вход по аккаунту



код для вставкиСкачать
A targeted drug delivery platform for assisting retinal surgeons for
treating Age-related Macular Degeneration (AMD).
M. A. Nasseri1 , M. Maier1 and C. P. Lohmann1
Abstract— In this paper we present our latest robotic setup,
which has been modified for sub-retinal interventions. The setup
consists of: 1) sub-retinal micro cannula with automatic pump;
2) Micromanipulator; 3) patient fixation mechanism and 4)
clinically compatible workstation. The primary objective of this
work is to allow ophthalmologists to improve administration
of substances such as drugs, stem-cells and gene cargos to
their desired targets in the sub-retinal microstructures. Such a
delivery method will enable effective treatment of Age-related
Macular Degeneration (AMD). AMD is the leading cause of
blindness in developed countries and as yet there is no efficient
treatment. To validate the precision of the system a successful
targeted delivery scenario with the proposed setup and using
an intra-operative OCT integrated microscope in a clinical
environment is presented in this paper.
Age-related Macular Degeneration (AMD) is the leading
cause of blindness in developed countries, particularly in
people older than 60 years [1], [2]. AMD causes damage
to the outer retina, choriocapillaris, retinal pigment epithelium (RPE), and Bruchs membrane (BM). The disease is
characterized by structural changes within BM which then
leads to cellular changes in the RPE including loss of RPE
cells and the eventual development of advanced forms of the
disorder. In many cases AMD progresses and lead to severe
vision loss (see Fig. 1). The projected number of people with
AMD in 2020 is 196 million and is projected to increase to
288 million in 2040 according to [3], due to demographic
change and an aging population. Despite the growing impact
and the importance of AMD that its consequences is growing
significantly, there is still no effective treatment. To date no
known therapeutic interventions exist for the most common
form (dry form AMD) even to slow the progression of the
macular detachment.
For the wet form drugs are available to slow the progression of the disease, however, to date there is no way to stop
1 M. Ali Nasseri, M. Maier and C. P. Lohmann are with the
ophthalmology department of klinikum rechts der Isar, Technische Universität München. ali.nasseri, mathias.maier,
chris.lohmann at
Fig. 1.
The indication of Aged-related Macular Degeneration (AMD).
978-1-5090-2809-2/17/$31.00 ©2017 IEEE
Fig. 2. Current injection method versus the targeted based injection method
for AMD treatment.
or reverse the disease effects, neither for dry nor wet AMD.
The conventional drug delivery approach involves repeated
injections into the anterior vitreous segment and the retina
absorbs the drug through diffusion (depicted in Fig. 2 - left).
However, the results, if any, are highly dissatisfactory and
unpredictable. Moreover, these multiple iterative injections
for each patient produce significant side effects. Recently, in
the area of regenerative medicine, clinical applications using
stem cells and gene therapies to regenerate RPE cells hold
the potential to be a powerful tool in the treatment of a wide
variety of diseases, in particular, disorders of the eye. American clinical trial database ( reveals a
number of stem cell-based therapies and gene therapies for
the treatment of AMD that have progressed to human clinical
trial [4], [5], [6]. Currently, the major technical challenge
in stem cell-based and gene therapy is safe delivery of
effective doses of substances to the target area. To perform
the subretinal injection, following conventional pars plana
vitrectomy, the surgeon penetrates a micro-cannula in the
retina and detaches the photoreceptor cell layer from the
RPE forming subretinal ’blebs’. Controlled volumes should
be injected and therapeutic effect is bound to areas proximal
to point of injection. Moreover, the subretinal surgery raises
a significant safety issue, as the microstructural anatomy
of the retina in AMD patients is fragile and the surgery
can induce mechanical damage, reactive gliosis, and loss
of function. Furthermore, route of administration should be
as less invasive as possible while studies show that the
transplanted stem cells tend to migrate only in the injured
retinal tissue, not in the normal retina [7].
