Deep Learning-Based Detection of Obtructive Sleep Apnea from ECG Using 1D-CNN

DOI

10.2991/978-94-6463-866-0_45How to use a DOI?

Keywords

Obstructive Sleep Apnea; ECG; Signal Processing; Automated Diagnosis; Feature Extraction; Non-Invasive Detection

Abstract

This study uses a One-Dimensional Convolutional Neural Network (1D-CNN) and deep learning to identify obstructive sleep apnea (OSA) from electrocardiogram (ECG) signals. The PhysioNet Apnea-ECG database's ECG recordings were divided into 60-s segments and categorized as either apnea or non-apnea. With 91.4% precision, 92.1% recall, 91.7 F1-score, and an AUC-ROC of 0.95, the 1D-CNN model, which was trained with balanced class weights and optimal hyperparameters, obtained a classification accuracy of 93.6%. Model development, evaluation, and preprocessing were all done with MATLAB. This method improves scalability and lowers preprocessing overhead by automatically learning characteristics from raw ECG data, in contrast to previous approaches that require manual feature extraction. For the early diagnosis of OSA, the suggested method provides a non-invasive, effective, and precise substitute for polysomnography. It has the potential to be integrated into wearable technology for ongoing home surveillance because of its near real-time inference capabilities.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Yalla Jeevan Kumar
AU  - Gajjala Abhinav Reddy
AU  - V. Arun
PY  - 2025
DA  - 2025/10/31
TI  - Deep Learning-Based Detection of Obtructive Sleep Apnea from ECG Using 1D-CNN
BT  - Proceedings of the International Conference on Intelligent Systems and Digital Transformation (ICISD 2025)
PB  - Atlantis Press
SP  - 539
EP  - 552
SN  - 2589-4919
UR  - https://doi.org/10.2991/978-94-6463-866-0_45
DO  - 10.2991/978-94-6463-866-0_45
ID  - Kumar2025
ER  -