The goal of this project is to improve markedly the physical signal-to-noise ratio (SNR) for the detection of small (about 7- 10mm) structures imaged with single-photon emission computed tomography (SPECT). The work will find importance primarily in studies of brain blood flow and receptor binding, as well as myocardial perfusion. For SPECT procedures such as these, the use of radiopharmaceuticals labeled with isotopes that are not considered ideal for SPECT (such as 1-123 and T1-201) renders a thorough physical design particularly important at this time. However, even when tracers labeled with a radionuclide of more suitable energy (e.g., Tc-99m) ultimately become available for routine clinical brain and heart studies, the expected improvement in the physical SNR resulting from this project will be 30%-90% for the detection of small lesions. The physical mechanisms which degrade contrast when imaging with today's state-of-the-art SPECT scanners include poor spatial resolution, radiation scattered in the patient, and collimator penetration an scatter (particularly for high energy isotopes). Several considerations indicate that all of these effects may be simultaneously minimized while increasing the physical SNR by altering judiciously the physical measurement process. The experimental approach will involve: (a) detailed computer simulation of the entire photon transport in SPECT so as to understand theoretically the geometric design considerations as a function of energy, and to predict optimal designs; (b) construction and testing of a novel, high pressure gas scintillation proportional camera (GSPC) which will improve both the intrinsic detector spatial resolution (to 2mm) and the energy resolution (to less than 2.5% at 140 keV); (c) fabrication of a collimator optimized for SPECT with the new GSPC; and (d) testing of all systems with physical and anthropomorphic phantoms.

Agency
National Institute of Health (NIH)
Institute
National Cancer Institute (NCI)
Type
Research Project (R01)
Project #
5R01CA046958-03
Application #
3190474
Study Section
Diagnostic Radiology Study Section (RNM)
Project Start
1987-05-15
Project End
1990-10-30
Budget Start
1989-05-01
Budget End
1990-10-30
Support Year
3
Fiscal Year
1989
Total Cost
Indirect Cost
Name
Worcester Polytechnic Institute
Department
Type
Biomed Engr/Col Engr/Engr Sta
DUNS #
City
Worcester
State
MA
Country
United States
Zip Code
01609
Moore, S C; deVries, D J; Nandram, B et al. (1995) Collimator optimization for lesion detection incorporating prior information about lesion size. Med Phys 22:703-13