The Theoretical And Experimental Study On The Dynamic Spectrum In Non-invasive Detection Of Blood Component

good will of the human being. The realization of this method will replace the traditional invasive method. It will release the pain of many patients. But it is very difficult to use this technology to the clinic, for the near infrared spectrum is disturbed by many factors, and the precision of the final result is restricted. In the NIR non-invasive detection, it is very hard to measure the influence of the individual to the spectrum, so it is the key of the NIR non-invasive detection whether it can eliminate the influence of the skin and other tissues.In this dissertation, a series of reasearch is taken which focus on eliminating the influence of the skin and other tissues:①In this dissertation, the approach of the Dynamic Spectrum in the frequency domain was first proposed, it is based on Photo-plethysmography (PPG) with fast Fourier transforms. Evaluating only the pulsatile part of the entire optical signal, this approach is rather independent of individual or time changes in scattering or absorption characteristics of the tissue. In this dissertation, a series of measures is taken, and high-precision Dynamic Spectrum in the frequency domain is got with the experiment. The approach of the Dynamic Spectrum is verified and this approch is very important in the non-invasive detection of blood.②The pulsatile spectrum is the base of the Dynamic Spectrum. In the course of collecting the pulsatile spectrum signal in vivo, it is inevitable to be interfused with yawp signals as respiration,high frequency interference, baseline drift and so on. Using the traditional adaptive filter, it is very difficult to collect the reference signal from the in vivo experiment. In this dissertation, Daubechies wavelet adaptive filter based on Adaptive Linear Neuron Networks is used to extract the signal of the pulse wave. This method can get better result than nonparametric results. This filter is found to be very effective in detection of symptoms from pulsatile part of the entire optical signal.③In this dissertation, the best experiment condition by which to extract the Dynamic Spectrum is discussed, it is discussed from the measurement position, integrate time, the contact press and the superficial condition respectively.④The outlier detection of the near infrared spectrum is reasearched deeply. Anew method to detect multiple outliers in multivariate data is first proposed in this dissertation. This method is the combination of Minimum subsets, Resampling and Self-organizing Map algorithm.⑤And in this dissertation we proposed another new method to detect multiple outliers in multivariate data in the NIR chemistry measurements research. That is the combination of Vector Length and Competitive Network. It is a robust way with Neural Network.The result of the experiment shows that the two method are simple, effective, intuitionistic and all the outliers in the spectrum can be detected in a short time. The two methods combined the advantage of the traditional robust method and the Neural Network.⑥In the NIR spectrum, the signal is overlapped, the peak of the spectrum is indistinct, and in this dissertation, the problem of overlapping peak is researched. For the influence of the wavelength selection to the model, the multiple of wavelet coefficient– box figure method is proposed. This method is simple and can improve the precision of the model.⑦In the research of non-invasive blood component concentration measurement, the influence of the scattering behavior of the blood on the measurement of the Dynamic Spectrum is discussed. In this dissertation, Monte Carlo method is used to analyses the scattering behavior of the blood, the influence of the scattering behavior of the skin tissue to the scattering behavior of the blood, and their influence to the Dynamic Spectrum.The Dynamic Spectrum method proposed a new idea to the NIR non-invasive detection.The feasibility of the technology is proved in theory. It has great value in this field. Theory and experimental results disclosed by this dissertation laid solid basis for the successful application of this method.