State-of-the-art Artificial Intelligence (AI) methods are progressively strengthened in Traditional Chinese Medicine (TCM), aiding physicians to make comprehensive clinical decisions. One of the well-known proven examinations in TCM i.e., hesitant pulse wave diagnosis, is a sign that the blood circulation of a person is sluggish and can be used to provide a preliminary diagnosis for physiological problems. Modern AI method such as artificial neural networks, achieves better performance than traditional methods, but the interpretability to understand the final decisions are hard to explain. In clinical situations, an easy-to-understand diagnosis is essential to be provided to the patients when selecting the appropriate clinical treatment. Therefore, this study presents feature extraction and clinical decision support systems based on Pulse-Line Intersection (PLI) and explainability AI (XAI) methods. The pulses were recorded from 34 patients in 6 different measurement points for 6 seconds. In addition, a comparison of several AI methods was provided to classify hesitant and normal waves. The contribution of each feature in the classification process was analyzed by unboxing each AI model parameter. The results revealed that all models performed comparably, evaluated using Area under the Curve (AUC) values on the test dataset: logistic regression AUC was 0.83; tree boosting AUC was 0.81; MLP AUC was 0.83. This work suggests that modern AI methods can provide comprehensive explainability which is compatible with traditional method rankings.
