Search

Objective Mathematical models for the delineation of regional myocardial perfusion and metabolism with PET require faithful reconstruction of arterial and myocardial time-activity curves following administration of radiotracers. High temporal resolution is often required in such measurements. Many commercially available tomographs exhibit long dead times that limit their count rate capabilities. To overcome these limitations, we developed and tested a whole-body tomographic device (Super PET 3000-E) with high count rate capabilities. The use of cesium fluoride scintillation detectors coupled with a one-to-one detector photomultiplier configuration reduces the system resolving and dead times. Materials and Methods The Super PET 3000-E was subjected to a series of tests with phantoms to determine its resolution, sensitivity, linearity, count rate capabilities, dead time, and random coincidence contribution. Results The system sensitivity is 136 kcounts/s/μCi/ml and its transverse and longitudinal resolutions are 8.5 and 10.5 mm full width at half-maximum, respectively. The system can easily record a total event rate of 2.0 Mcounts/s with minimal dead time loss and excellent linearity. Conclusion The system fulfills its design goals and allows the very high count rate performance needed for the application of the physiological models used in our cardiac studies.

A new computer system designed for Posi- tron Emission Tomography (PET) devices that utilize list- mode data acquisition and Time-of-Flight(ToF) information is presented. The approach uses commercially available subsys- tems, all supported by hardware and software industry stan- dards. The host supports a VME-based disk farm that provides inexpensive disk storage operating at high transfer rates capa- ble of easily supporting list-mode data collection. All software development as well as the production environment is suppomd entirely under UNIX. All code development uses the C language. Graphical user interfxes are developed under X-Windows, Version 11. Images and data which are produced by the scanner are easily transferred to other environments via etheme$ link and 'KP/lP. Benchmark results show an increase in list-mode data collection rates, while decreasing the loss associated with the collection protocol. Image reconstruction times are shown to decrease markedly. Overall cost is also decreased. This allows a very flexible approach to system selection, enabling the researcher to choose any standards-based plat- form which provides a satisfactory cost-performance ratio. 1. The Importance of Industry Standards

The benefits of block type detectors for PET applications are well understood and are provided by commercial BGO and NaI(Tl) systems. In order to extend the count rate capabilities for many applications, a block type detector using GSO(Ce) has been designed, constructed, and tested. GSO(Ce) crystals are shown to produce 2.9 times the effective light output of BGO and yield an energy resolution of 11% FWHM for 511 keV photons. An assembly of a square array of 36 crystals measuring 4.5/spl times/4.5/spl times/30 mm coupled to 4 cylindrical photomultiplier tubes produces signals which are sampled and recorded within 150 nsec. The computed Anger plot demonstrates individual detector delineation for both flood field and collimated beam illumination. The 56 nsec decay time of the emissions of GSO(Ce) coupled with the small size of the block should realize approximately an order of magnitude improvement in count rate over many current commercial BGO block detector modules based on dead time considerations. >

Specifications were developed for a PET (positron emission tomography) system that emphasizes high-count-rate performance and is specifically designed for the estimation of myocardial perfusion using oxygen-15 water and dynamic imaging protocols. The seven-slice CsF detector array is a one-on-one configuration, and has a reconstructed FOV (field of view) of 52 cm and an axial FOV of 10.7 cm. Wobble and axial translation are used. The gantry and associated electronics were interfaced to a new computer system, data collection and image reconstruction software was developed and the performance of the system was characterized with phantom, animal, and human studies. The system sensitivity is 136000 cts/s/ mu Ci/cm/sup 3/ and the reconstructed resolution is better than 9 mm. Imaging a 20-cm uniform phantom with 4.3 mu Ci/cm/sup 3/, the system collects a total rate of 1.57 Mcts/s with a loss fraction of 0.72 and a ratio of randoms/trues of 0.64. >

An accidental coincidence is defined as the erroneous registration of two photons, originating from separate positron annihilations, as having originated from the same positron annihilation. Previous analyses which did not consider accidental coincidences indicated that for a certain radioactivity distribution a gain in image signal-to-noise ratio of about 5 dB is achieved by the time-of-flight method over the conventional method. Subsequent experiments have validated this prediction in low counting rate situations. For higher, typical counting rates these experiments showed a significantly larger gain of about 9 dB, which was attributed to the way in which the time-of-flight method suppresses the degrading effects of accidental coincidences. We present an analytical model, extended from a previous model, which considers accidental coincidences. Calculations of signal-to-noise ratio, using this model, compare well with the experiments and show that the additional gain is indeed due to the treatment of accidental coincidences. An understanding of the model leads to an intuitive explanation of the gain mechanism and a determination of an effective coincidence-timing window that is achieved by the time-of-flight method.

