Search

The optimization of spatial resolution is a critical consideration in the design of small-diameter positron emission tomography (PET) scanners for animal imaging, and is often addressed with Monte Carlo simulations. As a faster and simpler solution, we have developed a new analytical model of the PET detector response function, and implemented the model for a small single-slice, multilayer PET scanner. The accuracy of the model has been assessed by comparison with both Monte Carlo simulations and experimental measurements published in the literature. Results from the analytical model agreed well with the Monte Carlo method, being noise free and two to three orders of magnitude faster. The only major discrepancy was a slight underestimation of the width of the point spread function by the analytical method as inter-crystal scatter is neglected. We observed good agreement between the predictions of the model and experimental measurements. For two large-diameter scanners additional discrepancies were seen due to photon acollinearity, which is not considered in the model. We have shown that the simple and fast analytical detector response function model can provide accurate estimates of spatial resolution for small-diameter PET scanners, and could be a useful tool for several applications, complementing or cross-validating other simulation methods.

Using our recently reported whole body PET imaging technique, we performed whole body PET studies of the skeletal system with [18F]fluoride ion in 19 patients with a range of malignant and benign skeletal conditions and in 19 normal male volunteers. The technique produces two-dimensional projection images of the entire skeletal system ("a PET bone scan"), in addition to coronal, sagittal, and axial tomographic images of the skeletal system. The tomographic images had a 13% higher lesion detection sensitivity than the projection images. Whole body PET skeletal imaging with [18F]fluoride ion is technically feasible, provides images of excellent quality, and may be coupled with more quantitatively precise kinetic PET [18F]fluoride ion studies (over limited regions of the body) when numerical estimates of skeletal [18F]fluoride ion uptake are desired. The method is potentially useful in clinical applications where the high resolution and numerical precision of PET are of particular value (e.g., in accurately defining the anatomic location and extent of lesions and in assessing changes in bone metabolism on serial studies).

We are developing an accelerated Monte Carlo simulation of positron emission tomography (PET) that can be used for scatter correction of three-dimensional (3-D) PET data. Our Monte Carlo technique accurately accounts for single, multiple, and dual Compton scatter events, attenuation through the patient bed, and activity from outside the field of view. We have incorporated innovative sampling techniques that are compatible with our simulation approach, increasing computational efficiency by a factor of seven while improving accuracy through more sophisticated stratification and by incorporating the true energy response of the scanner. The required execution time to acquire 10 million scatter coincidence events for a 3-D thorax PET scan is only 4 min on a 300-MHz Sun dual-processor workstation. We demonstrate that for a low-noise thorax phantom study, image data corrected using the Monte Carlo 3-D PET scatter correction demonstrate no statistically significant deviation from the true activity concentration provided corresponding input data are accurate. The speed and accuracy of our simulation makes it an efficient research tool for studying scatter effects in PET and a practical scatter correction for 3-D PET clinical imaging. Index Terms--Compton scatter, Monte Carlo simulation, photon tracking, positron emission tomography (PET), sampling techniques, scatter correction, three-dimensional (3-D) PET, variance reduction.

A key parameter in the practical application of filtered back-projection (FBP), the standard clinical image reconstruction algorithm for positron emission tomography (PET), is the choice of a low-pass filter window function and its cut-off frequency. However, the filter windows and cut-off frequencies for clinical reconstruction are usually chosen empirically, based on a small sample of images and filters. By considering the features of the signal and noise spectra in a sinogram, the desired image resolution, and the signal-to-noise ratio (SNR) of the filtered sinogram, a methodology for informed selection of a filter function and cut-off frequency for FBP was investigated. Simulations of sinogram data similar to whole body or cardiac studies provided information on the signal and noise frequency-domain spectra of noisy projection data. The improvements in SNR with different filter windows and cut-off frequencies were evaluated and compared. The projection spectrum SNR measure did not prove to be an accurate indicator of subjective image quality or lesion detectability with variations in Poisson noise and image resolution.

The performance of a phoswich array detector module for possible use in a combined single and coincidence photon imaging system has been evaluated. The detector consist of a linear array of discrete 4x4x15 [mm.sup.3] YSO elements coupled to a combined detector array/light guide of LSO, 10 mm thick. Since the scintillation light decay time is different in YSO and LSO (70 and 40 ns, respectively), events originating from the two detector materials can be separated by pulse shape discrimination. The front layer of YSO could then be used for detection of low energy, single photon events and the LSO layer for coincidence detection of annihilation radiation. The light collection of the PMTs coupled to the detector was found to be adequate to accurately identify each detector element in the array using the same positioning logic used in conventional BGO block detectors. The average energy resolution of the YSO elements at 140 keV for the block detector was found to be 14.5% FWHM, ranging from 13.8 to 15.4%. Spatial resolution of the detector block in single photon mode, using a high resolution collimator (geometric resolution 6.5 mm at 10 cm) was measured by scanning a 99mTc line source. The resolution at 5 and 10 cm from the collimator face was found to be 5.9 and 8.5 mm FWHM, respectively.

