SleepPPG-Net2: Deep learning generalization for sleep staging from photoplethysmography

TL;DR

SleepPPG-Net2 is a deep learning model that significantly improves sleep staging accuracy and generalization across diverse datasets using raw PPG signals, outperforming existing models.

Contribution

This paper introduces SleepPPG-Net2, a novel multi-source domain training deep learning model that enhances sleep staging accuracy and generalization from raw PPG data.

Findings

Higher performance over state-of-the-art models

Improved generalization with up to 19% better Cohen's kappa

Performance varies with age, sex, and sleep apnea severity

Abstract

Background: Sleep staging is a fundamental component in the diagnosis of sleep disorders and the management of sleep health. Traditionally, this analysis is conducted in clinical settings and involves a time-consuming scoring procedure. Recent data-driven algorithms for sleep staging, using the photoplethysmogram (PPG) time series, have shown high performance on local test sets but lower performance on external datasets due to data drift. Methods: This study aimed to develop a generalizable deep learning model for the task of four class (wake, light, deep, and rapid eye movement (REM)) sleep staging from raw PPG physiological time-series. Six sleep datasets, totaling 2,574 patients recordings, were used. In order to create a more generalizable representation, we developed and evaluated a deep learning model called SleepPPG-Net2, which employs a multi-source domain training approach.SleepPPG-Net2 was benchmarked against two state-of-the-art models. Results: SleepPPG-Net2 showed consistently higher performance over benchmark approaches, with generalization performance (Cohen's kappa) improving by up to 19%. Performance disparities were observed in relation to age, sex, and sleep apnea severity. Conclusion: SleepPPG-Net2 sets a new standard for staging sleep from raw PPG time-series.