Providing an Efficient Algorithm for Finding R Peaks in ECG Signals and Detecting Ventricular Abnormalities With...

Area of Science:

• Cardiology
• Biomedical Engineering
• Signal Processing

Background:

• Ventricular arrhythmias are a leading cause of global mortality, necessitating accurate early diagnosis.
• Electrocardiogram (ECG) signal analysis, particularly R-peak detection, is fundamental for assessing cardiac health.
• Identifying specific arrhythmias like ventricular tachycardia (VT) and premature ventricular contractions (PVCs) is critical for patient outcomes.

Purpose of the Study:

• To develop and validate a robust method for detecting various ventricular arrhythmias from ECG signals.
• To improve the accuracy and efficiency of diagnosing life-threatening heart rhythm abnormalities.

Main Methods:

• A hybrid algorithm combining Pan-Tompkins and state logic machine for R-peak detection in ECG signals.
• Analysis of morphological features within the QRS complex to differentiate between normal and abnormal heartbeats.
• Classification of arrhythmias using a Support Vector Machine (SVM) classifier with a Gaussian kernel.

Main Results:

• High classification accuracies were achieved for normal sinus rhythm (100%) and various ventricular arrhythmias: VT (95.8%), VFL (92.8%), VFI (94.5%), VEB (98.9%), and PVCs (91.5%).
• The combined algorithm effectively distinguished between healthy and abnormal heartbeats by comparing R-peak detection results.
• The SVM classifier demonstrated strong performance in categorizing different types of ventricular abnormalities.

Conclusions:

• The proposed R-peak detection and SVM-based classification method offers a reliable approach for diagnosing ventricular arrhythmias.
• Accurate identification of these arrhythmias can significantly aid in preventing mortality associated with cardiac conditions.
• This technique holds promise for improving real-time cardiac monitoring and diagnostic tools.