Yaqiang Liu

A new pileup-prevention front-end electronic design for high-resolution PET and gamma cameras

A new method for processing signals from Anger position-sensitive detectors used in gamma cameras and positron emission tomography (PET) is proposed for very high count-rate imaging. It has a same concept as high yield pileup-event recover (HYPER) method we introduced before by using 1) dynamically integrating a present event, the integrating will stop immediately before the next event is detected; 2) estimating a weighted-value to indicate the total energy inside the scintillation detector; and 3) remnant correction to remove the residual energy of all the previous events from the weighted-value. This paper introduces two improved practical techniques to get a better weighted-value with low noise sensitivity in order to improve the final pileup-free energy resolution. One applies a low-pass filter combined with multiple sampling to a weight-sum of the instantaneous signal and integrated signal. The other one is weighting the integration value of the income signal; the weighting also includes exponential distortion compensation. This paper also describes the application of the HYPER electronics in a high resolution low cost PET camera with 12 photomultipliers (PMTs)-quadrant-sharing (PQS) detector modules that can decode 38 016 bismuth-germinate (BGO) crystal elements using 924 PMTs. Each detector module has four Anger-HYPER circuits to further increase the count-rate. To use the HYPER circuit in coincidence imaging applications, there is a serious synchronization problem between the arrival time of an event and the end time of integration that is variable from event to event. This synchronization problem is solved by a field programmable gate array (FPGA) circuit with real time remnant correction and a high-resolution trigger delay unit with a small dead-time for recovering the synchronization of data and the event-trigger.

A HOTLink/networked PC data acquisition and image reconstruction system for a high-resolution whole-body PET with respiratory or ECG-gated performance

An ultra high resolution PET camera in whole-body scanning or gated imaging study needs super computer-processing power for creating a huge sinogram as well as doing image reconstruction. A real-time HOTLink attached to networked cluster personal computers (PC) has been developed for this special purpose. In general, the coincidence data from a PET camera is unidirectional; therefore an additional daisy-chain bus using high speed HOTLink (400Mbit/s, Cypress Semiconductor, inc.) transmitters and receivers is designed to carry the coincidence data to networked (LAN) computers (PCs). In whole-body scanning, each PC will acquire sinogram data for one bed position, the data from HOTLink is interfaced to a PC through a fast PCI I/O board (80Mbyte/s); and after completion of data acquisition the PC begins to reconstruct the image meanwhile another PC will start data acquisition for the next bed position. The overall architecture for the image acquisition and reconstruction computing system is a pipeline design. The image result from one PC will be sent to a master computer for final tabulation and storage through the standard network, and this PC will be free for processing a new bed position. In gated respiratory or gated ECG imaging study, each PC will be reconfigured for processing a specified time-section image of a respiratory or ECG cycle. We are developing a high resolution PET camera with 38,016 BGO crystal elements which needs 1 to 2 gigabytes sinogram memory; the HOTLink/networked structure design allows us to split the huge sinogram into several PCs in real-time and the image reconstruction can be done in parallel.

Signal characteristics of individual crystals in a high resolution BGO detector design using PMT-quadrant sharing

The PMT-quadrant sharing (PQS) detector design allows very high resolution detectors to be built with 70% fewer PMTs and lower cost. A common concern for the design is that there is a big gap (photo-insensitive area) between four circular PMTs and the photoelectron signal (pulse height) may be much lower for the central crystals. The concern increases with the use of smaller PMTs for high-resolution designs because small PMTs have relatively thicker walls and relatively larger tolerance spaces between them. The authors measured the pulse heights and energy resolution for each crystal in three different types of PQS blocks for 19 mm PMT. For a square 7 /spl times/ 7 block detector (2.66 mm /spl times/ 2.66 mm /spl times/ 18 mm BGO needles), the maximum photopeak signals occurred at the corner crystal of the block. The signals for the worst central five crystals (sitting on space with no PMT connection) had pulse heights 0.87 as high as that of the corner crystals. The 12 crystals (outside the central five) with coupling only to the glass wall but not to the photocathode had a relative pulse height of 0.92. The eight crystals with partial exposure to photocathodes had a 0.94 relative pulse height. The energy resolution for individual crystals was 22% - 30% with an average of 26%. Asymmetric photopeaks, especially for the corner crystals, were observed, and these were found to be the result of the depth-of-interaction effect. In the latest PQS design, extended blocks with asymmetric light distributions were used on the four edges and four corners of a large detector module so that the previously unused (wasted) half-row of peripheral PMT could be covered by crystals. An asymmetric block, single-extended (7 /spl times/ 8 crystals) was also tested. The pulse-height ratio between the worst and best group of crystals in the single-extended block was 0.72 and that of the double-extended block was also 0.72. In a more demanding, higher spatial resolution 8 /spl times/ 8 array (2.3 mm /spl times/ 2.3 mm /spl times/ 10 mm BGO) for mouse PET with shallower crystals, the pulse-height ratio was 0.73 with an average energy resolution of 20%. This study demonstrated that pulse height uniformity for the PQS design using circular PMT was excellent, better than the typical 3/1 pulse-height ratio in conventional block detectors.

