Collimator Evaluation for Tl-201 Myocardial Imaging

• Published in: The Journal of nuclear medicine (1978) Vol. 19; no. 9; pp. 1067 - 1073
• Main Authors: Nishiyama, Hiroshi, Romhilt, Donald W, Williams, Craig C, Adolph, Robert J, Sodd, Vincent J, Blue, James W, Lewis, Jeannine T, Gabel, Marjorie, van der Bel-Kahn, Johanna M
• Format: Journal Article
• Language: English
• Published: United States Soc Nuclear Med 01.09.1978
• Subjects: Animals; Dogs; Evaluation Studies as Topic; Heart - diagnostic imaging; Models, Structural; Radioisotopes; Radionuclide Imaging - instrumentation; Thallium
• ISSN: 0161-5505; 1535-5667

Three collimators--high-resolutions, converging, and pinhole--were evaluated for Tl-201 myocardial imaging. Line spread function, sensitivity measurements, and phantom and animal studies were used. Features common to all the collimators were: a) better resolution at a closer distance with higher count density, and b) higher infarct detection rate in the tangenital projection than in the en face view relative to the lesion. Furthermore, an infarct in the epicardial location was better visualized than one in the endocardial location. In terms of resolution and sensitivity, the high-resolution collimator was found to be satisfactory in most clinical imagings, but for visualization of an infarct, its size by weight must be over 10--12 g. The pinhole collimator could resolve an infarct as small as 7 g, and use of the pinhole yielded a diagnostic accuracy of over 90%, compared with 75-80% for the high-resolution collimator. Although the low sensitivity of the pinhole collimator precludes its routine clinical use, the selected view would increase diagnostic accuracy. The converging collimator performed poorly in terms of lesion detectability, and its routine clinical use is not encouraged. The conclusion drawn here is valid in the system we have studied, but the variety of converging collimators must be evaluated further for their specific purposes.