Sleep Apnea Detection with One-Dimensional Convolutional Neural Networks From Single-Lead Electrocardiogram
This repo is for the purpose of tracking progress in my Science Extension "Scientific Research Report" in which I intend to create a 1D-CNN for detecting Apneatic and Hypopneatic events in 1L-ECG data, sourced from the Apnea-ECG Database on Physionet.
This paper posits a method for diagnosing obstructive sleep apnea (OSA) based on a single-lead electrocardiogram (1L-ECG) recording taken during sleep. OSA is a common sleep disorder which is characterised by temporary cessations of breathing occuring due to obstructions in the upper airway. This study attempted to train a Convolutional Neural Network (CNN) to determine the presence of apneatic events, thus finding both the apnea hypopnea index (AHI) and hence the presence of OSA.