CT Liver Segmentation using Artificial Bee Colony Optimization

Banha University - faculty of Computers and Information

The automated segmentation of liver is an essential phase in liver diagnosis in medical images. In this paper, the artificial bee colony optimization algorithm (ABC) is used as a clustering technique to segment the whole liver. ABC calculates the centroids of clusters in CT liver image and extracts a binary image for each cluster. Using some morphological operations can help to remove small and thin regions, which represents parts of flesh around liver, sharp edges of organs and small lesions inside the liver. Then the large regions in each cluster binary image are filled. Summation of the binary images results in a considerable image of segmented liver. Finally the resulting image will be enhanced using simple region Growing (RG) technique. A set of 38 images, taken in pre-contrast phase, was used to segment the liver and test the approach. Similarity index is used to validate the success of the approach. The experimental results showed that the overall accuracy offered by the proposed approach, results in 93.73% accuracy.