Comparative analysis of machine learning techniques for brain tumor images

Medical advancement maintaining medical information by removing several irrelevant and unwanted features. Meta-heuristics are optimization strategies that offer an excellent approach by exploring and exploiting the whole search space to solve this problem and prove that the selection of the feature can be an efficient solution. However, the enormous data size increases dimensionality and reduces classification accuracy. A comparative analysis of metaheuristic algorithms is performed to diagnose the tumor in brain magnetic resonance images (MRI). The observation shows that particle Swarm Optimization (PSO) is giving the highest accuracy of 96.42% with Random Forest (RF) classifier and also obtaining the sensitivity, specificity, precision, negative predictive value (NPV), and F-score of 92.23%,100%, 100%,93.75% and 96.77 respectively.