An intelligent learning approach for improving ECG signal classification and arrhythmia analysis

Area of Science:

• Biomedical Engineering
• Signal Processing
• Cardiology

Background:

• Timely recognition of cardiac arrhythmias is crucial for preventing sudden cardiac death.
• Electrocardiogram (ECG) signal analysis is a cornerstone of cardiac arrhythmia diagnosis.
• Existing methods may face challenges with signal quality and classification accuracy.

Purpose of the Study:

• To develop a comprehensive framework for enhanced ECG signal analysis.
• To improve the accuracy and speed of cardiac arrhythmia classification.
• To investigate the application of Internet of Medical Things (IoMT) in arrhythmia recognition.

Main Methods:

• ECG signal quality enhancement using a dedicated filter combination for noise suppression.
• Feature extraction employing a specialized wavelet design.
• Cardiac arrhythmia classification using a Hidden Markov Model (HMM) for Normal (N), Right Bundle Branch Block (RBBB), Left Bundle Branch Block (LBBB), Premature Ventricular Contraction (PVC), and Atrial Premature Contraction (APC).
• Extraction of key statistical features: minimum, maximum, mean, standard deviation, and median.

Main Results:

• The proposed model achieved an overall accuracy of 99.7%.
• Sensitivity was recorded at 99.7%, with a positive predictive value of 100%.
• The detection error rate was remarkably low at 0.0004.

Conclusions:

• The developed framework provides a highly accurate and efficient method for cardiac arrhythmia detection.
• The study demonstrates the potential of advanced signal processing and machine learning techniques in cardiology.
• Integration with IoMT offers promising avenues for remote and real-time cardiac monitoring.