ML Models for Early Detection of Mental Health Disorders Using Wearable IoT Devices
DOI:
https://doi.org/10.63282/3050-9246.IJETCSIT-V2I2P106Keywords:
Mental Health, Machine Learning, Wearable IoT, Depression Detection, Deep Learning, Anxiety, Remote Monitoring, LSTM, SVM, Data PrivacyAbstract
The rampant increase in the levels of mental health disorders among people of all ages has now become one of the main health issues of concern in the world, with an estimated figure of over 450 million people affected globally according to the World Health Organization. Conventional methods of diagnosis of mental illnesses involve subjective analysis based on clinical interviews, which may result in late identification, incorrect diagnosis or poor reporting as a result of stigmatization by a community. Currently, wearable IoT devices and Machine Learning (ML) appear to be the most promising options for implementing real-time, continuous, non-invasive monitoring of physiological and behavioural measures that correspond to mental health disorders. The current paper explores ML models that could be used in predicting symmetric mental health disorders at an early stage through the application of wearable IoT-based data. Through the adoption of Internet of Things (IoT) in healthcare, it has become possible to have the constant monitoring of biometric data like Heart Rate Variability (HRV), sleeping habits, physical activity and skin temperature. Such data can be processed with the use of ML algorithms and used to reveal the markers of stress, anxiety, depression, and other mental disorders. We provide a comparative study of different supervised and unsupervised learning techniques such as Support Vector Machines (SVM), Decision Tree, Random Forest, K-Nearest Neighbors (KNN) and Deep Learning models, CNN and LSTM. With our research methodology, we have a structured data pipeline that includes preprocessing, feature selection, and model training and validation using benchmark data and data obtained from off-the-shelf wearables, such as Fitbit, Apple Watch, and Empatica E4. The results indicate that deep learning models, particularly LSTM networks, reflect higher performance in the process of extracting temporal patterns in physiological data, with more than 90 percent accuracy in detecting early factors of depression and anxiety.
Furthermore, we hypothesise a hybrid system that employs a hybrid data collection strategy (IoT-based) and processing (cloud-based ML), with real-time results provided to the user via a mobile health application. The research also talks about ethical, privacy, and security issues related to working with sensitive mental health data and suggests approaches to sharing such information through blockchain. In summary, this study highlights the potential of ML and wearable IoT devices in transforming mental care, enabling proactive measures that minimise the workload on healthcare systems and significantly improve the standards of living
Downloads
References
[1] Saeb, S., Lattie, E. G., Schueller, S. M., Kording, K. P., & Mohr, D. C. (2016). The relationship between mobile phone location sensor data and depressive symptom severity. PeerJ, 4, e2537.
[2] Ghandeharioun, A., Fedor, S., Sangermano, L., Ionescu, D., Alpert, J., Dale, C., ... & Picard, R. (2017, October). Objective assessment of depressive symptoms with machine learning and wearable sensor data. In the 2017 seventh international conference on affective computing and intelligent interaction (ACII) (pp. 325-332). IEEE.
[3] Burns, M. N., Begale, M., Duffecy, J., Gergle, D., Karr, C. J., Giangrande, E., & Mohr, D. C. (2011). Harnessing context sensing to develop a mobile intervention for depression. Journal of Medical Internet Research, 13(3), e1838.
[4] Canzian, L., & Musolesi, M. (2015, September). Trajectories of depression: unobtrusive monitoring of depressive states by means of smartphone mobility traces analysis, in Proceedings of the 2015 ACM international joint conference on pervasive and ubiquitous computing (pp. 1293-1304).
[5] Wang, R., Chen, F., Chen, Z., Li, T., Harari, G., Tignor, S., ... & Campbell, A. T. (2014, September). StudentLife: assessing mental health, academic performance and behavioral trends of college students using smartphones. In Proceedings of the 2014 ACM international joint conference on Pervasive and Ubiquitous Computing (pp. 3-14).
[6] Gjoreski, M., Gjoreski, H., Luštrek, M., & Gams, M. (2016, September). Continuous stress detection using a wrist device: in laboratory and real life. In Proceedings of the 2016 ACM international joint conference on Pervasive and Ubiquitous Computing: Adjunct (pp. 1185-1193).
[7] Smets, E., Rios Velazquez, E., Schiavone, G., Chakroun, I., D’Hondt, E., De Raedt, W., ... & Van Hoof, C. (2018). Large-scale wearable data reveal digital phenotypes for daily-life stress detection. NPJ digital medicine, 1(1), 67.
[8] Reece, A. G., & Danforth, C. M. (2017). Instagram photos reveal predictive markers of depression. EPJ Data Science, 6(1), 15.
[9] Sano, A., & Picard, R. W. (2013, September). Stress recognition using wearable sensors and mobile phones. In 2013, the Humaine association conference on affective computing and intelligent interaction (pp. 671-676). IEEE.
[10] Lane, N. D., Miluzzo, E., Lu, H., Peebles, D., Choudhury, T., & Campbell, A. T. (2010). A survey of mobile phone sensing. IEEE Communications magazine, 48(9), 140-150.
[11] Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural computation, 9(8), 1735-1780.
[12] Rieke, N., Hancox, J., Li, W., Milletari, F., Roth, H. R., Albarqouni, S., ... & Cardoso, M. J. (2020). The future of digital health with federated learning. NPJ digital medicine, 3(1), 119.
[13] Singh, G., Garg, V., Tiwari, P., & Goel, R. (2020). Emergence of IoT and Big Data: A Study in the Healthcare Industry. In Applications of Deep Learning and Big IoT on Personalized Healthcare Services (pp. 42-54). IGI Global.
[14] Ahmed, Z., Mohamed, K., Zeeshan, S., & Dong, X. (2020). Artificial intelligence with a multi-functional machine learning platform for better healthcare and precision medicine. Database, 2020, baaa010.
[15] Noorbakhsh, J., Chandok, H., Karuturi, R. K. M., & George, J. (2019). Machine learning in biology and medicine. Advances in Molecular Pathology, 2(1), 143-152.
[16] Chen, P. H. C., Liu, Y., & Peng, L. (2019). How to develop machine learning models for healthcare. Nature materials, 18(5), 410-414.
[17] Stiglic, G., Kocbek, P., Fijacko, N., Zitnik, M., Verbert, K., & Cilar, L. (2020). Interpretability of machine learning‐based prediction models in healthcare. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 10(5), e1379.
[18] Callahan, A., & Shah, N. H. (2017). Machine learning in healthcare. In Key advances in clinical informatics (pp. 279-291). Academic Press.
[19] Wang, S., Cao, J., & Philip, S. Y. (2020). Deep learning for spatio-temporal data mining: A survey. IEEE transactions on knowledge and data engineering, 34(8), 3681-3700.
[20] Chen, W., An, J., Li, R., Fu, L., Xie, G., Bhuiyan, M. Z. A., & Li, K. (2018). A novel fuzzy deep-learning approach to traffic flow prediction with uncertain spatial–temporal data features. Future Generation Computer Systems, 89, 78-88.
[21] Jha, D. K., Srivastav, A., & Ray, A. (2016). Temporal learning in video data using deep learning and Gaussian processes. International Journal of Prognostics and Health Management, 7(4).
[22] Amato, F., Guignard, F., Robert, S., & Kanevski, M. (2020). A novel framework for spatio-temporal prediction of environmental data using deep learning. Scientific reports, 10(1), 22243.
