EUVIMEDEUVIMED entdecken
Crossref

Automatic sleep scoring with LSTM networks: impact of time granularity and input signals

Alexandra-Maria Tăuțan iD, Alessandro C Rossi, Bogdan Ionescu

Biomedical Engineering / Biomedizinische Technik · 26.08.2022 · 67 (4) · 267-281

DOI: 10.1515/bmt-2021-0408 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biomedical Engineering / Biomedizinische Technik, online erschienen am 06.06.2022. Verfasst von Alexandra-Maria Tăuțan und 2 weiteren Autor:innen. Beteiligte Einrichtung: University Politehnica of Bucharest, Bucharest, Romania. Hauptthemen laut MEDLINE: Electroencephalography, Sleep Stages. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 59 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Supervised automatic sleep scoring algorithms are usually trained using sleep stage labels manually annotated on 30 s epochs of PSG data. In this study, we investigate the impact of using shorter epochs with various PSG input signals for training and testing a Long Short Term Memory (LSTM) neural network. An LSTM model is evaluated on the provided 30 s epoch sleep stage labels from a publicly available dataset, as well as on 10 s subdivisions. Additionally, three independent scorers re-labeled a subset of the dataset on shorter time windows. The automatic sleep scoring experiments were repeated on the re-annotated subset.The highest performance is achieved on features extracted from 30 s epochs of a single channel frontal EEG. The resulting accuracy, precision and recall were of 92.22%, 67.58% and 66.00% respectively. When using a shorter epoch as input, the performance decreased by approximately 20%. Re-annotating a subset of the dataset on shorter time epochs did not improve the results and further altered the sleep stage detection performance. Our results show that our feature-based LSTM classification algorithm performs better on 30 s PSG epochs when compared to 10 s epochs used as input. Future work could be oriented to determining whether varying the epoch size improves classification outcomes for different types of classification algorithms.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

ElectroencephalographySleep StagesNeural Networks, ComputerPolysomnographySleep

Stichwörter der Autor:innen

AASMAmerican Academy of Sleep MedicineECGEEGEMGElectrocardiogramElectroencephalographyElectromyographyLSTMLong Short Term MemoryMGHMassachusetts General HospitalNREMNon-Rapid Eye MovementPSG

Autor:innen und Institutionen

  • University Politehnica of Bucharest, Bucharest, Romania.
  • Onera Health, Eindhoven, The Netherlands.

ORCID-Profile: Alexandra-Maria Tăuțan

Zitieren

APA

Tăuțan, A. M., Rossi, A. C., & Ionescu, B. (2022). Automatic sleep scoring with LSTM networks: impact of time granularity and input signals. Biomedical Engineering / Biomedizinische Technik, 67(4), 267-281. https://doi.org/10.1515/bmt-2021-0408

Vancouver

Tăuțan AM, Rossi AC, Ionescu B. Automatic sleep scoring with LSTM networks: impact of time granularity and input signals. Biomedical Engineering / Biomedizinische Technik. 2022;67(4):267-281. doi:10.1515/bmt-2021-0408

BibTeX
@article{Tuan2022,
  author = {Tăuțan, Alexandra-Maria and Rossi, Alessandro C and Ionescu, Bogdan},
  title = {Automatic sleep scoring with LSTM networks: impact of time granularity and input signals},
  journal = {Biomedical Engineering / Biomedizinische Technik},
  year = {2022},
  volume = {67},
  number = {4},
  pages = {267--281},
  doi = {10.1515/bmt-2021-0408}
}
Kostenlos starten →