Customizing the Bull's-Eye to Improve the Clinician's Diagnostic Intuition

In 2002 the Cardiac Imaging Committee of the Council on Clinical Cardiology of the American Heart Association defined a standard for segmentation, nomenclature and display of the tomographic images of the heart. The "bull's-eye" plot defines how to divide the left ventricular (LV) myocardium in 17 segments, giving a name to each segment and assigning individual segments to the main coronary perfusion territories. Such display has been widely accepted by all imaging modalities, as evidenced by the huge number of scientific quotations. However, the graphical display of a threedimensional solid on a two-dimensional plane necessarily implies a distortion or a partial representation of the object. In the case of cardiac imaging such a distortion could affect the representation of the extent of a variety of pathological processes, such as myocardial perfusion deficits, wall motion abnormalities, left ventricular hypertrophy, or scars. With this in mind, we aimed to evaluate how the classical display in polar coordinates affects the visual assessment of myocardial infarct extension, and if this graphic display might be modified to provide a more likely information. Cardiac magnetic resonance was performed on 38 patients with mycardial infarction. The global and regional LV volumes, as well as the myocardial infarctum (MI) size, were measured using a validated semiautomatic method: boundaries were automatically traced and manually corrected when needed. All sizes were expressed as percent of LV mass, to be linked to the areas of the bull's-eye regions. The infarct size was assessed by the scores. To improve the clinician's perception of the heart, we propose some alternatives which take into account a more likely view of the percent distribution of the LV volumes on a twodimensional representation, while keeping the conventional 17 segments organization, although providing a customized bull's eye closer to the real LV of each patient.