Machine Learning Frameworks for Wearable-Based Stress Modeling in Naturalistic Settings: Scoping Review

Clinical Snapshot

Key Findings

• Effect Size: Not applicable — no pooled quantitative effect size reported; scoping review methodology

• Primary Outcome: Descriptive characterisation of machine learning modeling decisions (problem formulation, ground truth determination, algorithm selection) across 34 studies of wearable-based stress detection in naturalistic settings

• Nnt Or Sensitivity: Not applicable — individual study diagnostic performance metrics (sensitivity, specificity, accuracy) are not pooled; the review characterises methodological approaches rather than synthesising performance benchmarks

Clinical Application

The model card framework proposed is a research-facing tool rather than a clinical implementation guide. Adoption would require integration into research reporting standards by journals and funding bodies. Clinical deployment of wearable stress detection systems remains premature pending standardised validation frameworks, regulatory clearance, and clinical outcome evidence. No TGA-approved wearable stress detection devices are currently listed for clinical use in Australia. The RACGP does not yet provide specific guidance on wearable-based psychological stress monitoring. The proposed model card framework aligns with the Australian Digital Health Agency's emphasis on interoperability and data quality standards for digital health tools. PBS reimbursement for wearable-based mental health monitoring is not currently applicable. Australian researchers contributing to this field should note the MBS telehealth and digital health item review processes as potential future pathways. The findings are relevant to Australian occupational health, mental health nursing, and preventive cardiology contexts where stress monitoring interest is growing. Healthy adults in real-world settings; potential future applicability to clinical populations including those with anxiety disorders, occupational stress, cardiovascular risk, and burnout — though this review does not directly address clinical populations

Abstract

BACKGROUND: Stress, as commonly recognized, is an integral part of modern life and can significantly affect both mental and physical health. While substantial advancements have been made in measuring physical fitness through wearable devices, the detection and assessment of mental stress remain in their early stages. OBJECTIVE: The objective of this paper is to review recent studies of wearable-based stress detection in naturalistic settings, with a specific focus on characterizing machine learning frameworks inspired by the model card approach. METHODS: This review was conducted using the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) checklist. A total of 353 articles were identified through searches in databases such as PubMed, MEDLINE, ScienceDirect, IEEE, ACM Digital Library, Web of Science, and Embase. Studies were considered eligible if they collected data from healthy adults in naturalistic settings using wearable devices and used machine learning models for stress detection. RESULTS: A total of 34 articles met the eligibility criteria, including 11 conference papers, 22 journal articles, and 1 preprint published between 2017 and 2024. From these studies, we analyzed key machine learning modeling decisions such as problem formulation, ground truth determination, and machine learning algorithms. Additionally, we examined the major contributions of each study, focusing on the challenges they addressed and the solutions they proposed. Based on these findings, we proposed a model card framework for reporting machine learning-based, wearable-based stress detection. CONCLUSIONS: This scoping review highlights recent trends in machine learning models for stress detection and measurement using wearable signals. It underscores the need for improved standardization in reporting practices for datasets and key machine learning decisions, as well as the importance of addressing critical challenges associated with data collection in real-world settings. We hope this review will support and strengthen ongoing research efforts, promote knowledge sharing, and promote collaboration among researchers-ultimately advancing the field as a community.

References

1. 1.Sharma, S., Janakiraman, A. K., & Chen, L. K. (2026). Machine learning frameworks for wearable-based stress modeling in naturalistic settings: Scoping review. JMIR mHealth and uHealth. https://doi.org/10.2196/76632