Post-traumatic Stress Disorder (PTSD) Prediction in Covid-19 Survivors- A Clinical Questionnaire and Machine Learning Based Approach
Abstract
The Post-traumatic stress disorder (PTSD) can be denoted as a mental health issue which is caused due to the occurrence of a disturbing event, this event is either seen by a person or he has been involved in it. The Post-traumatic stress disorder (PTSD) is a psychological that disturbs the daily life of a person and halts his/her daily activities like job, work, study and much more. It also leads to other issues like depression, anxiety etc. The COVID-19 can be denoted as a transmissible disease that can harm the vital organs of a person and even cause death. However, the media hype of the disease being deadly has affected people psychologically more. Therefore, an onset of this disease in a person is a traumatic event and a person can get PTSD after he has beaten the disease. Thus, it is vital to detect the onset of PTSD early in a COVID-19 survivor to prevent any excessive health damage and allow for proper treatment of the patient. Machine learning methods have grown into a good alternative for analysis of health-related data and also detection of health issues. In this research, we suggest a Machine learning based model to create a response-based classifier which utilizes a reduced questionnaire to detect the early onset of PTSD in a COVID-19 survivor. The questions are precisely based on the Diagnostic and Statistical Manual of Mental Disorders, fifth edition (DSM-5). In short, this research utilizes the answers given by a COVID-19 survivor to detect PTSD utilizing the decision tree (CART) classifier with an accuracy percentage of 97.4%. The model is simple, computationally light and detects PTSD efficiently.
References
[2] Miseon Shim, Min Jin Jin, Chang-Hwan Im, Seung-Hwan Lee, Machine-learning-based classification between post-traumatic stress dis- order and major depressive disorder using P300 features, Neu- roImage: Clinical, Volume 24, 2019, 102001, ISSN 2213-1582, https://doi.org/10.1016/j.nicl.2019.102001.
[3] Mojjada RK, Yadav A, Prabhu AV, Natarajan Y. Machine Learning Models for covid-19 future forecasting. Mater Today Proc. 2020 Dec 9. doi: 10.1016/j.matpr.2020.10.962. Epub ahead of print. PMID: 33318952; PMCID: PMC7723767.
[4] Zoabi, Y., Deri-Rozov, S. & Shomron, N. Machine learning-based pre- diction of COVID-19 diagnosis based on symptoms. npj Digit. Med. 4, 3 (2021). https://doi.org/10.1038/s41746-020-00372-6.
[5] The University of Kashmir, Institute of Mental Health and Neurosciences (IMHANS), Muntazar Kashmir Mental Health Survey, (2015)
[6] Wshah S, Skalka C, Price M. Predicting Posttraumatic Stress Disor- der Risk: A Machine Learning Approach. JMIR Ment Health. 2019 Jul 22;6(7):e13946. doi: 10.2196/13946. PMID: 31333201; PMCID: PMC6681635.
[7] Galatzer-Levy, Isaac Ma, Sisi Statnikov, Alexander Yehuda, Rachel & Shalev, Arieh. (2017). Utilization of machine learning for prediction of post-traumatic stress: A re-examination of cortisol in the predic- tion and pathways to non-remitting PTSD. Translational Psychiatry. 7. 10.1038/tp.2017.38.
[8] Augsburger, M., Galatzer-Levy, I.R. Utilization of machine learning to test the impact of cognitive processing and emotion recognition on the development of PTSD following trauma exposure. BMC Psychiatry 20, 325 (2020). https://doi.org/10.1186/s12888-020-02728-4
[9] Zhang, J., Richardson, J.D. & Dunkley, B.T. Classifying post-traumatic stress disorder using the magnetoencephalographic connectome and ma- chine learning. Sci Rep 10, 5937 (2020). https://doi.org/10.1038/s41598-
020-62713-5
[10] Bashir A, McDonald M, Egan J, Hawk ME, Protocol of a Community- Based Intervention on Mental Health in Kashmir, Indian J Psychol Med, vol. 6 (2020)
