Arrhythmia Recognition and Classification Using ECG Morphology and Segment Feature Analysis.

Area of Science:

• Cardiology
• Biomedical Engineering
• Machine Learning in Healthcare

Background:

• Cardiac arrhythmias, indicated by abnormal heart electrical activity, require efficient recognition and classification.
• Accurate diagnosis of arrhythmias is crucial for effective patient management and treatment.
• Existing methods may have limitations in distinguishing between various types of arrhythmias.

Purpose of the Study:

• To propose a novel and accurate method for the recognition and classification of cardiac arrhythmias.
• To develop an automated system for diagnosing different types of heartbeats.
• To improve the detection sensitivity of supraventricular and ventricular ectopic beats.

Main Methods:

• Segmentation of P-QRS-T waves from electrocardiogram (ECG) waveforms.
• Extraction of morphological and ECG segment features using Principal Component Analysis (PCA) and Dynamic Time Warping (DTW).
• Classification of arrhythmias using Support Vector Machine (SVM) on extracted features.

Main Results:

• The proposed method achieved an overall accuracy of 97.80% in classifying four types of heartbeats: normal (N), supraventricular ectopic beats (SVEBs), ventricular ectopic beats (VEBs), and fusion (F).
• High sensitivity and positive predictivity were reported for each class, with notable improvements in detecting SVEBs and VEBs when combining proposed features.
• The algorithm demonstrated superior performance compared to four other peer algorithms in classifying cardiac arrhythmias.

Conclusions:

• The developed method provides an accurate and efficient approach for automated cardiac arrhythmia diagnosis.
• Combining morphological and segment-based ECG features enhances the detection of specific arrhythmia types.
• This novel technique shows significant potential for clinical application in arrhythmia monitoring and diagnosis.