Adaptive Filter-Based Grey Wolf Optimization Algorithm for Enhanced Medical Diagnosis

Abstract

Medical diagnostic systems often struggle with noise and data inconsistencies in physiological signals. This paper presents an Adaptive Filter-Based Grey Wolf Optimization (AF-GWO) algorithm that combines adaptive filtering for noise reduction and GWO for optimizing machine learning classifiers. The method was evaluated on biomedical datasets, including ECG and heart disease data, and compared with conventional techniques like GA and PSO. Results show that AF-GWO significantly improves classification accuracy, signal-to-noise ratio (SNR), and convergence speed. This hybrid approach provides an effective solution for real-time medical diagnostics, enhancing feature optimization and signal clarity. The framework demonstrates strong potential for AI-driven medical applications. Future work will explore its application in multimodal medical datasets.