To increase the efficacy of current treatments and open
doors for future interventions, a promising method is to
deliver drugs, stem-cells or genetic cargo directly into the
degenerated retinal segment where blood vessels or drusen
are located. This method is suggested by ophthalmologists
including clinical partners of this work. Due to physiological
limitations of human surgeons, the feasibility of controlled
sub-retinal interventions has not yet been tested. Intraoperative visibility of sub-retinal layers together with microscale precision of manipulation (e.g. injection) are challenges
beyond current human surgeons’ physical capabilities that
reduce the safety and intuitiveness of the sub-retinal interventions and therefore, need to be addressed first.
In the present work we design and develop a targeted
drug delivery system, which combined with the state of the
art in robotic surgical tools, helps overcome the physical
shortcomings of human surgeons. Intra-operative Optical
Coherence Tomography (iOCT) is an emerging technology
for visualizing sub-retinal layers in real-time [8]; Robotic
technology enhances surgeons’ intra-ocular dexterity [9],
[10], [11], [12], [13], [14] and a modular-based workstation
integrates these technologies with a surgical planner architecture to enable targeted based delivery into the degenerated
retinal segment (e.g. macula).
To achieve the aforementioned objectives, we developed
a platform called Cognitive Retinal Generator (CRG). CRG
consists of: 1) Robotic setup (the Cognitive Tool); 2) Subretinal needle: a micro cannula to be attached to the robot;
3) Automatic drug pump for hands-free drug release; 4)
Patient fixture to comfortably and properly fix the patient to
the operation table and 5) Workstation to connect all these
elements with ophthalmic surgery equipment such as iOCT,
Microscope, etc. In this paper we present the components
of our setup and integration of these components into a
clinically compatible platform. Furthermore, we report on
a precise intra-retinal operation scenario, performed by the
proposed setup and under iOCT supervision. Robotic retinal
operation under intra-operative OCT was not reported before.
In this section we introduce each component of the setup.
A. Sub-retinal Cannula
To successfully perform sub-retinal injection, having a
proper micro-needle is the first requirement. Due to the fact
Fig. 4. Injection Mechanism: a) Sub-retinal Cannula; b)Drug Reservoir;
c)Reservoir-Pump tube; d)Luer Lock interface; e)Robot Gripper; f)Robot.
that there is no reliable definition of a “proper” needle for
sub-retinal injection, we tried several types of micro-cannulas
to investigate which is more suitable for our application,
these micro-cannulas in the current practice are used for
retinal vessel cannulation. As shown in Fig. 3, four cannulas
were evaluated: A micro cannula, 23G (0.56mm) with 80µ m
straight-bevel tip [INCYTO Co., Ltd.]; three micro cannulas,
25G (0.46mm) with 50µ m bent-bevel tips (35◦ , 40◦ and
45◦ )[Geuder AG.]. To optimize the micro injection, we
developed a novel concept, Instrumental Injection System
(IIS), to use a laser-integrated needle for the operation. The
tool consists of a 23G cannula for inserting a catheter in
the region of interest. The cannula comprises of two parallel
channels: the first channel is for receiving and guiding the
catheter (Channel: 200µ m in diameter Catheter: 40µ m in
diameter) and to provide intuitive access, this channel has a
bent tip; the second channel or the pointing channel (200µ m
in diameter) has an optical fiber that emits a visible laser
beam (Safety class I, 40µ W) to indicate the catheter insertion
point on the retina. Using this mechanism, surgeons are able
to visualize the final hitting point of the needle on the surface
of the retina and therefore precisely locate it [17]. (see Fig. 3I).
B. Drug reservoir and automatic pump
To avoid bringing surgeons’ tremor – one of the main
sources of imprecision – into the surgical site, we designed
and developed an injection pump connected to the drug
reservoir. A bidirectional mechanism to be used for injection
and suction. The pump is controlled by a foot pedal. The
injection/suction speed is proportional to the force applied
by the surgeon and therefore, they are able to control their
desired injection/removal dozes.
