Atrial Fibrillation (AFiB) is the most prevalent kind of cardiac rhythm problem, and it is closely correlated with other heart-related disorders, including a high mortality rate and an increased risk of strokes. The early diagnosis and treatment of AFiB are of critical importance from a clinical standpoint. Automatic identification of AFiB based on electrocardiogram data assists in accurate and timely detection of the condition, which is a difficult undertaking. In recent years, with the growth of artificial intelligence, the Deep Learning models have gained initial success in ECG data processing, particularly in the identification of AFiB. In the proposed approach, the objective is to devise a Deep Learning approach, based on CNN, with data augmentation technique to deal with imbalance data, as well as the Savitzky-Golay filter used for noise removal, in order to accurately predict AFiB from ECG signals. By using this approach, the ECG signals are classified into three classes, AFiB, Normal, and other with an average accuracy of 97.87%. In addition, the performance of the suggested model has also been evaluated using several evaluation matrices to determine how well the model performed. The suggested study has the potential to assist clinicians in the detection of common AFiB in real time on routine electrocardiogram monitoring.
Shoukat et al. (Sun,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: