Comparative Analysis of Particle Swarm Optimization and Whale Optimization Algorithm for Medical Image Enhancement

Abstract

Medical images play a critical role in diagnosing and treating various diseases, making their quality a crucial factor in clinical decision-making. However, raw medical images often suffer from issues such as low contrast, noise, blurring, and poor illumination, which can erase important anatomical details and reduce diagnostic accuracy. Histogram equalization and unsharp masking are the classic methods of enhancing images, which can be effective, although they frequently do not effectively trade-off between contrast enhancement and noise reduction. Therefore, advanced and intelligent image enhancement methods are crucial for ensuring more precise visualization of critical features, such as tumours, blood vessels, and tissue structures. Metaheuristic optimization methods (MOAs) are frequently applied for image optimization task, especially, in the circumstance of medical image enhancement (MIE). Nevertheless, as the rate of newer MOAs proposed in the literature continues to escalate, a question arises and whether there are any meaningful differences between these various MOAs and specifically in terms of MIE. This paper will compare two metaheuristic algorithms, particle swarm optimization (PSO) and whale optimization algorithm (WOA) in the process of medical image enhancement. In the study, we utilize a practical evaluation function and transformation function on both MOAs. Medical images were then taken from the Medpix dataset where the representative samples were taken across the various parts of the body to carry out MIE evaluation. Results show that WOA algorithm performed slightly better than PSO in the selected performance metrics, for example for number of edges WOA produced 4724 while PSO produced 4623. Findings further show that both metaheuristic algorithms enhance medical images, which will improve medical diagnosis and decision-making, thereby saving more lives.