ABSTRACT
Accurate liver segmentation in a computed tomography (CT) images are essential for various medical applications. This study presents a novel liver segmentation method that combines shape prior features and a modified Chan-Vese (CV) model. The proposed algorithm extracts shape characteristics from a training set using statistical shape modeling. A comprehensive comparison of the proposed approach with existing methods is conducted based on performance parameters like maximum symmetric surface distance (MSD), relative volume difference (RVD), average symmetric surface distance (ASD), root mean square symmetric surface distance (RMSD), and volumetric overlap error (VOE). Experimental results on SLIVER and IR- CAD datasets showcase the superiority of proposed method. The algorithm demonstrates enhanced segmentation accuracy and efficiency, making it a valuable asset in medical image analysis.
