Cardiac Arrhythmia Classification by Multi-Layer Perceptron and Convolution Neural Networks

Area of Science:

• Cardiology
• Biomedical Engineering
• Artificial Intelligence in Medicine

Background:

• Electrocardiogram (ECG) is crucial for diagnosing cardiovascular diseases by recording heart electrical activity.
• Arrhythmia, an irregularity in heart rhythm, encompasses conditions like tachycardia and bradycardia, necessitating accurate detection methods.
• Existing diagnostic tools for arrhythmia can be improved with advanced computational approaches.

Purpose of the Study:

• To develop and evaluate deep neural network (DNN) algorithms for distinguishing various types of cardiac arrhythmias from ECG signals.
• To implement Multi-layer Perceptron (MLP) and Convolutional Neural Network (CNN) models using TensorFlow for arrhythmia classification.
• To compare the performance of the proposed DNN algorithms against existing methods in terms of accuracy, sensitivity, and precision.

Main Methods:

• Utilized ECG datasets from PhysioBank.com and Kaggle.com for model training, testing, and validation.
• Developed an MLP model with four hidden layers, including weights and biases.
• Designed a four-layer Convolutional Neural Network (CNN) architecture for ECG signal analysis.
• Employed Python and the TensorFlow library for implementing the deep learning models.

Main Results:

• The proposed MLP and CNN algorithms effectively mapped ECG samples to different arrhythmia classes.
• The developed deep learning models achieved high accuracy in classifying various arrhythmias.
• The algorithm's performance, in terms of sensitivity and precision, exceeded that of current state-of-the-art methods.

Conclusions:

• Deep neural networks, specifically MLP and CNN, offer a powerful approach for automated arrhythmia detection from ECG data.
• The proposed TensorFlow-based algorithms provide a promising tool for improving the accuracy and efficiency of cardiovascular disease diagnosis.
• This research highlights the potential of AI in enhancing the capabilities of cardiologists for better patient outcomes.