Extreme Learning Machine for Heartbeat Classification with Hybrid Time-Domain and Wavelet Time-Frequency Features

Area of Science:

• Biomedical Engineering
• Artificial Intelligence in Healthcare
• Signal Processing

Background:

• Cardiovascular diseases are a leading cause of mortality globally.
• Electrocardiogram (ECG) analysis is crucial for diagnosing heart conditions.
• Existing heartbeat classification methods struggle with high dimensionality and slow processing.

Purpose of the Study:

• To develop an efficient and accurate automated heartbeat classification system.
• To address the limitations of existing feature extraction and classification techniques.
• To improve the speed and accuracy of diagnosing cardiac arrhythmias using ECG.

Main Methods:

• Hybrid feature extraction combining RR interval (time-domain) and wavelet time-frequency features.
• Utilizing Extreme Learning Machine (ELM) for multi-class heartbeat classification.
• Implementation and testing on the public MIT-BIH arrhythmia dataset for 16-class classification.

Main Results:

• The proposed ELM approach achieved superior classification accuracy.
• Demonstrated significantly faster training and recognition speeds compared to existing algorithms.
• Successfully performed 16-class heartbeat classification on the MIT-BIH dataset.

Conclusions:

• The hybrid feature extraction and ELM classification method offers an efficient solution for automated heartbeat classification.
• This approach holds promise for timely diagnosis and prevention of cardiovascular diseases.
• The method provides a balance of high accuracy, fast processing, and good generalization ability.