In order to achieve high sensitivity and maintain good uniform spatial resolution over the field of view in high resolution PET systems, adequate depth of interaction information must be extracted from the crystal. A phoswich detector can supply one solution to the depth of interaction problem. In this approach, two or more scintillators exhibiting different light decay times are positioned on top of each other and separated by pulse shape discrimination. Initially, our experiments focused on separating different types of scintillators such as LSO and GSO or LSO and YSO. These combinations were all well separated as expected. During the investigation, a shift in the time distribution of different samples of LSO was noticed. Further investigation showed two groups of LSO. The shift in the zero cross time was more than twice the FWHM of the time distribution. A single photon experiment revealed that the decay time of the 'fast' crystal was 33.4 nanoseconds while the decay of the 'slow' crystal was 42.2 nanoseconds. A spectral plot revealed that the spectral output of the 'slow' crystal was skewed to the longer wavelengths as compared to the 'fast' crystal. Further investigation on other crystal samples revealed decay times between the two extremes, suggesting a continuum in the light decay.

The performance of a YSO/LSO phoswich detector module is evaluated for potential use as in a combined PET/SPECT imaging system. The high light output of LSO ([approximately]5-6 times BGO) would allow the construction of a detector block that would have similar intrinsic resolution characteristics at 140 keV as a conventional high resolution BGO block detector at 511 keV ([approximately]4 mm FWHM). However, the intrinsic radioactivity of LSO prevents the use of this scintillator in single photon counting mode. YSO is a scintillator with higher light output than LSO but worse absorption characteristics than LSO at higher energies. In a phoswich detector block the two detector materials are combined, where YSO is placed in a front layer and is used for single photon imaging and LSO in a second layer is used for PET imaging. Since the scintillation decay times in YSO and LSO are significantly different (70 and 40 ns, respectively), events in the two detector materials can be separated by pulse shape discrimination. In measurements of the system resolution using a high resolution collimator, the spatial resolution at 10 cm collimator distance was 8.5 mm for the phoswich detector compared to 7.6 mm for a conventional scintillation camera. However, in SPECT simulations of the two detector systems of a 20 cm diameter hot spot phantom imaged at different collimator distances using a high resolution collimator, no appreciable difference in resolution was seen in the reconstructed images between the two camera systems, including the ideal situation of zero distance between collimator and phantom.

New positron emission tomography (PET) scanners have higher axial and in-plane spatial resolutions but at the expense of reduced per plane sensitivity, which prevents the higher resolution from being fully realized. Normally, Gaussian-weighted interplane axial smoothing is used to reduce noise. In this study, we developed a new algorithm that first elastically maps adjacent planes, and then the mapped images are smoothed axially to reduce the image noise level. Compared to those obtained by the conventional axial-directional smoothing method, the images by the new method have improved signal-to-noise ratio. To quantify the signal-to-noise improvement, both simulated and real cardiac PET images were studied. Various Hanning reconstruction filters with cutoff frequency = 0.5, 0.7, 1.0 x Nyquist frequency and Ramp filter were tested on simulated images. Effective in-plane resolution was measured by the effective global Gaussian resolution (EGGR) and noise reduction was evaluated by the cross-correlation coefficient. Results showed that the new method was robust to various noise levels and indicated larger noise reduction or better image feature preservation (i.e., smaller EGGR) than by the conventional method.

A hybrid attenuation correction technique (ACT) is under development for 18F-FDG total body positron emission tomography (PET). With a short transmission scan of the thorax, any time within a few days of the imaging session, this technique can correct for attenuation in the entire body. Segmentation, registration, and active contour finding techniques are applied to both emission and short transmission images to locate and map the major attenuating structures in the body. This technique eliminates the need for the patient to remain still from the start of the transmission scan to the end of the emission scan without the added noise of simultaneous or post emission transmission scan measurements. The results of volunteer studies are comparable to standard measured ACT, both visually and quantitatively. The efficient use of scanner time and improved patient comfort make this technique particularly suitable for clinical imaging.

Measured attenuation correction in PET is routinely performed using transmission scans. Acquisition time and noise considerations necessitate low-pass filtering of the transmission data before generating the attenuation correction matrix. This smoothing operation reduces noise propagation from transmission into emission data but also introduces image artifacts that are mostly pronounced around areas of strongly varying attenuation coefficients. The source of these artifacts, which lies in the mismatch of the spatial resolutions of emission and transmission data, was investigated in this study. The effect of different transmission and emission sinogram smoothing protocols on the emission images was also investigated. A method is proposed that addresses the problem, in conjunction with the filtering step during reconstruction. Instead of the standard in-plane low-pass filtering of the emission data during reconstruction, emission and transmission sinograms ran be volumetrically filtered to the desired image resolution prior to attenuation correction while reconstruction is performed with no smoothing (Ramp filter). This operation reduces or eliminates resolution mismatch and consequent image artifacts, especially in cardiac studies. The proposed method improves the accuracy of the activity distribution in emission images with minimal computational and signal-to-noise ratio (SNR) cost.

