally heat small volumes in mice while maintaining normothermic skin and core body temperatures. Note that the physical dimensions of the prototype applicator designed for heating murine bladder can be scaled with frequency to adjust the diameter and depth of heating. Therefore, the design procedure described in this work may be applied to design similar devices for other small‐animal hyperthermia studies. 38 3.1.1 Methods
3.1.1.1 Antenna Design Optimization The first step in the development of a simulated murine bladder heating model was the creation of an anatomically accurate, 3D CAD model of a mouse. Micro‐MR images of a female mouse [acquired using a 7T Magnex Scientific magnet] were obtained from Duke University’s Center for In Vivo Microscopy. The images used to create a 3D model have a resolution of 6.25μm resolution in all 3 dimensions. Using Avizo® software package (Visualization Sciences Group, Burlington, MA), the micro‐MR images were segmented to yield 3D surfaces for the mouse’s body, bladder, vagina, and rectum (see Appendix A for more detail). These surfaces were assumed to be the outer boundary of a uniform volume of tissue. Once imported into HFSS™ electromagnetic simulation software, frequency‐dependent tissue properties, acquired from literature (57), were assigned to each volume. Tissue properties at 2.45GHz are listed in Table 3.1. The mouse was oriented to lie on its back, with the applicator positioned over the pelvis. For the purposes of optimizing the antenna, no significant differences were seen with the addition of typical thickness skin and fat layers between the applicator and bladder. 39 Table 3.1 Properties of murine tissues at 2.45GHz(57) Density, ρ Specific Thermal Electrical Dielectric Conductivity, (kg/m3) Constant , ε Conductivity, Heat, Cp k (W/mºC) σ (S/m) (J/kgºC) Body (muscle) 52.729 1.7388 3546 0.53 1041 †
Bladder (urine) 71.238 7.4257 4200 0.561 1000 Uterus 57.814 2.2465 3580 0.50 1052 Colon 53.879 2.0383 3653 0.56 1044 †Urine and water have similar thermal conductivity and density values, so we assume that the specific heat of urine is also similar to water. To achieve adequate power deposition in the bladder with an integrated surface‐
cooling system, we used a deionized (DI) water‐loaded circular waveguide design. The largest diameter tubular waveguide which allows for only a single mode to propagate was selected by sweeping through diameter sizes in 1mm increments. The length of the coaxial monopole feed protruding inside the tube and the distance between the backshort (end plate of waveguide tube) and the coaxial feed were design variables, and a parametric sweep of these variables was performed to optimize the applicator load impedance (S11) when coupled to the pelvis of the mouse and to maximize specific absorption rate (SAR) in the bladder (see Appendix B for a detailed explanation of the simulation setup in HFSS). Convergence was met when a maximum change of 5% in the S‐parameters was achieved between two consecutive adaptive mesh passes. The S11 was optimized at less than 1% reflected power for 2.45GHz. Following optimization of the antenna design 40 parameters for load impedance, the local SAR pattern was also determined with HFSS™. Similar to previous work in which an array of patch antennas was optimized to maximize power deposition in a tumor target (71), bladder heating efficiency was defined by integrating the local SAR in the bladder volume and dividing by total SAR in the mouse (Eqn 11). This efficiency value was optimized to obtain maximum differential heating of the bladder. ∗
∗
Eqn 11 Because DI water efficiently absorbs electromagnetic power at 2.45GHz, the overall height of the waveguide was minimized to optimize transmission of power into the mouse (S21) and minimize direct heating of the coupling fluid. The water input channel and coaxial connector size required that the height of the waveguide from the center of the coaxial input feed to the circular waveguide output port be at least 15mm. In a circular waveguide, the coaxial feed length and backshort distance determine the resistance and reactance of the waveguide. In a circular waveguide, the optimum backshort distance should be around 0.25λ (3.4mm in our water‐loaded waveguide), but the waveguide was constructed with a 1 to 10mm‐range adjustable backshort distance so the feed‐point reactance could be manually adjusted as needed to impedance match the applicator to the mouse load. This was realized in construction by soldering a copper 41 disk (backshort) to the end of the water output port tube which had a watertight sliding fit in the back end of the circular waveguide. The backshort can then be positioned closer to or further from the coax feed by sliding the water output port in or out of the applicator (see Figure 3.1). 3.1.1.2 Thermal Modeling To complete the optimization of a murine bladder heating system, a heat transfer study was implemented in COMSOL Multiphysics® (see Appendix C for a detailed explanation of simulation setups for coupled electromagnetic (EM)‐Thermal models). This study was based on the Pennes Bioheat Transfer Equation (60) which accounts for all major thermal inputs and outputs of the energy balance in biological tissue, namely, blood perfusion, thermal storage, heat conduction, metabolic heat production and externally applied SAR (calculated in HFSS™). In the hyperthermia temperature range, both perfusion and metabolism rates increase in response to external applied heating of the targeted tissues. These variations are significant and are accounted for as shown in Eqn 12 and Eqn 13. 1.9
200 ∗ 10 Eqn 12(72, 73) Eqn 13(63, 74) In these expressions temperature (T) is measured in degrees Kelvin, Ea = 0.67eV is 42 the activation energy, kb is the Boltzmann constant and Qm0 = 10760W/m3 is the basal heat metabolic rate (64). Anesthesia induces hypothermia which adversely affects the normal thermal regulation of mice, inducing a general drop of 30% in basal metabolism (65). For this reason mice were placed on a heated pad with a constant temperature of 37oC. The rest of the body was exposed to room temperature (25oC) with an assumed convection coefficient of 5W/m2K. The final thermal boundary condition is water bolus temperature, which is expected to play an important role in avoiding the overheating of skin from the externally applied microwave field. All of these considerations were accounted for in this thermal model. 3.1.1.3 S Parameter and SAR Validation With a latex membrane sealed over the front opening of the waveguide to contain the circulating water, the applicator was placed adjacent to skin over the pelvis of C57BL/6 mice (Harlan Laboratories, Indianapolis, IN) and S11 was measured using an Agilent E5071C network analyzer. The applicator’s adjustable backshort‐to‐coaxial feed distance was varied until an S11 of approximately ‐20dB was obtained for the desired operating frequency of 2.45GHz. To confirm the validity of SAR patterns simulated by HFSS™, the SAR pattern of the applicator was measured with an electric field probe 43 (APREL Laboratories, Ottawa, Ontario, Canada) that was scanned in a tank containing tissue‐equivalent liquid with 1mm step size in 3D using procedures that have been published previously (75). The tank contained a mixture of Tween 80 and DI water to create a liquid that simulates muscle tissue at 2.45GHz. The measured dielectric constant was within 5% and the electrical conductivity within 1% of published values (57). 3.1.1.4 In Vivo Murine Bladder Heating Studies C57BL/6 mice were anesthetized with an IP injection of 65mg/kg Nembutal and placed on a thin acrylic platform. Underneath the platform, heated water was circulated to help maintain 37°C core body temperature. Heating performance of the applicator was tested in mice with full bladder volumes of approximately 0.15mL at 2‐3mm depth. Temperature distributions in the mouse pelvis were characterized by mapping Luxtron® fiber‐optic temperature sensors (LumaSense Technologies, Santa Clara, CA) inside three parallel 20Ga polyurethane catheters inserted into the perineum of the mouse and extending 20mm superiorly into the pelvis. The catheters were inserted at depths of 0‐1mm, 2‐3mm, and 4‐5mm below the abdominal wall, with the mid‐depth catheter passing immediately adjacent to the bladder. The sensors were connected to a motor that pulled the sensors out, stopping to allow for a temperature measurement every 1mm and then pushing the sensors back 1.5cm deep again, collecting 15 44 temperature measurements in each 1 minute cycle. Systemic temperature was measured with a stationary fiber‐optic temperature probe inserted orally. The temperatures were collected and displayed in real‐time using LabVIEW software (National Instruments, Austin, Texas), which allowed the user to manually control applicator power as needed to maintain the desired temperature distribution in the mouse at all times. All murine in vivo studies described in this manuscript were conducted in accordance with the Institutional Animal Care & Use Committee at Duke University. 3.1.2 Results
With the fixed backshort distance of 0.25λ, a parameter sweep in HFSS™ determined the optimal coaxial probe length to be 1.5mm past the center line of the waveguide (6.25mm from sidewall feed‐point). Figure 3.1 shows pictures of the prototype 2.45GHz applicator, the HFSS™ model, and the applicator being used in a murine bladder hyperthermia study. 45 (a
a) Optimization
n Antenna O
Parameters (b) (c) (e) (d) Figurre 3.1 a) Para
ameters variied to optim
mize SAR an
nd S11; b) Fro
ont‐view piccture of 2.45GH
Hz applicato
or; c) Bottom
m‐view pictu
ure of 2.45GH
Hz applicato
or; d) HFSS™
™ model of ap
m
plicator on m
mouse; e) Sn
napshot of eexperimentaal setup for murine blad
dder hypertherrmia study w
with applica
ator position
ned on the sskin overlyin
ng the pelviis. At 2.4
45GHz, the d
dielectric pro
operties of D
DI water decrrease the waavelength fro
om 12
2.2cm to 13.6
6mm, with a
a correspond
ding decreasse in wavegu
uide diameteer required ffor effficient wavee propagatio
on. The wave propagatio
on constant was calculatted at 1mm in
ncrements ov
ver a range iin diameter from 4‐16mm
m, and a diaameter of 9.55mm was seelected in orrder to limit wave propa
agation to th
he dominant mode. After waveguide construction
n, S11 and SA
AR were meaasured in ho
omogeneous muscle tissue
m
e simulating material and compared
d to simulateed results. Th
he S11 simulaation 466 results are most accurate (≤ 5%) below 2.6GHz. Figure 3.2 demonstrates less than 5% difference between simulated and measured S11 around the critical 2.45 GHz design frequency, with approximately 1% power reflected at 2.45GHz. 0
HFSS Model
Measured
-5
S11 (dB)
-10
-15
-20
-25
2
2.1
2.2
2.3
2.4
Frequency (GHz)
2.5
2.6
2.7
2.8
Figure 3.2 S11: HFSS™ simulation compared to measurements using an Agilent E5071C network analyzer. Figure 3.3 shows that the SAR pattern simulated in HFSS™ matches well to the SAR data that is measured in muscle‐equivalent liquid using an E‐field scanning probe. The E‐field probe contains three orthogonal dipoles centered 3mm from the tip; 47 therefore, SAR patterns are recorded with the closest measurements already 3mm from the antenna front face (Figure 3c). The error between the simulated and measured SAR does not exceed 10%, and applying a moving average window to the simulated data suggests that errors in the side lobe regions may be due to the fact that the diameter of the electric field probe used for SAR measurement is 70% as large as the waveguide aperture. 48 a)) b)
c)) d)
Figurre 3.3 SAR: H
HFSS™ simulation com
mpared to meeasurementts with all vaalues normalized
d to the maxiimum value
e at 3mm dep
pth: a) proffile across x‐‐axis, b) pro
ofile across y‐axis, c) depth
h profile, an
nd d) 3D SAR
R measurem
ments in mu
uscle phanto
om Simullations of SA
AR in the mo
ouse model ((Figure 3.4) iindicate thatt maximum power deposition will occur in the sk
kin and subccutaneous tisssue directly
y above the d subcutaneo
ous tissue m
may be lowerred by bladder. The temperaturee of skin and
ppropriate teemperature w
water coolan
nt through tthe applicato
or. This mak
kes ciirculating ap
499 fo
ortuitous usee of the high
h dielectric constant wate
ter‐loading o
of the waveg
guide which is allso used to rreduce waveeguide diameter for operration at 2.455GHz. a) b) S™ SAR sim
mulation in m
mouse modeel a) x‐axis p
profile and b
b) y‐
Figure 3.4 HFSS
axis profi le Figure
e 3.5 presentts the effect of the waterr bolus at diffferent temp
peratures on the reesulting tem
mperature pro
ofile. It is rea
adily seen th
hat as the su
urface tempeerature is decreased thee hot spot m
moves away ffrom the skin
n surface to the bladder region and the percentage off bladder heated from 42
2‐44oC is inccreased. 500 (a) (b)) (c) (d)) Bladder Vag
gina Reectum Figu
ure 3.5 2D sllice tempera
ature profilees in mouse body, bladd
der, vagina aand o
re
ectum with bolus tempe
erature (a) u
unregulated (b) constan
nt 41 C, (c) co
onstant 38oC
C and o
(d
d) constant 35 C and RF power set aat (a) 5W and
d (b), (c), (d)) 12W. To stu
udy the relattive heating of all mousee tissues, thee correspond
ding tissue‐
av
verage temp
peratures aree presented iin Figure 3.66. To restrictt the analysis to just the heated region
n, average reectal and vag
ginal temperratures weree calculated from tissue lo
ocated within a cylinder of the diam
meter of the aantenna and extending d
down under the an
ntenna. Figu
ure 3.6 demo
onstrates tha
at the rectum
m is always 11‐2oC below the temperaature of the vagina and the vag
gina is alway
ys 1‐2oC belo
ow the temp
perature of th
he bladder fo
or bolus temperratures between 35‐41oC. These resu lts also indiccate that a bolus of 36‐38oC w
will yield thee desired tem
mperature raange of 42.5‐‐43.5oC in thee teemperature o
bladder. For tthermosensiitive liposom
mes loaded w
with mitomy
ycin‐C, the aassumed targ
get 511 teemperature iis between 4
41‐43oC, thou
ugh studies will be cond
ducted with higher and lo
ower bladdeer temperatu
ures as shown possible in
n Figure 3.6.. Figurre 3.6 COMS
SOL simulattion of averaage temperaatures as a fu
unction of b
bolus temperatu
ure In vivoo murine bla
adder hyperthermia stud
dies were co
onducted to validate resu
ults seeen in simullation. As seeen in the Mo
ouse 1 experriment (Figu
ure 3.7), RF p
power (15W)) was tu
urned on at tthe 5‐minutee time point and steady state was reeached at thee 10 minute time point. At stea
ady state, thee subcutaneo
ous tissue w
was maintain
ned around 440ºC due to tthe ooling effectts of 37ºC wa
ater circulatiing inside th
he applicatorr. Tissues around the co
bladder and rrectum weree elevated to
o 42‐43ºC wh
hile the core body tempeerature was 522 maintained a
m
round 38ºC throughout the heating period. Thee RF power w
was turned o
off at th
he 18‐minutee time point and the tisssue temperatture returne d to normotthermic co
onditions by
y the 25‐minu
ute time poiint. The sam
me heating prrotocol was u
used for thee fo
ollowing fou
ur mice, with
h minor variiations in coo
oling bolus ttemperaturee, RF power leevel, and hea
at start and sstop times. Fig
gure 3.7 Tem
mperature da
ata taken fro
om five indeependent m
murine bladd
der hypertherrmia studiess. Temperatu
ure increasees occur wheen RF power is turned o
on (around tthe 2‐5minu
ute mark) an
nd decreasess occur when
n RF power is turned offf (around the 18‐20minute
e mark). The dotted lin
ne signifies 442˚C. 533 3.1.3 Discussion
A miniature water‐loaded waveguide applicator operating at 2.45GHz was optimized, built and tested. The diameter of the applicator scales with frequency and affects the volume and depth of heating. A 2.45GHz applicator was appropriately sized for heating murine bladder, while lower frequency (larger diameter) applicators would be more ideal for heating larger or deeper regions and higher frequency (smaller diameter) applicators could be designed to heat smaller and more superficial targets. However, when heating deep targets in larger animals such as rabbits and rats, applicator miniaturization is less necessary and existing hyperthermia systems may be considered. For example, a readily‐available low frequency (433MHz) horn antenna has been shown to effectively heat rat bladder (76). The 2.45GHz murine bladder‐heating applicator heats small regions at depth (2‐
