Electroencephalography-based Emotional Analysis Using Virtual Reality Therapy for Drug Addiction Diagnosis and Rehabilitation

Abstract Context: Drug abuse affects the human brain and damages several neurons, causing people to lose their sensing and learning abilities. To understand and acquire the human brain, use the electroencephalogram (EEG) modality, which helps to understand the brain’s activities and behavior according to the respective frequencies of the signals. Virtual reality (VR) therapy is being explored as a potential method to aid in the rehabilitation of drug-addicted individuals by stimulating and improving brain function. Aims: The primary objective of this research is to perform a comparative analysis of EEG data signals obtained from healthy participants and drug-addicted patients within a controlled VR environment. Additionally, the study aims to evaluate the effectiveness of VR therapy in enhancing brain function in drug-addicted individuals by comparing EEG signals before and after the therapy sessions. Subjects and Methods: This study involved 80 participants—40 diagnosed with drug addiction and 40 healthy individuals. EEG data were collected using a five-electrode EMOTIV Insight wearable device. Recordings were taken before and after VR therapy sessions for all participants. The data were analyzed using the Statistical Package for the Social Sciences (SPSS) and various machine learning (ML) techniques including k-nearest neighbors (k-NN), XGBoost, Naïve Bayes, and logistic regression. Performance metrics such as accuracy, precision, recall, and F1-score were used to evaluate the classification models. The Mann–Whitney U test was employed to identify statistical differences in EEG frequency bands between groups. Results: The Mann–Whitney U test revealed significant differences in EEG frequency bands ( P < 0.001) between drug-addicted patients and healthy. Among the ML techniques used, k-NN and XGBoost achieved the highest accuracy in distinguishing between EEG signals recorded before and after VR therapy. Conclusions: The findings suggest that EEG is effective in detecting neurological differences between healthy and drug-addicted individuals. Furthermore, VR therapy has a positive impact on brain activity in drug-addicted patients. Machine learning models, particularly k-NN and XGBoost, proved to be effective tools for classifying and tracking these neural changes.