Treadmill Exercise Test (TET) application in terms of detecting Obstructive Coronary Artery Disease (OCAD) is defined to be limited due to low sensitivity and specificity rates.1 Despite elaborative knowledge in this situation, TET still protects its position in daily clinical practice owing to its widespread utilization and significant data content.2 Yet, 1 of the main limitations of TET is the unreasonable high false positive rates.3 To improve TET's utility and accuracy in the diagnosis and follow-up of CAD patients, new algorithms need to be proposed. In recent years, artificial intelligence (AI) based systems have come to be recognized as promising models for detecting cardiovascular-related diseases.4, 5, 6 In a recent study, a machine learning (ML) system was presented to improve TET success for CAD testing.7 They presented 5 models exhibiting diagnostic performance compared to conventional TET. ST-segment depression has been the main electrocardiographic (ECG) finding applied in the model of this study. One of the most common ECG changes in daily clinical practice is an ST-segment change. A total of 93 features were gathered and reduced to 30 features using feature selection methods in the study. Comparing the most successful model to the traditional TET, performance improved by 13% and specificity increased by 20% when clinical features were added. However, more accurate, sensitive, and specific AI-based algorithms are needed in order to take maximal advantage of TET in daily clinical practice. Furthermore, new parameters other than ST-segment depression alone should be proposed as a marker of significant ischemia with higher accuracy.8 A deep learning based study shows that AI-based systems can learn the relation between diseases and ECG signal features.6 However, there is a need for a large sample size for deep learning, and they produce stochastic results. ML-based systems generate more deterministic and trust-worthy outputs than deep learning-based systems. Currently, there is a lack of evidence on the feasibility of using AI-based algorithms for OCAD detection using TET results. Thus, we aimed to evaluate the performance of a new algorithm which was based on using ECG signal features such as time difference and amplitude values of P, QRS and T wave. P, QRS and T wave directly follow the cardiac cycle of the heart. For this reason, time and amplitude information of P, QRS, and T waves can improve success of TET results for OCAD detection.

To the best of our knowledge, this is the first ML study to predict OCAD using time and amplitude features of P, QRS, and T waves. In this study, our main contributions to the literature: (1) ECG signal dataset in JPEG format with extracted time and amplitude features of P, QRS, and T waves from obstructive and non-obstructive CAD patients, (2) Five ML models that were trained to predict OCAD using features extracted from the V5 signal, (3) Comparison of these models' performance against the performance of 17 cardiologists.