Research On Noninvasive Blood Glucose Measurement With Simulate Sample By NIR Spectroscopy

Blood sugar is glucose in body blood which is an important substance in human body. Concentrations of glucose reflect people's health. So, the veracity of glucose measurement in human body is very important. Blood glueose measurement technique is one of the popular lessons in instrument and life science fields. Near Infrared Spectroscope analytic technique(NIRS),has some quality characters such as high speed,good precision,reagent-free,non-invasive,naturally near infrared non-invasive blood glucose measurement is one of the important issues in this filed. It will make aqualitative progress in the method of diabetes measurement,and also has a practical meaning to the treatment of this disease.In our paper, first, we have made an investigation about the character of glucose in near infrared spectrum and used diffuse reflection spectrum (DRS).Put the glucose powder in sample cell, we can get the near infrared diffuse reflection spectrum plot .From the plot, we can see that the reflect of glucose powder was higher in 1000～1410nm wave band, and the reflective index was more than 20%; the reflective index in 1500～2000nm wave band was fluctuated; if it was inverted in to absorbance plot, the absorbance in 2000～2500nm wave band would been higher. Above we found were same to other studies. The determination in near infrared spectrum of glucose powder would build up many practical experience for glucose solution and blood glucose. The determination in near infrared spectrum of glucose powder made we known glucose has absorption peak in near infrared spectrum,but the position we can not been sure. So, next studied the different concentration of glucose solution.In this paper, The NIR character of glucose powder and the removing water absorption ways are analyzed. The spectroscopy range, the spectrum data pretreatment and the chemomtrics method are discussed in detail for the glucose solution. 1400-1800nm and 2000-2500nm containing glucose absorption are selected for the spectrum analysis region, baseline correction, smoothing and differentiate are chosen as the pretreatment ways and Partial Least Squares(PLS) is the method to build calibration model. The models are built for glucose solution and protein mixed solution, and the region for the model built and the method for the data pretreated are confirmed again. The results suggested that on the wave bands from 2 000 to 2 500 nm, after smoothing, baseline and derivative, The best correlation values for glucose were 0.998 and the best RMSEP values were 0.045. The calculated collection samples were used to detect the protein mixed solution model. The result reaches the same level that received by using near-infrared spectrum in others countries. It makes a huge progress in the region of use near-infrared spectrum to measure people's blood sugar. A fundamental foundation of noninvasive measurement of blood sugar concentration has been established.