A Hybrid CNN-LSTM Model for Accurate Prediction of Cardiac Arrhythmia using ECG Signals

Cardiac arrhythmia is a significant health concern, necessitating timely and precise diagnosis to prevent life-threatening complications. This study investigates the potential of machine learning and deep learning techniques for arrhythmia classification using the MIT-BIH Arrhythmia dataset, which includes five types of annotated electrocardiogram recordings. The Synthetic Minority Over-sampling Technique (SMOTE) was used to handle the class imbalance in the dataset.We propose a hybrid model that integrates Convolutional Neural Networks (CNN) with Long Short-Term Memory (LSTM) networks and benchmark its performance against traditional classifiers, such as K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Decision Trees, and Stochastic Gradient Descent (SGD). The CNN-LSTM hybrid model outperformed the comparative classifiers, achieving an accuracy of 98% and a Matthews Correlation Coefficient (MCC) of 94%. Detailed evaluation through confusion matrices and classification reports demonstrates the model's robustness and suitability for arrhythmia detection. The promising results suggest that this CNN-LSTM model may be viable for real-time arrhythmia monitoring, potentially enabling continuous heart monitoring in wearable devices or integration into clinical applications.