On the Generalizability of ECG-based Stress Detection Models

Prajod, Pooja; André, Elisabeth

doi:10.48550/arxiv.2210.06225

Cited by 1 publication

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Fairly good classification scores are obtained in the independent and cross-dataset analysis with better generalisation capabilities demonstrated by the RF model (62% F1 score), although it performed slightly less well than the LR model on MMSD independent evaluation. These results align well with those obtained in a similar study found in literature using the same standards (LOSO and cross-dataset analysis) [ 41 ].…”

Section: Discussionsupporting

confidence: 92%

“…An RF model applied the MMSD dataset without using a LOSO cross-validation, resulted in an f1 score greater than 80% [ 39 ]. Studies using the same procedure also tend to yield similar results [ 40 , 41 ]. The main advantage of LOSO is that it makes the most efficient use of the available data.…”

Section: Results and Discussionmentioning

confidence: 83%

“…In [ 41 ] as well, the RF model was found to have good generalizability across datasets. The results showed that while RF achieved slightly lower F1-score and Accuracy than other HRV based models on within dataset evaluations for binary stress classification, it outperformed them in the cross-dataset analysis.…”

Section: Results and Discussionmentioning

confidence: 99%

See 2 more Smart Citations

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

Benchekroun

Velmovitsky

Istrate

et al. 2023

Sensors

View full text Add to dashboard Cite

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.

show abstract

Section: Discussionsupporting

confidence: 92%