C. Robot
Fig. 3. Sub-retinal cannulas: I: Lasser guided cannula is in prototyping
phase. II: Conventional sub-retinal cannulas with minor modification: a)
23G (0.56mm) with 80µ m straight-bevel tip; b)23G (0.56mm) with 70µ m
35◦ tip; c)23G (0.56mm) with 70µ m 40◦ tip; d)23G (0.56mm) with 70µ m
45◦ tip.
The sub-retinal cannula is attached (with Luer Lock) to
the drug reservoir and the reservoir is connected to the pump
with tubes (See Fig. 4). The next component is the robot that
holds the needle and reservoir. The robot is similar to [16],
that is adjusted to be used for sub-retinal injection. The key
features of the robot is the small size: 185 × 44 × 226mm and
Fig. 5. (a) Serial robot: A and B are parallel coupled joint elements. C
is the tool gripper consisting of a prismatic actuator and an optional tool
rotator, (b) relevant and simplified model of the serial robot [16].
low weight of only 300g. These properties make the system
portable and comparatively easy to bring into the operating
room. To achieve such a small construction, five piezo
prismatic actuators are used to manipulate the end-effector.
The original configuration of the robot (see Fig. 5) consists
of three serial segments; two Parallel Coupled Joint Mechanisms (PCJMs), the tool gripper comprising a prismatic
actuator for tool translation and an optional revolute actuator
for tool rotation. The forward kinematics of the end effector
ensures that the working volume required for vitreo-retinal
procedures observed in [?] can be satisfied. The precision of
each actuator with nano-meter optical encoders is 0.5µ m and
the measurement result of the endeffector precision using
a Coordinate Measurement Machine (CMM) in x, y and z
directions are 14µ m, 10µ m and 4µ m, respectively.
D. Patient Fixator
In the latest platform, prototyped for this work, two
Degrees Of Freedom (DOF) are added to the setup (see
Fig. 8 - segment 1). These DOFs are linear sliders (100mm
displacement for each) to move the base of the robot in
two directions (in a working volume of 100mm × 100mm).
Additional DOFs, by increasing the working volume of the
robot, give the surgeon the possibility of bringing the robot
Fig. 6.
Patient’s head fixation mechanism.
to the optimum position with respect to the patient’s eye
and easily remove the robot from the surgical site when
the robotic operation is no longer needed. The optimum
position is the entry point to an additional trocar inserted
for robotic surgery, the location of the trocar is related to
the procedure and the target position and is defined by the
surgeon. These 2 DOFs are kinematically decoupled and
affect neither kinematics nor control of the robot. Furthermore, the mounting mechanism is changed, the gripper is
adapted to the application and the robot base has the ability
to be fixed to the eye. One of the challenges in vitreo-retinal
surgery is unwanted movement of the operating table, patient
and the patient’s head. These motions that could be in the
order of centimiters are serious obstacles to achieve high
precision operations. In conventional vitreo-retinal surgery
these movements are partially controlled by the surgeon.
Surgeons put their hands (abductor digiti minimi region)
on patient’s forehed to control patient’s motion and also to
stabilize their own motion. In robotic vitreo-retinal surgery,
however, this condition was not considered before, and yet it
is a challenge to be solved. To address the unwanted motions
our next step was to build a rapid, non-traumatic fixation
of the head during surgery, suitable for all head shapes. A
direct, non-traumatic fixation of the skull is viable only via
the upper jaw and requires individual adaptation of the fixing
unit (eg. by a jaw impression) which must be performed prior
to operation. To avoid this step, the skull is fixed indirectly
via the scalp. In order to prevent movement between the
skin of the scalp and the skull, the fixing mask has to fit as
accurately as possible and preferably full-faced, similar to a
foam-packaging. The operation of the head fixation can be
described as follows: The patient is supine and the occiput
rests on a head ring.
