Identification of Bundle Branch Block Using Wave Analysis of ECG Signals

Electrocardiogram (ECG) is the most common method to detect heart diseases/disorders. In this research, we have employed ECG signals to detect left and right bundle branch blocks in patients by analyzing certain factors that affect and differentiate them from the ECG report of a normal person. A web-based application can make it easier and more accurate than manual or other means to detect the disorder. In this research, we evaluate all the influencing factors to detect Bundle Branch Blocks with the help of Python making it more efficient and precise due to its vast extent of signal analysis libraries (SciPy, NumPy, WFDB, Matplotlib, and Seaborn, etc.). By the use of these signal processing methodologies feature extraction and inference models, we can easily identify major markers of Bundle Branch Block which are: QRS complex duration detection (between 80–100 ms or >120 ms), analyzing peaks in the signal waves, ‘M’ and ‘W’ notches detection in the leads and absent or inverted T wave. Besides, in this analyzing part of this research we convert the ECG signal image (e.g. .png) file to .hea, .dat, and .atr files for detection so, that if anyone has even a little bit of trouble breathing or any type of restlessness people can have their ECG tested for at least these disorders in a real-time environment. Results indicate that this approach achieves high accuracy and computational efficiency, providing a scalable solution for clinical use. This paper outlines the workflow, experimental findings, and potential applications in modern healthcare.