We present a new automated method for SPECT quantification and display, and its validation using computer simulations. The computer simulations provide validation for calculating myocardial defect severity. The short axial slices of the phantoms were divided into four walls (Anterior, Septal, Inferior and Lateral).Simulated myocardial defects with various degrees of severity were created in each wall from apical to basal slices. Twenty-nine computerized phantoms each with twelve SPECT short axial slices were quantified and analyzed using the Yale-CQ (Yale Circumferential Quantification) software. Two dimensional (2D) and three dimensional (3D) quantitative profiles and defect scores from the simulated short axial slices were obtained. The calculated defect score correlated highly with simulated defect severity (R = 0.99 both in 2D and 3D).Thus, the quantitative defect score generated with our SPECT quantification algorithm can be used as a reliable index for detecting severity of myocardial perfusion defect, 70% to 100% of maximum (simulating heterogeneity as observed in patients) and ( f ) abnormal regions being defined as count densities below 70% of maximum. METHODSA pixel size of 5 mm was simulated for the images. A a x 6 4 image matrix representing a field of view of 320 mm x 320 mm was used. For each computerized phantom, the center-to-epicardial radius of the short axial slices ranging from 3 pixels at the apex to 8 pixels at the base, with a myocardial thickness of 2 pixels. To simulate a normal myocardial perfusion image, the computerized phantom was made in such a way that relative count densities of lateral (Lat) anterior (Ant), septal (Sep) and inferior (Inf) walls were 100, 90, 80 and 70 % of maximum, respectively. A schematic of our simulated normal basal slice is shown in Fig. 1. To simulate myocardial perfusion defects of varying 320mm ____))

No abstract