Abstract
A new method for computer-assisted quantitative analysis of a thallium (Tl)-201 myocardial image ("corrected" circumferential profile method) was described. Since the Tl-201 myocardial image of a normal subject is not homogeneous, an attempt was made to correct for this non-homogeneity. Three groups of subjects were studied, including 10 normal volunteers (group A, mean age of 27.5 years), 14 patients with atypical chest pain and normal coronary arteriogram (group B, mean age of 56.7 years) and 16 patients with first transmural myocardial infarction (group C, mean age of 71.0 years). The myocardial images were acquired at rest at anterior, left anterior oblique (30° and 60°) and left lateral projections. With a scintigram, the left ventricle was outlined and divided into 24 radial segments by radii drawn from the center of the left ventricle. Average radioactivity per pixel in normal subjects was obtained in each segment and normalized to the highest segment (=100%). The ratio of 100% to the mean of normalized radioactivity in percent in each segment from 10 normal volunteers was calculated and designated as a correction factor for each segment. After the correction of average radioactivity per pixel in each segment using this correction factor and after relating it to the highest radioactive segment, the mean (=100%) and standard deviation (SD) were calculated and (100-4SD)% was defined as the normal lower limit. In the scintigram of groups B and C, "corrected" circumferential profiles were obtained from the regional radioactivity multiplied by the correction factor, normalized to the highest segment (100%) and compared with the lower normal limit. The ratio (%) of the area of patient's circumferential profile curve below the normal limit to the total area below the normal limit was obtained at each view and the sum of them was called the total "corrected" defect score. Infarct size analyzed by the "corrected" circumferential profile method correlated well with that analyzed by visual interpretation, and was useful in differentiating group C form group B.