3mm) while maintaining normothermic surface and core body temperatures. Other designs were considered, such as the open end of a 6mm diameter semi‐rigid coaxial cable; however, the water‐loaded circular waveguide was chosen for its overall performance. With a water bolus naturally integrated into the device itself, this design eliminated the need to construct a separate miniature water bolus for surface cooling. 54 Bones were not used in the numerical model. They were not visible for segmentation in the micro‐MR images and because the pelvic bones in mice are far from the heated region, it was not necessary to include them in this case. Micro‐CT images should be considered for future small‐animal antenna design and planning studies when bones are in proximity to the treatment region. Measured S11 of the prototype applicator matches the simulated S11 within 5% at 2.45GHz. The small difference may be explained by the convergence criteria used in simulation; convergence was met when a maximum change of 5% in the S‐parameters was achieved between two consecutive adaptive mesh passes. Greater than 5% error at other frequencies is due to the fact that a dispersive model used for the different tissues in the HFSS™ model and convergence of S parameters was only required at the frequency of interest (2.45GHz). Simulated SAR and measured SAR reveal the same pattern of power deposition in muscle‐equivalent liquid. The largest difference between simulated and measured SAR is apparent in the side lobe regions likely due to the diameter of the electric field probe used to measure SAR. However, both thermal simulations and measurements confirm that the water cooling eliminates temperature rise in tissues superficial to the bladder and thus negates any deleterious effect that side lobes would have contributed. 55 In pre‐clinical and clinical hyperthermia studies, calculating local SAR is not a good predictor for steady state temperature in the body because it does not account for thermal effects, such as convection and perfusion, which play such an important role in the thermoregulation of rodents. Therefore, thermal models where these effects are taken into account are paramount for predicting treatment outcome. Accurate thermal models require knowledge of the real SAR and power level in the system, as well as the thermal characteristics of all involved biological tissues. Recently, the relation between model and real‐world SAR and power levels was successfully realized (77). The effect of heat on perfusion is not well‐characterized in highly perfused organs and tissues, and therefore more work is still required before thermal models can be used for clinical hyperthermia treatment planning in various organ and tissue sites. For less perfused organs, such as the bladder, thermal modeling is more achievable. Figures 3.4 and 3.5 were the results of a first attempt to characterize the thermal distribution during a murine bladder hyperthermia study with a thermal model. A difference of 1‐2oC between the bladder and the rectum was predicted by this thermal model and validated in in vivo experiments (Figure 3.6), demonstrating that a thermal model may be used to accurately predict internal temperatures during murine 56 hyperthermia studies. Therefore, electromagnetic and thermal simulations can be used as a reliable system design and treatment planning methodology. A similar difference of 1‐2oC between the bladder and the vagina was predicted (Figure 3.5). Temperature measurements could not be taken inside the bladder, but the results of thermal simulations of water cooled microwave antenna heating indicate that a probe in the vagina can be used as a reliable surrogate for the bladder with an accuracy of ±1˚C, where the bladder is always at a higher temperature than what is measured in the vagina. Aside from the results of these thermal simulations, the bladder was assumed to be at temperatures close to or higher than the vagina for two reasons: 1) urine is a more efficient absorber of microwave power than the surrounding tissues, and 2) surrounding tissues experience the cooling effects of blood perfusion more than urine inside the bladder which has no perfusion. As seen in Figure 3.7, temperature profiles induced by the bladder hyperthermia treatment looked similar in all five mice. Minor differences in the amount of RF power used and temperature of cooling bolus are seen, likely due to how each mouse’s body temperature regulation mechanisms respond under anesthesia. After the hyperthermia treatments, visual inspection of the mouse skin directly under the applicator revealed no burns, as expected from the low measured surface temperatures. In these studies, water 57 bolus temperature was measured at the input port and is likely to be slightly lower than actual skin surface temperature. In future studies, a more accurate measure of skin‐bolus interface temperature will be obtained by measuring the temperature of the water flowing from the output port and averaging input and output port temperatures. We kept body core temperature above 30˚C with a temperature controlled water pad placed under the mouse throughout the heat treatment. Due to the fact that mice depend on their tails for thermoregulation, it should be noted that the entire tail of the mouse was placed in direct contact with the temperature‐controlled pad. However, the core temperature variations in Figure 3.6 demonstrate that the temperature‐controlled water pad was not a reliable way to regulate body temperature. In future studies, a feedback‐controlled constant temperature mat such as the CMA 450 Temperature Controller (CMA Microdialysis AB, Solna, Sweden) will be used to better maintain core body temperature throughout the heat treatments. A constant temperature boundary under the mouse was used in the thermal simulations. A more reliable method for achieving a constant body temperature experimentally by using a feedback‐controlled temperature mat would potentially yield a more reliable method for attaining the internal temperatures seen in Figure 3.7. 58 Although varying antenna operating frequency has the most effect on power deposition pattern and volume of heating, another method to control the depth distribution of heating is to vary temperature of the cooling bolus (Figure 3.5). A lower temperature bolus removes more heat from surface tissues, effectively increasing relative heating at depth. Therefore, this microwave antenna can be used either for heating sub‐surface tumors while maintaining cool temperatures in overlying tissue, or by increasing bolus temperature can heat tumors effectively from the surface down to 1.5‐2cm depth. 3.1.4 Conclusions and Future Work
Until now, there has not been a reliable, cost‐effective method to non‐invasively heat deep seated small animal targets like murine bladder while maintaining normothermic skin and body core temperatures. To create an RF heating device specifically for this purpose, we used electromagnetic simulations to optimize the design of miniature water‐loaded circular waveguide antenna operating at 2.45GHz. We built an optimized prototype and validated its utility for heating murine bladder with laboratory and pre‐clinical measurements. The results from these studies confirm that we have built a low‐cost microwave applicator that effectively, and non‐invasively, 59 heats murine bladder while maintaining normothermic surface and core body temperatures. The device described and implemented in this study can form the backbone of a preclinical bladder hyperthermia program. Temperatures achieved in the mouse bladder are quite similar to what we have achieved clinically in patients using the BSD2000 deep regional heating system. The similarity in temperatures achieved will permit pre‐clinical studies to be performed that recapitulate the temperatures achieved clinically. Thus the development of this hyperthermia device will allow us to screen drugs in bladder tumor bearing mice for their efficacy in combination with heat (78, 79). Additionally, we will be able to study the pharmacokinetics and pharmacodynamics of novel chemotherapeutic drugs packaged in thermally sensitive liposomes thereby refining the tools available for clinical study in humans (80‐82). 60 3.2 Murine Brain Hyperthermia
In brain cancer therapy, many tumoricidal agents that are highly effective in vitro are ineffective in vivo because the blood‐brain barrier prevents therapeutic concentrations of the tumoricidal agents from reaching the tumor site (83). Microwave‐
induced hyperthermia is an effective method for increasing the permeability of the blood‐brain barrier (84). However, this increase in permeability is not necessarily specific to the tumor site so additional methods are needed to protect healthy brain tissue from the dangerous effects of chemotherapeutic drugs. The sequestration of drugs within liposomes is one method that is currently being explored as a means for administering a higher dose that is preferentially delivered at the tumor site. Liposomes can also be engineered to release encapsulated drugs at specific temperature thresholds (i.e., temperature‐sensitive liposomes (TSLs)). In preclinical flank tumor models, heating the tumor to 42°C followed by injection of doxorubicin‐loaded TSLs resulted in a 30‐fold increase in drug delivery to the tumor compared with free drug, and a fivefold increase in drug delivery compared with non‐thermally sensitive liposomes (85). In the case of brain tumor treatment, hyperthermia could potentially serve to both increase permeability of the 61 liposomes across the blood–brain barrier and also act to promote release of liposome‐
encapsulated therapeutic agents into the tumor. Up to now, interstitial microwave antennas have been used as a reliable method for heating murine brain (84). The work described in this chapter demonstrates a reliable non‐invasive microwave heating device that is designed to heat murine brain tumors through a layer of Mylar® fixed over a cranial window. Being able to heat through the cranial window is an especially important feature that enables optical imaging of the brain tumor directly post‐heating, allowing for the observation and quantification of fluorescent dye or drug (such as doxorubicin) extravasation from the liposomes in both tumor and healthy tissues. 3.2.1 Methods
The antenna optimization methods are described in 3.1.1‐3.1.3. However, the target is now brain tumor rather than bladder and a higher operating frequency, 4.2GHz, was also selected due to the relative size and location of murine brain tumors. The tissue properties used in electromagnetic and thermal simulations were obtained from literature (57, 86) and are listed in Table 3.2. 62 Table 3.2 Properties of murine tissues at 4.2GHz(57, 86) Specific Thermal Dielectric Electrical Conductivity, Constant , ε Conductivity, Heat, Cp k (W/mºC) σ (S/m) (J/kgºC) 40.281 2.7781 3648 0.815 3824 0.51 10.43 0.7737 1289 0.4 Brain† Blood Skull (cortical bone) Tumor(87, 88) 45.54 6.21 3500 0.6 †Values are an average between gray matter and white matter. Density, ρ (kg/m3) 1041 1060 1990 1020 3.2.1.1 Accounting for the Thermal Effects of the Superior Sagittal Sinus Unlike bladder, which is a non‐perfused target, the brain contains large blood vessels that play a role in its thermoregulation. Therefore, a thermal model of brain tumor hyperthermia must account for the addition of large veins nearby that act as a heat sink. In this case, the superior sagittal sinus helps regulate brain temperature which will counteract hyperthermia. This is accounted for in the thermal model by the addition of a 3.5mm diameter blood vessel with a brain‐blood vessel boundary (defined by Eqn 14) that serves as a heat sink. Eqn 14 In Eqn 14, (K) is the initial temperature of the brain, (W m‐1 K‐1) is thermal conductivity of brain, (W m‐2 K‐1) is the convection coefficient of blood vessels, and (K) is blood temperature inside the vessel. Experimental measurements of 63 anesthetized murine brain with cranial window revealed temperatures a few degrees lower than body temperature, so T was defined as 308.15K. Tbl was defined as core body temperature, 310.15K, rather than accounting for the effects of anesthesia because the body temperature was maintained in these experiments with the addition of a feedback‐
controlled mat placed under the mouse. was defined as 1800W m‐2 K‐1 (89). 3.2.1.2 Phantom Studies for Thermometry Inserting Luxtron® fiber‐optic temperature sensors (LumaSense Technologies, Santa Clara, CA) into the murine brain proved to be fatal, so a phantom brain was created to assess antenna performance via thermometry. The brain phantom had properties similar to those described in the literature (57) and was composed of a jellified mixture of salt water and oil as described by Yuan et al (90) (see Appendix D for a detailed explanation of phantom development methods). The gel phantom was contained in a 4.6x5.6cm box that was sealed with a 0.125mm Mylar® window (the same window used in murine brain heating experiments). Channels for the temperature sensors were created by inserting 20ga catheters across the box at 1mm, 2mm and 3mm depths with the configuration shown in Figure 3.8a. Windows were cut out of the catheters to ensure direct contact between the temperature sensors and the heated phantom (Figure 3.8b). Ultrasound gel was used between the antenna and the Mylar® 64 window to en
w
nsure good ccontact (the same proced
dure is used
d in murine b
brain heating
g ex
xperiments). (a
a) (c) (b)
Figure 8 a) Schem
matic of braiin phantom setup with temperaturee sensor chaannel placement, b
p
b) picture off temperaturre sensor chaannels with
h window cu
utouts, c) piccture of final b
box with brain phantom
m, Mylar® w
window and
d antenna pllaced on top
p The teemperature ssensors weree connected to a motor tthat moved the sensors allong the cha
annels in 0.5m
mm incremeents. The mo
otor was useed to position
n the teemperature ssensors with
hin the anten
nna’s “hot sp
pot,” after w
which the sen
nsors remain
ned in
n their fixed positions. T
The temperattures were ccollected and
d displayed iin real‐time using LabVIE
EW softwaree (National In
nstruments,, Austin, Texxas), which aallowed the user to
o manually ccontrol appliicator powerr as needed to achieve th
he desired teemperature rise in
n the phanto
om. In both tthe phantom
m and in vivo studies, thee applicator w
was tuned fo
or op
peration at 3
3.4GHz for improved am
mplifier perfformance at >20W. 655 3..2.1.3 In Vivo Murine Brrain Tumor Heating Stu
udies Model. A recta
angular cran
nial window
w is made usiing a low‐sp
peed Craniaal Window M
dental drill w
with irrigatio
on. The wind
dow extendss from the brregma to lam
mbdoid sutures an
nd is centereed on the miid‐sagittal su
uture. The d
dura is puncttured and ex
xcised and 1ˣ 10 D270 glioma cellss are implantted at a deptth of 3mm u
using a Hamilton syringee. A hin piece of M
Mylar® is pllaced over th
he cranial w
window and is glued to tthe surround
ding th
bone (Figure 3.9). The miice are then g
given 12 day
ys to recoverr and allow tumor grow
wth. Figure 3.9
9 Cranial Wiindow Mod
del Heatin
ng Protocol. T
The mouse iss anesthetizeed with an IP
P injection o
of Nembutall and ntrolled heatting pad, a C
CMA 450 Tem
mperature C
Controller (C
CMA placed on a feeedback‐con
Microdialysis
M
s AB, Solna, Sweden), to
o maintain bo
ody temperaature at 37˚C
C. The head is fiixed in a sterreotaxic fram
me and the a
antenna is pllaced over th
he tumor on the cranial 666 window with ultrasound gel to ensure good contact. The brain temperature is given 5 minutes to equilibrate to the effects of warm circulating water in the antenna. The amplifier is then turned on to deliver 25W of energy for 3 minutes, at which time an injection of 6mg/kg Dox LTSLs is administered via the central retinal vein. The heating is continued for another 10 minutes. The tumor is then imaged through the cranial window using a Zeiss 780 confocal microscope. 3.2.2 Results
With a fixed backshort distance of 0.25λ, a parameter sweep in HFSS™ determined the optimal coaxial probe length to extend across the entire width of the waveguide. At 4.2GHz, the dielectric properties of DI water decrease the wavelength from 7.1cm to 7.9mm resulting in a waveguide aperture of 4.5x2.5mm. Figure 3.10 shows pictures of the 4.2GHz applicator with an adjustable backshort for frequency tuning. 67 Water circulation
n tub
bes Adjjustable bacckshort Figure 3.1
10 Prototypee 4.2GHz ap
pplicator After waveguide construction
n, S11 was meeasured and
d compared tto simulated
d reesults. The S
S11 simulation
n results aree most accuraate (≤ 5%) arround 4.2GH
Hz. Figure 3.11 demonstratess less than 5%
% differencee between sim
mulated and
d measured S11 around th
he crritical 4.2GH
Hz design freequency, witth less than 55% power reeflected at 4.2GHz. Fiigure 3.11 S11
1 : HFSS™ simulation c ompared to measureme
ents using aan Agilent E5071C netw
work analyzeer. 688 Figure
e 3.12 showss the SAR pa
attern simulaated in HFSS™. Simulattions of SAR
R in th
he mouse mo
odel indicatee that maxim
mum power deposition w
will occur in
n the tumor. However, eff
H
ects of perfu
usion will alsso dictate th
he resulting ttemperaturees in the tum
mor an
nd surround
ding normal brain tissuee. Figure 3.113a presents the effect off the largest nearby blood
d vessel, the ssuperior sag
gittal sinus, o
on the resultting temperaature profilee. It is reeadily seen tthat the vesssel will do itss part in therrmoregulatiion and coun
nteract the efffects of the deposited m
microwave en
nergy. This iis useful in llocalizing th
he region of hyperthermia
a. For examp
ple, a tumor located in th
