Cross Dataset Analysis for Generalizability of HRV-Based Stress Detection Models

Abstract

Stress is an increasingly prevalent mental health condition across the world. In Europe, for example, stress is considered one of the most common health problems, and over USD 300 billion are spent on stress treatments annually. Therefore, monitoring, identification and prevention of stress are of the utmost importance. While most stress monitoring is carried out through self-reporting, there are now several studies on stress detection from physiological signals using Artificial Intelligence algorithms. However, the generalizability of these models is only rarely discussed. The main goal of this work is to provide a monitoring proof-of-concept tool exploring the generalization capabilities of Heart Rate Variability-based machine learning models. To this end, two Machine Learning models are used, Logistic Regression and Random Forest to analyze and classify stress in two datasets differing in terms of protocol, stressors and recording devices. First, the models are evaluated using leave-one-subject-out cross-validation with train and test samples from the same dataset. Next, a cross-dataset validation of the models is performed, that is, leave-one-subject-out models trained on a Multi-modal Dataset for Real-time, Continuous Stress Detection from Physiological Signals dataset and validated using the University of Waterloo stress dataset. While both logistic regression and random forest models achieve good classification results in the independent dataset analysis, the random forest model demonstrates better generalization capabilities with a stable F1 score of 61%. This indicates that the random forest can be used to generalize HRV-based stress detection models, which can lead to better analyses in the mental health and medical research field through training and integrating different models.

Can heart rate variability data from the Apple Watch electrocardiogram quantify stress?

“…Data from the same study, this time from 27 participants, was used by Benchekroun et al ( 20 ), although in this case the HRV data was derived from the Empatica E4 device rather than from the Apple Watch ECG. The Empatica E4 device collects data continually as opposed to cross-sectionally, providing larger datasets.…”

Comprehensive Review and Proposed Model for Multimodal Depression Detection

“…1) Random Forests: Random Forest is another popular machine learning algorithm used in depression detection models that use Heart Rate Variability (HRV) data. One such study [13] used Random Forest to construct a depression detection model based on HRV parameters. The study used a dataset of 60 participants, including 30 patients with depression and 30 healthy controls.…”

A Convolutional Neural Network model for stress classification based on electrocardiogram data

“…Among these disorders, stress has received a lot of attention from researchers because, according to Hickey et al [2] and Long et al [3], stress causes negative impacts on the individual's quality of life and can trigger serious illnesses that can seriously compromise physical health. Another justification for the concern about stress disorders is that, according to Benchekroun et al [4], stress is considered one of the most prevalent mental health problems worldwide and the estimated expenditure on stress treatments is around 300 billion dollars per year in Europe.…”

“…According to Becker [12], abnormalities detected in the phase sequence or in the time interval between phases may be indicators of certain disorders such as atrial fibrillation, atrial flutter, and atrioventricular block; and all these disorders cause cardiac arrhythmias and their consequences. According to Benchekroun et al [4] and Arquilla et al [13], the detection of stress in ECG records is usually performed through the analysis of Heart Rate Variability (HRV). The HRV is the measure of the time interval between successive R peaks (highest wave of the QRS complex) and is a metric that reflects the balance between the SNS and PNS systems.…”

“…The HRV is the measure of the time interval between successive R peaks (highest wave of the QRS complex) and is a metric that reflects the balance between the SNS and PNS systems. According to Benchekroun et al [4], a decrease in HRV indicates an increase in SNS activity and consequent increase in stress, and an increase in HRV indicates an increase in SNP activity and a consequent reduction in stress. Fig.…”