Abstract
Objective: The objective is to develop a new deep learning method for the estimation of respiratory effort from a chest-worn accelerometer during sleep. We evaluate performance, compare it against a state-of-the art method, and assess whether it can differentiate between sleep stages. Methods: In 146 participants undergoing overnight polysomnography data were collected from an accelerometer worn on the chest. The study data were partitioned into train, validation, and holdout (test) sets. We used the train and validation sets to generate and train a convolutional neural network and performed model selection respectively, while we used the holdout set (72 participants) to evaluate performance. Results: A convolutional neural network with 9 layers and 207,855 parameters was automatically generated and trained. The neural network significantly outperformed the best performing conventional method, based on Principal Component Analysis; it reduced the Mean Squared Error from 0.26 to 0.11 and it also performed better in the detection of breaths (Sensitivity 98.4 %, PPV 98.2 %). In addition, the neural network exposed significant differences in characteristics of respiratory effort between sleep stages (p < 0.001). Conclusion: The deep learning method predicts respiratory effort with low error and is sensitive and precise in the detection of breaths. In addition, it reproduces differences between sleep stages, which may enable automatic sleep staging, using just a chest-worn accelerometer.
Original language | English |
---|---|
Article number | 104726 |
Number of pages | 12 |
Journal | Biomedical Signal Processing and Control |
Volume | 83 |
DOIs | |
Publication status | Published - May 2023 |
Bibliographical note
Publisher Copyright:
© 2023 The Author(s)
Funding
This work has been performed in the IMPULS framework of the Eindhoven MedTech Innovation Center (e/MTIC, incorporating Eindhoven University of Technology, Philips Research, and Sleep Medicine Center Kempenhaeghe). The funders had no role in the study design, decision to publish, or preparation of the manuscript. Fons Schipper, Pedro Fonseca, and Angela Grassi are employed by Philips Research. The employer had no influence on the study and on the decision to publish. Ruud van Sloun is employed by both Philips Research and by the Eindhoven University of Technology. The other authors declare no competing interests.
Funders | Funder number |
---|---|
Sleep Medicine Centre Kempenhaeghe | |
Eindhoven University of Technology |
Keywords
- Accelerometer
- Adam stochastic optimization
- Convolutional neural network
- Deep learning
- Principal component analysis
- Receptive field
- Respiratory effort
- Sleep staging
Fingerprint
Dive into the research topics of 'A deep-learning approach to assess respiratory effort with a chest-worn accelerometer during sleep'. Together they form a unique fingerprint.Research areas
-
Sleep Medicine
van Gilst, M. M. (Content manager) & van der Hout-van der Jagt, M. B. (Content manager)
Impact: Research Topic/Theme (at group level)