he left hemissphere can b
be heated to 42‐
43
3oC while th
he right hemisphere is att lower temp
peratures. Applicator Coverslip Tumor Figurre 3.12 HFSS
S™ SAR sim
mulation in m
mouse mod
del a) x‐axis p
profile and b) y‐
axis profi le 699 (b) (a
a) ure 3.13 COM
MSOL simu
ulation of 2D
D temperatu
ure profiles iin mouse brrain Figu
and brain tum
a
mor underg
going hypertthermia treaatment a) wiith the effeccts of the sag
gittal sinus a
and b) witho
out the effeccts of the sag
gittal sinus In Fig
gure 3.13b, itt is apparentt that withou
ut the sagittaal sinus, both
h the tumor and th
he non‐tumo
or‐bearing heemisphere w
would achiev
ve slightly h
higher tempeeratures with
h the sa
ame input power. The reesults in Fig
gure 3.13b arre analogouss to a brain p
phantom stu
udy where the sag
w
gittal sinus is not presen
nt. Figure 3.114 demonstrrates the resu
ults of a phantom stud
dy in which 22W was ussed to achiev
ve a 5‐6.5°C rise in temp
perature in th
he reegion of inteerest. Based o
on these resu
ults, we can assume thaat temperatures of 41‐42..5°C would be ach
w
hieved in thee region of in
nterest if thee initial muriine brain tem
mperature iss 36
6°C. However (as shown
n in Figure 3
3.13), once th
he effect of tthe sagittal ssinus is taken
n in
nto account, it is evidentt that a higheer input pow
wer would b
be needed to
o counteract the 700 efffect of the ssagittal sinuss; therefore, 25W was ussed for the in
n vivo murin
ne brain hyperthermia
a studies. Figu
ure 14 Chan
nge in tempe
erature of th
he brain phaantom as meeasured in fo
our lo
ocations before and afteer 22W is ap
pplied In vivoo murine bra
ain tumor hy
yperthermiaa studies werre conducted
d to assess th
he feeasibility of u
using this deevice to deliver liposom
mes across thee blood‐braiin barrier an
nd to promote the rrelease of lip
posome‐enca
apsulated do
oxorubicin in
nto the tumo
or. As seen iin he combinattion of heat a
and doxorub
bicin‐loaded
d TSLs delivers a higher Figure 3.15, th
oncentration
n of doxorub
bicin than eitther of the o
other two casses: 1) no heaat with co
711 doxorubicin‐loaded TSLs or 2) no heat with free doxorubicin. This is evident in both tumor and normal vasculature. 72 FIT
TC Doxoru
ubicin Combin
nation a) Normal Vasculature V
No T
TSLs,
No Heat Tumor Vasculature V
b)) Normal Vasculature V
With T
TSLs,
No Heat Tumor Vasculature V
733 c)) Normal Vasculature V
With T
TSLs,
With
Heat Tumor Vasculature V
Fiigure 3.15 Zeiss 780 con
nfocal micro
oscope imagees taken of m
mouse brain
n throug
gh cranial wiindow model a) after frree doxorub
bicin injectio
on, b) after doxorubicin‐
d
‐loaded TSL
L injection, a
and c) after doxorubicin
n‐loaded TS
SL injection with heat ttreatment. Scale on imag
ges indicatees 100μm. 3.2.3
3
Discu
ussion
A miiniature wa
ater‐loaded waveguide applicator operating at 4.2GHz was op
ptimized, bu
uilt and testted for its feasibility f
in
n facilitating
g the deliverry and releaase of doxorubicin‐lloaded TSLss in murine brain tumorrs. The appllicator heatss small regio
ons at depth (2‐3mm
m) while ma
aintaining no
ormothermiic core body
y temperaturres with thee help of a feedback
k‐controlled
d heating pa
ad placed u
under the m
mouse. Otherr designs caan be co
onsidered. The T water bo
olus that is naturally in
ntegrated intto the devicce itself was only used to keep the applicattor from oveer‐heating, an
nd did not sserve to cool or warm murine tiissues in thiis experimen
ntal setup. A A design th
hat eliminatees the need for a circullating 744 water system would decrease the complexity of future heating systems for studies in murine window chamber models. Measured S11 of the prototype applicator matches the simulated S11 within 5% at 4.2GHz. The small difference may be explained by the convergence criteria used in simulation; convergence was met when a maximum change of 5% in the S‐parameters was achieved between two consecutive adaptive mesh passes. Greater than 5% error at other frequencies is due to the fact that a dispersive model was used for the different tissues in the HFSS™ model and convergence of S parameters was only required at the frequency of interest (4.2GHz). Simulated SAR indicates that highest power deposition occurs within the tumor. SAR could not be measured and validated in experimental studies because the electric field probe (APREL Laboratories, Ottawa, Ontario, Canada) utilized to collect SAR data is too large (5mm diameter) relative to the extremely small antenna aperture (4.5x2.5mm). A simple thermal model was created to simulate the thermal effects of the murine brain hyperthermia treatment. It is evident from this model that temperatures can reach up to 42.5˚C in the brain tumor in one hemisphere of the brain while the other hemisphere remain at lower temperatures due to the heat sink effects of the superior 75 sagittal sinus. Temperatures could not be measured and validated in vivo for two reasons. First, the size of the Luxtron® fiber‐optic temperature sensors (LumaSense Technologies, Santa Clara, CA) that were inserted into the brain tumor would result in mouse death before the RF applicator was turned on. Second, near‐infrared images were inadequate for thermometry due the inability to image through a Mylar® window. Therefore, thermometry was performed in a phantom model. The main difference between the phantom and in vivo models is that the phantom model lacks perfusion. The effects of this difference were assessed via thermal modeling, and the results demonstrate the ability to raise the region of interest in murine brain from 36°C to 41‐
42.5°C with an input power of 25W. A visual analysis of the mice that successfully underwent heat treatments revealed no burning. Furthermore, confocal imaging of tumor‐bearing murine brain treated with DOX‐loaded TSLs revealed drug extravasation in the tumor post‐heating while no drug extravasation was present in the hemisphere opposite that of the tumor‐
bearing hemisphere. This suggests that the device is able to heat the mouse tumor to at least 41.5˚C (the temperature at which the TSLs will release doxorubicin) while maintaining relatively lower temperatures in regions far from the tumor. 76 3.2.4 Conclusions and Future Work
Until now, there has not been a reliable method to non‐invasively heat murine brain tumor through the cranial window model. To create an RF heating device specifically for this purpose, we used electromagnetic simulations to optimize the design of miniature water‐loaded rectangular waveguide antenna. We built an optimized prototype and validated its utility for heating murine brain with laboratory and pre‐
clinical measurements. The results from these studies confirm that we have built a low‐
cost microwave applicator that effectively and non‐invasively heats murine brain tumor to temperatures that facilitate the delivery and extravasation of doxorubicin‐loaded TSLs while maintaining lower temperatures in the non‐tumor‐bearing hemisphere. Some future work is required to determine reproducibility of the results seen in Figure 3.15. Once reliability has been assessed, this device will form the backbone of a preclinical brain tumor hyperthermia program. With this device, scientists are now able to non‐invasively (through a cranial window) treat brain tumor‐bearing mice with drugs and drug‐loaded TSLs to assess their efficacy in combination with heat. 77 4. Metamaterial Antennas for Pediatric Bladder
Hyperthermia
Vesicoureteral reflux (VUR) is a dysfunction of the urinary system that is prevalent in 30% of children diagnosed with urinary tract infection (91). It is characterized by retrograde flow of urine from the bladder to the kidneys. It is currently diagnosed by injection of a contrast agent into the bladder via invasive Foley catheter followed by x‐ray imaging of the kidneys. Because of the long‐term effects of undergoing ionizing radiation as well the psychological trauma associated with this procedure, a non‐invasive, non‐toxic modality for VUR detection is highly desirable. One novel approach for detecting VUR involves gently warming urine in the bladder while monitoring kidney temperature with a microwave radiometer. For a more detailed explanation of VUR and how it’s possible to predict refluxed urine volume from radiometric power measurements, please see the Master’s thesis written by Fabio D’Isidoro (92). This chapter describes the optimization and testing of a small metamaterial antenna for pediatric bladder heating. To heat the urine non‐invasively, a small microwave heat antenna was optimized to maximally deposit energy into pediatric bladder. The antenna can easily be integrated into a belt that would comfortably fasten around a child’s pelvis. 78 It is allso importan
nt to note tha
at these mettamaterial an
ntennas may
y also be useed for pre‐clinical blladder canceer treatmentt studies in laarger animaal (i.e., canine) models (d
discussed fu
urther in secttion 4.3). 4.1
4 Metam
material Antenna
A
Design
D
Figure
e 4.1 is a sch
hematic of th
he metamateerial antennaa. The antenn
na has four la
ayers of dieleectrics and ttwo layers off metal. Onee of the metaal layers is th
he microstrip
p feeed line (loca
ated above L
Layer 2) and
d the other m
metal layer co
ontains the g
ground plan
ne, dielectric slotts and coppeer rectangula
ar patch reso
onators (locaated above L
Layer 1). Figure 16 Sc
F
chematic of Metamateriial Antenna The ch
haracteristicc impedance of the micro
ostrip feed liine is determ
mined by its efffective widtth (
), acctual width ( ), dielectriic constant o
of substrate (( ), thicknesss (“
“height”) of substrate ( ), thickness of strip metaallization ( )), and imped
dance of freee sp
pace (
). Fo
or this anten
nna, the dieleectric constan
nt of the sub
bstrate is nott only 799 determined by the substrate layer in between the microstrip line and the ground (i.e., the metal layer containing the dual rectangular patch loaded slot), but also by the two superstrate dielectric layers above the microstrip line. Some care must be given to the selection of the superstrate layers (93) for optimal directivity. Setting the characteristic impedance of the microstrip to 50‐ohm (to match the SMA input), the effective width is calculated using Eqn 14 and then the actual width is determined by using Eqn 15. The line width, as well as the dielectric constants and thicknesses of the layers involved in the impedance calculation are then fixed to maintain the 50‐ohm impedance match. In other words, the only layer with variable dielectric constant and thickness for optimization is the layer directly below the ground plane (Dielectric Layer 1). 2
2 1
ln 1
Eqn 14 4
14
11
8
4
14
8
4
11
1
1
2
Eqn 15 80 1
1
2
ln
4
1
1
11
10
The microstrip line length is important in determining the resonant frequencies. Phase delays along the line create the backwards wave effect. The Z‐shape of the line allows for the phase shift necessary for constructive or destructive behavior between the coupled lines that feed into the upper and lower edges of the slot. Using a simple microstrip line length calculator, phase delays of 58°, 56°, and 51° respectively will provide approximately 180° phase addition (at multiple frequencies) between the two slot edges. Note that the line length is dependent upon the effective permittivity of substrate and superstrate layers which is less trivial in this multi‐layered antenna. Therefore, it is best to optimize these parameters as if they are variables in the design. The slot width and height and rectangular patch width and height can be optimized for a resonant frequency as well. Figure 4.2 shows some rule‐of‐thumb ratios for the slot and rectangular patch dimensions. The ratios stated are good starting points, but should be variable parameters in the design optimization to achieve optimum performance for the desired antenna specifications and applications. 81 Figurre 17 Antenn
na slot and rrectangular patch dimeensions in teerms of effecctive wavelengtth As apparent in these ratios, th
he physical aantenna leng
gth (L) is dessigned with slot siizes smaller than λ/5. Att this dimenssion, the ante
tenna itself ccan be defineed as a mateerial with its own w
effective ma
aterial propeerties. A morre mathemattically rigoro
ous analysiss that defines this a
antenna as a metamateria
al can be dettermined by
y the disperssive responsee of th
he antenna. O
One dispersive function
n, group delaay ( ), is deefined by Eqn 16(94). From
m Eqn 16, it is eevident that a negative g
group delay results in a n
negative ind
dex of refracttion. 822 By causality, negative index of refraction can only occur when both effective permittivity and permeability are negative. Eqn 16 (where c is the speed of light in a vacuum)
Figure 4.3 shows the phase of S21 (our transmitted wave) for a 433MHz metamaterial antenna. The slope changes from negative to positive in the region around 860MHz, indicating a backward wave. In that same region, group delay also becomes negative, indicating an effective negative index of refraction. A material with negative index of refraction (metamaterial lens) focuses energy in the near field. Note that the electric field pattern does not change dramatically across frequencies (i.e., the near field focus is observed across a large frequency band). For hyperthermia, good transmission (impedance match) is necessary and a focus point that is farther away from the antenna surface is desirable. Therefore, the metamaterial resonance frequency is used to create the focusing effect but also acts as a band‐stop. The antenna is operated at lower frequencies (at half resonance) for deep tissue heating. 83 Figure 18 At the m
magnitude S
S21 minimum
m (860MHz)), a positive phase slopee and negative
e group dela
ay are obserrved (highlig
ghted in red
d), while thee antenna iss imped
dance match
hed across th
he 200‐500M
MHz band (h
highlighted in green). 844 Based on these design parameters it is easy to see that this antenna can be scaled to operate at any frequency of interest. Increasing the dielectric constant of the multi‐
layered antenna will allow for a decrease in the effective wavelength and therefore a miniaturization of the overall antenna size. Both of these features are used to design small, high dielectric antennas that operate at 915MHz and 180MHz. This chapter will discuss the design optimization and validation of the 915MHz antenna and the next chapter will focus on the 180MHz antenna. 4.1.1 Benefits of this Design
Other than the ease at which we can scale this simple design and the fact that it can be constructed with flexible dielectric materials, an important advantage that this metamaterial antenna has over the conventional gold standard hyperthermia antennas is its ability to focus energy at depth. As seen in Figure 4.4a, the 915MHz metamaterial antenna has a focus point 2mm at depth in a homogenous muscle tissue load while the equivalent 915MHz waveguide has an exponential drop‐off from the surface. This is especially important in the pediatric bladder heating application where the bladder is located approximately 5‐10mm at depth. From Figure 4.4a it is evident that the metamaterial antenna would provide a dramatic increase in power delivered to the target region. 85 (a
a) (b
b) (cc) ure 19 (a) The
e metamaterrial antennaa can have im
mproved po
ower deposition Figu
at depth wh
a
hen compare
ed to the equ
uivalent wav
veguide app
plicator; (b) the focus point at depth can be manip
pulated by v
varying the antenna geo
ometry – in this examplle, Layer 1 th
hickness hass been varied
d to move th
he location o
of the focuss point; (c) th
he equivalent of (b) excep
pt that these images reveeal the SAR
R pattern insstead of E‐fieeld. 866 Unlike the conventional metamaterial community that seeks to have a focus point as close to the lens as possible (i.e., for the case of sub‐wavelength, high resolution imaging), a metamaterial antenna designed for hyperthermia may want the focus point “far” from the antenna surface – though still within what is considered to be the “near field” (e.g., at 915MHz, the near field at depth in tissue is within 4.5cm). Therefore, this antenna is designed such that the backward wave phenomenon acts as a band‐stop with a wideband impedance match at lower frequencies. The antenna is then operated at lower frequencies (around one‐half the resonance frequency where the band‐stop is located) where the focus is farther from the surface of the antenna. 4.2 Antenna Optimization for Pediatric Bladder Hyperthermia
Although the design of the metamaterial antenna is vastly different from the waveguide‐based antenna in chapter 3, similar optimization methods are used. Various geometric parameters of the antenna are optimized to obtain an impedance match across the frequency band of interest, and to maximize the percent power deposited in the bladder. However, the metamaterial antenna has an additional point of interest, and that is to optimize the position of the near‐field focus. Ideally, this focus point would be located in the bladder or at the very least beyond the skin and/or fat layers. To 87 determine the ideal range of depth for the focus point and thereby design an optimal antenna for pediatric bladder heating, an average patient model was developed. 4.2.1 Average Patient Model
