Transformer-Based Multimodal Framework for Accurate Sleep Apnea Detection Using ECG and SpO2 Signals

Sleep apnea is a common sleep disorder that not only impacts the quality of sleep but also poses risks for severe health conditions such as cardiovascular diseases. This paper proposes a comprehensive detection algorithm for sleep apnea based on the fusion of electrocardiogram (ECG) and blood oxygen saturation (SpO2) signals. By integrating ECG and SpO2 data and leveraging deep learning techniques, the proposed method achieves more accurate detection of sleep apnea. This study enhances detection accuracy and improves system practicality, offering a low-cost and efficient solution for sleep monitoring in clinical settings.