Research on TCM pulse condition identification using probabilistic neural networks

The key of objectivity of traditional Chinese Medicine (TCM) pulse condition lies on the objective test and correct recognition for all kinds of pulse conditions. In view of the ambiguity, variety and complexity of TCM pulse conditions, and the shortcomings of the traditional recognition methods and back propagation (BP) neural network method, a kind of TCM pulse-condition recognition method based on probabilistic neural networks (PNN) is put forward. In this research three experienced practitioners of Xi´an TCM Hospital gave us their expert advice and determined the percentage of correct recognition of our PNN method. For the twelve kinds of TCM pulse conditions, we attained an average recognition accuracy of about 93% with PNN method, better than the average recognition accuracy of about 75% attained with the traditional fuzzy cluster method and 87.1% with BP neural network method. The contrastive experiments of the BP neural network method and the PNN method are given. The results show that recognition accuracy of PNN method is out and away higher than that of BP neural network method in high noise circumstances.