[Experimental considerations of myocardial single photon emission CT image with a cardiac phantom]

We studied the quantity of thallium-201 myocardial single photon emission CT (SPECT) image with a cardiac phantom. The myocardial SPECT image is influenced by several causes, i.e., the absorption of gamma rays, collimator aperture, statistical noise and limited angle of rotation in data acquisition. The attenuation correction under the assumption of uniform distribution is not sufficient, so we have to consider the actual distribution of absorption. The effects of the nonuniform attenuation are most prominently appeared when we evaluate the quantity by the integral method. The aperture characteristics of a collimator are decided by the distance between the rotational center of the gamma camera and the collimator surface. Simultaneously scattered gamma rays affect the reconstructed image in the low frequency component, so the effective attenuation coefficient varies gradually. The statistical noise intermixed in the projection data generate artifacts like as lump-shaped pattern. The noise are clearly appeared in the spatial frequency upper than 0.25 cycle, so the low-pass filter are required with that of cut-off frequency. Moreover to enhance the details of the radionuclide distribution of cardiac muscle, the Wiener filter should be applied. The scan area also affects the appearance of the artifact in the myocardial SPECT image. The 180-degree scan usually generates false defects just nearby the center of reconstructed cross section. And the positions are changed with the conditions of gamma ray absorber and its relative position to the myocardial area. The 360-degree scan should be used from the standpoint of reconstructing more quantitative image.