Optimizing Brain Tumor Classification: A Comparative Analysis of Nature Inspired Algorithms with GLCM Features

Abstract

Advancements in medical imaging have led to an increasing demand for accurate and efficient methods of brain tumor classification. This study delves into the realm of nature-inspired optimization algorithms, with a focus on their application in the field of medical image analysis. We examine the performance of three distinct algorithms: Firefly, Cat Swarm Optimization (CSO), and Artificial Fish Swarm Optimization Algorithm (AFSA), in the context of brain tumor classification. Among these, CSO emerges as the star performer, achieving an impressive accuracy rate of 96.36%. The study employs Gray-Level Co-occurrence Matrix (GLCM) features, a widely recognized set of texture features for medical image analysis. Through a rigorous comparative analysis, we explore the capabilities of these algorithms in accurately classifying brain tumors, shedding light on their potential to enhance diagnostic precision.