Author_Institution :
Dept. of Electr. Eng., Nat. Central Univ., Jhongli, Taiwan
Abstract :
This paper proposes a simple and new method to analyze the electrocardiogram (ECG) signals for diagnosing cardiac arrhythmias utilizing the Mahalanobis distance method. It can accurately classify and distinguish the difference between normal heart beats (NORM) and abnormal heart beats. The illnesses with abnormal heart beats may include the following: left bundle branch block (LBBB), right bundle branch block (RBBB), ventricular premature contractions (VPC), and atrial premature contractions (APC). Analysis of the ECG signals consists of three major stages: (1) detecting the QRS waveform; (2) the qualitative features extraction; and (3) illness case determination. In these experiments, the sensitivity achieves 98.28%, 90.35%, 86.97%, 92.19%, and 94.86% for NORM, LBBB, RBBB, VPC, and APC, respectively. The average accuracy rate of all experiments is about 93.57%.
Keywords :
biomechanics; diseases; electrocardiography; feature extraction; medical signal detection; medical signal processing; APC; ECG beats; LBBB; RBBB; VPC; abnormal heart beats; atrial premature contractions; diagnosing cardiac arrhythmias; electrocardiogram signals; feature extraction; left bundle branch block; mahalanobis distance; normal heart beats; right bundle branch block; ventricular premature contractions; waveform detection; Cardiac disease; Cardiology; Electrocardiography; Feature extraction; Heart beat; Heart rate variability; Humans; Information analysis; Signal analysis; Spatial databases;