To maximize the total power that the antenna deposits into the bladder for the greatest number of children, an average patient model was constructed based on measurements from CT scans and ultrasound images of 40 children (Figure 4.5a). The resulting “average child” has skin thickness of 1mm, fat thickness of 3mm, and bladder isocenter depth of 3.8cm. Bladder volume was highly variable, so for the purposes of this study 150cc was selected based on the amount of contrast agent injected for the x‐
ray imaging studies. This means the extent of the full bladder is located between 5mm and 7.1cm deep. A 3D simulation model was created using VSG Avizo® by segmenting the CT data set of the constructed “average child” anatomy (Figure 4.5b). The model was imported into Ansys HFSS™ and each tissue was assigned dielectric properties. 88 (b) (a
a) 180
160
140
Bllue: Children
n <3yrs old
Red: C
Children betw
ween 3‐7yrs old 120
100
80
60
40
20
0
Fat Layer (mm) Isoceenter Depth
(mm)
Volume (mL)
Fat layer thiickness, blad
dder isocenter depth an
nd volume aas Fiigure 20 (a) F
determined
d by 40 patie
ent CT and U
Ultrasound images; (b) 3D averagee patient mo
odel created
d by image segmentatio
on in VSG A
Avizo® and imported in
n HFSS™. 4.2.2
4
EM-T
Thermal Siimulation--Based Op
ptimizatio
on
The in
nitial metam
material anten
nna was dessigned to opeerate at 433M
MHz. Using the ru
ule‐of‐thumb
b ratios desccribed in secction 4.1, the antenna waas scaled to o
operate at 91
15MHz and miniaturizeed by selectin
ng high diellectric constaant substratees. The miniaturized m
915MHz an
ntenna was th
hen placed o
on the child model to ch
haracterize im
mpedance m
match and po
ower deposittion into thee bladder. As can
n be seen in Figure 4.6, tthe frequenccy band overr which the antenna is im
mpedance m
matched to th
he child mod
del (850‐12000MHz) narro
ows with thee addition of a water bolus (
w
825‐1050MH
Hz). As a safety precautiion, a water bolus is stilll used to keeep 899 surface tissues cool while heating at depth, and the not too significantly narrowed band of operation is not a great concern. Frequency (GHz)
0
0
0.5
1
1.5
2
2.5
‐5
S11 (dB)
‐10
With Bolus
‐15
No Bolus
‐20
‐25
‐30
Figure 21 Reflected power of the 915MHz antenna on the child model, with and without a water cooling bolus The power deposition pattern in the pediatric model is shown in Figure 4.7. The efficiency, as defined by Eqn 11, determines the percent of power deposition in the bladder and was calculated to be 25.2%. 90 Figure 22 S
SAR pattern
n in pediatriic bladder From simulated eefficiency alo
one, it is not possible to d
determine w
whether the an
ntenna can d
deliver the a
amount of po
ower needed
d to raise thee urine temp
perature by 1°°C/minute w
while keeping the surfacee normotherrmic. Time iss needed before the anteenna ca
an be tested for its abilitty to heat pediatric bladd
der urine in clinical trialls. Thereforee, to validate the a
antenna’s peerformance, a
a virtual phaantom study
y was condu
ucted using Ansys HFSS™
A
™ and COM
MSOL Multip
physics ®. Th
he results off the virtual phantom stu
udy ca
an be valida
ated via laboratory meassurements an
nd would prrovide the in
nsight needeed to fu
urther optim
mize the anteenna for pediatric bladdeer heating. 911 Figure 4.8a shows a cross‐section of the phantom study set‐up with an SAR plot overlay. The distance between the bladder phantom and the tank’s Mylar window was increased (Figure 4.8b) until the same percentage of the total power is deposited in the bladder phantom as was seen in the child model (25.2%). This occurred when the front of the bladder phantom is located around 1.45cm away from the Mylar window. (a) (b) 92 % Power Deposition in Bladder
28
Child Model 27
26
25
24
23
13
13.5
144
14.5
15
Depth in T
D
Tank (mm)
Fig
gure 23 (a) SAR pattern in the virtu
ual phantom
m study model; (b) Perceent power depo
osited in the
e bladder ph
hantom as d
distance betw
ween the ph
hantom and the Mylarr window is increased After the approprriate position
n was determ
mined, the H
HFSS model and SAR weere trransferred in
nto COMSOL
L for EM‐Th
hermal simu lations (i.e., a “virtual phantom stud
dy”). The flow rate
T
e of the liquid
d muscle ph
hantom in th
he tank was sset to maxim
mum to main
ntain 37
7°C. This wa
as done to m
more easily validate the rresults seen iin simulatio
on. Once the siimulation reesults proved
d to be an acccurate meassure of anten
nna perform
mance, the mu
uscle phantom can be set to have the same temperaturre‐dependen
nt perfusion properties aas ans for a relia
able predictiion of expeccted pediatriic bladder heeating seeen in huma
performance.. 933 As seeen in Figure 4.9, when th
he muscle ph
hantom is ciirculated to maintain 377°C he cooling bo
olus was sett to 33°C, thee fat phantom
m layer was maintained
d at temperatture th
between 33‐37°C. With on
ne antenna o
operating at 45W, 90% o
of the urine iin the bladdeer was brought w
up to 42°C a
at steady statte (T90 = 42°C
C). ure 24 Plot o
of temperatu
ures in the v
virtual phan
ntom study m
model. The fat Figu
phantom tem
p
mperature iss maintained between tthe cooling b
bolus and m
muscle phan
ntom te
emperaturess while the b
bladder pha
antom is heaated to 41.8°°C. The arrow
ws indicate flow patttern in the ccirculated m
muscle phanttom that is m
maintained aat 37°C. 944 4.2.2.1 One vs. Two Antennas Taking normal perfusion characteristics into account, it is highly unlikely that the temperature of urine in the bladder can be raised to 42°C without some heating in the intervening layers of fat and muscle. To prevent overheating in those layers, two antennas should be used to heat the bladder. The volumes of fat and muscle under each antenna are independent of each other while the volume of urine in the bladder is shared. By doing this, the thermal dose in non‐target tissue volumes is reduced. Figure 4.10 demonstrates that the same T90 can be achieved with less power in each of the two antennas. For example, a T90 of 42°C at steady state requires 45W from one antenna or 30W each from two antennas. 95 47
Temperature (°C)
46
45
44
43
1 antenna
42
2 Antennas
41
40
39
15
25
35
45
55
65
Power (W)
Figure 25 Plot of T90 in the bladder phantom as a function of power for the case of 1 antenna vs. 2 antennas 4.2.3 Experimental Validation of Metamaterial Antennas
10) The 915MHz metamaterial antennas were constructed on Rogers 3010 (
substrate with Emerson and Cuming ECCOSORB custom high dielectric (
15, 30) flexible materials as superstrate layers (Layers 3 and 4) and coverlay (Layer 1). The non‐
flexible Rogers substrate was used because it was easier to align the microstrip feed line across the slots properly when they were printed on a rigid substrate. When the structures are larger (i.e, in the case of a lower frequency antenna), it is easier to manufacture the antenna using the flexible materials for every layer. This is 96 demonstrated
d in chapter 5. Fortunateely, it is morre importantt for larger antennas to b
be made with fle
m
exible materrials than sm
maller antenn
nas like the 9915MHz. Simullated results for S11 weree compared tto measured
d results on tthree in
ndependent subjects (Fig
gure 4.11). Itt is importan
nt to note that the antenn
na is impedaance matched over
m
r a broader b
band than siimulated ressults, but this is expected
d. When the model contain
m
ns dispersiv
ve media (i.e., tissue load
ds), simulateed results aree most accurrate att the center rresonant frequency. Thee variation seeen between
n subjects is the motivatiion behind design
ning a broad
der impedan
nce matched
d antenna. If the trough m
moves to slig
ghtly lo
ower or high
her frequencies, the centeer of the ban
nd is still hass good S11. Figu
ure 26 Comp
parison of siimulated refflected pow
wer and meassured refleccted power on
n 3 independ
dent subjectts 977 The results from the virtual phantom study were validated in laboratory measurements. Figure 4.12a shows a picture of the phantom study setup that is based on the virtual phantom study setup. The muscle phantom was circulated to maintain 37°C and the water cooling bolus was circulated to maintain a temperature that was slightly above room temperature 27°C. Results of the temperature measurements from Luxtron fluoroptic temperature sensors that were placed between the cooling bolus and the fat phantom, between the fat phantom and the muscle phantom, and inside the bladder phantom are seen in Figure 4.12b. As expected from the virtual phantom study, the fat phantom maintained temperatures between the cooling bolus and muscle phantom while the bladder phantom temperature increased from 37°C to 42°C. 98 (a
a) (b
b) Figu
ure 27 (a) Piccture of labo
oratory phan
ntom study setup; and (b) data from
m a phantom stu
p
dy where po
ower was tu
urned on at tthe 8 minutee mark and turned off aat the nute mark (a
approximate
ely 10 minuttes after steaady state waas reached)
45 min
Minorr differencess exist betweeen the phan
ntom study aand experim
mental resultss, but th
hey can all b
be accounted
d for and corrrected. In th
he experimen
nt, the ampliifiers were sset to 999 output 38W into each antenna, while in simulation only 30W was needed in each antenna to achieve a T90 of 42°C in the bladder. S21 measurements revealed a 10W loss in the cables between the amplifiers and the antenna, indicating that only 28W was transmitted into the phantom. In the experimental study, a point measurement was taken in the bladder phantom, so there is no way to know if 90% of the phantom was at 42°C. When the simulation is run with only 28W, temperatures of 40.93‐42.65°C are seen in the bladder and T90 is 41.3°C. Therefore, it is possible that 30W may be needed for T90=42°C while 28W was used because the point measurement taken indicated that it was enough to achieve 42°C in the bladder. Note that (according to simulated results) when T90 for the bladder phantom is 42°C (achieved with 30W into each antenna), a point measurement of temperature in the bladder could be anything between 41.76‐
43.31°C. 4.3 Expected Antenna Performance for Pediatric Bladder Heating
Because the differences between experimental and simulation results are in agreement, it is not unreasonable to suggest that a EM‐Thermal model can be utilized to provide useful insight into what temperatures can be expected in the various tissue layers when heating pediatric bladder with these antennas. Therefore, next steps were taken in simulation to predict pediatric bladder heating performance. The average child 100 model was im
m
mported into
o COMSOL a
and temperaature‐depen
ndent perfusiion ch
haracteristiccs were assig
gned to each tissue (Tablle 5.1). The rresults from
m two‐antenn
na heating with a cooling bo
olus tempera
ature of 30°C
C are seen in
n Figure 4.133. ure 28 Stead
dy‐state temperature pro
ofile in two cross‐sectio
ons (xy and xz) Figu
across the center of tthe average child modell bladder affter two‐anteenna heating n, the power outputted th
hrough each
h antenna was only 8W In thiss simulation
when the coo
w
oling bolus w
was set to 20°°C. This low
w power leveel may seem peculiar wh
hen co
ompared to the power n
needed in thee phantom sstudies, but iin fact this iss a reasonab
ble prediction; heeat removal is much mo
ore efficient iin a high‐flo
ow circulatin
ng phantom than in
n this case w
where realistiic, less‐efficient perfusio
on is counterracting the h
heat. As seen
n in Figure 4.13, h
heating the p
pediatric bladder to T90=442°C (tempeerature range in bladderr: 1.68‐42.7°C) with the sam
me two‐anteenna configu
uration used
d in the phan
ntom study iis 41
1001 feasible without overheating the intervening tissues (<44°C). This makes a strong case for future studies to explore the use of these antennas for pediatric bladder heating. 4.4 Conclusions and Future Work
A 915MHz metamaterial antenna was optimized for heating pediatric bladder. The results from simulations accurately predicted antenna performance, and is therefore a reliable method for designing future metamaterial antennas. Chapter 4, Section 2.2.1 describes the benefits of using two antennas rather than one for decreasing thermal dose in fat and muscle tissues while maintaining thermal dose in bladder urine (Figure 4.10). For further sparing of intervening tissues, a transient study should be performed to study the effects of alternating on/off between the two antennas at various intervals. This would decrease the thermal dose in muscle and fat tissues even further, allowing the tissues under one antenna to recover while the other antenna is still heating the bladder. Once the risk of overheating fat and muscle layers has been minimized, the antennas are ready to be used in a pediatric bladder heating clinical study. Devices to heat pediatric bladder would be classified as a Class III medical device through the FDA. Data that ensures the safety of patients is needed for Class III medical devices; therefore, it is likely that this device will be tested in animals first. 102 A reassonable anim
mal model fo
or both pediiatric bladdeer heating an
nd cancer th
herapy testin
ng is the can
nine bladder model. Therre is interestt in using canine modelss to sttudy hyperth
hermia as an
n adjuvant th
herapy in blladder canceer (95‐100). F
Figure 4.14 sh
hows examp
ple models o
of medium an
nd small breeed dogs. Th
he anatomy is not siignificantly d
different fro
om pediatric bladder, so it may servee a dual purrpose to run pre‐
cllinical studiees in canine models to asssess the perrformance o
of these anten
nnas for heaating bladder as weell as to asseess the effica
acy of combin
nation hypeerthermia an
nd chemotheerapy to
o treat bladd
der cancer. (a
a) (b) ure 29 (a) Av
vizo® mode
el of a golde n retriever; (b) HFSS m
model of beaagle Figu
wiith tumor (rred) on blad
dder wall (yeellow) 1003 5. Metamaterial Antennas for Bladder Cancer
Hyperthermia Treatments
Chapter 1, section 4 details the current devices being used for bladder cancer hyperthermia research. Ideally, the current devices would be replaced by a non‐
invasive, relatively inexpensive and portable device in order to facilitate multi‐
institutional clinical studies. Metamaterial antennas may be the key developing such a device. This chapter details the first steps taken to explore the use of metamaterial antennas for heating adult bladder. 5.1 Single Antenna vs Phased Array Heating
The same design principles detailed in Chapter 4 were used to create a metamaterial antenna that operates at 110‐210MHz. The antenna was constructed using Emerson and Cuming high‐Dk ECCOSORB material. These materials are flexible, have high dielectric constants (15 and 30) and exhibit extremely low‐loss. Figure 5.1 is a picture of the constructed 915MHz and 180MHz antennas. 104 1800MHz Anten
nna Fig
gure 30 Pictu
ure of 915MH
Hz (4x5cm) aand 180MH
Hz (12x20cm)) metamaterrial antennas w
with US$0.25
5 (to indicate
e scale) and strap to seccure antennaas around to
orso Figure
e 5.2 comparres simulateed and measu
ured S11. Sim
milar to the 9915MHz an
ntenna, S11 iss measured on three ind
dependent su
ubjects and aagrees well w
with simulated reesults. Once again, the b
broadband im
mpedance m
match is a desirable qualiity that allow
ws fo
or optimal power transm
mission desp
pite any anattomical variaations seen aacross patien
nts. If th
he antennas are operated
d at 180MHz
z, the center of the band
d is protected
d across diffeerent patients and w
with or with
hout a bolus (which may
y be explored
d in the futu
ure since these an
ntennas hav
ve a power d
deposition m
maximum on the surface)). In the casee where MR th
hermal imag
ging is used iin conjunctio
on with thiss device, opeerating at 1800MHz will aalso av
void interferrence with 3
3T MRI whicch has a reso
onant frequen
ncy of 128M
MHz. 1005 Figu
ure 31 Comp
parison of siimulated refflected pow
wer and meassured refleccted power on
n 3 independ
dent subjectts The an
ntenna’s SAR pattern was also meassured and co
ompared to simulation nstrates good
d agreementt between sim
mulated and
d measured data. reesults. Figurre 5.3 demon
This reaffirm
T
s the assump
ption that simulated SA
AR is a reliab
ble heat source in the theermal siimulations u
utilized to asssess whetheer these anteennas can su
uccessfully heat adult bladder. 1006 Figure 32 Simu
ulated vs. me
easured datta for the low
w frequency
y metamaterrial antenna
a show good
d agreementt Becau
use the anten
nna is made of flexible m
materials, it w
will conform
m around thee human torso.. A scalable, conformal m
model was ccreated in HF
FSS to study
y the field patterns when the antenn
na is curved.. The SAR frrom the confformal modeel was impo
orted n COMSOL ffor EM‐Therrmal analysiis. Table 5.1 details the tthermal prop
perties assig
gned in
to
o the human
n tissues in th
his EM‐Therrmal model.
