Obstructive Sleep Apnea Detection Using Machine Learning

OSA (Obstructive Sleep Apnea) presents a significant challenge to sleep quality, described by disruptions in breathing and the hallmark of irregular airflow during sleep. Prevalent among middle-aged individuals, statistics indicate that around 2% of middle-aged women and 4% of middle-aged men are impacted by this respiratory disorder. The diagnostic journey for OSA involves a two-step process overseen by physicians, initially focusing on the review of medical records attained through the polysomnography system. Acknowledging the limitations of existing diagnostic procedures, the study introduces innovative approaches leveraging HRV (Heart Rate Variability) and ECG (Electrocardiography) signals.This involves collecting ECG records from both patient and control groups and subsequently utilizing an optical filter to refine ECG signals and HRV variables.The intricacies of the electrocardiographic data are further explored as the ECG signal is applied to compute the HRV parameter, and functionality is extracted from both signals.