Laterally to the head two vertical plates (Fig. 6 - 1)
are fixed in position. Between the plates and the head,
two vacuum-pads (Fig. 6 - 2) are attached on either side.
The vacuum-pads are filled with a fine-particle granulate
and, at the start, also with air. Subsequently two additional,
unfilled air-cushions (Fig. 6 - 3) are placed between the
vacuum-pads and the side plates. The vacuum-pads are now
deflated (evacuated), while the air-cushions will be inflated
simultaneously. The surface of the vacuum-pads adapts to
the individual shape of the patients head and becomes rigid,
while the air-cushions create the required pressure to lock the
head into position. When the operation is over, the fixation
can be released by opening the air valves.
E. Eye Fixator
An optional rigid arm with an air suction ring tip, similar
to Corneal vacuum trephine system [18], is used to fix the
patient’s eye to the base of the robot. However, in the wet
labs with pig eyes this arm is not necessary, this is due to
the limited unwanted eye motions in ex-vivo experiments in
which the pig eye is fixed to the dummy head.
G. Workstation
Fig. 7.
Robotic setup is mounted on the patient fixation system.
F. Control Strategy
The surgical workflow of the injection, assuming that the
pars plana vitrectomy was performed before, is as follows:
I) Phase I (see Fig. 8-1): Introducing the needle into the
eye chamber through the additional incision point. This is
performed by the robot base actuators (first 2 DOFs) - this
is controlled directly by the surgeon;
II) Phase II (see Fig. 8-2): Approach. Here, the surgeon,
using robot, fine-localizes the intersection of the tooltip
and the injection point - this is controlled directly by the
surgeon while the robot provides virtual fixtures such as
remote center of motion (RCM) to pivot the tool around the
incision point;
III) Phase III (see Fig. 8-3): Intra-retinal insertion. It is
done by the insertion or z axis motion only when all the
other DOFs are stabilized - this is controlled directly by the
IV) Phase IV: Sub-retinal needle tip stabilization. It is
performed under intra-operative OCT supervision;
V) Phase V: Drug Injection (in some cases suction). It is
carried out by the automatic-pump when the surgeon’s hand
is out of the surgical site - this is done by the surgeon using
a food pedal, surgeon observes the amount of injected drug
from both the iOCT and en-face image.
Fig. 8.
Robot assisted surgical work flow.
In order to increase the precision of the micro- manipulation even more it is necessary to give the surgeon
control over important parameters of the system. Access
to the robotic translational and rotational velocities allows
for higher precision during crucial tasks and higher speeds
when saving time is the priority. Other components should
also be able to give access to their important parameters
like the injection speed or injection dose for an injection
module. And all of this should be available in a single
place, with a single GUI where everything is simply and
quickly accessible. The workstation (surgical cockpit) acts
as a central control unit to link, control and manage internal
components of CRG as well as external components that
need to be linked to the setup. The first surgical cockpit
prototype of this work consists of a high performance PC
with a touchscreen monitor. Currently the components that
are linked to it are: robot controllers, surgeon’s input unit,
moderator’s input unit, injection pump, microscope and the
safety camera. The surgical cockpit has been designed with
feedback from surgeons and technical staff in order to ensure
its usability in a surgical environment.
Fig. 9. CRG Surgical Setup: a) Cognitive Tool; b) Surgical Microscope;
c) Surgical Cockpit; d) Training/Simulator; e) Patient Fixation: wooden
prototype; f) Input Devices; g) Air-Cushion; h) Vacuum-Pad.
H. Input Device
A custom made input device was manufactured for controlling the micromanipulator. The device consists of two
SpaceNavigators [3Dconnexion]. Two users, for example an
expert surgeon and a medical student, can easily switch
giving input. This could prove useful for training as an
experienced surgeon can intervene anytime and is able to
override commands of the other user. Additionally, there are
two buttons, one for each SpaceNavigator device. The users
can conveniently press them by resting their wrist or forearm
on it. Only when this button is being pressed commands from
the corresponding SpaceNavigator are being forwarded to the
robot. This is a safety feature as it ensures that the robot does
not move by mistake.
