Automated quantification of myocardial ischemia and wall motion defects by use of cardiac SPECT polar mapping and 4-dimensional surface rendering.

SPECT of cardiac perfusion and blood pools provides ungated 3-dimensional and gated 4-dimensional (4D) datasets of the ventricular myocardium. Modern reconstruction and review software is used to reorient the transverse thoracic slices into cardiac short-axis slices. Several validated algorithms are used to analyze these data. These programs segment out the left ventricle, determine the apical and basal limits, and then contour the endo- and epicardial surfaces. From these, 4D images of cardiac function that enable a dynamic review of wall motion are obtained. Global function in gated studies can be quantified by automatic computation of stroke volume and ejection fraction. Remapping of myocardial perfusion and wall motion into polar maps enables standardized quantification of the extent and severity of heart disease after comparison with databases of healthy hearts (normal databases). Several validated software packages make processing of these SPECT datasets comparatively easy and operator independent. The objectives of this review article are to describe the steps in the processing of a cardiac SPECT dataset for viewing and quantification, to explain the underlying algorithms used for automated processing, to compare the features of various software packages, to demonstrate how to read polar maps, and to identify and correct artifacts resulting from errors in automated processing.