Detection of Heart Arrhythmia on Electrocardiogram using Artificial Neural Networks

Area of Science:

• Cardiology and Biomedical Engineering.
• Utilizes advanced signal processing techniques for cardiac diagnostics.

Background:

• The electrocardiogram (ECG) is a crucial tool for assessing heart structure and function.
• Cardiac arrhythmias, or rhythm disorders, can stem from environmental and genetic factors.
• ECG signals reflect cardiac irregularities, enabling arrhythmia detection.

Purpose of the Study:

• To develop and evaluate a high-accuracy method for diagnosing heart arrhythmias using ECG signals.
• To leverage signal processing and machine learning for improved cardiac rhythm disorder identification.

Main Methods:

• ECG signals from healthy individuals and those with arrhythmias were segmented into 10-minute intervals.
• Feature vectors were constructed using the arithmetic mean of wave and interval data.
• A multilayer perceptron neural network classifier was employed to identify arrhythmias based on feature vectors.

Main Results:

• The proposed method demonstrated high classification accuracy in diagnosing arrhythmias from ECG data.
• ROC analysis and contrast matrices confirmed the effectiveness of the ECG-based classifier.

Conclusions:

• The developed technique effectively diagnoses heart arrhythmias using ECG signals and a multilayer perceptron neural network.
• This approach offers a reliable tool for arrhythmia diagnosis, enhancing cardiac care.