Fig. 10. Microscopic view and crosshair iOCT B-scan of a ex-vivo experiment. These images are taken from a robotic intra-retinal vessel injection
experiment. A) shows the location of the needle tip to the injection point (yellow arrow); B) shows the insertion of the needle tip to a retinal blood vessel
(red arrow in OCT shows the cross section of the retinal vessel which is punctured); 3) shows injection of brilliant blue dye - as an indicator agent - into
the retinal blood vessel.
I. Training
Due to the fact that CRG is introducing new classes
of ophthalmic surgical procedures, training is a vital step
towards bringing the method and the technology to the
eye clinics. A simulator that enables controlling a virtual
representation of the micro-manipulator has been developed
as a basis for the training platform. The user is able to use
the very same input device as with the real robot and the
movements of the robot are being displayed realistically.
The scenery consists of a mannequin head with a model
of an eye as well as the ophthalmic robot equipped with a
micro needle. The virtual camera can be manipulated freely
using the mouse. In addition, there is a second viewpoint,
which shows the eye directly from above. Using this view
the trainee is able to see the retina through the pupil, just
like a surgeon using an operating microscope in real surgery.
Our training strategy, inspired by pilot training, has two
modes of operation: 1) A simulation/training environment
designed and implemented for CRG setup. With the CRG
Simulator surgeons are able to perform multi-level tasks by
loading games or various real unhealthy retinal images; 2)
CRG Simulator similar to the actual setup have two input
devices. One of them is used by the trainee and the other by
the trainer. Using the double-controller method, the trainer
is able to guide, teach and correct the trainee.
We assembled and evaluated a functional prototype of the
setup at the Ophthalmology department of Klinikum rechts
der Isar. Our focus in the preliminary wetlabs was to evaluate
the intuitiveness of the system and its compatibility with
conventional retinal surgery setup. The wooden mock-up of
the head fixation system was tested with 7 (5 male and 2
female) human heads with variable shapes and sizes; these
tests, where the head was held in position for 15 minutes,
show that the head fixation mechanism is not uncomfortable
for the duration of the retinal surgery (less than 15 minutes),
especially when the patients with topical anesthesia were able
to fix the last stage by themselves i.e. by manually pumping
the air into the bags. Furthermore, using vacuum pads with
less stiffness than human skull, avoids unwanted damages
or pains. With this kind of fixation the average relative 3D
motions of the dummy head (without skin) and human head
(with skin) with respect to the robot base are 4mm (max:
6mm) and 6mm (max: 9mm) (with deviation of less than 1mm
in each axes for each case) respectively. This level of the head
fixation, that was measured by an optical tracking system
[Polaris - Northern Digital Inc.], guarantees that fixing the
eye ball to the robot base using a suction ring will not cause
anatomical damage. We furthermore performed all surgical
work flow phases successfully, with full compatibility with
conventional surgical configuration and devices such as ophthalmic microscope and Phaco machine; no changes in the
makeup or position of the operating room staff were needed.
The relative precision of the tool tip motion with respect to
the robot base is 14×10×4µ m in x, y and z axes respectively.
To evaluate the entire setup in a clinical environment we
asked our surgeons to perform retinal vein cannulation, one
of the most challenging ophthalmic surgeries, using the CRG
setup under an OCT integrated microscope. The aim was to
inject a liquid in a retinal vessel. For this experiments our
clinicians decided to perform the operation on pig eye that
has almost similar anatomy to human eye. We collected pig
eyes directly from a local slaughterhouse approximately 30
minutes before the operation; to keep the eyes fresh and
avoid having blurred cornea and degenerated retina. Directly
after collection we put the eyes in Balanced Salt Solution
(BSS) container. After fixing the pig eye on a dummy head
and fixing the dummy head utilizing the fixation system,
surgeons initiated the experiment by pars-plana procedure.