Evaluation of the effect of filter apodization for volume PET imaging using the 3-D RP algorithm

We investigated the influence of filter apodization and cutoff frequency on the image quality of volume positron emission tomography (PET) imaging using the three-dimensional reprojection (3-D RP) algorithm. An important parameter in 3-D RP and other filtered backprojection algorithms is the choice of the filter window function. In this study, the Hann, Hamming, and Butterworth low-pass window functions were investigated. For each window, a range of cutoff frequencies was considered. Projection data were acquired by scanning a uniform cylindrical phantom, a cylindrical phantom containing four small lesion phantoms having diameters of 3, 4, 5, and 6 mm and the 3-D Hoffman brain phantom. All measurements were performed using the high-resolution PET camera developed at the M.D. Anderson Cancer Center (MDAPET), University of Texas, Houston, TX. This prototype camera, which is a multiring scanner with no septa, has an intrinsic transaxial resolution of 2.8 mm. The evaluation was performed by computing the noise level in the reconstructed images of the uniform phantom and the contrast recovery of the 6-mm hot lesion in a warm background and also by visually inspecting images, especially those of the Hoffman brain phantom. For this work, we mainly studied the central slices which are less affected by the incompleteness of the 3-D data. Overall, the Butterworth window offered a better contrast-noise performance over the Hann and Hamming windows. For our high statistics data, for the Hann and Hamming apodization functions a cutoff frequency of 0.6-0.8 of the Nyquist frequency resulted in a reasonable compromise between the contrast recovery and noise level and for the Butterworth window a cutoff frequency of 0.4-0.6 of the Nyquist frequency was a reasonable choice. For the low statistics data, use of lower cutoff frequencies was more appropriate.

Event-by-Event Continuous Respiratory Motion Correction for Dynamic PET Imaging

Existing respiratory motion-correction methods are applied only to static PET imaging. We have previously developed an event-by-event respiratory motion-correction method with correlations between internal organ motion and external respiratory signals (INTEX). This method is uniquely appropriate for dynamic imaging because it corrects motion for each time point. In this study, we applied INTEX to human dynamic PET studies with various tracers and investigated the impact on kinetic parameter estimation. Methods: The use of 3 tracers—a myocardial perfusion tracer, 82Rb (n = 7); a pancreatic β-cell tracer, 18F-FP(+)DTBZ (n = 4); and a tumor hypoxia tracer, 18F-fluoromisonidazole (18F-FMISO) (n = 1)—was investigated in a study of 12 human subjects. Both rest and stress studies were performed for 82Rb. The Anzai belt system was used to record respiratory motion. Three-dimensional internal organ motion in high temporal resolution was calculated by INTEX to guide event-by-event respiratory motion correction of target organs in each dynamic frame. Time–activity curves of regions of interest drawn based on end-expiration PET images were obtained. For 82Rb studies, K1 was obtained with a 1-tissue model using a left-ventricle input function. Rest–stress myocardial blood flow (MBF) and coronary flow reserve (CFR) were determined. For 18F-FP(+)DTBZ studies, the total volume of distribution was estimated with arterial input functions using the multilinear analysis 1 method. For the 18F-FMISO study, the net uptake rate Ki was obtained with a 2-tissue irreversible model using a left-ventricle input function. All parameters were compared with the values derived without motion correction. Results: With INTEX, K1 and MBF increased by 10% ± 12% and 15% ± 19%, respectively, for 82Rb stress studies. CFR increased by 19% ± 21%. For studies with motion amplitudes greater than 8 mm (n = 3), K1, MBF, and CFR increased by 20% ± 12%, 30% ± 20%, and 34% ± 23%, respectively. For 82Rb rest studies, INTEX had minimal effect on parameter estimation. The total volume of distribution of 18F-FP(+)DTBZ and Ki of 18F-FMISO increased by 17% ± 6% and 20%, respectively. Conclusion: Respiratory motion can have a substantial impact on dynamic PET in the thorax and abdomen. The INTEX method using continuous external motion data substantially changed parameters in kinetic modeling. More accurate estimation is expected with INTEX.

