Improvement of Classification and Clustering of Images using Evolutionary Techniques

Abstract

Segmentation of medical images is significant as it aids in mining the region of interest, such that the body part under analysis is extracted. Medical image segmentation helps in treatment of diseases, in surgeries and also aids in medical diagnosis. Various performance factors like Volumetric Overlap Error, Relative Volume Difference, Average Symmetric Surface Distance, Root Mean Square Symmetric Surface Distance, Maximum Symmetric Surface Distance. were evaluated which shows that outlier detection technique provides better results as compared to the implementation done without using this technique.