Transformer Model Detects Antidepressant Use From a Single Night of Sleep, Unlocking an Adherence Biomarker

2025-10-11

AI Summary

Antidepressant nonadherence is highly prevalent, substantially increasing risks of relapse, hospitalization, and suicide, while generating substantial avoidable healthcare expenditures. Current monitoring methods suffer from either invasiveness (e.g., therapeutic drug monitoring) or low accuracy (e.g., pill counts, prescription refill records). To address this gap, we propose the first noninvasive, biomarker-based approach for assessing antidepressant adherence using single-night, multimodal sleep physiological signals—acquired via consumer-grade wearable or contactless wireless sensors—and a Transformer-based model enabling daily, unobtrusive evaluation in home settings. Validated on real-world data from 21,000 participants across 62,000 nights, our model achieves an AUROC of 0.84. It demonstrates robust generalizability across antidepressant classes, sensitivity to dose-dependent effects, and resilience to confounding from concomitant psychotropic medications.

Technology Category

• Cognitive Modeling & Cognitive Systems: Adaptive Behavior
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Knowledge Representation and Reasoning: Nonmonotonic Reasoning

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Security and Privacy: Large-scale security measurements
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI

Abstract

Antidepressant nonadherence is pervasive, driving relapse, hospitalization, suicide risk, and billions in avoidable costs. Clinicians need tools that detect adherence lapses promptly, yet current methods are either invasive (serum assays, neuroimaging) or proxy-based and inaccurate (pill counts, pharmacy refills). We present the first noninvasive biomarker that detects antidepressant intake from a single night of sleep. A transformer-based model analyzes sleep data from a consumer wearable or contactless wireless sensor to infer antidepressant intake, enabling remote, effortless, daily adherence assessment at home. Across six datasets comprising 62,000 nights from >20,000 participants (1,800 antidepressant users), the biomarker achieved AUROC = 0.84, generalized across drug classes, scaled with dose, and remained robust to concomitant psychotropics. Longitudinal monitoring captured real-world initiation, tapering, and lapses. This approach offers objective, scalable adherence surveillance with potential to improve depression care and outcomes.

Problem

• Detecting antidepressant nonadherence from sleep data
• Providing noninvasive biomarker for adherence monitoring
• Enabling remote daily adherence assessment at home

Innovation

• Transformer model analyzes sleep data
• Detects antidepressant use from single night
• Uses consumer wearable or wireless sensor