A NOVEL FUZZY ADAPTIVE Q-LEARNING MODEL FOR AUTOMATED COVID-19 DIAGNOSIS USING CHEST X-RAY IMAGES

Abstract

The rapid spread of infectious diseases such as COVID-19 has highlighted the importance of intelligent medical diagnosis systems capable of providing accurate and timely predictions. This research presents a Fuzzy Adaptive Q-Learning Model for intelligent decision optimization in COVID-19 prediction using chest X-ray images. The proposed system integrates fuzzy logic with adaptive Q-learning techniques to enhance diagnostic accuracy and support efficient clinical decision-making under uncertain medical conditions. Fuzzy logic is employed to handle ambiguous and imprecise medical image features by transforming them into meaningful linguistic representations, while the adaptive Q-learning mechanism continuously learns optimal classification strategies through reward-based learning. The proposed model dynamically adjusts learning parameters to improve convergence speed, prediction accuracy, and system adaptability. Feature extraction from chest X-ray images is performed to identify critical infection patterns associated with COVID-19, and the hybrid learning framework optimizes the decision process for reliable disease prediction. Experimental analysis demonstrates that the proposed approach achieves improved performance in terms of classification accuracy, sensitivity, and decision stability when compared with traditional machine learning techniques. The developed model can support healthcare professionals in rapid screening, early diagnosis, and intelligent medical decision-making, particularly in large-scale pandemic situations. Overall, the integration of fuzzy reasoning and adaptive reinforcement learning provides an efficient and scalable framework for intelligent COVID-19 prediction using medical imaging data.