Segmentation of Liver Abnormality based on Label Connected Component Algorithm

Medical imaging uses a non invasive method to view the internal structure of human body. This helps the radiologists to study and visualize the region of interest for treatment planning and surgery. Computer tomography imaging gives detail view of soft tissues of abdominal organ. The objective of the proposed work to segment the liver from abdominal CT image using segmentation technique. The method uses bilateral filter for preprocessing and for finding the liver region, label connected component algorithm is proposed. The performance of the algorithm is compared with ground truth from experts. The performance measures includes Dice coefficient, misclassification rate and the volume of liver.