A Blood Glucose Detecting System Based On Three Wavelengths In Near-infared Region

The Near-infrared region of optical spectrum extended from 0.7 to 2.5μm and can be used for quantitative measurement of organic functional groups, especially C-H,O-H,N-H and C=O. Analyzing sample concentration using near infrared spectra focused on the designing of optical sensor and the spectral data processing. The near-infrared continuous non-invasive method made a beam of light irradiated the blood region of human, then extracted the information of blood glucose from the spectrum. The key was how to improve the ratio of signal to noise then we can be distinguished the very low glucose absorption. In this thesis, we studied the system design and the application based on the system.According to the absorption of glucose, a system based on three wavelengths in near-infrared region was designed. The system included the light source, optical chopper, the detector and the lock-in amplifier several important parts. We chose the signal wavelength 1610nm by glucose absorption peak in the overtone band and the reference wavelengths 1200/1350nm was to eliminate the interference effect. The detector used the ring light bundle, in this case, the intensity of incidence light was improved largely, and the ratio of S/N was improved too. To minimize the effect of background scattering, the optimum source-detector distance 3-5mm was given by the equation of photon diffusion.In the normal human physiological blood glucose range: 0-500mg/dL, we have done three different concentration interval experiments, the result were: when the interval was 100mg/dL, the partial least square method to analysis, the relative coefficient (R) was 0.998, and the root mean square error of prediction (RMSEP) was 17.08mg/dL; the interval is 50mg/dL, the R and RMSEP were 0.907 and 54.52mg/dL respectively; lastly, the interval was 20mg/dL, the R and RMSEP were 0.959 and 23.22mg/dL.