Studying sleep: towards the identification of hypnogram features that drive expert interpretation
SLEEP · 2024
DOI: 10.1093/sleep/zsad306 ↗Kurz gefasst
Fachartikel in der Zeitschrift SLEEP, 2024. Verfasst von Caspar van der Woerd und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department Mathematics and Computer Science, Eindhoven University of Technology. Hauptthemen laut MEDLINE: Sleep Deprivation, Electroencephalography. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 27 Literaturangaben. Gefördert von: Eindhoven MedTech Innovation Center, Dutch Ministry of Economic Affairs and Climate.
Abstract (Originalsprache)
Study Objectives Hypnograms contain a wealth of information and play an important role in sleep medicine. However, interpretation of the hypnogram is a difficult task and requires domain knowledge and “clinical intuition.” This study aimed to uncover which features of the hypnogram drive interpretation by physicians. In other words, make explicit which features physicians implicitly look for in hypnograms.
Methods
Three sleep experts evaluated up to 612 hypnograms, indicating normal or abnormal sleep structure and suspicion of disorders. ElasticNet and convolutional neural network classification models were trained to predict the collected expert evaluations using hypnogram features and stages as input. The models were evaluated using several measures, including accuracy, Cohen’s kappa, Matthew’s correlation coefficient, and confusion matrices. Finally, model coefficients and visual analytics techniques were used to interpret the models to associate hypnogram features with expert evaluation.
Results
Agreement between models and experts (Kappa between 0.47 and 0.52) is similar to agreement between experts (Kappa between 0.38 and 0.50). Sleep fragmentation, measured by transitions between sleep stages per hour, and sleep stage distribution were identified as important predictors for expert interpretation.
Conclusions
By comparing hypnograms not solely on an epoch-by-epoch basis, but also on these more specific features that are relevant for the evaluation of experts, performance assessment of (automatic) sleep-staging and surrogate sleep trackers may be improved. In particular, sleep fragmentation is a feature that deserves more attention as it is often not included in the PSG report, and existing (wearable) sleep trackers have shown relatively poor performance in this aspect.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department Mathematics and Computer Science, Eindhoven University of Technology.
- Remote Patient Management and Chronic Care, Philips Research, Eindhoven, The Netherlands.
- Department Electrical Engineering, Eindhoven University of Technology, Eindhoven, The Netherlands.
- Sleep Medicine Center, Kempenhaeghe, Heeze, The Netherlands.
- Academic Sleep Clinic, CIRO, Horn, The Netherlands.
ORCID-Profile: Caspar van der Woerd, Hans van Gorp, Fokke van Meulen, Stef van den Elzen, Sebastiaan Overeem, Pedro Fonseca
Förderung
- Eindhoven MedTech Innovation Center
- Dutch Ministry of Economic Affairs and Climate
Zitieren
APA
van der Woerd, C., van Gorp, H., Dujardin, S., Sastry, M., Garcia Caballero, H., van Meulen, F., van den Elzen, S., Overeem, S., & Fonseca, P. (2024). Studying sleep: towards the identification of hypnogram features that drive expert interpretation. SLEEP. https://doi.org/10.1093/sleep/zsad306
Vancouver
van der Woerd C, van Gorp H, Dujardin S, Sastry M, Garcia Caballero H, van Meulen F, et al. Studying sleep: towards the identification of hypnogram features that drive expert interpretation. SLEEP. 2024. doi:10.1093/sleep/zsad306
BibTeX
@article{vanderWoerd2024,
author = {van der Woerd, Caspar and van Gorp, Hans and Dujardin, Sylvie and Sastry, Manuel and Garcia Caballero, Humberto and van Meulen, Fokke and van den Elzen, Stef and Overeem, Sebastiaan and Fonseca, Pedro},
title = {Studying sleep: towards the identification of hypnogram features that drive expert interpretation},
journal = {SLEEP},
year = {2024},
doi = {10.1093/sleep/zsad306}
}
EUVIMED