A pulse signal processing system for analysis of tumor sickness and syndrome

Pulse signal checking is one of the four Chinese traditional medical diagnosis methods. But few of doctors can use these techniques now in practical diagnosis because of difficulty of understanding. In this paper, a modern pulse signal processing system based on TMS320C6713 DSP for tumor sickness diagnosis and syndromes analysis is described. Pulse signal is sampled with 300 Hz frequency and 8 bits quantization, then saved in a data buffer with length of 10 Kbytes to be processed by DSP. Linear predictive cepstrum coefficient is selected as features and vector quantization is hired for model training. The system is designed to be used as a tumor sickness diagnosis and syndrome analysis assistant in practical application.