Development of a Data-Driven Digital Phenotype Profile of Distress Experience of Healthcare Workers During COVID-19 Pandemic

Due to the constraints of the COVID-19 pandemic, healthcare workers have reported acting in ways that are contrary to their moral values, which may result in moral distress and injury. The purpose of this work is to create a digital phenotype profile (DPP) tool to automate and evaluate the distress of participants in the COVID-19 VR Healthcare Simulation for Moral Distress dataset (NCT05001542). The dataset collected passive physiological signals and active mental health questionnaires. This paper focuses on correlating electrocardiogram, respiration, photoplethysmography and galvanic skin response with moral injury outcome scale (Brief MIOS). The DPP tool uses data-driven techniques to create a robust evaluation of distress among participants. To accomplish this, we applied pre-processing techniques which involved normalization, data sanitation, segmentation, and windowing. During feature analysis, we extracted statistical and heart rate variability features, followed by feature selection techniques to rank the importance of features. Prior to classification, we used k-means clustering to cluster the Brief MIOS scores to low, moderate, and high moral distress as the Brief MIOS is yet to have an established cut-off scores. We then classified the separation of the Brief MIOS scores using weighted support vector machine with leave-one-subject-out-cross-validation to achieve an accuracy of 98.67±0.87%. Various machine learning ablation tests were performed to support our results and further enhance the understanding of the predictive model. Our findings suggest that the proposed DPP tool could be used as an automated tool to examine moral distress. With further replication and validation, it could be used as a potential AI-based clinical decision-making tool to expedite and validate mental health treatments.