Deep Convolutional Neural Networks for Classifying Body Constitution

Body constitution is a classification of individuals into different types of physical condition in order to prevent disease and promote health. The problem of standardizing constitutional classification has become a constraint on the development of Chinese medical constitution. Traditional recognition methods, such as questionnaire and medical examination have the shortcoming of inefficiency and low accuracy. We present an advanced deep convolutional neural network (CNN) to simulate the function of pulse diagnosis, which is able to classify an individuals constitution based only his or her pulse. The CNN model employed the latest activation unit, rectified linear unit and stochastic optimization. This model takes the lead in trying to classify individual constitution using CNN. During the experiment, the CNN model attained a recognition accuracy 95% on classifying 9 constitutional types.