Open Tool-kit for AI-based Sleep Apnea Scoring

Abstract

Sleep is the most essential and fundamental to an individual’s well-being and vitality because it provides for the restoration and re-energizing of both the body and mind. Sleep apnea is a common sleep problem that affects millions of individuals worldwide. It is distinguished by breathing pauses or shallow breathing during sleeping, which can result in a variety of health problems such as heart disease, stroke, and diabetes [1]. Sleep apnea diagnosis and management can be difficult because the cost and time-consuming polysomnography (PSG) testing requires specialized equipment and trained personnel. To address these concerns, we created and validated an artificial intelligence (AI)-based sleep apnea scoring system that analyses electrocardiogram (ECG) signals to predict the severity of sleep apnea. The system analyses ECG signals using 1D-CNN to predict the Apnea-Hypopnea Index (AHI), a measure of the severity of sleep apnea. The tool is organized into three sections: data exploration, data visualization, and prediction, and it is aimed at accurately forecasting patients’ risk of sleep apnea. Our research demonstrates the potential of AI-based approaches for the diagnosis and management of sleep apnea, and we believe that our system can help improve patient outcomes and quality of life.