Developed Convolutional Neural Network for Mental Health Evaluation using Wearable Devices on WESAD Dataset

Monitoring of mental health at an early stage allows for the intervention and treatment that suits the individual’s needs, thus enhancing successful outcomes with less stigma associated with mental illness. This paper presents a comparative analysis of the following algorithms-Convolutional Neural Network (CNN), Cyber-Human Systems (CHS), Deep Stacked Generalization Ensemble Learning (DeSGEL), Support Vector Machine (SVM), Random Forest (RF), K-Nearest Neighbor (KNN), and VGG16 performance on data from 15 participants, aged 27.5 ± 2.4 years, 12 males, and 3 females. After feature selection and pre-processing, the algorithm CNN gave the best results achieving 98% Accuracy, 98% Precision, 94% Recall, and an F1-score of 95%. This has made the search for wearable technology to serve real-world mental health applications even more promising and motivating towards research in this field. This paper contributes to Sustainable Development Goal 3, Good Health and Well-being. This study reveals a sustainable approach to monitoring mental health with the help of technology, namely wearable devices and AI-driven models, which promotes early detection and intervention so as not to burden healthcare systems in the long run.