вход по аккаунту


Dynamic cerebral positron emission tomographic studies.

код для вставкиСкачать
Dynamic Cerebral Positron Emission
Tomographc Studies
Michel M. Ter-Pogossian, PhD, David C . Ficke, MS, Mark A. Mintun, MD,
Peter Herscovitch, MD, Peter T. Fox, MD, and Marcus E. Raichle, MD
It is well established that a number of variable factors
related to the regional, functional integrity of the brain,
such as blood flow, metabolism, and blood volume,
change rapidly as a function of time, probably with a
time scale appreciably shorter than 1 minute. Under
the circumstances, it appears desirable for the investigation of such variables by positron emission tomography (PET) to apply methods that mirror these events
in the time dimension as closely as is practical. Yet
many PET examinations of the brain, including glucose
metabolic studies carried out with fluorine- 18-labeled
fluorodeoxyglucose and oxygen metabolic studies carried out by equilibrium inhalation are performed with
data acquisition times from many minutes to close to an
hour. Such prolonged data acquisition times are dictated in part by design features of the tomograph used
as well by the tracer model. The measure of regional
cerebral oxygen metabolism and blood flow by the
bolus administration of oxygen-15-labeled oxygen and
water circumvents the limitations of rapid change and
also exhibits the advantage of permitting repeated measurements in the same subject in a relatively short period of time. However, the utilization of these bolus
techniques imposes stringent design considerations
upon the tomographic systems to be used. In particular, such PET devices must be designed with an optimized temporal resolution. In general, PET tomographic systems with better temporal resolution are
highly desirable for at least four reasons:
Maintenance of a physiological steady state in the
brain over prolonged time periods is indeed
As mentioned previously, some studies, such as
measurement of cerebral blood flow by the Kety
autoradiographic technique, require data acquisition
times shorter than 1 minute to be valid.
Serial measurements are often desirable in the same
patient within a short interval.
Some quantitative tracer models (such as measure-
From the Edward Mallinckrodt Institute of Radiology, Department
of Neurology and Neurosurgery, Washington University School of
Medicine, St Louis, M O 63 110.
ments of receptor binding) require the rapid collection of sequential data sets.
Higher temporal resolution in PET studies of the
brain requires the use of an imaging device capable of
rapid data acquisition with a signal-to-noise ratio compatible with the required spatial and contrast resolution
desired; it is also preferable that such a device be capable of list-mode data acquisition for dynamic studies.
We developed a PET imaging device called Super
PEl'T I, which is specifically designed to achieve a temporal resolution of a fraction of a minute. This speed
was achieved through the use of fast scintillators
(cesium fluoride and barium fluoride) and through the
application of photon time-of-flight measurements with
a coincidence timing resolution of less than 500 ps.
Such a device has not only permitted the development
and implementation of strategies for the measurement
of blood flow and oxygen utilization with oxygen-15labeled water and oxygen based on a temporal resolu-.
tion of less than 1 minute; it has also made it possible to
use more detailed parameter estimation schemes to
measure tissue glucose kinetics and receptor pharmacology and to assess physiological phenomena that are
extremely brief in duration, such as cortical evoked
Some preliminary studies are being carried out with
the purpose of determining the ability of Super PE'IT I
to measure extremely brief events in the human brain
by examining changes in local cerebral blood volume in
the visual cortex in response to visual stimuli. In particular, we are examining the local response in cerebral
blood volume to bilateral, full-field, pattern-flash visual
stimulation, which is known to produce increased metabolic activity as well as blood flow in the visual cortex
of a variety of species, including humans. In these experiments, data are obtained using list-mode data collection for a total duration of approximately 10 minutes. From these data, images are reconstructed for the
40 ms preceding and the 40 ms immediately following
the visual stimulation. Each image therefore consists
Address reprint requests to Dr Ter-Pogossian
of the summation of about 6,500 such increments of
40 ms each. Preliminary data with this experimental
paradigm indicate that with a stimulus rate of 11.5 cycles per second, blood volume in the occipital cortex
varies approximately 20% in a cyclic and predictable
fashion. These data indicate that a fast device such as
Super PE'IT I has the capacity to follow the metabolic
changes associated with very transient electrical events
in the human brain. The capacity of such fast PET
devices opens up a broad horizon of new applications
for PET in the evaluation of physiological changes in
the human central nervous system.
Ter-Pogossian et al: Dynamic Cerebral PET Studies
Без категории
Размер файла
129 Кб
tomography, emissions, dynamics, positron, studies, cerebral
Пожаловаться на содержимое документа