Automatic Liver Segmentation Using a Statistical Shape Model With Optimal Surface Detection

Abstract

In this letter, we present an approach for automatic liver segmentation from computed tomography (CT) scans that is based on a statistical shape model (SSM) integrated with an optimal-surface-detection strategy. The proposed method is a hybrid method that combines three steps. First, we use localization of the average liver shape model in a test CT volume via 3-D generalized Hough transform. Second, we use subspace initialization of the SSM through intensity and gradient profile. Third, we deform the shape model to adapt to liver contour through an optimal-surface-detection approach based on graph theory. The proposed method is evaluated on MICCAI 2007 liver-segmentation challenge datasets. The experiment results demonstrate availability of the proposed method.