An Explainable Anomaly Detection Framework for Monitoring Depression and Anxiety Using Consumer Wearable Devices

2025-05-05

AI Summary

Early dynamic monitoring of depressive and anxiety symptoms remains challenging due to the lack of clinically grounded, interpretable digital phenotyping methods. Method: We propose the first wearable-data analytics framework integrating clinical significance criteria—specifically, PHQ/GAD score changes ≥5 points as a dynamic anomaly threshold—using unsupervised LSTM autoencoders trained on consumer-grade device data (sleep, steps, resting heart rate) from 2,023 baseline-healthy participants. Pathological interpretability is achieved via SHAP-based feature attribution. Contribution/Results: The framework successfully detected 393 clinical deterioration events among 341 participants, achieving an adjusted F1-score of 0.80 overall; performance improved to 0.84 for comorbid deterioration and 0.85 for PHQ/GAD changes ≥10 points. Quantitative SHAP analysis identified resting heart rate as the most discriminative physiological driver. This work establishes a clinically interpretable, standard-validated paradigm for digital phenotype–based abnormality detection.

Technology Category

• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Humans and AI: Brain-Sensing and Analysis
• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Continuous monitoring of behavior and physiology via wearable devices offers a novel, objective method for the early detection of worsening depression and anxiety. In this study, we present an explainable anomaly detection framework that identifies clinically meaningful increases in symptom severity using consumer-grade wearable data. Leveraging data from 2,023 participants with defined healthy baselines, our LSTM autoencoder model learned normal health patterns of sleep duration, step count, and resting heart rate. Anomalies were flagged when self-reported depression or anxiety scores increased by>=5 points (a threshold considered clinically significant). The model achieved an adjusted F1-score of 0.80 (precision = 0.73, recall = 0.88) in detecting 393 symptom-worsening episodes across 341 participants, with higher performance observed for episodes involving concurrent depression and anxiety escalation (F1 = 0.84) and for more pronounced symptom changes (>=10-point increases, F1 = 0.85). Model interpretability was supported by SHAP-based analysis, which identified resting heart rate as the most influential feature in 71.4 percentage of detected anomalies, followed by physical activity and sleep. Together, our findings highlight the potential of explainable anomaly detection to enable personalized, scalable, and proactive mental health monitoring in real-world settings.

Problem

• Detect worsening depression/anxiety via wearable data patterns
• Identify clinically significant symptom severity increases
• Explain anomalies using physiological/behavioral feature importance

Innovation

• LSTM autoencoder model detects health anomalies
• SHAP analysis identifies key physiological features
• Consumer wearables enable real-time mental monitoring