Exploring the potential of Machine Learning for diagnosis of Atrial Fibrillation
Paper Key : IRJ************488
Author: Mohsin Imam,Sufiyan Adam
Date Published: 04 Jan 2023
The most prevalent type of arrhythmia (Greek a-, loss + rhythmos, rhythm loss of rhythm) that results in hospitalisation in the United States is called atrial fibrillation (AF). Although atrial fibrillation can occasionally go unnoticed, it is linked to a higher risk of heart failure and stroke in individuals, as well as a lower quality of life in terms of overall health (HRQOL). The yearly cost of treating AF is estimated between $6.0 to $26 billion for the American healthcare system 1. Early detection of atrial fibrillation (AF) and therapeutic intervention can help patients have fewer symptoms and have better health-related quality of life (HRQOL), all while saving money on medical expenses. However, an electrocardiogram (ECG) that was only recorded at one moment in time is used as the standard test for identifying atrial fibrillation (AF). This approach provides no information on the relationship between the symptoms and AF or cardiac rhythm. Due to the democratisation of health monitoring 2 and the introduction of powerful computers in the last ten years, Machine Learning algorithms have been demonstrated to be helpful in identifying AF from the ECG of patients. The symptoms of atrial fibrillation (AF), its diagnosis, and the possibility for further study on the subject are all summarised in this article.