Authors
Technical College of Management, Mosul, Northern Technical University, Mosul, 41001, Iraq
[email protected]
Abstract
The aim of the study was to create a tool for the segmentation of images, which was shown to be effective in the processing of medical images for tumor detection. The major challenge that arises in the segmentation of medical images, especially radiographic images, lies in the poor contrast and noise that are present, which leads to inaccurate results for such images. To overcome these challenges, the research incorporates a combination of genetic algorithms and ant colony optimization, especially through the concept of spatial routing for precise MRI tumor detection. In this case, the genetic algorithm helps in the determination of the location of the tumor through the concept of spatial routing, where the pheromones are used to direct the ants to the micro-edges through local sifting.
The results of the experiment showed that the tool was effective, with a classification accuracy of 95.8% and a Dice coefficient of 0.93.