Machine Learning Techniques for Prediction of Stress-Related Mental Disorders: A Scoping Review

• Published in: Proceedings of the Human Factors and Ergonomics Society Annual Meeting Vol. 66; no. 1; pp. 300 - 304
• Main Authors: Razavi, Moein, Ziyadidegan, Samira, Sasangohar, Farzan
• Format: Journal Article
• Language: English
• Published: Los Angeles, CA SAGE Publications 01.09.2022
• ISSN: 2169-5067; 1071-1813; 2169-5067
• DOI: 10.1177/1071181322661298

The goal of this paper is to review the literature on machine learning (ML) and big data applications for mental health, emphasizing current research and practical implementations. To explore the field of ML in mental health, we used a scoping review process. The literature identified application domains of detection and prediction of stress as a contributor to mental health disorders. We evaluated the articles and data on the mental health application, machine learning approach, type of data (sensor, survey, etc.), and type of sensors. Most studies extracted features before developing AI-based stress detection algorithms. Findings revealed that heart rate, heart rate variability, and skin conductance features are the key indicators of stress. Moreover, among AI stress-detection methods, Random Forest and Neural Networks show promising results.