Atrial fibrillation is a cardiac disorder usually identified by rapid heart rhythm and irregular beating of a heart which leads to an increased risk of heart failure and cardiac stroke. In this paper, we propose the classification of normal and abnormal ECG signals using a Recurrent Neural Network and Long Short-Term Memory Network in Deep Learning. PhysioNet Challenge 2017 dataset is used for training and testing Neural Network. We train signals by applying various filters to attain high prediction accuracy. The total accuracy achieved by the method we used is 91% which is greater than the previous research/methods used to classify Atrial fibrillation signals.
₹10000 (INR)
NON IEEE-2022