
Baraah Al Dwa
Sechenov University, Russian FederationTitle: Recognition algorithm of cardiac arrhythmias based on a single-channel electrocardiogram and photoplethysmography using machine learning
Abstract
Arrhythmias are very common with a prevalence of 2-5% worldwide. Early detection of arrhythmias can prevent complications such as thromboembolic events and even death. Screening for abnormal heart rhythms has been recommended, but population implementation is yet to be carried out. Nowadays, there has been a rising interest from researchers around the world to develop machine learning based algorithms to detect diseases in cardiology and especially in cardiac electrophysiology. The most common task of ML models in cardiology is to detect cardiac arrhythmias. The availability of large amount of digital data and more powerful computing technologies have made autonomous ECG interpretation using Ml more applicable. Moreover, ML algorithms can be embedded in smart portable ECG devices which will help more people have early screening of arrhythmias. Algorithms that relay on PPG only have limited specificity due to noise and ECG is the gold standard for diagnosis of arrhythmias. This study aims to develop ML based algorithms to detect atrial fibrillation, premature supraventricular beats and premature ventricular beats using both single lead ECG and PPG.
Biography
Baraah Al Dwa has first degree in general medicine, Doctor of medicine and as an international student has started her postgraduate studies in Cardiology at Institute of Personalized Cardiology, Sechenov University. She is an ECFMG certificate candidate and a newly growing researcher.