Computational Framework for Prediction of Cardiac Disorders by Analyzing Ecg Signals Using Machine Learning Technique

The clinical diagnosis of heart disorders relies heavily on electrocardiograms (ECGs). Numerous abnormalities in heart are being identified with a record of heart signal throughout intervals. Monitoring and diagnosing ECGs signals in daily life are appearing recently due to an increase in healthcare equipment. This paper presents a novel computational framework for detecting heart disorders by analyzing the ECG signals using machine learning technology. The manual interpretation of ECGs by medical experts is time-consuming and subject to human error. This study presents a comprehensive review and implementation of an automated ECG classification system utilizing artificial neural networks (ANNs) for efficient and accurate cardiac condition identification. The proposed ECG classifier employs a multilayered ANN architecture, trained and validated using a diverse dataset of annotated ECG recordings. Preprocessing techniques, including signal denoising, feature extraction, and data augmentation, enhance the quality and relevance of input data for the ANN. The selected features encompass morphological and temporal attributes, essential for distinguishing normal and abnormal ECG patterns. In this paper, noise removal from input ECG dataset is performed by adaptive filter technique and baseline wander is also removed. Machine learning in ECG classification is done by ANN that allows to use less energy while still providing accurate classification. MATLAB software is employed to carry out this work and corresponding outputs are obtained for ECG classification.