Cardiac Arrhythmia - Prediction and Classification using Support Vector Machine

Untreated, arrhythmia is a potentially life-threatening condition that can lead to severe illness in patients. Early detection of heart arrhythmias can save lives. Based on ECG recordings. New advances in various types of equipment and machine learning models provide an opportunity to use computer-aided diagnosis (CAD) to make ECG classification procedures applicable in a real clinical setting. This allows classification of electrocardiographic arrhythmias between patients. The objective of this paper is to design and evaluate an innovative cross-patient approach for analyzing extensive electrocardiogram datasets, with the goal of enhancing the identification of rare arrhythmia We study four supervised machine learning models: support vector machine (SVM), k-nearest neighbor (KNN), random forest (RF), and a suite of three methods. We will test the effectiveness of these strategies in classifying: normal beaters (NOR), left leg blockers (LBBB), right leg blockers (RBBB), failure of atrial complexes (PAC) and premature ventricular complexes (PVC), next. segmenting and normalizing the data, obtaining patients' actual ECG data from MIT-DB, and measuring four parameters: precision, accuracy, recall, and f1 score. The results of the experiment show that, without the use of complex data preprocessing or engineering methods, the SVM classifier outperforms other methods using our patient interviews in all cases. An achieved Calculation value accuracy is of 0.84; is still very important. ECG is important in using the arrhythmia classification model. This approach is more accurate in clinical settings. where multiple ECG signals are collected from different patients.