In positron emission tomography (PET), the annihilation radiation is usually detected as a coincidence occurrence that localizes the position of the annihilation event to a straight line joining the detectors. The measure of the difference between the time of flight (TOF) of the annihilation photons between their inception and their detection permits the localization of the position of the annihilation event along the coincidence line. The incorporation of TOF information into the PET reconstruction process improves the signal-to-noise ratio in the image obtained. The utilization of scintillation detectors utilizing cesium fluoride scintillators, fast photomultiplier tubes, and fast timing circuits allows sub-nanosecond coincidence timing resolution needed for the effective use of TOF in PET. Mathematical considerations and pilot experiments show that with state-of-the-art electronic components and through the application of proper reconstruction algorithms, the combination of TOF and PET positional data improves severalfold the signal-to-noise ratio with respect to conventional PET image reconstruction at the cost of increasing the amount of data to be processed. The construction of a TOF-assisted PET device is within the capability of state-of-the-art technology.

The physical characteristics and some imaging capabilities of Super PETT I, a positron emission tomograph utilizing time-of-flight (TOF) in its image reconstruction process were assessed experimentally by means of measurements carried out in phantoms and clinical imaging studies. The performance characteristics assessed included sensitivity, spatial resolution, image improvements resulting from time-of-flight information utilization, system dead time, and linearity. The clinical examples included imaging of the brain, the heart, the liver, and a demonstration of Super PETT I's capability of achieving cardiac gating.

Recent advances in PET designs have shown that a gain in signal-to-noise ratio can be expected by incorporating time-of-flight data in positron emission tomography over the conventional PET mode. It has also been shown that cesium fluoride (CsF) offers the potential of faster timing and high detection efficiency which would be required for a clinical scanner utilizing time-of-flight information. Our research with CsF and the results of a feasibility study of time-off-light positron emission tomography reconstruction have shown that, indeed, a significant improvement in image quality results from such an approach and that coincidence resolving times of less than 500 psec FWHM are easily achievable with CsF detectors. However, the design of fast tomographic systems with multiple detectors which maintain this fast coincidence timing poses a challenging technical problem. The solution to this problem requires a departure from the conventional mode of PET designs to a fast on-line microprocessor based system which is capable of compressing and correcting the data for timing differences, normalization and image function. Such a system is described in this paper and its advantages and disadvantages are discussed.

The first positron computed tomography system with cesium fluoride scintillation detectors, PETT VI, has been developed. The system provides 7 slice images with 4 detector rings (57 cm). Performances of the system are discussed and clarified based on experimental data. Adoption of CsF detectors decrease the random coincidence rate by achieving a short coincidence timing resolution. The timing resolution of a pair of detector heads is 1.5 nsec FWHM and 3.0 nsec FWTM. After a preliminary timing alignment of 288 detectors, the coincidence window width (2?) of 11 nsec or wiider has yielded maximum coincidence sensitivity, and operation at 5.9 nsec has given 91% of maximum. The full sensitivity for a cyclindrical uniform phantom (20 cm dia × 13 cm) is 354 kcps/microCi/cc/7 slices in a low resolution mode (intrinsic resolution at center is 11.7 mm FWHM). In a high resolution mode (7.1 mm), it is 31% of the full sensitivity. Coincidence rates ratio, [random/(true + scattered)], in the low resolution mode, is 0.16 ? with 5.9 nsec window width inside the phantom images reconstructed without a random correction, where ? is activity density (microCi/cc). The ratio is 1/3 of that obtained when operating at 20 nsec. Scattered coincidence fraction at the center of the phantom images, without the random correction process, is 9% of the [true + scattered].

As part of the Positron-Emission Tomography (PET) [1] reconstruction process, annihilation counts are normalized for photon absorption, detector efficiency and detector-pair duty-cycle. Several normalization methods of time-of-flight and conventional systems are analyzed mathematically for count bias and variance. The results of the study have some implications on hardware and software complexity and on image noise and distortion.

Various high-resolution scintillation detectors are analyzed for resolution and sensitivity. In particular, the effects of using a wedge face and inter-crystal tungsten-septa are studied for both BaF2 and BGO crystals. The results indicate that usage of septa alone causes an improvement in resolution in areas away from the center of the image, but also causes a degradation in sensitivity in those areas, and that usage of a wedge face alone causes no improvement in resolution, while usage of septa and a wedge face in combination causes a similar resolution improvement, but with a lesser degradation in sensitivity.

Incorporation of differential time-of-flight (TOF) information in positron emission tomography offers significant performance improvement at the cost of increasing the quantity of data to be processed. A tomographic unit utilizing TOF data, called Super-PETT, is currently nearing the final stage of construction. Design and implementation aspects of the data acquisition system are described with an emphasis on instrument operation in an on-line preprocessing mode.

A new image-reconstruction algorithm for time-of-flight (TOF) assisted positron-emission tomography is presented. This linear algorithm differs from previously published linear algorithms in the way the angular information of the measurements is used. The needed computations are reduced by coalescing data having angles within a range into a smaller number of "group" angles. This angle-reduction is made possible by the additional TOF measurement. It is shown that in the limit as the numer of groups is very large this reduced-angle weighting algorithm is the same as the confidence-weighting algorithm [1]. Similarly, if the data are grouped into a single angular range, the new algorithm is the same as the most-likely-position algorithm [1].

