Automatic left and right heart segmentation using power watershed and active contour model without edge

Abstract

Conclusions: The proposed method extracts the left and right heart accurately, demonstrating that this approach can assist the cardiologist.Purpose: In this paper, we present an automatic method to segment a whole heart and separate left and right heart regions in cardiac computed tomography angiography (CTA) efficiently.Methods: First, we smooth the images by applying filters to remove noise. Second, the volume of interest (VOI) is detected by using k-means clustering. In this step, the whole heart is coarsely extracted, and it is used for seed volumes in the next step. Third, we detect seed volumes using a geometric analysis based on anatomical information and separate the left and right heart with power watershed. Finally, we refine the left and right sides of the heart using active contour model without edge, which used region-based information for a more accurate segmentation.Results: In experimental results using twenty clinical datasets, the average segmentation error was less than 5%. The average processing time was 51.66±3.35 s. © 2014, Korean Society of Medical and Biological Engineering and Springer.