Liver segmentation in computed tomography abdomen images based on particle swarm optimization and morphology

Abstract

Segmenting liver from abdominal images is a thought-provoking task. A method for segmenting liver region from CT abdomen is proposed in this paper. Particle Swarm Optimization (PSO) method is employed for segmenting multiple regions in the abdomen image. Morphological operation such as Erosion and Dilation is used for segmenting exact portion of liver. Largest connected component and filling holes operation are applied as supporting techniques for image corrections. Experiment on our proposed segmentation approach is carried out and the results are discussed. The quantitative validation was performed with Dice similarity co-efficient metric.