Automatic quantification of epicardial fat volume on non-enhanced cardiac CT scans using a multi-atlas segmentation approach

PURPOSE: There is increasing evidence that epicardial fat (i.e., adipose tissue contained within the pericardium) plays an important role in the development of cardiovascular disease. Obtaining the epicardial fat volume from routinely performed non-enhanced cardiac CT scans is therefore of clinical interest. The purpose of this work is to investigate the feasibility of automatic pericardium segmentation and subsequent quantification of epicardial fat on non-enhanced cardiac CT scans. METHODS: Imaging data of 98 randomly selected subjects belonging to a larger cohort of subjects who underwent a cardiac CT scan at our medical center were retrieved. The data were acquired on two different scanners. Automatic multi-atlas based method for segmenting the pericardium and calculating the epicardial fat volume has been developed. The performance of the method was assessed by (1) comparing the automatically segmented pericardium to a manually annotated reference standard, (2) comparing the automatically obtained epicardial fat volumes to those obtained manually, and (3) comparing the accuracy of the automatic results to the inter-observer variability. RESULTS: Automatic segmentation of the pericardium was achieved with a Dice similarity index of 89.1 ± 2.6% with respect to Observer 1 and 89.2 ± 1.9% with respect to Observer 2. The correlation between the automatic method and the manual observers with respect to the epicardial fat volume computed as the Pearson's correlation coefficient (R) was 0.91 (P < 0.001) for both observers. The inter-observer study resulted in a Dice similarity index of 89.0 ± 2.4% for segmenting the pericardium and a Pearson's correlation coefficient of 0.92 (P<0.001) for computation of the epicardial fat volume. CONCLUSIONS: The authors developed a fully automatic method that is capable of segmenting the pericardium and quantifying epicardial fat on non-enhanced cardiac CT scans. The authors demonstrated the feasibility of using this method to replace manual annotations by showing that the automatic method performs as good as manual annotation on a large dataset.