Traditional Chinese medicine(TCM) has existed for more than two thousand years and one of the important diagnostic methods is the pulse diagnosis. It generally takes decades of training for a practitioner to master this skill as pulse acquisition and diagnosis require long-term experiences and are very subjective. The project aims to use the combination of advanced sensor technology and artificial intelligence to emulate the TCM practice for health monitoring. A flexible piezoelectric film is designed to record the wrist pulse data. A group of representative pulse features has been extracted from the volunteer data and fitted into a machine learning algorithm. The trained machine learning model has been proved to be effective in telling different health conditions of volunteers.