The pulse diagnosis is one of the most important method for diagnosing diseases in the traditional Chinese medicine. In the past, the development of pulse diagnosis tools was first by single-point sensing, but there was only a single amount of information that a lot of information was missed. At present, although the amount of information is large with a multi-point sensing method, the multi-point sensing pulse instrument is expensive and is not portable. In combination with the above, combined with the advantages of single-point sensing and multi-point sensing, we want to know the way of single-point sensing that is it possible to increase the amount of information by other sensing methods to improve the correctness of the pulse? However, there is a product on the market that measures the disease by measuring the skin resistance (Galvanic Skin Response, GSR). We know that the pulse diagnosis and skin resistance measurement can judge the disease. In this paper we assume that there is a correlation between pulse diagnosis and skin resistance. Therefore, we want to classify the pulse by a photoplethysmography (PPG) signal and a skin resistance sensing method. Because the mobile phone is very popular, the camera lens on the mobile phone can capture the PPG signal. Therefore, using the mobile phone as a platform that the PPG signal is taken by the original camera lens, and the additional GSR information is used to make the pulse diagnosis instrument based on these two signals, and try to classify the pulse image. In our experiments, the feature of the resonance theory C1-C10 by Professor Wang W.K, the feature of time domain and the theory of meridian feature through the support vector machine (SVM) can effectively classify the normal pulse and the wiry pulse from our collect data. The recognition rate is as high as 94.5456%. Therefore, we can see from the results that the correlation between these two signals is indeed existing, and each has its own characteristics to improve the correctness of the positive pulse.


Prototype Demo

Display the .csv feature file

Download .csv file Link : https://lens.csie.ncku.edu.tw/~Platform/UserInformation.php>