Optimizing And Enhancing The Glucose Measurement Sensitivity For The Differential Near Infrared Spectroscopy

Differential near infrared diffuse reflectance spectroscopy based on position reference,which has been proposed in recent years by our research group,is one promising method for the long-term blood glucose monitoring since it can reduce the influence from the light source drift.For the glucose measurement using this method,it is necessary to acquire enough measurement sensitivity for the week glucose signals.This thesis focuses on investigating the rule of the measurement sensitivity changing with the used positions,i.e.the light source-detector distances(SDD),and seeking two optimum SDDs to do the glucose sensing.Moreover,we also propose a new method to improve the sensitivity furtherly by using multiple SDDs near the optimum SDD.The main work of this thesis is following:1)Referring the theory of optimum optical pathlength,the theory of optimum SDD for the differential near infrared diffuse reflectance spectroscopy based on position reference is proposed.The optimum two SDDs,including the measurement SDD and the reference SDD,were acquired to get an optimum measurement sensitivity.Using the diffuse equation for the infinite media,the distance of two optimum SDDs can be derived at being the reciprocal of the effective attenuation coefficient of the medium.The Monte Carlo simulation and phantom experiment were used to test the optimum SDD by evaluating the glucose sensitivity for all the used SDDs.Also,we acquired the optimum SDDs for the wavelengths in the range of 1000-1350 nm.2)At the acquired optimum SDDs,we measured glucose in some scattering media and got the glucose sensitivity.Based on the sensitivity,we estimated the minimum SNR requirement of the instrument for the glucose measurement target.In the experiment,we measured the glucose in multiple media,which are with different scattering property or different hemoglobin concentrations,and the measurement errors are evaluated.3)We proposed a new data processing method based on multi-position signals and derived the relevant formulas about measurement sensitivity and measurement limit.The method is validated by theoretical calculation and experiment in that the accumulated signals from the selected SDDs by our proposed rule can exhibit a higher measurement sensitivity than the signal at one SDD.Then the method is applied to the glucose measurement to further improve the glucose measurement limit.In summary,in this thesis we systematically introduced the theory of the differential near infrared diffuse reflectance spectroscopy based on position reference,and we proposes the optimum measurement positions for this method to acquire the optimum sensitivity.The optimized measurement position was applied to non-invasive blood glucose measurement,and the optimum measurement sensitivity was realized.Furthermore,we propose an effective data processing method using multi-position signals to improve the glucose measurement sensitivity furtherly.The research results in this thesis would be beneficial in the noninvasive blood glucose measurement based on NIR diffuse spectroscopy.