A novel segmentation algorithm and 3d rendering of liver lesion and major blood vessels in abdomen CTA images

Abstract

The American Cancer society for hepatic cancer and intrahepatic bile duct cancer estimated that throughout the world more than 700000 people are diagnosed with liver cancer every year More than 600000 deaths occur each year because of liver cancer making it a leading cause for cancer deaths worldwide Computed tomography CT images are widely used by physicians for accurate diagnosis and detection of liver lesions due to its high spatial resolution fast imaging speed wide availability and low cost when compared to MRI This work presents the development and implementation of algorithms for 3D liver lesion and intra hepatic vessel segmentation from tri phase CTA images The proposed algorithm requires minimal human intervention without compromising the accuracy of the segmented results The geometric properties and intensity distributions of liver vary to a large extent in interpatient and intra-patient groups The presence of noise and ambiguous boundaries makes the segmentation of the liver from Computed Tomography Angiogram CTA images to be a challenging task Moreover the shape of liver varies in every individual and the presence of tumors or cysts will make the tissue to have various morphologies A method for liver segmentation has been implemented by integrating the Spatial Fuzzy c Means Clustering process SFCM and localized region based active contour segmentation algorithm The standard FCM algorithm wrongly classifies the noisy pixel in noisy images because of its abnormal feature data Hence the spatial fuzzy c means algorithm is used which corrects the misclassified pixels in noisy images newline