Article, 2024

A deep transfer learning approach for sleep stage classification and sleep apnea detection using wrist-worn consumer sleep technologies

IEEE Transactions on Biomedical Engineering, ISSN 1558-2531, 0018-9294, Volume PP, 99, Pages 1-12, 10.1109/tbme.2024.3378480

Contributors

Olsen, Mads 0000-0002-9071-6760 [1] Zeitzer, Jamie Marc 0000-0001-6174-5282 [1] Nakase-Richardson, Risa [2] Musgrave, Valerie H [1] Sorensen, Helge B. D. [3] Mignot, Emmanuel Jean-Marie 0000-0002-4894-2254 [1] Jennum, Poul Jørgensen [4]

Affiliations

  1. [1] Stanford University
    2. [NORA names: United States; America, North; OECD];
  2. [2] University of South Florida
    3. [NORA names: United States; America, North; OECD];
  3. [3] Technical University of Denmark
    4. [NORA names: DTU Technical University of Denmark; University; Denmark; Europe, EU; Nordic; OECD];
  4. [4] Rigshospitalet
    5. [NORA names: Capital Region of Denmark; Hospital; Denmark; Europe, EU; Nordic; OECD]

Abstract

Obstructive sleep apnea (OSA) is a common, underdiagnosed sleep-related breathing disorder with serious health implications Objective - We propose a deep transfer learning approach for sleep stage classification and sleep apnea (SA) detection using wrist-worn consumer sleep technologies (CST). Methods - Our model is based on a deep convolutional neural network (DNN) utilizing accelerometers and photo-plethysmography signals from nocturnal recordings. The DNN was trained and tested on internal datasets that include raw data from clinical and wrist-worn devices; external validation was performed on a hold-out test dataset containing raw data from a wrist-worn CST. Results - Training on clinical data improves performance significantly, and feature enrichment through a sleep stage stream gives only minor improvements. Raw data input outperforms feature-based input in CST datasets. The system generalizes well but performs slightly worse on wearable device data compared to clinical data. However, it excels in detecting events during REM sleep and is associated with arousal and oxygen desaturation. We found; cases that were significantly underestimated were characterized by fewer of such event associations. Conclusion - This study showcases the potential of using CSTs as alternate screening solution for undiagnosed cases of OSA. Significance - This work is significant for its development of a deep transfer learning approach using wrist-worn consumer sleep technologies, offering comprehensive validation for data utilization, and learning techniques, ultimately improving sleep apnea detection across diverse devices.

Keywords

REM, REM sleep, Results Training, accelerometer, apnea, apnea detection, approach, arousal, associated with arousal, association, breathing disorders, cases, cases of obstructive sleep apnea, classification, clinical data, comprehensive validation, conclusions, consumer sleep technologies, convolutional neural network, data, data input, data utility, dataset, deep convolutional neural network, deep transfer learning approach, desaturation, detection, development, device data, devices, disorders, diverse devices, enrichment, event associations, events, health, health implications, hold-out test dataset, implications, improvement, input, international dataset, learning, learning approach, learning techniques, method, model, network, neural network, nocturnal recordings, obstructive sleep apnea, oxygen, oxygen desaturation, performance, photo-plethysmography signals, potential, raw data, raw data input, records, results, screening solution, signal, significance, sleep, sleep apnea, sleep apnea detection, sleep stage classification, sleep technology, sleep-related breathing disorders, solution, stage classification, stream, study, system, technique, technology, test dataset, training, transfer learning approach, undiagnosed cases, utilization, validity, wearable, wearable device data, wrist-worn device

Data Provider: Digital Science

LINKS
-

Matching Records in NORA

  • No Matches Found

SUBJECTS
+

METRICS
+