Using 23G (0.64mm in diameter) trocar system, they created
three incision points on the sclera of the pig eye. They
connected the infusion line to the first incision point in order
to control the pressure of the eye during the operation to
avoid collapsing. They located the microscope [Lumera 700
Rescan, Carl Zeiss AG.] on the eye, then they introduced
the illuminator through the second incision point and the
vitrectomy tool through the third incision point. Then they
performed vitrectomy to cut the vitreous from the eye and
replace it by a clear liquid.
The next step was to create an additional incision point,
this time using a 25G (0.56mm in diameter) trocar. During
the experiment, this trocar was used to insert the cannula
mounted on the robot.
Meanwhile, beside the operation, the assistant prepared the
robot. She filled the drug reservoir with MembraneBlue Dual
Syringe [D.O.R.C International BV.]. This is a blue substance
that normally is used for staining (e.g. for membrane peeling) in this experiments, however, the surgeons suggested
using MembraneBlue as an agent to be injected in order to
visualize the procedure. Furthermore, the assistant placed a
25G cannula with bent-bevel tip (40◦ , 50µ m)[Geuder AG.]
on the reservoir and placed the reservoir on the robot.
Next, the surgeon located the robot, using the first 2 DOFs,
in a position where he could access the needle through the
4th incision point. It is worth mentioning that during this
experiment, to minimize surgeon’s failure, the dual input
device was activated meaning that an additional surgeon was
supervising the procedure. However, due to the fact that the
main surgeon had more than 2 hours of simulation training,
the experiment was performed without emergency interrupt.
In the main part of the operation, the surgeon: a) adjusted
and fixed the robot base; b) located the needle on the entry
point of the 4th trocar; c) activated the RCM mode; [from
this point he used the microscope] d) pushed the cannula
inside the eye; e) defined an exact injection point; f) finelocated the needle tip to the target and locked all DOFs but
the last (insertion) actuator; g) inserted the needle tip into the
blood vessel when he was observing the intra-vessel location
using the realtime OCT image; h) stabilized the robot and
consequently the needle tip; i) injected the MembraneBlue
Dual Syringe into a retinal vessel (see Fig. 10).
This experiment was performed successfully. On the first
attempt, the surgeon was able to locate the tool tip in x, y
plane at the target point which was a circular point with a
diameter of 70µ m (see Fig. 10-A). The time he needed to
perform fine-locating the needle from intra-ocular cannula
insertion to stabilizing the tip on the target was approximately 230 seconds. Under iOCT, in another 25 seconds
manipulation, he inserted the cannula into a blood vessel
along z axis with less than 10µ m precision (see Fig. 10B). Finally he stabilized the cannula-tip, with cross section
diameter of 50µ m, in a retinal blood vessel, with cross
section diameter of 70µ m, for duration of 7seconds to inject
50µ L of MembraneBlue (see Fig. 10-C).
iOCT is an emerging technology that has recently found
routes to the ophthalmic operating rooms. Today with this
technology, eye surgeons are able to intra-operatively see
underneath the retina: a region where the causes of various
eye diseases and sources of vision loss can be found. Combining iOCT with an assistive tool that improves the surgical
skills of the surgeons will enable sub-retinal interventions
such as targeted intra- and sub-retinal drug delivery. AMD
is a disease with high prevalence among elderly people that
benefits greatly from such an intervention. In this work,
we designed, developed and evaluated a setup to enhance
intraocular maneuver of the surgeons and enabled targeted
administration of drugs in retinal micro-structural anatomies.
Here, we reported a successful retinal vein cannulation,
which is one of the most challenging retinal procedures,
using a robotic platform, clinically grade fixation systems
and under supervision of iOCT integrated microscope. This
work is a step towards addressing AMD treatment. A team
of surgeons and engineers is involved and cooperate in
this research to ensure the compatibility of the results with
clinical requirements, patient benefits and clinical safety
The next step of the project is to link the setup with an
iOCT integrated ophthalmic microscope to perform semiautonomous drug injection in a degenerated macula.
