Automatic Assessment System of Pulse Diagnosis for Treatment Effect Evaluation

Pulse Diagnosis (PD) is a well-established diagnostic technique used in Traditional Chinese Medicine (TCM). The positions for PD are the radial arteries in the left and right wrists, and the pulse assessment is done by placing index, middle, and ring fingers on each wrist. The pulse of patients is then measured in the superficial, middle, and deep levels by TCM doctors to obtain more information regarding energy or Qi imbalance of the patients. For modern medical researches, PD is also used for indicating some status of patients. However, it is a challenging issue for all medical staff to properly assess the pulse of various patients. Therefore, this work proposes a systematic scheme to measure patient pulse by Photoplethysmography (PPG) sensors. An evaluation framework of some treatments, e.g., acupuncture therapy, is then built on top of the PPG sensing system with machine learning solutions. In our implementation, sensor data will be collected and transferred from Arduino platforms to our server for the PD assessment. The effect of the evaluated treatments will be estimated by the analysis of the PD results with machine learning supports. The major performance index is on the accuracy of the assessment for the treatments.