The Research On Recognition Of Atherosclerosis Based On Pulse Wave

Cardiovascular and cerebrovascular diseases caused by arteriosclerosis have become the main threat to the health of human beings. For there is a lot of information about the whole blood circulation system in the pulse wave, it is significant to study the pulse signal for the early prevention of arteriosclerosis. The arteriosclerosis analysis system designed in this paper can obtain the pulse signal of user, and estimate the degree of arteriosclerosis ultimately by processing the sampled signal.The whole system consists of pulse signal sampling hardware and arteriosclerosis analysis software. The pulse signal sampling hardware is basically composed by band-pass filter, amplifier and AD converter. The sampled pulse data is transmitted from the MCU to the upper computer by serial port for farther processing.The arteriosclerosis analysis software is programmed by LabVIEW, it is composed of four functional modules: pulse signal sampling, waveform pretreatment, feature extraction and arteriosclerosis state analysis. Pulse signals and user's information can be saved and displayed in the pulse signal sampling module. The waveform pretreatment module completes the baseline drift calibration of the pulse signal based on UWT algorithm, and than the single period signal can be located automatically. In the feature extraction module, all the parameters obtained from time domain analysis and frequency domain analysis are clustered by fuzzy ISODATA algorithm, then the parameters with better clustering result are chosen as the characteristic parameters for arteriosclerosis classification. A fuzzy classification system was designed, and the degree of arteriosclerosis can be estimated in the arteriosclerosis state analysis module. The fuzzy classification system is tested by the 50 samples in the built pulse database, and a good recognized result is achieved.