The Analysis Of Biomedical Signal Pattern Based State Representation Methodology

the analysis of biomedical signal pattern has been an important medical signalprocessing problems. Along with the development of mobile medical technology, digitaltechnology, in the past few years, problems and methods to the study of biomedical signalare increased. More and more big companies pay their attention to its new applicationplatform and new use way. In the near future it will be more and more applied in communityhealth care, family health care, clinical care, etc. However, the characteristics of biomedicalsignal determines the complexity of the application and the difficulty of the popularizationand application. In a medical information system, medical signal processing including manyapproximate continuous periodic signal, this kind of Biomedical Signal Pattern (BSP)including such as ecg, pulse, heart sounds, breathing, and so on. In recent years, manyscholars have always been studying the medical signal acquisition and analysis system thatcan be used to the general family. In this type of computer aided medical system[1], there arethree basic tasks:（1）The location and segmentation of signal in time domain or frequency domain（2）The classification and recognition of signal based on signal analysis and study（3）The clinical diagnosis and evaluation based on medical statistical analysisIn the frequency domain, in practical applications often need to determine the frequencyand frequency variation, frequency distribution, and other issues. And in the time domain,judgment of periodic time domain and the statistics of each cycle have commoncharacteristics, to find out some characteristics of the signal and the characteristics ofclassified information. Determine, for example, the complexity of the QRS waves in ecg,arrhythmia, and analysis of abnormal heart sounds is often necessary. Among them, the signalsegmentation is an important basic work. In practice, the signal, after the completion ofsegmentation model description are clearly the next is an important task, and then patternevaluation can be completed. Based on the comparative analysis of traditional evaluation method in this paper, at thesame time, raises the method (SRM) with a state analysis of a new method of BSP. ThoughSRM began as a form of assessment of structural health monitoring system (SHMS) putforward by the new evaluation method, but this method also can be applied to BSP. Firstsimply introduces several traditional methods of comparison, also compared with somemainstream methods of SRM is similar, mainly including support vector machine (SVM),response surface method (RSM), etc. This paper focuses on how to analyze BSP in SRM,descriptions of BSP is one of the most important question. Heart sounds, for example, thearticle uses the method of[2]can realize the segmentation of heart sound signals, then,characteristics analysis of heart sound signals can be sliced frequency wavelet transform(FSWT), extraction time and frequency characteristics of the signal. Through SRM modelingmethod, which can be further evaluation model is set up, from the description of the model wecan compare the gap between people. This paper lists many problems and methods of thestudy, at the same time for the main steps in the application of SRM to BSP are introduced,combining with FSWT SRM can be used as analysis provide a new method of BSP.Results show that with the help of the method to describe the system status of the kernelfunction tools and measure tools, describe BSP angle of system to analyze, the classificationresults and clinical evaluation methods have good consistency, shows that the system statedescription method can be effectively applied to the evaluation of BSP, for the future of BSPevaluation provides a new thinking and method of the evaluation.