1007 Table 5.1 Thermal properties of human tissues(101, 102) Skin Fat Muscle ‐3
Density (kg m ) 1109 911 1090 ‐1 ‐1
Specific heat (J kg K ) 3390 2348 3421 ‐1 ‐1
Thermal conductivity (W m K ) 0.37 0.21 0.49 ‐1 ‐3
0.52 Basal blood perfusion (kg s m ) 2.06
0.75 ‐3
Metabolic heat generation (Wm ) 1827 462 1052 Blood 1050 3617 ‐ ‐ ‐ In Table 5.1, the following expressions were used to calculate basal blood perfusion (103, 104): 1
9.2exp
1
1
44
10
45
12
exp
7.9exp
45
12
A bolus temperature of 20°C and input power of 210W was used in the single antenna case (Figure 5.4b) and 140W was used in each antenna of the four‐antenna array (Figure 5.4c). When multiple antennas are used, a phase addition of each of their radiated fields increases the SAR in the bladder. In the four‐antenna array case, the maximum temperature in the intervening tissue layers is 44°C when the average temperature in the bladder is 42°C. This is an improvement over using a single antenna, 108 where the maximum in the intervening tissues layers is 46°C when an average temperature of 42°C is reached in the bladder. 109 (a
a) (b
b) (cc) Fig
gure 33 (a) C
Conformal m
model of 1800MHz metam
material anttenna array in HFSS (sin
ngle antenn
na model nott shown, bu
ut placed dirrectly above bladder); (b
b) temperature profile of ssingle antenna heating iin COMSOL
L; (c) temperature profiile of fo
our‐antenna
a array heatiing in COM
MSOL 1110 5.2 Shielding
One of the most important qualities of this design is the ability to shield backward radiation so the device can be used outside of a shield room (i.e., making the device truly portable and therefore more accessible for multi‐institutional clinical trials). A Wandel & Golternmann EM Radiation Meter was used to measure the field strength and power density back‐radiated from the antenna with and without an absorber and shield cloth (Table 5.2). This preliminary study demonstrates the ability to use this antenna with readily available absorbers (ARC Technologies, Inc) and shielding cloth (Ripstop Silver Fabric, Less EMF, Inc). It is important to note that absorbers will eventually heat up with power levels significantly higher than the 10W used in this preliminary study, so it is possible that more absorbers (>1cm thick) may be necessary to carry energy (i.e., heat) away from the back of the antenna. Thermal management of the back‐radiation can be done by stacking the absorbers in a less absorbing‐to‐more absorbing scheme for graded attenuation. Note that S11 does not change with the additional layers unless a shield cloth is used without an absorbing layer first. Table 5.2 Backward radiation levels measured 5cm from the back of the antenna with input power of 10W compared to FCC Requirements Nothing With With Absorber and Shield FCC Requirements (2007) Absorber Cloth 2
mW/cm 0.3 0.15 0.002 <1 V/m 26 15 2.2 <61.4 111 The antennas will require at least 100W of input power, but the data in Table 5.2 suggests that even with a ten‐fold increase, the FCC requirements for maximum exposure (over a 6 minute period) and broadcasted fields would still be met. Therefore, a thin shield cloth like the one used in this experiment should be sufficient. The potential that metamaterial antennas have for being used outside of a shielded room is reason enough to continue this work. Multi‐institutional clinical trials are more realizable with a device that doesn’t require a shielded room for operation. 5.3 Conclusions and Future Work
A low‐frequency metamaterial antenna array was designed to heat adult bladder. The results from simulations accurately predicted antenna performance, and is therefore a reliable method for designing future metamaterial antennas. An accurate adult bladder model is still required. The effect of bones and other structures were not taken into account in this model and will affect the SAR pattern and therefore heating efficiency of the antenna array. The average and extremes (very small and very large) patient models can easily be created from existing MR and CT scan data. The current HFSS model with conformal antennas described in this chapter easily scales around any patient size. 112 It is evident from the EM‐Thermal results using the simple model that using one antenna is not sufficient for heating adult bladder, and that four or more antennas are needed for phased array heating. The current antenna size is too large to accommodate more than four antennas around an average adult torso. Fortunately, the antenna can be further miniaturized by using higher dielectric substrate layers. A conformal antenna is desirable, and Emerson and Cuming’s high‐Dk ECCOSORB materials are currently the only conformal high dielectric substrate layers available on the market. The maximum off‐the‐shelf dielectric was used in this dissertation (
15); therefore, miniaturization will require custom‐made dielectric substrates, and Emerson and Cuming currently offers custom made high‐Dk ECCOSORB for a dielectric constant up to 30. Once the antenna is miniaturized so that more than four antennas can be placed around an adult torso, the same EM‐Thermal simulation procedures can be used predict heating performance. To validate the results, a simple phantom study should be conducted. Ideally, an adult‐sized, gelatin‐filled torso phantom that contains bone, muscle and bladder phantoms should be heated in conjunction with MR thermometry. Once safety and efficacy are validated in phantom and large animal pre‐clinical studies, the antenna array will be ready for use in clinical studies. 113 Concluding Remarks
Concluding Remarks
We have yet to establish the optimum combination of hyperthermia with radiation and/or chemotherapy for effective treatment of bladder cancer. Convenient and affordable microwave applicators capable of well‐localized non‐invasive heating of murine, canine and human bladder cancers is essential for logical progression of studies from pre‐clinical to multi‐institution clinical trials, as needed to investigate the effects of hyperthermia as an adjuvant treatment for bladder cancer. The primary objective of this research was to utilize state‐of‐the art segmentation and simulation software to optimize target‐specific microwave antennas for more uniform heating in pre‐clinical and clinical investigations of bladder hyperthermia. The development of a reliable antenna design methodology was tested in each of the specific aims detailed in this dissertation. By first identifying and characterizing the target tissue with accurate 3D imaging and segmentation, and then proceeding with rigorous multi‐physics simulation‐based design optimization centered on target‐defined specifications, the product development cycle was dramatically shortened, making it possible to create many successful antennas in a short time. In addition to the four 114 antennas described in this dissertation, two log‐spiral antennas were designed, built and tested for deep tissue thermal sensing applications in kidney and brain. Using the new target‐specific simulation based design methodology, a microwave applicator was developed for heating murine bladder to 40‐43˚C while maintaining surface and core temperatures normothermic, to enable pre‐clinical bladder hyperthermia studies (described in Chapter 3). It is clear that hyperthermia enhances the effects of chemo‐ and radio‐ therapies, and this device will allow scientists to investigate the basic principles underlying this phenomenon more systematically. Since the creation of this device, many research groups have shown interest in purchasing it for their own use. The requests also led to the development of a miniature microwave applicator for heating murine brain tumors (also described in Chapter 3) and preliminary work using a slightly larger antenna to heat mouse flank tumors. Results from all these studies confirm that we have built low‐cost microwave applicators that effectively heat murine tumors non‐invasively with minimal increase in core temperature. The family of murine tumor heating devices described and implemented in this dissertation can form the backbone of a preclinical bladder hyperthermia program. Temperatures in the tumors are readily controlled to match what is achieved clinically in patients undergoing hyperthermia treatments. Thus the development of these 115 hyperthermia devices has provided the ability for researchers to screen drugs in tumor bearing mice for their efficacy in combination with heat. Additionally, these pre‐clinical studies will facilitate the characterization of the pharmacokinetics and pharmacodynamics of novel chemotherapeutic drugs packaged in thermally sensitive liposomes, thereby expediting the translation of new therapeutic approaches into clinical trials. Another significant contribution of this work is the development of metamaterial lens antennas for deep tissue hyperthermia. Up to now, deep heating has always been achieved with far field phase addition of multiple antennas coupled to a patient through a very thick layer of water that many patients cannot endure for an hour‐long treatment due to the pressure exerted by the thick water bolus against the patient’s abdomen. Focusing energy in the near field (as is possible with metamaterial antennas) to increase heating efficiency beneath the tissue surface is a complete paradigm‐shift in the field of hyperthermia. Metamaterial antennas have the potential to decrease the cost and complexity of hyperthermia systems, and increase the portability of the devices because the addition of lightweight absorbers and shield cloths enable their use outside of a shield room (i.e., to be used directly in the chemo suite). The ability to design them with conformal materials and use them with a thinner cooling bolus allows for the entire 116 antenna array and coupling bolus to be integrated into a flexible belt that facilitates patient comfort. Most importantly, these improvements over previous devices will enable the multi‐institutional clinical trials required to complete FDA approvals and obtain insurance reimbursement of deep‐tissue thermal therapy. In summary, this dissertation outlines a robust method for electromagnetic and thermal design optimization of microwave antennas for heating tissue. The method was used to design novel antennas optimized specifically for four tissue sites: murine bladder, murine brain, pediatric bladder, and adult bladder. The antennas were fabricated and tested in both phantom models and in vivo mouse studies to verify computer simulated results in terms of impedance matching to tissue, SAR patterns in homogeneous muscle material, and heating patterns in realistic murine tissues. It is expected that the antennas designed in this work will find suitable applications in pre‐
clinical investigations of hyperthermia, and the methodology of site‐specific antenna design will find future applications in the design of applicators for other challenging tissue sites. This work is intended to facilitate further spread in the use of hyperthermia as an adjuvant to current cancer therapies.