A modular low dead-time coincidence system for high-resolution PET cameras

A high-resolution-oncologic-transformable PET (HOTPET) is currently under development. The detector ring of the camera consists of 12 detector modules. Because the High-Yield-Pile-Up-Event-Recovery (HYPER) front-end electronics is applied in this new generation PET camera, a low dead-time coincidence circuitry is needed to take advantage of the efficiency improvement from this new technology. The coincidence matching of events coming from different detector modules is performed by an AND-logic on the arriving edges of the module output timing-pulses. A flexible modular architecture has been adopted to facilitate the use of the coincidence circuitry for different detector module configurations as well as different electronic and mechanical implementation. The application of backplane/plug-in architecture and programmable devices (EPLD/FPGA) and DSP (Digital Signal Processor) provide enough reprogrammable flexibility and expandability, ranging from animal and breast PET to whole-body PET. There are 42 possible pair combinations of modules derived from maximum 12 detector modules in coincidence with 7 opposite modules. Both the total (true + accidental) and accidental coincidences are simultaneously collected in real time; the accidental timing shift is /spl ges/200ns relative to true events. The timing-gate window for the coincidence AND-logic can be dynamically digitally adjusted during data acquisition between 6.5-16ns to optimize signal/noise in the data. The prototype circuit showed that the timing accuracy is far better than 0.5ns and the coincidence dead-time is less than 21ns.

Scatter and crosstalk corrections for (99m)Tc/(123)I dual-radionuclide imaging using a CZT SPECT system with pinhole collimators.

PURPOSE The energy spectrum for a cadmium zinc telluride (CZT) detector has a low energy tail due to incomplete charge collection and intercrystal scattering. Due to these solid-state detector effects, scatter would be overestimated if the conventional triple-energy window (TEW) method is used for scatter and crosstalk corrections in CZT-based imaging systems. The objective of this work is to develop a scatter and crosstalk correction method for (99m)Tc/(123)I dual-radionuclide imaging for a CZT-based dedicated cardiac SPECT system with pinhole collimators (GE Discovery NM 530c/570c). METHODS A tailing model was developed to account for the low energy tail effects of the CZT detector. The parameters of the model were obtained using (99m)Tc and (123)I point source measurements. A scatter model was defined to characterize the relationship between down-scatter and self-scatter projections. The parameters for this model were obtained from Monte Carlo simulation using SIMIND. The tailing and scatter models were further incorporated into a projection count model, and the primary and self-scatter projections of each radionuclide were determined with a maximum likelihood expectation maximization (MLEM) iterative estimation approach. The extracted scatter and crosstalk projections were then incorporated into MLEM image reconstruction as an additive term in forward projection to obtain scatter- and crosstalk-corrected images. The proposed method was validated using Monte Carlo simulation, line source experiment, anthropomorphic torso phantom studies, and patient studies. The performance of the proposed method was also compared to that obtained with the conventional TEW method. RESULTS Monte Carlo simulations and line source experiment demonstrated that the TEW method overestimated scatter while their proposed method provided more accurate scatter estimation by considering the low energy tail effect. In the phantom study, improved defect contrasts were observed with both correction methods compared to no correction, especially for the images of (99m)Tc in dual-radionuclide imaging where there is heavy contamination from (123)I. In this case, the nontransmural defect contrast was improved from 0.39 to 0.47 with the TEW method and to 0.51 with their proposed method and the transmural defect contrast was improved from 0.62 to 0.74 with the TEW method and to 0.73 with their proposed method. In the patient study, the proposed method provided higher myocardium-to-blood pool contrast than that of the TEW method. Similar to the phantom experiment, the improvement was the most substantial for the images of (99m)Tc in dual-radionuclide imaging. In this case, the myocardium-to-blood pool ratio was improved from 7.0 to 38.3 with the TEW method and to 63.6 with their proposed method. Compared to the TEW method, the proposed method also provided higher count levels in the reconstructed images in both phantom and patient studies, indicating reduced overestimation of scatter. Using the proposed method, consistent reconstruction results were obtained for both single-radionuclide data with scatter correction and dual-radionuclide data with scatter and crosstalk corrections, in both phantom and human studies. CONCLUSIONS The authors demonstrate that the TEW method leads to overestimation in scatter and crosstalk for the CZT-based imaging system while the proposed scatter and crosstalk correction method can provide more accurate self-scatter and down-scatter estimations for quantitative single-radionuclide and dual-radionuclide imaging.

