The objective of our study is to enhance the accuracy of the Q-wave detection to handle abnormal ECG signal patterns more accurately. Q-wave detection is a significant factor for segmenting the cardiac vibrational signal for coronary artery disease (CAD) detection. By identifying the abnormal ECG signal segment (e.g., caused by Afib, premature contraction, movement artifacts, and other unknown sources), we can condition the CAD detection algorithm accordingly.
Towards that objective, we developed a novel and efficient method to reveal ECG signal characteristics, such as the P-wave, QRS, T-wave and PVC. Different from most conventional machine learning approaches (“black box” approach), our method is transparent and the results can be interpreted intuitively by experts in the fields. We believe such methods are novel and practical which can make an impactful contribution to the science community. After completion of our study, we will submit a scientific paper in late spring / early summer of 2023.