117 Appendi
A
x A: 3-D
Dimensio
onal Pattient and
d Anima
al Models Sttep 1: Collecct MR/CT Da
ata Set The MR/CT d
T
data set must be in RAW
W or DICOM
M formats. To
o create an av
verage patieent model, it is be
m
est to collectt medical im
mages from aa number of p
patients witthin the disease population (i.e., in this diissertation, w
we collected
d 20 image seets from chilldren with V
VUR o create the a
average child model for VUR). to
Sttep 2: Take M
Measuremen
nts to Determ
mine the “Av
verage” Model Open the firs
O
t image set iin Avizo by clicking the “Open Dataa” button an
nd selecting aall th
he files that ccreate your 3
3D image. 1118 The images w
T
will be stored
d in one buttton that has now been crreated in thee right side‐b
bar. Click on the b
C
button that ccontains all tthe images, aand select th
he option “O
Orthoslice” w
which ap
ppears as an
n orange buttton at the to
op of the righ
ht side‐bar. You’ll be able
Y
e to scroll thrrough the im
mages using the slide‐baar located in the lower h
half of th
he right sidee‐bar and lab
beled Slice N
Number. Clicck on the Perrspective/Orrthographic
bu
utton on thee top tool barr to select an
n Orthograp
phic view. No
ow when yo
ou scroll thro
ough th
he images th
he aspect ratiio remains cconstant. Thiis is importaant when you
u are interessted in
n taking mea
asurements a
across multiiple slices. 1119 Click on the M
C
Measuremen
nt button on
n the top tooll bar and sellect 3D Leng
gth to take measurement
m
ts of details like fat thick
kness, bladd
der depth, etcc. Draw a lin
ne by clickin
ng on tw
wo points off interest in tthe image an
nd Avizo willl automaticcally calculatte the distan
nce between the ttwo points a
and display tthe measureements on th
he lower halff of the rightt siide‐bar (the measuremen
nts are storeed in the Me asurement b
button that’ss been createed in th
he right sidee‐bar). 1220 Once you’ve O
taken the m
measurements you need aacross all thee patient image sets, ca
alculate the average and
d determine which patien
nt best matcches the averrage. You w
will crreate the 3D patient mod
del out of th
his image sett. Sttep 3: Imagee Segmentatiion to Createe the 3D Mo
odel Right‐click on
R
n the button that stores tthe image seet and select Image Segm
mentation>>>>Edit New Label Fi
N
ield. The Image Segmen
ntation Edito
or will open. Under the M
Materials secction lo
ocated on thee right side‐bar, add all the organs/ttissues you w
would like to segment b
by cllicking the b
button that allows you to
o Add a new
w material an
nd then righ
ht click to Rename the a
R
additional m
material. 1221 Now you are
N
ready to seg
gment the im
mages. Notee: start from the outside and work y
your way in (i.e: st
w
tart by segm
menting the b
body contou r, then segm
ment organs inside the bo
ody, th
hen segmentt structures iinside the orrgans, etc.). T
Try the diffeerent Tools p
provided forr seegmentation
n to see whicch you prefer. I like the L
Lasso (modee: Freehand, options: Au
uto‐
Trace) for cre
T
eating the bo
ody contour. 1222 When you are
W
e done seleccting the bod
dy, click the material in tthe Materialls list (in thiss ca
ase: Body) an
nd then click
k the + butto
on to add thee selection in
nto the Body
y label. 1223 To make sure
T
e that the surrfaces of you
ur segmente d objects aree smooth, it is best to in
nterpolate in
nstead of labeling the body in every image indiv
vidually. To do this, click
k “C
Current slicee” in the Selection sectio
on. Then clicck the “selectt” button in the Materiaals list for the ma
aterial you a
are segmentiing (in this ccase: Body). T
Then scroll aahead 10‐20 im
mages in thee stack (the n
number of im
mages you caan skip depeends on how
w much these im
mages are ch
hanging as y
you scroll – y
you want yo
our interpolaation to be acccurate betw
ween th
he two imag
ges you are selecting). Ussing the Lassso tool, selecct the Body iin the new im
mage th
hat is 10‐20 iimages away
y from the im
mage you’vee already sellected. When
n you are do
one press Ctrl+I a
and you will see all the im
mages in be tween will aalso be selectted. Add all of 1224 th
he images in
nto the label. Always intterpolate bettween a slicce you’ve alrready added
d to th
he label and
d a new slice
e that’s furth
her along in
n the image sstack so the transitions are sm
mooth. When
n you are all done labelin
ng the differrent organs/ttissues that y
you want to seegment, click
k the Projectt View tab att the top of tthe right side‐bar. Right‐click on thee new bu
utton that iss located belo
ow the butto
on that contaains all the images and eends with th
he word “labels”
w
” and select Generate Su
urface. A new Gener
A
rate Surface b
button will a
appear. Clicck on the buttton. In the P
Properties window, click
w
k the Apply button and the 3D surfaaces will be generated. 1225 A new button
A
n that ends w
with “labels.surf” will ap
ppear in the right side‐b
bar. Right cliick on
n the button
n and select S
Surface View
w. Remove th
the Ortho Sliice view by cclicking on tthe sm
mall orange box located in the orang
ge Ortho Slicce button on
n the right siide‐bar. 1226 You need to s
Y
save each segmented surrface as sepaarate files. In
n the Surfacee View Properties wiindow, selecct All from th
he Materialss drop down
n menu and tthen click Remove butto
R
on from the Buffer optio
ons. 1227 Choose the fi
C
rst material of interest frrom the Matterials drop down menu
u (in this case: Right Kidney
R
y), and click the Add buttton from thee Buffer opttions. Right‐cclick on the yellow Surfacce View buttton and select Extract Su
urface. 1228 Click the Extr
C
ract Surface button and then click th
he apply buttton in the Ex
xtract Surfacce Properties wiindow to creeate a green button calleed ExtractedS
Surface. Rig
ght‐click this green button and select S
Surface View
w to create a Surface View
w 2 button. T
Turn off thee other Surfacee View by cliicking the orrange squaree inside the y
yellow Surfaace View bu
utton. Now you wil
N
ll only see th
he surface that’s in Surfaace View 2. If you want aa more siimplified mo
odel, click on
n the ExtracttedSurface b
button and th
hen select th
he Simplificaation Editor from th
he propertiees window. IInstead of 188000 faces, y
you can simp
plify to 1000 aces (or less)). Click the S
Simplify Now
w button to see the simp
plified modeel. fa
1229 When you are
W
e done simp
plifying, righ
ht‐click on th
he ExtractedS
Surface buttton and Savee Data As an .S
D
STL file for im
mporting intto other CAD
D software p
packages. 1330 Appendix B: Electromagnetic Simulations Ansys HFSS™ was used in this dissertation for all electromagnetic simulation‐
based solutions. Our lab currently has a Strategic Partnership with Ansys which provides us with top‐notch customer support accessible online via Ansys Customer Portal (https://www1.ansys.com/customer/default.asp). The Customer Portal also gives students access to many tutorials which help form a basic understanding of how to use HFSS to set up and run electromagnetic simulations. To avoid redundancies, this Appendix will not provide a How‐To for setting up electromagnetic simulations, but will demonstrate a few features that are specifically relevant for hyperthermia and radiometry projects. Importing 3D Animal and Patient Models into HFSS After inserting an HFSS Design into your new project, select Modeler>>>Import to import .STL files. It is important to make sure that the dimensions of your model “make sense.” HFSS has Ruler to determine the size of your model. If the Ruler is not visible select View>>>Visibility>>>Ruler. If your model scale is incorrect, then right‐click on the imported material located in the Solids list and select Edit>>>Scale to open the scaling tool. This is also useful if you want to define a scaling factor as an optimization parameter. 131 If your imported material is not a Solid, you may need to fix it in Avizo. Models that are not solids usually contain holes. If your imported material is a Solid but has some Mesh errors, use the tools under Modeler>>>Model Analysis, Model Preparation to “Heal” the mesh errors. If you want HFSS to ignore the Mesh errors because they are not important to your simulation results, select Modeler>>>Validation Settings and change the Entity Check Level from Strict to Warning Only. Defining Dispersive Materials The most important dispersive materials in hyperthermia and radiometry projects are bodily tissue and organs. To get the dispersive properties of body tissues (based on publications by Gabriel), see: http://niremf.ifac.cnr.it/tissprop/htmlclie/htmlclie.htm. To get dispersive properties of other materials (such as lab phantoms), use the dielectric probe to collect the data on the network analyzer. Save dielectric constant and conductivity data as separate files with extension .TAB (ex: tissue_epsr.tab and tissue_sigma.tab). Each file should have a column for frequency and a column for dielectric constant or conductivity. In HFSS, select Project>>>Datasets to import these .TAB files as defined variables (ex: $tissue_epsr, $tissue_sigma). Then right‐click on the material you would like to define and Add a New Material. Name your new Material 132 (ex: Tissue) and under Relative Permittivity define: pwl($tissue_epsr, Freq) and under Bulk Conductivity define: pwl($tissue_sigma, Freq). Using the Fields Calculator to Define Optimization Parameters In the Project Manager list, right‐click on Field Overlays and select the Calculator. Use the calculator buttons to define your equation. Once you have defined the equation, click the Add button in the Library options and name the equation. It will be added to the Named Expressions list. To see a plot with this newly defined parameter, right‐click on Results in the Project Manager list and select Create Fields Report>>>Rectangular Plot. The newly defined parameter will be an option on the Named Expressions list. You can also create new mathematical expressions using the newly defined parameter with other existing parameters by clicking the Output Variables button when creating a plot of the parameter. Exporting Field (SAR) Data In the Project Manager list, right‐click on Field Overlays and select the Calculator. Select the field data you are interested in from the list of Named Expressions and then click Copy to Stack. Then under the Output column, click the Export button. Type in an Output File Name and click the radio button for Calculate Grid Points. Fill in the grid points of interest and spacing to export a matrix with the field data of interest. 133 Appendi
A
x C: Coupled EM-Therm
mal Simulations
s COMS
SOL Multiph
hysics® wass used in thi s dissertatio
on for all elecctromagnetic (E
EM)‐Therma
al simulation
n‐based solu
utions. Many
y COMSOL ttutorials exisst online to h
help fo
orm a basic u
understandiing of how to
o use COMSSOL to set up
p and run EM
M‐Thermal siimulations. T
To avoid red
dundancies, this Append
dix will not provide a H
How‐To for seetting up mu
ultiphysics ssimulations, but will dem
monstrate a few featuress that are sp
pecifically reelevant for h
hyperthermia and radiom
metry projeccts. Sh
haring CAD
D Geometry F
Files betweeen Ansys HF
FSS™ and CO
OMSOL Mu
ultiphysics®
1. Exporrt the geomeetry from HF
FSS™ a. Use step eextension 
OMSOL, turn
n on “Selection List” 2. In CO
1334 a. View  “S
Selection Lisst” 3. Imporrt the geomeetry to COM
MSOL a. Right click
k “Geometry
y”Import i. Bro
owseSelecct fileImpo
ort geometry 4. Edit g
For geometriees created in
n HFSS, it is possible thaat no furtherr corrections need to be ad
dded. Howeever, it is com
mmon to hav
ve problemss at the boun
ndaries of diifferent bodiies. a. Check if bodies interseect. 1335 YES, perform
m a Boolean addition bettween the in
ntersection aand i. If Y
onee of the bodiies, and then
n perform th
he Boolean subtraction from thee other body
y. Iff the geomettries were geenerated in A
AVIZO the rresult will bee a surface raather than a volume. Wheen importing
g surfaces into COMSOL
L, the softwaare will breaak one surfacce up in
nto several. IIt will be neccessary to un
nite all of thee surfaces. Example 1: The “Selectio
T
n List” featu
ure is very useful becausse it is possib
ble to see alll of the impo
orted su
urfaces and objects (and
d their namess). In this casse we imporrted the bod
dy of a mouse an
nd it was su
upposed to h
have only on
ne object but we have sev
veral. To solve this prob
blem, we select all o
w
objects and u
unite them. Example 2: 1336 H
Hole 1 Hole 2
This is a mou
T
use body witth holes wheere the bladd
der should b
be. One way to remove th
he holes is to creeate a spheree that encom
mpasses them
m and then u
unite the sph
here with thee body. Note th
hat the spherre should no
ot be so largee that it passses through the outer su
urface of the mouse!! Using SAR fr
U
rom HFSS ass a Heat Source in COMSSOL 1. Open the Fields calculator in HFSS gn  Field O
Overlays  Calculator HFSS Desig
1337 2. Exporrt SAR data a. In “Named
d expression
ns” select “V
Volume_Losss_Density” ((has units W
Wm‐3). b. Click on “Copy to Stacck” c. Click on “Export…” 3. Exporrt data a. Click on th
he yellow folder to add a name, ex: sarXYZanteenna. b. Add the reegion of inteerest under ““Calculate g
grid points” i. Use unit “meteers” (or the u
unit used in COMSOL) 1338 ii. Tip
p: Create a box in HFSSS and fit iit to the geeometry. Use the coo
ordinates of the box as m
minimum an
nd maximum
m of the grid
d. But ma
ake sure thatt the box is l ocated on th
he Global coordinate sysstem. iii. Ch
hoose spacin
ng accordin
ng to the g
geometry. T
The data wiill be intterpolated in
n COMSOL.. If the maxximum in CO
OMSOL is llower tha
an the one in
n HFSS a fineer grid spaciing is requirred. iv. Cliick “Ok” to eexport. 4. Prepa
are the file fo
or import intto COMSOL
L a. The exported file has a
a FLD exten
nsion. Use Exxcel to open it. b. Delete the first two rows. c. Select the first row ataText to Columns i. Da
