Volume 20, No. 1, 2023
Automatic Segmentation Of Medical Images And Medical Diagnoses Using Genetic Algorithm
Wasan Abdallah Alawsi, Zainab Fahad Alnaseri, Wisam Ch.Alisawi, Panov Alexander Vladimirovich
Abstract
Segmentation in magnetic resonance imaging (MRI) is a challenging problem. This research provides a new method for determining the proper number of segments and automatic segmentation of human normal and pathological MR brain pictures. The goal of automatic segment diagnosis is to determine the number of segmented image regions in an image based on its entropy. Accurate segment diagnosis also increases segmentation precision. We used the evolutionary algorithm with the fuzzy c-means (FCM) approach to tackle the problem of automatically estimating the number of image segments and the center of segments, which takes numerous algorithm tests to achieve high accuracy. In order to perform picture segmentation more correctly, it has been attempted to modify the FCM approach as a fitness function for combination with genetic algorithms. In contrast to existing methods, our experiment demonstrates that the suggested strategy significantly improves the accuracy of image segmentation.
Pages: 50-62
Keywords: Genetic A , MR , FCM , LSGA , segmentation , noisy pixel .