Detection of Sleep Apnea Using Machine Learning Algorithms Based on the ECG Signal

This project focuses on detecting sleep apnea using artificial intelligence and ECG signals. Traditional sleep apnea diagnosis methods are costly and uncomfortable, requiring specialized facilities. The project employs machine learning techniques like KNN, DT, NB, RBF and ANN on the Apnea-ecg dataset from Physionet. The results reveal that the ANN algorithm significantly enhances accuracy compared to traditional methods, achieving an accuracy rate of 76.01%. This method is suitable for implementation on affordable, portable embedded devices due to its low computational complexity. This innovative approach holds potential for various applications in sleep medicine due to its adaptability and cost-effectiveness.