Uncertainty-Driven Anomaly Detection for Psychotic Relapse Using Smartwatches: Forecasting and Multi-Task Learning Fusion

TL;DR

This paper presents smartwatch-based frameworks using uncertainty-driven anomaly detection and multi-task learning to improve early psychotic relapse detection, achieving significant performance gains.

Contribution

It introduces two novel, complementary frameworks employing Transformer encoders and ensemble uncertainty estimation for robust relapse prediction from wearable data.

Findings

Fused model improves relapse detection accuracy by 8% over baseline.

Both frameworks independently demonstrate strong predictive power.

Integration of cardiac, sleep, and motion data enhances detection robustness.

Abstract

Digital phenotyping enables continuous passive monitoring of behavior and physiology, offering a promising paradigm for early detection of psychotic relapse. In this work, we develop and systematically study two smartwatch-based frameworks for daily relapse detection. The first forecasts cardiac dynamics and flags deviations between predicted and observed features as indicators of abnormality. The second adopts a multi-task formulation that fuses sleep with motion and cardiac-derived signals, learning time-aware embeddings and predicting measurement timing. Both pipelines use Transformer encoders and output a daily anomaly score, derived from predictive uncertainty estimated via an ensemble of multilayer perceptrons to improve robustness to real-world wearable variability. While each framework independently demonstrates strong predictive power, we show that they capture complementary physiological signatures. Consequently, we propose a late-fusion strategy that synergistically combines the anomaly signals from both architectures into a unified decision score. We benchmark our methodology on the 2nd e-Prevention Grand Challenge dataset, where our fused model achieves a 8% relative improvement over the competition-winning baseline. Our results, supported by extensive ablation studies, suggest that the integration of diverse digital phenotypes, cardiac, motion, and sleep, is essential for the high-fidelity detection of psychotic relapse in real-world settings.