What question did this study set out to answer?

The aim is to develop a deep learning model for accurate detection of obstructive sleep apnea using ECG signals.

January 10, 2026IET Signal Processing0 citationsOpen Access

A Multi‐Branch Attention‐Enhanced Architecture for OSA Detection Using ECG Signals

ACAlireza CheshmberahGolestan University MZMajid ZiaratbanGolestan University

Key Result

The proposed deep learning model achieved 91.97% accuracy, 91.28% sensitivity, and 92.41% specificity for OSA detection from single-lead ECG signals.

Key Points

The aim is to develop a deep learning model for accurate detection of obstructive sleep apnea using ECG signals.
Developed a multi-branch architecture combining convolution, LSTM, STFT, and attention mechanisms.
Integrated parallel branches for comprehensive feature extraction from ECG signals.
Implemented channel attention-based feature fusion to enhance signal relevance and reduce noise.
Achieved segment-level classification performance with 91.97% accuracy.
Demonstrated 91.28% sensitivity and 92.41% specificity in detecting OSA.
Proposed model is efficient with a small number of parameters for real-time applications.

Structured PICO

Does a multi-branch attention-enhanced deep learning architecture accurately detect obstructive sleep apnea using single-lead ECG signals?

Population

PhysioNet Apnea-ECG dataset

Intervention

Multi-branch attention-enhanced deep learning architecture (convolution, LSTM, STFT, attention, and transformer modules)

Outcome

Segment-level classification performance (accuracy, sensitivity, specificity)

A novel deep learning architecture using single-lead ECG signals demonstrated high accuracy (91.97%) for automated obstructive sleep apnea detection, suggesting potential for wearable monitoring applications.

Main Result

Absolute Event Rate: 0% vs 0%

Abstract

Obstructive Sleep Apnea (OSA) is a prevalent and underdiagnosed sleep disorder that can lead to serious cardiovascular and cognitive complications if left untreated. This study presents a novel deep learning architecture based on convolution, LSTM, short‐time Fourier transform (STFT), attention, and transformer modules for automated OSA detection using single‐lead electrocardiogram (ECG) signals, aiming to improve diagnostic accuracy. The proposed model integrates six parallel branches for feature extraction, combining convolutional layers, recurrent units, STFT, and residual connections to capture multiscale temporal and frequency‐domain patterns. A three‐path feature refinement that incorporates sequential, convolutional, and transformer encoder is considered as the second part of the model. Channel attention‐based feature fusion modules are employed in the first and second parts to enhance feature relevance and suppress noise. Experimental evaluations on the PhysioNet Apnea‐ECG dataset demonstrate that the proposed model achieves superior segment‐level classification performance with 91.97% accuracy, 91.28% sensitivity, and 92.41% specificity. These findings suggest that the proposed method offers a robust, and scalable solution. Regarding the small number of parameters of the model, it can potentially be considered for real‐time and wearable‐based OSA monitoring applications. All codes and the trained model are released at https://github.com/mziaratban/OSA .

Ask AI

Helpful

Bookmark

View Full Paper