Bipolar disorder relapse detection and prediction using smartwatches. A pilot study for machine learning models using anomaly detection methods
Journal of Affective Disorders · 2026-08 · 407 · 121731
DOI: 10.1016/j.jad.2026.121731 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Affective Disorders, 2026-08. Verfasst von Arnaud Pouchon und 6 weiteren Autor:innen. Beteiligte Einrichtung: Univ. Grenoble Alpes, Inserm, CHU Grenoble-Alpes, GIN, 38000, Grenoble, France; Department of Psychiatry, Grenoble Alpes University Hospital, 38000, …. Hauptthemen laut MEDLINE: Bipolar Disorder, Machine Learning, Wearable Electronic Devices. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 45 Literaturangaben. Gefördert von: Fondation de l'Avenir pour la Recherche Médicale Appliquée.
Abstract (Originalsprache)
BACKGROUND: Bipolar disorder (BD) is a chronic mental illness with recurrent mood episodes, with up to 70% of patients relapsing within two years. Early detection of prodromal symptoms is critical for timely intervention but remains challenging. Wearable devices and digital phenotyping allow monitoring of physiological and behavioral changes in real time. This pilot study evaluated the feasibility of using machine learning models on smartwatch-derived data to detect and predict mood relapses and prodromal phases in BD. METHODS: Ten BD patients were monitored over six months using smartwatches recording heart rate, heart rate variability (HRstd), sleep stages, and physical activity. Monthly clinical and psychometric assessments labeled periods as healthy, prodromal, or relapse. Unsupervised anomaly detection models were trained on healthy data to detect relapses and predict prodromes across multiple time windows. RESULTS: Across 1395 days of usable data, 15 depressive and 8 hypomanic relapses occurred. Physiological changes were detected during prodromal and relapse phases. HRstd and steps best detected depressive relapses (Precision-Recall Area Under the Curve [PR AUC] = 0.67), while light sleep and mean HR were optimal for hypomanic relapses (PR AUC = 0.33). Prodromal phases were predicted with lower but above-chance performance (PR AUC = 0.34 for prodromal depression). CONCLUSIONS: Smartwatch-derived physiological signals combined with unsupervised anomaly detection can detect mood relapses and, to a lesser extent, prodromal states in BD. These results support the integration of wearable monitoring and AI into clinical practice to anticipate relapses and enable timely intervention.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Univ. Grenoble Alpes, Inserm, CHU Grenoble-Alpes, GIN, 38000, Grenoble, France; Department of Psychiatry, Grenoble Alpes University Hospital, 38000, Grenoble, France. Electronic address: apouchon@chu-grenoble.fr.
- Univ. Grenoble Alpes, CEA, Leti, F-38000, Grenoble, France.
- Univ. Grenoble Alpes, CHU Grenoble-Alpes, 38000, Grenoble, France.
- Univ. Grenoble Alpes, Inserm, CHU Grenoble-Alpes, GIN, 38000, Grenoble, France; Department of Psychiatry, Grenoble Alpes University Hospital, 38000, Grenoble, France.
Förderung
- Fondation de l'Avenir pour la Recherche Médicale Appliquée
Zitieren
APA
Pouchon, A., Aloui, S., Mayer Dalverny, L., Djennaoui, L., Guyon-Gardeux, C., Godin, C., & Polosan, M. (2026). Bipolar disorder relapse detection and prediction using smartwatches. A pilot study for machine learning models using anomaly detection methods. Journal of Affective Disorders, 407, 121731. https://doi.org/10.1016/j.jad.2026.121731
Vancouver
Pouchon A, Aloui S, Mayer Dalverny L, Djennaoui L, Guyon-Gardeux C, Godin C, et al. Bipolar disorder relapse detection and prediction using smartwatches. A pilot study for machine learning models using anomaly detection methods. Journal of Affective Disorders. 2026;407:121731. doi:10.1016/j.jad.2026.121731
BibTeX
@article{Pouchon2026,
author = {Pouchon, Arnaud and Aloui, Saifeddine and Mayer Dalverny, Luca and Djennaoui, Lydia and Guyon-Gardeux, Claire and Godin, Christelle and Polosan, Mircea},
title = {Bipolar disorder relapse detection and prediction using smartwatches. A pilot study for machine learning models using anomaly detection methods},
journal = {Journal of Affective Disorders},
year = {2026},
volume = {407},
pages = {121731},
doi = {10.1016/j.jad.2026.121731}
}
EUVIMED