Bundle branch block detection using statistical features of QRS-complex and k-nearest neighbors

Electrocardiogram (ECG) is a non-invasive method primarily used by physicians for the diagnosis of various heart disorders. Bundle Branch Block (BBB) is a cardiac disorder, occurs due to blockage in the flow of electrical impulses of the heart. This irregularity can be observed on the ECG. The proposed work presents a technique to automatize the detection of BBB. The proposed technique uses the statistical features calculated from the extracted QRS-complex to classify between a normal, left bundle branch block and a right bundle branch block beat. Performance of the proposed technique analyzed using the standard MIT-BIH Arrhythmia Database has an overall Se=98.48 %, Sp=99.30 %, +P= 98.48 %, and Ac=99.05 %. The results show that the proposed technique is an accurate and reliable method for automatic detection of BBB.