1339 ii. Sellect “Delimitted” and clicck NEXT iii. Sellect “Space” and unselecct “Tab” and
d click NEXT
T iv. Cliick finish d. Delete all N
Nan (not a n
number) i. If tthe geometry
y is not regu
ular, HFSS w
will try to dettermine SAR
R outside the mo
odel and thaat will result in a Nan. ii. Fin
nd and Repla
ace all Nan w
with 0 iii. Ch
heck if it mak
kes sense forr each Nan to
o be zero. If the zero is in bettween two h
high values ((>1), perhapss it’s reasonaable to use th
he aveerage of thosse values insstead. e. Save the fiile as: “Text (tab delimitted)”. 5. Createe an interpollation functiion: a. Model  D
Definitions 
Functionss  Interpolaation 1440 6. Interp
polation settiings a. Change “D
Data Source”
” to File b. Change “N
Number of a
arguments” tto 3 (x,y,z)
c. Change fu
unction namee from “int11” to sar (sug
ggestion). a heat sourcee 7. Add a
a. Model  R
Right click o
on Heat tran
nsfer Heat Source 1441 ng heated b. Select the geometries tthat are bein
x,y,z) under “User defin
ned” c. Type sar(x
all necessary
y import/exp
port function
ns have been
n made from
m HFSS into 8. Now a
COMS
SOL for EM‐Thermal sim
mulations. 1442 Appendix D: Phantom Formulations Solid (Gel) Phantoms Materials: 1. Distilled water 2. Canola Oil 3. Gelatin 4. Soap 5. Vacuum and vacuum beaker with vacuum grease 6. Thermocouple 7. Water bath at 70‐90°C 8. Aluminum foil 9. Formaldehyde (36.5%) 10. Protective wear (i.e., gloves) 11. Stirring rods 12. Beakers 13. Container for final phantom Procedure: 1. Wear protective‐wear throughout procedure. 2. Preheat water bath to at least 70°C (preferably 80 or 90°C) 3. Calculate the water:fat volume ratio that is desired and the total volume of solution needed. Oil‐in‐gelatin dispersions Saline concentration Volume % Volume% Material type g / 1000 ml DI water aqueous gelatin oil Tumor 95 5 7.5 Muscle 90 10 6.0 Fat (bone marrow) 15 85 0.24 4. Using those values, calculate the amount of water and oil needed for the solution. 5. Measure and pour the amount of each liquid in two separate beakers, one for water and one for oil. 6. Calculate the amount of gelatin needed by multiplying the amount of water by 0.1596g/mL. 143 7. Add the gelatin to the water in small increments (<50g), continually stirring as gelatin is added. As you approach the end of the gelatin, it may be hard to stir/mix. This is normal. 8. Once the gelatin is thoroughly mixed, place the beaker into the water bath with the top covered by aluminum foil. Let the gelatin mixture heat, periodically stirring the solution to break up clumps. Once the gelatin has reached 70˚C or higher, it is ready (the solution should become semi‐transparent). 9. Take the gelatin solution out of the water bath and set aside to cool. 10. Place the beaker of oil into the water bath. Heat the oil to approximately 50˚C as you let the gelatin solution cool to approximately 50˚C. Once both solutions are near 50˚C, take the oil beaker out of the water bath. If making a phantom with ≤50% fat: 11. Pour the oil into the water beaker. Now add approximately 1mL of soap for every 100mL of the total oil that will be in the phantom. Once the soap is added, stir vigorously until the solution is completely mixed and is now a creamy white color. Add additional soap if needed to improve homogeneity. If making a phantom with >50% fat: 11. Measure a volume of oil that matches the total volume of the water and pour it into the water beaker, making a 50:50 solution. Now add approximately 1mL of soap for every 100mL of the total oil that will be in the phantom. Once the soap is added, stir vigorously until the solution is completely mixed and is now a creamy white color. 12. With the solution thoroughly mixed, begin to add the rest of the oil in 50‐100mL increments. After adding oil, vigorously stir the solution until homogenous again. Continue adding oil until there is no more remaining. Do not add additional soap after the creation of the 50:50 solution. 13. Once the solution is fully mixed, attach the vacuum to the top of the beaker, using vacuum grease to secure the vacuum. Turn the vacuum up slowly to prevent the solution from being sucked into the vacuum. Once a stable vacuum is established, vigorously stir the solution by rotating the beaker. Repeat this several times until air bubbles have been removed. Between applications of vacuum return the beaker to the water bath to keep the solution around 70˚C. Do not let the beaker become too hot however, since it will weaken the glass, potentially breaking the beaker when it is placed under vacuum. For all solutions: 144 14. Take the beaker to the fume hood. For every 100mL of water in the solution, pour 1mL of the Formaldehyde solution into the beaker. Once it has been poured, quickly stir the solution with a stirring rod and pour your phantom into your container of choice. 15. While wearing your protective wear, clean all beakers with soap and water. Liquid Phantoms The recipe for liquid phantoms is less precise. Basic materials/procedure: 1. To decrease the dielectric constant: add ethylene or propylene glycol, or Tween (for high frequency phantoms) 2. To increase the dielectric constant: add distilled water 3. To increase conductivity: add salt 4. Mix thoroughly 145 Works Cited
1. Bladder Cancer2011 October 10, 2011:[50 p.]. Available from: http://www.cancer.org/acs/groups/cid/documents/webcontent/003085‐pdf.pdf. 2. Wallner KE, Banda M, Li GC. Hyperthermic enhancement of cell killing by mitomycin C in mitomycin C‐resistant Chinese hamster ovary cells. Cancer Res. 1987;47(5):1308‐12. Epub 1987/03/01. 3. Breyer BN, Whitson JM, Carroll PR, Konety BR. Sequential intravesical gemcitabine and mitomycin C chemotherapy regimen in patients with non‐muscle invasive bladder cancer. Urol Oncol. 2010;28(5):510‐4. Epub 2009/01/28. 4. Colombo R, Da Pozzo LF, Lev A, Freschi M, Gallus G, Rigatti P. Neoadjuvant combined microwave induced local hyperthermia and topical chemotherapy versus chemotherapy alone for superficial bladder cancer. J Urol. 1996;155(4):1227‐32. Epub 1996/04/01. 5. Gofrit ON, Shapiro A, Pode D, Sidi A, Nativ O, Leib Z, et al. Combined local bladder hyperthermia and intravesical chemotherapy for the treatment of high‐grade superficial bladder cancer. Urology. 2004;63(3):466‐71. Epub 2004/03/19. 6. Ott OJ, Rodel C, Weiss C, Wittlinger M, St Krause F, Dunst J, et al. Radiochemotherapy for bladder cancer. Clin Oncol (R Coll Radiol). 2009;21(7):557‐65. Epub 2009/07/01. 7. Colombo R, Salonia A, Da Pozzo LF, Naspro R, Freschi M, Paroni R, et al. Combination of intravesical chemotherapy and hyperthermia for the treatment of superficial bladder cancer: preliminary clinical experience. Crit Rev Oncol Hemat. 2003;47(2):127‐39. 8. Vujaskovic Z, Craciunescu O, Stauffer P, Carroll M, Lan L, Dewhirst M, et al., editors. Pilot study of external hyperthermia and intravesical mitomycin‐C to treat recurrent bladder cancer after failed standard therapy. ASCO Annual Meeting; 2010: Journal of Clinical Oncology. 146 9. Grbic A, Eleftheriades GV. Overcoming the diffraction limit with a planar left‐
handed transmission‐line lens. Phys Rev Lett. 2004;92(11). 10. Pendry JB. Negative refraction makes a perfect lens. Phys Rev Lett. 2000;85(18):3966‐9. 11. Fang N, Lee H, Sun C, Zhang X. Sub‐diffraction‐limited optical imaging with a silver superlens. Science. 2005;308(5721):534‐7. 12. Melville DOS, Blaikie RJ. Super‐resolution imaging through a planar silver layer. Opt Express. 2005;13(6):2127‐34. 13. Jeppesen C, Nielsen RB, Boltasseva A, Xiao S, Mortensen NA, Kristensen A. Thin film Ag superlens towards lab‐on‐a‐chip integration. Opt Express. 2009;17(25):22543‐52. 14. Papahadjopoulos D, Gabizon A. Liposomes designed to avoid the reticuloendothelial system. Prog Clin Biol Res. 1990;343:85‐93. Epub 1990/01/01. 15. Sharpe M, Easthope SE, Keating GM, Lamb HM. Polyethylene glycol‐liposomal doxorubicin: a review of its use in the management of solid and haematological malignancies and AIDS‐related Kaposiʹs sarcoma. Drugs. 2002;62(14):2089‐126. Epub 2002/09/25. 16. Allen TM, Hansen CB, de Menezes DEL. Pharmocokinetics of long‐circulating liposomes. Advanced Drug Delivery Reviews. 1995;16(2‐3):267‐84. 17. Hosokawa T, Sami M, Kato Y, Hayakawa E. Alteration in the temperature‐
dependent content release property of thermosensitive liposomes in plasma. Chem Pharm Bull (Tokyo). 2003;51(11):1227‐32. Epub 2003/11/06. 18. Amano T, Kumini K, Nakashima K, Uchibayashi T, Hisazumi H. A combined therapy of hyperthermia and tumor necrosis factor for nude mice bearing KK‐47 bladder cancer. J Urol. 1990;144(2 Pt 1):370‐4. Epub 1990/08/01. 147 19. Tacker JR, Anderson RU. Delivery of antitumor drug to bladder cancer by use of phase transition liposomes and hyperthermia. J Urol. 1982;127(6):1211‐4. Epub 1982/06/01. 20. Marmor JB, Hilerio FJ, Hahn GM. Tumor eradication and cell survival after localized hyperthermia induced by ultrasound. Cancer Res. 1979;39(6 Pt 1):2166‐71. Epub 1979/06/01. 21. Marmor JB, Hahn N, Hahn GM. Tumor cure and cell survival after localized radiofrequency heating. Cancer Res. 1977;37(3):879‐83. Epub 1977/03/01. 22. Maehara T, Konishi K, Kamimori T, Aono H, Naohara T, Kikkawa H, et al. Heating of ferrite powder by an AC magnetic field for local hyperthermia. Jpn J Appl Phys 1. 2002;41(3A):1620‐1. 23. Nomura S, Mukasa S, Yamasaki H, Maehara T, Aono H, Kikkawa H, et al. Inductive heating of mg ferrite powder in high‐water content phantoms using AC magnetic field for local hyperthermia. Heat Transfer Eng. 2007;28(12):1017‐22. 24. Sonvico F, Mornet S, Vasseur S, Dubernet C, Jaillard D, Degrouard J, et al. Folate‐
conjugated iron oxide nanoparticles for solid tumor targeting as potential specific magnetic hyperthermia mediators: Synthesis, physicochemical characterization, and in vitro experiments. Bioconjugate Chem. 2005;16(5):1181‐8. 25. Dewhirst MW, Jones EL, Samulski TV, al. e. Hyperthermia. In: D.W. Kufe ea, editor. Cancer Medicine. Hamilton: BC Decker, Inc; 2003. p. 623‐36. 26. Overgaard J, Gonzalez DG, Hulshof MCCH, Arcangeli G, Dahl O, Mella O, et al. Hyperthermia as an adjuvant to radiation therapy of recurrent or metastatic malignant melanoma. A multicentre randomized trial by the European Society for Hyperthermic Oncology. Int J Hyperther. 1996;12(1):3‐20. 27. Valdagni R, Amichetti M. Report of Long‐Term Follow‐up in a Randomized Trial Comparing Radiation‐Therapy and Radiation‐Therapy Plus Hyperthermia to Metastatic Lymph‐Nodes in Stage‐Iv Head and Neck Patients. Int J Radiat Oncol. 1994;28(1):163‐9. 148 28. Vernon CC, Hand JW, Field SB, Machin D, Whaley JB, vanderZee J, et al. Radiotherapy with or without hyperthermia in the treatment of superficial localized breast cancer: Results from five randomized controlled trials. Int J Radiat Oncol. 1996;35(4):731‐44. 29. Wust P, Hildebrandt B, Sreenivasa G, Rau B, Gellermann J, Riess H, et al. Hyperthermia in combined treatment of cancer. Lancet Oncol. 2002;3(8):487‐97. 30. Israel L, Chahinian P, Depierre A. Response of 65 measurable epidermoid bronchogenic tumors of known spontaneous doubling time to four different chemotherapeutic regimens‐‐strategic deductions. Med Pediatr Oncol. 1975;1(2):83‐93. Epub 1975/01/01. 31. Gong Y, Wang G. Superficial Tumor Hyperthermia with Flat Left‐Handed Metamaterial Lens. Prog Electromagn Res. 2009;98:389‐405. 32. Karathanasis KT, Karanasiou IS, Uzunoglu NK. Enhancing the focusing properties of a prototype non‐invasive brain hyperthermia system: a simulation study. Conf Proc IEEE Eng Med Biol Soc. 2007;2007:218‐21. Epub 2007/11/16. 33. Wang G, Gong Y. Metamaterial lens applicator for microwave hyperthermia of breast cancer. Int J Hyperther. 2009;25(6):434‐45. 34. Snyder WS, Fisher HL, Ford MR, Warner GG. Estimates of Absorbed Fractions for Monoenergetic Photon Sources Uniformly Distributed in Various Organs of a Heterogeneous Phantom. J Nucl Med. 1969;10(3S):7‐&. 35. Kramer R, Zankl M, Williams G, Drexler G. The calculation of dose from external photon exposures using reference human phantoms and Monte Carlo methods. Munich: Gesellschaft f¨ur Strahlenund Umweltforschung; 1982. 36. Cristy M, Eckerman KF. Specific Absorbed Fractions of Energy at Various Ages from Internal Photon Sources. Laboratory ORN, editor. Oak Ridge, TN: ORNL; 1987. 149 37. Han EY, Bolch WE, Eckerman KF. Revisions to the ORNL series of adult and pediatric computational phantoms for use with the mird schema. Health Phys. 2006;90(4):337‐56. 38. Stabin M, Watson E, Cristy M, Ryman J, Eckerman K, Davis J, et al. Mathematical models and specific absorbed fractions of photon energy in the nonpregnant adult female and at the end of each trimester of pregnancy. Laboratory ORN, editor. Oak Ridge, TN: ORNL; 1995. 39. Caon M. Voxel‐based computational models of real human anatomy: a review. Radiat Environ Bioph. 2004;42(4):229‐35. 40. Dimbylow P. Development of the female voxel phantom, NAOMI, and its application to calculations of induced current densities and electric fields from applied low frequency magnetic and electric fields. Phys Med Biol. 2005;50(6):1047‐70. 41. Kramer R, Vieira JW, Khoury HJ, Lima FRA, Fuelle D. All about MAX: a male adult voxel phantom for Monte Carlo calculations in radiation protection dosimetry. Phys Med Biol. 2003;48(10):1239‐62. 42. Kramer R, Khoury HJ, Vieira JW, Loureiro ECM, Lima VJM, Lima FRA, et al. All about FAX: a female adult voxel phantom for Monte Carlo calculation in radiation protection dosimetry. Phys Med Biol. 2004;49(23):5203‐16. 43. Lee C, Lee C, Park SH, Lee JK. Development of the two Korean adult tomographic computational phantoms for organ dosimetry. Med Phys. 2006;33(2):380‐
90. 44. Lee C, Lee C, Williams JL, Bolch WE. Whole‐body voxel phantoms of paediatric patients ‐ UF Series B. Phys Med Biol. 2006;51(18):4649‐61. 45. Nagaoka T, Watanabe S, Sakurai K, Kunieda E, Watanabe S, Taki M, et al. Development of realistic high‐resolution whole‐body voxel models of Japanese adult males and females of average height and weight, and application of models to radio‐