A Monte Carlo simulation has been developed to simulate and correct for the effect of Compton scatter in 3-D acquired PET brain scans. The method utilizes the 3-D reconstructed image volume as the source intensity distribution for a photon-tracking Monte Carlo simulation. It is assumed that the number of events in each pixel of the image represents the isotope concentration at that location in the brain. The history of each annihilation photon's interactions in the scattering medium is followed, and the sinograms for the scattered and unscattered photon pairs are generated in a simulated 3-D PET acquisition. The calculated scatter contribution is used to correct the original data set. The method is general and can be applied to any scanner configuration or geometry. In its current form the simulation requires 25 hours on a single Sparc10 CPU when every pixel in a 15-plane, 128x128 pixel image volume is sampled, and less than 2 hours when 16 pixels (4x4) are grouped as a single pixel. Results of the correction applied to 3-D human and phantom studies are presented.

Whole body PET images suffer from relatively high noise levels due to inherently poor counting statistics in the emission data. Optimization of acquisition parameters is essential, to minimize any additional noise contamination. It has previously been shown that by using a continuous or redundant axial sampling scheme, a reduction in statistical noise and improvement in image quality are possible. In this work, the continuous axial sampling technique is further characterized and compared to the conventional step-and-shoot technique. The main source of additional noise contamination with conventional sampling is the detector sensitivity normalization procedure which is applied to the emission data. With continuous axial sampling and a single normalization matrix for all planes, the statistical noise in the normalization factors is reduced by factor close to the number of planes in the scanner. The continuous sampling technique is shown to be less sensitive to small patient movements (

Transmission scans for attenuation correction of whole body PET studies are not acquired routinely, since noise considerations impose acquisition time constraints that make conventional scanning techniques infeasible. The aim of this work is to optimize data acquisition and processing parameters and improve the SNR of whole body transmission scans, in order to achieve within a reasonable time frame, an attenuation correction of the same quality as in cardiac PET. Methods to improve the scanner sensitivity by using additional coincidence planes, as well as smoothing methods for the randoms and the transmission data were investigated. Phantom and patient studies show that transmission scans are feasible for whole body PET studies, with equivalent noise introduced in the chest area as in a typical cardiac PET attenuation correction, at fractions of the time required by conventional transmission scan protocols.

Computationally intensive techniques that offer potential clinical use have arisen in nuclear medicine. Examples include iterative reconstruction, 3D PET data acquisition and reconstruction, and 3D image volume manipulation including image registration. One obstacle in achieving clinical acceptance of these techniques is the computational time required. This study focuses on methods to reduce the computation time for 3D PET reconstruction through the use of fast computer hardware, vector and parallel programming techniques, and algorithm optimization. The strengths and weaknesses of i860 microprocessor based workstation accelerator boards are investigated in implementations of 3D PET reconstruction.

In the present work we will demonstrate that the nuclear dynamics have a strong influence on the delocalization of an exciton in a dimer, even if they do not effect the excitonic interaction. It will be shown that the internal nuclear conformation of the molecules forming the dimer depends critically on the delocalization of the exciton state in the dimer and vice versa. The resulting closed loop enforces a localization of the lower excitonic state, but, contrary to the commonly accepted view, a delocalization of the upper one. Qualitatively different time-evolution of the delocalization length for the lower and upper excitonic state will be shown. Besides, it will turn out that the nuclear motions inhibit a complete delocalization of the excitonic state in any case. To accomplish nuclear and exciton dynamics, the nonadiabatic coupling between the two excitonic states will be deduced. This causes a relaxation from the upper to the lower excitonic state, which limits the maximum reachable exciton delocalization. [ABSTRACT FROM AUTHOR]

Several measures of exciton delocalization and their variation with temperature, inhomogeneous broadening and time are compared and the time dependent localization of the B850 exciton as reflected in experimental transient absorption and superradiance measurements is stimulated. It is observed that the exciton wave packet prepared by a 40 fs pulse is initially delocalized over a substantial portion of the ring and contracts to a significantly smaller size on the time scale of a few hundreds of femtoseconds.

We have been developing Monte Carlo Techniques for calculating primary and scatter photon distributions in PET. Our first goal has been to accelerate the Monte Carlo Code for fast PET simulation. Our second goal has been to use the simulation to analyze scatter effects in PET and explore the potential for use in scatter correction of clinical 3D PET studies. We have reduced the execution time to about 30 minutes or ~1 million coincidences per minute on a dual 300MHz processor UltraSparcII workstation. The short execution time makes it feasible to use this technique for 3D PET scatter correction in the clinic. Fast simulation also allows us rapid feedback for the close examination of the accuracy of the method. We present techniques used to improve computational efficiency of Monte Carlo PET simulations. We use the simulation to analyze how scatter from within the body, outside the FOV, and from scanner shielding as well as the chosen energy threshold affect 3D PET sinograms.