The long-term objective of this proposal is to develop a collimating insert for positron emission tomography (PET) scanners for improving spatial resolution and quantification. This collimating insert may be useful at the scale of small-animal imaging and/or clinical imaging, so it will be studied at both scales. Collimation will reduce the efficiency of PET since the collimator will absorb some of the emitted 511-keV gammas. However, the detected gammas will have better spatial resolution since the collimation will be designed to reduce the widths of the lines of response while also decreasing the scatter fraction. In particular, a resolution improvement of almost a factor of two will be sought by covering half of each crystal in the transaxial direction at any time. Factors such as penetration of the collimator will prevent the full factor of two gains in resolution. Since two crystals are needed to form a coincidence and each crystal has two different exposures, there will be a factor of four improvements in sampling, achieved by repositioning the collimator or patient during the scan. The collimator design will allow all of the new lines of response to be measured for a fraction of the scanner's field of view. That fraction depends on the acceptance angle of the collimator, which also determines the amount of penetration. It is likely the fraction will encompass a mouse or rat for the small-animal scanner and about a 20-cm diameter central region for the whole-body scanner. It is also likely that the collimation may be most useful when some of the sensitivity loss can be overcome by a longer scan time for the bed positions that use collimation or when the structure size of interest is smaller than the scanner's resolution capabilities. Applications where the resolution enhancement would be beneficial and that are often a single bed position include brain, breast, and prostate imaging. Radiotherapy planning may also be aided by improved resolution, for example, if hypoxia agents such as EF5 can be imaged to determine if the tumor's core is hypoxic, which may affect the treatment plan.
The specific aims of this proposal include (1) developing accurate models of sensitivity and resolution with penetration;(2) designing, building, and integrating an experimental prototype collimator for the small-animal scanner by developing the appropriate hardware for mounting and positioning and the appropriate software for synchronizing with the scanner, which is needed in order to acquire the 4-fold increase in sampled lines;(3) developing iterative reconstruction that accurately models the sensitivity and resolution of the collimation;and (4) evaluating the prototype collimator experimentally on the small-animal scanner and with detailed simulations for the whole-body scanner. Collimation, if successful, could be used as an upgrade to existing scanners or be directly integrated in future designs. It is also possible that collimation could change the design of future scanners since larger crystals, which have better energy and timing resolution and are less likely to result in inter-crystal scatter, could be used in fabrication at reduced cost. Future efforts could also involve collimation to improve axial resolution or for application-specific devices.

Public Health Relevance

This project develops a new collimation technique for improving spatial resolution in PET at the cost of reduced collection efficiency. Advantages are likely for both small-animal and clinical imaging. It is likely that collimation may be clinically most beneficial for imaging the breast, prostate, and brain, and for radiotherapy planning, since improved spatial resolution is expected to improve the detail and clarity and because the region of interest is within a single bed position to allow for recovering count statistics with a longer scan.

National Institute of Health (NIH)
National Institute of Biomedical Imaging and Bioengineering (NIBIB)
Exploratory/Developmental Grants (R21)
Project #
Application #
Study Section
Special Emphasis Panel (ZRG1-SBIB-J (80))
Program Officer
Sastre, Antonio
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
University of Pennsylvania
Schools of Medicine
United States
Zip Code
Li, Yusheng; Matej, Samuel; Karp, Joel S et al. (2017) Model-Based Normalization of a Fractional-Crystal Collimator for Small-Animal PET Imaging. IEEE Trans Radiat Plasma Med Sci 1:262-267
Li, Yusheng; Matej, Samuel; Karp, Joel S et al. (2015) LOR-interleaving image reconstruction for PET imaging with fractional-crystal collimation. Phys Med Biol 60:647-70
Li, Yusheng; Matej, Samuel; Metzler, Scott D (2014) Image reconstructions from super-sampled data sets with resolution modeling in PET imaging. Med Phys 41:121912
Metzler, Scott D; Matej, Samuel; Karp, Joel S (2013) Resolution Enhancement in PET Reconstruction Using Collimation. IEEE Trans Nucl Sci 60:65-75