ECG-Based RBD Classification for Alleviating PSG Burden: Potential as a Screening Tool 


Vol. 51,  No. 4, pp. 778-791, Apr.  2026
10.7840/kics.2026.51.4.778


PDF Full-Text
  Abstract

REM sleep behavior disorder (RBD) is a parasomnia characterized by abnormal behaviors during REM sleep and is a precursor to neurodegenerative syndromes such as Parkinson's disease. Accurate RBD detection is essential for early diagnosis and intervention, typically conducted via polysomnography (PSG). However, PSG is costly and uncomfortable for patients due to logistical complexities, which has led to increasing interest in selective diagnostic approaches. This study investigates an ECG-based method using heart rate variability (HRV) features to classify RBD patients, aiming to mitigate the limitations of PSG and to verify its potential as a screening tool for RBD. We utilized data from 24 patients (9 RBD, 15 healthy controls) collected at Severance Hospital, and extracted 91 HRV features using Pan-Tompkins algorithms. These features were analyzed using 20 machine learning and deep learning models, including 1D-CNN, ANN, and GRU. The 1D-CNN model demonstrated the highest classification accuracy (97.58%), with 36–39 HRV features contributing most significantly to performance. The proposed method offers a non-invasive, cost-effective approach to RBD diagnosis, achieving comparable or superior performance to traditional PSG-based studies. This study highlights the potential for HRV-based ECG analysis to streamline RBD detection. Future research will extend to larger datasets and single-lead ECG applications.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Related Articles
  Cite this article

[IEEE Style]

J. Han, H. Song, W. Lee, Y. Kim, J. Park, "ECG-Based RBD Classification for Alleviating PSG Burden: Potential as a Screening Tool," The Journal of Korean Institute of Communications and Information Sciences, vol. 51, no. 4, pp. 778-791, 2026. DOI: 10.7840/kics.2026.51.4.778.

[ACM Style]

Ju-Hyuck Han, Ho-Jung Song, Won-Woo Lee, Yong-Suk Kim, and Jong-Uk Park. 2026. ECG-Based RBD Classification for Alleviating PSG Burden: Potential as a Screening Tool. The Journal of Korean Institute of Communications and Information Sciences, 51, 4, (2026), 778-791. DOI: 10.7840/kics.2026.51.4.778.

[KICS Style]

Ju-Hyuck Han, Ho-Jung Song, Won-Woo Lee, Yong-Suk Kim, Jong-Uk Park, "ECG-Based RBD Classification for Alleviating PSG Burden: Potential as a Screening Tool," The Journal of Korean Institute of Communications and Information Sciences, vol. 51, no. 4, pp. 778-791, 4. 2026. (https://doi.org/10.7840/kics.2026.51.4.778)
Vol. 51, No. 4 Index

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880