MULTI-CONTEXTUAL FUSION OF PHYSIOLOGICAL SIGNALS FOR ROBUST PAIN RECOGNITION SYSTEMS

Abstract

Accurate pain recognition is critical for effective patient care and management, yet it remains a challenging task due to the subjective nature of pain. This study proposes a novel approach that leverages multi-contextual fusion of physiological signals— including heart rate, skin conductance, and electromyography—to enhance the robustness of automated pain recognition systems. By integrating multi-level context information such as environmental, temporal, and individual variability, the model improves discrimination between pain and non-pain states. Experimental evaluation on benchmark datasets demonstrates significant improvements in accuracy and reliability compared to traditional single-context models. The results highlight the potential of context-aware physiological signal fusion in developing practical and sensitive pain monitoring tools, particularly beneficial in clinical and remote healthcare settings.