Recent evidence indicates that pericardial fat may be a significant cardiovascular risk factor. Although pericardial fat is routinely imaged during computed tomography (CT) for coronary calcium scoring, it is currently ignored in the analysis of CT images. The primary reason for this is the absence of a tool capable of automatic quantification of pericardial fat. Recent studies on pericardial fat imaging were limited to manually outlined regions-of-interest and preset fat attenuation thresholds, which are subject to inter-observer and inter-scan variability. In this paper, we present a method for automatic pericardial fat burden quantification and classification. We evaluate the performance of our method using data from 23 subjects with very encouraging results.