frequency electromagnetic‐field dosimetry. Phys Med Biol. 2004;49(1):1‐+. 150 46. Lee C, Lodwick D, Hasenauer D, Williams JL, Lee C, Bolch WE. Hybrid computational phantoms of the male and female newborn patient: NURBS‐based whole‐
body models. Phys Med Biol. 2007;52(12):3309‐33. 47. Jones DG. A realistic anthropomorphic phantom for calculating organ doses arising from external photon irradiation. Radiat Prot Dosim. 1997;72(1):21‐9. 48. Kramer R, Khoury HJ, Vieira JW. Comparison between effective doses for voxel‐
based and stylized exposure models from photon and electron irradiation. Phys Med Biol. 2005;50(21):5105‐26. 49. Lee C, Lee C, Bolch WE. Age‐dependent organ and effective dose coefficients for external photons: a comparison of stylized and voxel‐based paediatric phantoms. Phys Med Biol. 2006;51(18):4663‐88. 50. Staton RJ, Pazik FD, Nipper JC, Williams JL, Bolch WE. A comparison of newborn stylized and tomographic models for dose assessment in pediatric radiology. Health Phys. 2003;84(6):S258‐S9. 51. Zankl M, Fill U, Petoussi‐Henss N, Regulla D. Organ dose conversion coefficients for external photon irradiation of male and female voxel models. Phys Med Biol. 2002;47(14):2367‐85. 52. Chao TC, Xu XG. Specific absorbed fractions from the image‐based VIP‐Man body model and EGS4‐VLSI Monte Carlo code: internal electron emitters. Phys Med Biol. 2001;46(4):901‐27. 53. Jones DG. A realistic anthropomorphic phantom for calculating specific absorbed fractions of energy deposited from internal gamma emitters. Radiat Prot Dosim. 1998;79(1‐4):411‐4. 54. Petoussi‐Henss N, Zankl M. Voxel anthropomorphic models as a tool for internal dosimetry. Radiat Prot Dosim. 1998;79(1‐4):415‐8. 55. Stabin MG, Yoriyaz H. Photon specific absorbed fractions calculated in the trunk of an adult male voxel‐based phantom. Health Phys. 2002;82(1):21‐44. 151 56. Zankl M, Petoussi‐Henss N, Fill U, Regulla D. The application of voxel phantoms to the internal dosimetry of radionuclides. Radiat Prot Dosim. 2003;105(1‐4):539‐48. 57. Gabriel C. Compilation of the dielectric properties of body tissues at RF and microwave frequencies. In: Occupational and environmental health directorate RRD, editor. Brooks Air Force Base, Texas (USA)1996. 58. Gabriel C, Gabriel S, Corthout E. The dielectric properties of biological tissues .1. Literature survey. Phys Med Biol. 1996;41(11):2231‐49. 59. Gabriel S, Lau RW, Gabriel C. The dielectric properties of biological tissues .2. Measurements in the frequency range 10 Hz to 20 GHz. Phys Med Biol. 1996;41(11):2251‐
69. 60. Pennes HH. Analysis of Tissue and Arterial Blood Temperatures in the Resting Human Forearm. J Appl Physiol. 1948;1(2):93‐122. 61. Diller K, Valvano J, Pearce J, editors. Bioheat Transfer. Second Edition ed. Boca Raton: CRC press; 2005. 62. Song Cea, editor. Microvasculature and perfusion in normal tissues and tumors. Berlin, New York: Springer‐Verlag; 1995. 63. Gillooly JF, Brown JH, West GB, Savage VM, Charnov EL. Effects of size and temperature on metabolic rate. Science. 2001;293(5538):2248‐51. 64. Barclay C, Woledge R, Curtin N. Effects of UCP3 genotype, temperature and muscle type on energy turnover of resting mouse skeletal muscle. Pflug Arch Eur J Phy. 2009;457(4):857‐64. 65. Sessler DI. Temperature monitoring and perioperative thermoregulation. Anesthesiology. 2008;109(2):318‐38. 66. Joiner MC, Steel GG, Stephens TC. Response of two mouse tumours to hyperthermia with CCNU or melphalan. Br J Cancer. 1982;45(1):17‐26. Epub 1982/01/01. 152 67. Peller M, Schwerdt A, Hossann M, Reinl HM, Wang T, Sourbron S, et al. MR characterization of mild hyperthermia‐induced gadodiamide release from thermosensitive liposomes in solid tumors. Invest Radiol. 2008;43(12):877‐92. Epub 2008/11/13. 68. Hill SA, Denekamp J. The effect of vascular occlusion on the thermal sensitization of a mouse tumour. Br J Radiol. 1978;51(612):997‐1002. Epub 1978/12/01. 69. Meyer DE, Shin BC, Kong GA, Dewhirst MW, Chilkoti A. Drug targeting using thermally responsive polymers and local hyperthermia. J Control Release. 2001;74(1‐
3):213‐24. 70. Mukhopadhaya A, Mendecki J, Dong X, Liu L, Kalnicki S, Garg M, et al. Localized hyperthermia combined with intratumoral dendritic cells induces systemic antitumor immunity. Cancer Res. 2007;67(16):7798‐806. Epub 2007/08/19. 71. Trefna HD, Vrba J, Persson M. Evaluation of a patch antenna applicator for time reversal hyperthemia. Int J Hyperther. 2010;26(2):185‐97. 72. Song C, Choi I, Nah B, Sahu S, Osborn J. Microvasculature and perfusion in normal tissues and tumors. In: Seegenschmiedt MH FP, Vernon CC, editors, editor. Thermoradiotherapy and thermochemotherapy. Berlin, New York: Springer‐Verlag; 1995. p. 139‐59. 73. Diller K, Valvano J, Pearce J. The CRC Handbook of Mechanical Engineering. Second Edition ed. DYG FK, editor: CRC Press; 2005 2005. 74. Gordon C. Quantifying the instability of core temperature in rodents. Journal of Thermal Biology. 2009;34(5):213‐9. 75. Neuman DG, Stauffer PR, Jacobsen S, Rossetto F. SAR pattern perturbations from resonance effects in water bolus layers used with superficial microwave hyperthermia applicators. Int J Hyperther. 2002;18(3):180‐93. 153 76. Haveman J, Smals OAG, Rodermond HM. Effects of hyperthermia on the rat bladder: a pre‐clinical study on thermometry and functional damage after treatment. Int J Hyperther. 2003;19(1):45‐57. 77. de Bruijne M, Samaras T, Chavannes N, van Rhoon GC. Quantitative validation of the 3D SAR profile of hyperthermia applicators using the gamma method. Phys Med Biol. 2007;52(11):3075‐88. 78. Gunther JH, Jurczok A, Wulf T, Brandau S, Deinert I, Jocham D, et al. Optimizing syngeneic orthotopic murine bladder cancer (MB49). Cancer Res. 1999;59(12):2834‐7. 79. Yamada H, Matsumoto S, Matsumoto T, Yamada T, Yamashita U. Murine IL‐2 secreting recombinant Bacillus Calmette‐Guerin augments macrophage‐mediated cytotoxicity against murine bladder cancer MBT‐2. J Urol. 2000;164(2):526‐31. 80. Chen Q, Krol A, Wright A, Needham D, Dewhirst MW, Yuan F. Tumor microvascular permeability is a key determinant for antivascular effects of doxorubicin encapsulated in a temperature sensitive liposome. Int J Hyperthermia. 2008;24(6):475‐82. Epub 2008/07/09. 81. Lindner LH, Eichhorn ME, Eibl H, Teichert N, Schmitt‐Sody M, Issels RD, et al. Novel temperature‐sensitive liposomes with prolonged circulation time. Clin Cancer Res. 2004;10(6):2168‐78. Epub 2004/03/26. 82. Tacker JR, Anderson RU. Delivery of antitumor drug to bladder cancer by use of phase transition liposomes and hyperthermia. J Urol. 1982;127(6):1211‐4. Epub 1982/06/01. 83. Jain RK. Barriers to Drug‐Delivery in Solid Tumors. Sci Am. 1994;271(1):58‐65. 84. Moriyama E, Salcman M, Broadwell RD. Blood‐Brain‐Barrier Alteration after Microwave‐Induced Hyperthermia Is Purely a Thermal Effect .1. Temperature and Power Measurements. Surg Neurol. 1991;35(3):177‐82. 154 85. Kong G, Anyarambhatla G, Petros WP, Braun RD, Colvin OM, Needham D, et al. Efficacy of liposomes and hyperthermia in a human tumor xenograft model: Importance of triggered drug release. Cancer Res. 2000;60(24):6950‐7. 86. Recommended tissue parameters for the Virtual Family [database on the Internet]. Stanford University. [cited January 11, 2012]. Available from: http://stanford.edu/~sujason/MR/TissueParameters_VirtualFamily.pdf. 87. DʹAmbrosio V, Dughiero F, Giri M, Maluta S. Numerical Model for RF Capacitive Hyperthermia for Brain Tumors2006 January 11, 2012. Available from: http://www.esho.info/esho2006/pdf/p047_ambrosio.pdf. 88. Yoo DS. The dielectric properties of cancerous tissues in a nude mouse xenograft model. Bioelectromagnetics. 2004;25(7):492‐7. 89. Jain MK, Wolf PD. Finite element analysis predicts dose‐response relationship for constant power and temperature controlled radiofrequency ablation. P Ann Int Ieee Embs. 1997;19:165‐8. 90. Yuan Y, Wyatt C, Maccarini P, Stauffer P, Craciunescu O, Macfall J, et al. A heterogeneous human tissue mimicking phantom for RF heating and MRI thermal monitoring verification. Phys Med Biol. 2012;57(7):2021‐37. Epub 2012/03/21. 91. Sargent MA. Opinion ‐ What is the normal prevalence of vesicoureteral reflux? Pediatr Radiol. 2000;30(9):587‐93. 92. DʹIsidoro F. Signature techniques for diagnosis of vesicoureteral reflux with microwave radiometry. Rome: Universita degli Studi di Roma “Tor Vergata”; 2011. 93. Alexopoulos NG, Jackson DR. Fundamental Superstrate (Cover) Effects on Printed‐Circuit Antennas. Ieee T Antenn Propag. 1984;32(8):807‐16. 94. Eleftheriades GV, Balmain KG. Negative‐refraction metamaterials : fundamental properties and applications. Hoboken, NJ: J. Wiley; 2005. xv, 418 p. p. 95. Rigotti E, Fontana D, Negri GL, Palestro G, Randone DF, Borgno M. Results of Hyperthermia on Bladder Carcinomas of Dog. J Urol Nephrol. 1977;83(3):175‐84. 155 96. Viglianti BL, Lora‐Michiels M, Poulson JM, Lan L, Yu DH, Sanders L, et al. Dynamic Contrast‐enhanced Magnetic Resonance Imaging as a Predictor of Clinical Outcome in Canine Spontaneous Soft Tissue Sarcomas Treated with Thermoradiotherapy. Clinical Cancer Research. 2009;15(15):4993‐5001. 97. Lora‐Michiels M, Yu D, Sanders L, Poulson JM, Azuma C, Case B, et al. Extracellular pH and P‐31 magnetic resonance spectroscopic variables are related to outcome in canine soft tissue sarcomas treated with thermoradiotherapy. Clinical Cancer Research. 2006;12(19):5733‐40. 98. Hauck ML, Larue SM, Petros WP, Poulson JM, Yu DH, Spasojevic I, et al. Phase I trial of doxorubicin‐containing low temperature sensitive liposomes in spontaneous canine tumors. Clinical Cancer Research. 2006;12(13):4004‐10. 99. Thrall DE, Larue SM, Pruitt AF, Case B, DeWhirst MW. Changes in tumour oxygenation during fractionated hyperthermia and radiation therapy in spontaneous canine sarcomas. Int J Hyperther. 2006;22(5):365‐73. 100. Thrall DE, LaRue SM, Yu DH, Samulski T, Sanders L, Case B, et al. Thermal dose is related to duration of local control in canine sarcomas treated with thermoradiotherapy. Clinical Cancer Research. 2005;11(14):5206‐14. 101. Hasgall P, Neufeld E, Gosselin M, Klingenböck A, Kuster N. ITʹIS Database for thermal and electromagnetic parameters of biological tissues. www.itis.ethz.ch/databaseSeptember 26th, 2011. 102. Drizdal T, Togni P, Visek L, Vrba J. Comparison of Constant and Temperature Dependent Blood Perfusion in Temperature Prediction for Superficial Hyperthermia. Radioengineering. 2010;19(2):281‐9. 103. Tompkins DT, Vanderby R, Klein SA, Beckman WA, Steeves RA, Frye DM, et al. Temperature‐Dependent Versus Constant‐Rate Blood Perfusion Modeling in Ferromagnetic Thermoseed Hyperthermia ‐ Results with a Model of the Human Prostate. Int J Hyperther. 1994;10(4):517‐36. 156 104. Song CW. Effect of Local Hyperthermia on Blood‐Flow and Microenvironment ‐ a Review. Cancer Res. 1984;44(10):4721‐30. 157 Biography
Sara Salahi Born February 10, 1985 Chicago, Illinois USA Education 

Duke University, Durham, NC PhD Biomedical Engineering, May 2012 MS Electrical and Computer Engineering, May 2010 Northwestern University, Evanston, IL BS Biomedical Engineering, June 2007 BS Computer Engineering, June 2007 Honors and Affiliations 





National Academies Award, USNC‐URSI National Radio Sciences Meeting (2012) New Investigator Award, Society of Thermal Medicine (2010) Graduate Research Fellowship, National Science Foundation (2007‐2010) Honorable Mention Outstanding Woman Leader Award, Duke WiSE (2008) IEEE Member Society of Thermal Medicine Member Publications S Salahi, PF Maccarini, DB Rodrigues, K Arunachalam, W Etienne, C Landon, BA Inman, MW Dewhirst, and PR Stauffer. ʺMiniature Microwave Applicator for Murine Bladder Hyperthermia Studies.ʺ International Journal of Hyperthermia; In Press. 158 Abstracts and Conference Proceedings S Salahi, PF Maccarini, K Arunachalam, W Etienne, C Landon, BA Inman, MW Dewhirst and PR Stauffer. ʺMicrowave applicator for localized heating in small animals.ʺ Proceedings of the Society of Thermal Medicine; 2009 April 23‐26; Clearwater FL Abstract No. 0067. S Salahi, PF Maccarini, MW Dewhirst and PR Stauffer. ʺFamily of miniature microwave applicators for focused heating in small animals.ʺ Proceedings of the Society of Thermal Medicine; 2010 April 29‐May 2; New Orleans LA Abstract No. 0018. S Salahi, PF Maccarini, A Boico, PR Stauffer. “Optimization Approach for Microstrip Spiral Antennas Used in Deep Tissue Radiometry.” Proceedings of the USNC‐URSI National Radio Sciences Meeting; 2012 January 4‐7; Boulder, CO Invited Abstract No. 1155. S Salahi, PF Maccarini, PR Stauffer. “Wideband Conformal Metamaterial Antenna for Pediatric Bladder Urine Hyperthermia.” Proceedings of the Society of Thermal Medicine; 2012 April 13‐16; Portland OR Abstract No. 0037. PR Stauffer, PF Maccarini, K Arunachalam, V De Luca, S Salahi, A Boico, Ø Klemetsen, Y Birkelund, SK Jacobsen, F Bardati, P Tognolatti, and B Snow. ʺMicrowave Radiometry for Non‐Invasive Detection of Vesicoureteral Reflux (VUR) Following Bladder Warming.ʺ Proceedings of SPIE, Volume 7901; 2011 January 23; San Francisco CA Invited Paper. C Wilson, S Salahi, PF Maccarini, S Li, MW Dewhirst, and G Grant. ʺIn vivo feasibility study of mild hyperthermia in brain tumors to enhance drug delivery.ʺ Proceedings of the Society of Thermal Medicine; 2010 April 29‐May 2; New Orleans LA Abstract No. 0063. PR Stauffer, PF Maccarini, K Arunachalam, V De Luca, S Salahi, A Boico, Ø Klemetsen, Y Birkelund, SK Jacobsen, F Bardati, P Tognolatti, and B Snow. ʺRadiometric Monitoring of Kidneys During Bladder Warming for Non‐Invasive Detection of Vesicoureteral Reflux (VUR).ʺ Proceedings of the Society of Thermal Medicine; 2010 April 29‐May 2; New Orleans LA Abstract No. 0045. 159 
Документ
Категория
Без категории
Просмотров
1
Размер файла
6 631 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа