Classification of Sleep Apnea Event Using Heuristic 1D-CNN Based Deep Learning Approach

Obstructive Sleep Apnea (OSA) is a common sleep-related respiratory disorder characterized by recurrent interruptions of breathing during sleep, resulting in considerable health hazards. Globally, an increasing number of individuals are experiencing OSA. The precise and prompt categorization of sleep apnea occurrences is essential for efficient diagnosis and therapy. This research introduces a heuristic 1D Convolutional Neural Network (1D-CNN) deep learning method for classifying sleep apnea events from Electrocardiogram (ECG) information. The proposed method utilizes heuristic optimization techniques to augment the feature extraction process followed by feature extraction with a 1D-CNN architecture designed to capture the temporal properties of the signals, hence improving the efficacy of the model. The model is trained and evaluated using a publicly accessible sleep apnea dataset. The experimental findings underscore the efficacy of the proposed heuristic 1D-CNN model in detecting sleep apnea, with an overall accuracy of 93.58% and a decrease in false-positive rates. This method presents a viable solution for the automated categorization of sleep apnea events, with potential uses in real-time sleep monitoring and diagnostic systems.