Processing of ECG Signal Using Optimized Feature Selection for the Prediction of Irregular Heart Beat

The categorization of Electro Cardio Gram (ECG) trials into significant arrhythmia programs to identify abnormal heartbeats is presented in this study as an effective hybridized method. The most frequently used and recognized detection technology to display heart fitness is the physiological detection utilizing ECG data. Arrhythmia beat categorization also plays a significant part in ECG analysis that is intended to clarify cardiac well-being position when investigating heartbeat. Artificial Neural Networks (ANN) is deployed for the segmentation of input ECG images. The proposed work uses Fourier transform to remove the intrinsic sound of ECG signals during the preprocessing step in order to properly categorize ECG samples into main arrhythmia classifications. Chicken Swarm Optimization (CSO) is employed for the effective feature selection which is followed by CNN based classification. This demonstrates how the suggested ECG classification model is successful in correctly classifying ECG signals for abnormality detection along with IoT based