Continuous monitoring and detection of post-traumatic stress disorder (PTSD) triggers among veterans: A supervised machine learning approach

Anthony D. McDonald, Farzan Sasangohar, Ashish Jatav, Arjun H. Rao

Research output: Contribution to journalArticle

7 Scopus citations

Abstract

Post-traumatic stress disorder (PTSD) is a prevalent mental health condition among United States combat veterans, associated with high incidence of suicide and substance abuse. While PTSD treatments exist, such methods are limited to in-person therapy sessions and medications. Tools and technologies to monitor patients continuously, especially between sessions, are largely absent. This article documents efforts to develop predictive algorithms that utilize real-time heart rate data, collected using commercial off-the-shelf wearable sensors, to detect the onset of PTSD triggers. The heart rate data, pre-processed with a Kalman filter imputation approach to resolve missing data, were used to train five algorithms: decision tree, support vector machine, random forest, neural network, and convolutional neural network. Prediction performance was assessed with the Area Under the receiver operating characteristic Curve (AUC). The convolutional neural network, support vector machine, and random forests had the highest AUC and significantly outperformed a random classifier. Further analysis of the heart rate data and predictions suggest that the algorithms associate an increase in heart rate with PTSD trigger onset. While work is needed to enhance algorithm performance and robustness, these results suggest that wearable monitoring technology for PTSD symptom mitigation is an achievable goal in the near future.

Original languageEnglish (US)
Pages (from-to)201-211
Number of pages11
JournalIISE Transactions on Healthcare Systems Engineering
Volume9
Issue number3
DOIs
StatePublished - Jul 3 2019

Keywords

  • Convolutional neural networks
  • human physiology
  • post-traumatic stress disorder
  • random forest
  • wearable technology

ASJC Scopus subject areas

  • Safety, Risk, Reliability and Quality
  • Safety Research
  • Public Health, Environmental and Occupational Health

Fingerprint Dive into the research topics of 'Continuous monitoring and detection of post-traumatic stress disorder (PTSD) triggers among veterans: A supervised machine learning approach'. Together they form a unique fingerprint.

Cite this