One recent development in image reconstruction utilizing time-of-flight information yields a compression of the three dimensional array used in forming the unfiltered "preimage." A reduction from 96 angles to 16 angles is shown to produce equivalent quality images and may form a basis for future architectural strategies. A second development in image reconstruction can strongly affect the quality of results when constraints such as cardiac gating are imposed, while not inherently sacrificing spatial resolution due to inadequate sampling. A correction for sampling density obtained by analyzing list mode data is shown to greatly reduce circular artifact production. This technique is extendable to systems without time-of-flight measurements and to scanning motions other than wobbling.

We designed and built a positron emission transverse tomograph (PETT VI), designed specifically for fast dynamic studies in the human brain, and for cardiac studies in experimental animals. The scintillation detectors incorporated into this device are fitted with cesium fluoride crystals. Cesium fluoride was selected for this purpose because its short fluorescence decay allows the use of a short coincidence resolving time with a concomitant reduction of unwanted random coincidences. PETT VI utilizes four rings of 72 detectors simultaneously yielding seven tomographic sections. The system can be operated in either a low or high resolution mode with intrinsic geometrical resolutions in the plane of section of 7.1 to 11.7 mm full width at half maximum (FWHM), for a slice thickness with a resolution at the center of 13.9 mm FWHM. The maximum sensitivity of the system for seven slices in the low resolution mode is 322,000 cps/microCi/cc in a 20 cm diameter phantom. The contribution of random coincidences before subtraction in PETT VI was found to be approximately 14% of the counts in the phantom image with a source of approximately 3.5 mCi of a positron emitting radionuclide dispersed in a 20 cm diameter tissue equivalent phantom with a concentration of 1 microCi/cc. The short coincidence resolving time of the system permits rapid data acquisition for attenuation corrections and clinical dynamic studies with data acquisition times of less than a minute.

The ideal scintillation detector for positron tomography would be a very high Z material, nonhygroscopic, have a fast decay time, and copious light output. NaI(Tl) has good light output, and even though it is hygroscopic has found acceptance as a good scintillator. Recently BGO has been shown to satisfy the first two requirements for tomography, and has replaced NaI(Tl) as the detector of choice in tomography. However, its low light yield and poor coincidence timing make it unsuitable for fast scanners. Cesium fluoride (CsF) has been investigated by us as a possible scintillation detector for positron tomography, and shows great promise even though it is extremely hygroscopic and exhibits low scintillation efficiency as compared to NaI(Tl). Its very fast decay time, good detection efficiency, and fast coincidence timing make it an ideal detector for tomographs designed for fast dynamic studies. Moreover, it permits the incorporation of time of flight information with conventional tomography for improved signal to noise ratio in the image.

Imaging of the brain by positron emission tomography can be optimized for sensitivity by dedicating the design of the tomograph to this application. We have designed a multislice positron emission tomograph (PETT V) for imaging the human brain and the whole body of small experimental animals. The detector system of PETT V consists of a circular array of 48 NaI(Tl) scintillation detectors, each fitted with two photomultiplier tubes, with one dimensional positioning capability. Suitable sampling is achieved by rotation of the circular array of detectors and by a wobbling motion of the detector circle. The proposed system is capable of providing seven slices simultaneously, with a spatial resolution in the plane of the slice from 7 to 15 mm and with slice thicknesses of 7 and 14 mm. The minimum scanning time is 1 sec. The estimated overall sensitivity of PETT V is 350,000 counts/sec/mCi in a 20 cm diameter phantom for a resolution of approximately 1.5 x 1.5 cm. The system is under construction.

The gain achieved in image quality by utilizing, in the image forming process, the time-of-flight information (TOF) of positron annihilation photons between their inception and detection was measured experimentally by means of a positron emission tomograph (PET)-Super PETT I. The measurements were carried out by imaging a 35 cm cylindrical uniform phantom containing different positron activity concentrations. The gain achieved through the incorporation of TOF information, defined as the ratio of variances in images reconstructed with and without TOF information, was found to be approximately 3 at the lowest activity concentration and 5-8 in the activity concentration range typically encountered in clinical studies especially in fast or dynamic studies. This increase in gain with activity was interpreted as resulting from the reduction of random coincidences when TOF information is used. Further image improvement is yielded by incorporating TOF information into the PET attenuation correction provided by the measurement of transmission of annihilation photons in the object imaged.

The effect of the software coincidence timing window (SCW) has been studied and demonstrated experimentally using Super PETT I, a time-of-flight (TOF) assisted positron emission tomograph. This method utilizes the TOF information to eliminate unwanted random and scattered coincidence events outside of a region of interest. The effect is substantial for high count rate studies, and can dramatically improve the noise contribution from measured attenuation factors. The effective coincidence window is reduced to 3.3 nsec for reconstructions of the field of view, and reduced to only 2 nsec for measurement of the attenuation factors.

PETT VI is a circular ring, multislice positron emission tomograph designed for small field studies. Features such as wobble and rotate motions, bimodal resolution, cesium flouride scintillators, fast timing circuits and dual buffer histogram memories allow a flexible data acquisition environment. Efficient modularized software concepts are implemented to facilitate concurrent tasks of data acquisition, processing and display. The ability to effectively manage high count rate situations and collect continuous time frames makes PETT VI ideally suited for dynamic studies.