A significant proportion of worldwide mortality is caused by cardiac diseases and atrial fibrillation (AFIB) is one of the most common cardiac arrhythmias among elderly population.
The demographics of western countries is alarming this health issue. It should be mentioned, that the consequences of AFIB is much more serious than AFIB itself since it may end up in serious heart failures and strokes. From an economic point of view, treatment of AFIB can be very expensive as well as challenging.
One of the most common ways for physicians to diagnose AFIB is through visual examination of the electrocardiogram (ECG) recordings. However, it is not always easy and in most cases cumbersome to analyse these big amounts of ECG data. Therefore, it is required to develop analytic software in order to automatically analyse the heart beats and rhythms which helps accelerating the process of detecting AFIB.
In this research, a novel algorithm to detect short episodes of AFIB using Machine/Statistical Learning and Signal Processing methods is being developed which paves the way to extend to real-time AFIB detection applications.
This project is supported by the Danish Innovation Foundation.