[1] L. S. Lim et al., “Age-related macular degeneration,” The Lancet,
379(9827), 2012, pp. 1728-1738.
[2] R. Ratnapriya and E. Y. Chew, “Age-related macular degenerationclinical review and genetics update,” Clin Genet 84, 2013, pp. 160166.
[3] W. L. Wong et al., “Global prevalence of age-related macular degeneration and disease burden projection for 2020 and 2040: a systematic
review and meta-analysis,” The Lancet Global Health, 2(2), 2014, pp.
[4] S. D. Schwartz et al., “Human embryonic stem cell-derived retinal
pigment epithelium in patients with age-related macular degeneration
and Stargardt’s macular dystrophy: follow-up of two open-label phase
1/2 studies,” The Lancet 385.9967, 2015, pp. 509-516.
[5] S. D. Schwartz et al., “Embryonic stem cell trials for macular
degeneration: a preliminary report,” The Lancet 379.9817, 2012, pp.
[6] M. J. Song and B. Kapil, “Looking into the future: Using induced
pluripotent stem cells to build two and three dimensional ocular tissue
for cell therapy and disease modeling,” Brain research 1638, 2016, pp.
[7] S. D. Nicoaraet al., “Novel Strategies for the Improvement of Stem
Cells Transplantation in Degenerative Retinal Diseases,” Stem Cells
International 2016.
[8] J. P. Ehlers et al. “Intraoperative optical coherence tomography using
the RESCAN 700: preliminary results from the DISCOVER study,”
British Journal of Ophthalmology, 2014.
[9] W. Wei et al., “Enabling technology for microvascular stenting in
ophthalmic surgery,” ASME Journal of Medical Devices vol.4, 2010,
pp. 1-6.
[10] T. Nakano et al., “A parallel robot to assist vitreoretinal surgery,” In:
Int. Journal of Computer Assisted Radiology and Surgery. V. 4, I. 6,
2009, pp. 517-526.
[11] S. Charles and T. Ohm, “Dexterity-enhanced tele-robotic microsurgery,” In The Proc. of the IEEE Int. Conf. In advanced robotics,
[12] W. T. Ang et al., “Active tremor compensation in microsurgery,” In
Eye, 2004, pp. 2738-2741.
[13] A. Uneri et al., “New steady-hand eye robot with micro-force sensing
for vitreoretinal surgery,” In Proc. of the 2010 3rd IEEE RAS and
EMBS Int. Conf. on Biomedical Robotics and Biomechatronics, 2010,
pp. 814-819.
[14] P. Caers et al., “Precision experiments on a comanipulated robotic
system for use in retinal surgery,” In Proc. of the 2011 SCATh
Joint Workshop on New Technologies for Computer/Robot Assisted
Surgery, 2011, pp. 1-7.
[15] A. Barthel et al. “Haptic interface for robot-assisted ophthalmic
surgery,” In Proc. of the 37th IEEE Int. Conf. of Engineering in
Medicine and Biology Society (EMBC), 2015, pp. 4906-4909.
[16] M. Ali Nasseri et al., “The introduction of a new robot for assistance
in ophthalmic surgery,” In Proc. of the 35th IEEE Int. Conf. of
Engineering in Medicine and Biology Society (EMBC), 2013, pp.
[17] M. Ali Nasseri et al., “Laser Guided Catheter for Sub-retinal Injection,” In Investigative Ophthalmology and Visual Science June 2016.
[18] G. N. Krasner and J. W. Berkman, “Corneal vacuum trephine system,”
U.S. Patent No. 4,796,623. 10 Jan. 1989.
Без категории
Размер файла
6 392 Кб
8037815, 2017, embc
Пожаловаться на содержимое документа