A Fast Accuracy Crystal Identification Method Based on Fuzzy C-Means (FCM) Clustering Algorithm for MicroPET

A high resolution detector is being developed for our small animal position emission tomography (MicroPET). The detector unit consist of 8x8 crystal blocks, coupled to four photomultiplier tubes (PMTs). Each scintillation event is mapped in a two dimensional (2-D) position through the relative ratio of the output signals of the PMTs. Crystal Look-up table (CLT) used in ThuMicroPET scanner defines the matching relation between signal position of a detected event to a corresponding detector pixel location. It has a direct impact on imaging quality and brings significant influence to the gantry overall performance. However, the currently used method involves a lot of human interaction for CLT corrections, and cannot be implemented as a general process due to its complexity. This paper introduces a fast accuracy method based on Fuzzy C-Means (FCM) Clustering Algorithm for crystal identification. In the FCM, a cluster center and a fuzzy partition matrix of individual events in the 2-D position are defined. By iteratively updating the cluster centers and the membership grades for each event, we can move the cluster center to the right location in a short time, based on minimizing objective function that represents the distance from any given events to a cluster center weighted by its membership grade. The preliminary result shows that FCM can be used effectively in CLT construction, which significantly reduces the time, and brings excellent accuracy than we expected.

A simple smart time-to-digital convertor based on vernier method for a high resolution LYSO MicroPET

We proposed a new and low cost design of time-to-digital converter (TDC) based on Vernier method using only one FPGA EPF10K30ATI144-3 with 1.3 ns timing resolution performance. Neither ECL (emitter-coupled logic) circuit nor high frequency clock was used in this design, which greatly reduced the complication and the power supply. We used two oscillators with slightly different frequencies to measure small time interval. All the time-to-digital converter functions were implemented on only one low-cost Altera FLEX II Family device. Our preliminary results showed: 1) Vernier TDC had a less than 1.3 ns timing resolution that met the demand for coincidence measurement for LYSO MicroPET. 2) No inacceptable degradation was observed in time resolution as of the number of the sliding jaw clock circulations was increased before coincidence. 3) Vernier TDC accomplished on FPGA had good stability with temperature. In a sum, we made a sufficient proof of high resolution and good stability of the proposed Vernier TDC design. Now, we are planning to achieve higher time resolution and higher stability by high-performance FPGA using this smart Vernier TDC method. And such simple design is being applied to our animal PET.

Gantry design with accurate crystal positioning for a high-resolution transformable PET camera

A positron emission tomography (PET) camera capable of transforming its geometric configuration is being developed. This high-resolution oncologic transformable PET (HOTPET) can be modified from a large detector ring of 83 cm to a small diameter ring of 54 cm. The system consists of 12 rectangular detector modules arranged in a polygon. The detector gap between modules remains constant in both configurations because each module is rotated around its own axis and displaced radially, bringing together adjacent modules. HOTPETs detectors are highly pixilated (crystal pitch 2.6 mm), requiring accurate placement of the modules relative to each other to ensure alignment of crystals within the same detector ring. We have designed a precise detector bank holder with keyways and complementary keys built onto its sides to allow interlocking with each other to form a polygon and maintain crystal coplanarity. Consequently, we were able to design the gantry supporting the modules using wider tolerances and so reduce its construction cost. The module provides support to 77 photomultiplier tubes (PMTs), the analog front-end electronics, and an automated PMT-gain control, all enclosed within a controlled environment. Potential development of light leaks was minimized with only two parting surfaces throughout the modules box, and tortuous-path air ducts inside the walls. Internal airflow allows temperature control. Simple removal of a back cover and a motherboard gives access to any part of the electronic components or a PMT with minimal disturbance to other components.