Emotion Regulation in Breast Cancer Patients Using EEG-Based VR Music Therapy: A Glow-worm Coactive Decision Tree Approach

Abstract

Virtual reality (VR) technology is currently being used in emotion management and musical environment modeling to improve mental and emotional wellness through psychological advantages and a flexible musical environment. The purpose of the study is to utilize the Glow Worm Coactive Decision Tree (GW+DT) classifier to develop a technique for controlling feelings and creating authentic musical situations. An electroencephalogram (EEG) wave signal is collected in participant when they listen to VR-based music. Recursive Feature Elimination (RFE) is an extraction technique for extracting the collected EEG recording signals from the patients. Then the Improved Glow Worm Swarm Optimization (IGSO) method has been employed to determine an optimal set of characteristics for accurate emotion classification. Emotion is classified using the Decision Tree (DT) method depending on the feature selected in the EEG wave signal. The valence and arousal levels were measured using the self-assessment manikin (SAM). The GW+DT method achieved a greater accuracy (95%), recall (82.10%) and F1-Score (80.52%), significantly outperforming traditional methods. The findings highlight the probable involvement of VR and music therapy as a therapeutic approach to enhance mental health and emotional stability in clinical settings.