Hybrid PSO-ACO Optimized CNN: A Novel Approach for Pneumonia Detection

Pneumonia detection is a critical task in medical imaging that involves accurately classifying chest X-ray images to diagnose the presence of pneumonia. Deep Convolutional Neural Networks (CNNs) have demonstrated remarkable success in this domain by automatically extracting and learning relevant features from X-ray images, making them highly effective for medical diagnosis. However, optimizing CNN architectures and identifying ideal hyperparameters remains a complex and challenging process that requires extensive tuning and evaluation. This study proposes a novel approach that integrates Particle Swarm Optimization (PSO) and Ant Colony Optimization (ACO) with CNNs to enhance image classification performance. By leveraging PSO’s global exploration capabilities and ACO’s local exploitation strengths, the proposed PSO-ACO framework effectively balances the search process within the hyperparameter space. The seamless integration of these metaheuristic algorithms into CNN architecture optimization results in significant improvements in classification accuracy and a reduction in training time. Beyond enhancing CNN efficiency, this hybrid strategy contributes to improved diagnostic accuracy while mitigating time constraints in clinical settings.