Wearable Sleep Detection System Based on Piezoelectric Signals and Convolutional Neural Network Analysis

Clinical Snapshot

Key Findings

• P Value: Not reported

• Effect Size: Cohen's Kappa coefficient = 0.949 (indicating near-perfect agreement by Landis & Koch criteria, though reference standard is unclear)

• Primary Outcome: Five-state sleep stage classification accuracy of 95.90% on a 1000-sample processed held-out test set using a CNN model trained on piezoelectric eye movement signals

• Nnt Or Sensitivity: Per-class sensitivity and specificity not reported; overall accuracy 95.90% and Kappa 0.949 are the only performance metrics available from the abstract; these are insufficient for diagnostic test evaluation without per-stage breakdown and PSG comparison

• Confidence Interval: Not reported — a critical omission

Clinical Application

Eyelid-mounted sensor application requires user training and may be poorly tolerated across full-night recordings; long-term adhesion, displacement during sleep, and skin tolerance are unvalidated; manufacturing scalability and cost are unknown; no regulatory approval pathway described This device would require TGA registration as a Class IIa or IIb medical device (depending on intended diagnostic claims) before clinical use in Australia. Current evidence is insufficient for TGA submission. The RACGP and Australasian Sleep Association (ASA) guidelines recommend PSG or validated home sleep apnoea testing (HSAT) for sleep disorder diagnosis — this device does not yet meet those standards. PBS reimbursement for sleep studies requires accredited laboratory or validated home monitoring devices; this system is not positioned for PBS consideration at this stage. Australian sleep medicine practice would require prospective clinical validation against PSG in a diverse Australian cohort before adoption. Potentially applicable to adults seeking ambulatory sleep quality monitoring in non-clinical settings; clinical applicability to patients with suspected sleep disorders (OSA, insomnia, narcolepsy, REM behaviour disorder) cannot be established from current evidence

Abstract

Long-term and comfortable sleep monitoring is a core requirement for daily health management. However, existing sleep monitoring technologies fail to simultaneously achieve accurate sleep stage classification, wearing comfort, and direct monitoring of characteristic physiological signals. Here, we report a wearable sleep monitoring system (SMS) based on piezoelectric sensing and convolutional neural network (CNN) analysis. The system assesses sleep structure and quality by capturing characteristic eye movement signals and provides actionable recommendations for long-term sleep health management. The sensor employs a highly stereoregular polyacrylonitrile (PAN) composite fiber film modified with PDA@ZnO as the sensing unit, providing high sensitivity, excellent wearing comfort, and long-term durability. It conforms closely to the upper eyelid, enabling effective acquisition of direct physiological signals associated with sleep. Furthermore, we developed a CNN model for five-state sleep classification and validated it using original-recording-level 5-fold validation and a 1000-sample processed held-out test set, achieving an accuracy of 95.90% with a Cohen's Kappa coefficient of 0.949. The system efficiently presents sleep quality information through a simple and user-friendly human-computer interaction interface. The proposed system introduces a novel approach to convenient health monitoring and provides a feasible technical solution for accurate and long-term sleep health assessment in daily life.

References

1. 1.Zhang, L., Cao, P., Jiang, J., Zhang, J., Tan, S., & Zhang, H. (2026). Wearable sleep detection system based on piezoelectric signals and convolutional neural network analysis. Langmuir: The ACS Journal of Surfaces and Colloids. https://doi.org/10.1021/acs.langmuir.6c01554