Review of Set Theoretic Approaches to Magnetic Resonance Brain Image Segmentation

Abstract

Image segmentation is a vital step in image processing and has attracted many researchers towards its potential applications like object recognition, pattern recognition, computer vision and artificial intelligence. The medical image segmentation is challenging and complex due to the artifacts or noise present in the image. Many segmentation techniques are proposed in the literature, using mathematical set theoretic approaches like k-means, fuzzy c-means, rough c-means, intutionistic fuzzy c-means, and hybrid clustering algorithms to extract the brain tissues from magnetic resonance brain images. The set theoretic approaches easily model the clustering techniques to extract the brain tissues without the operator intervention. The fuzzy sets, rough sets, and intutionistic sets proved to handle the vagueness, noise and uncertainty present in the medical images, whereas the soft sets can easily parameterize the rough sets for better performance. In this paper, a summary of all brain image segmentation methods based on set theoretic approach is described in detail and also categorized accordingly. Set theoretic based-image segmentation provides a better framework for medical image segmentation.