Detection of Cardiac Arrhythmias in Patients with Obstructive Sleep Apnea using Machine Learning and Deep Learning Algorithms

This paper presents a pioneering effort to understand the complex relationship between Obstructive Sleep Apnea (OSA) and cardiac arrhythmias, utilizing Machine Learning (ML) and Deep Learning (DL) techniques. OSA, a widespread sleep disorder, is known to influence cardiac health. The approach involves designing ML algorithms and DL models trained on extensive Electrocardiogram (ECG) datasets. These models detect subtle patterns in data, helping us understand the specific traits of arrhythmias in patients with OSA. The paper is validated through rigorous testing, evaluating the accuracy, precision, F1 score, and recall of the developed models. It was seen that the Long Short-Term Memory (LSTM) model gives the best accuracy of 0.85 when compared to a random forest with an accuracy of 0.84 for the combined data of cardiac arrhythmia and OSA. Ultimately, the goal is to refine diagnostic precision, providing clinicians with a deeper understanding of the cardiac implications associated with OSA.