Study On Measurement Conditions For Non-invasive Spectral Analysis Of Human Body Composition

The various chemical components present in human body carry important information on health status. Such chemical information also serves as an important indicator to a number of clinical diagnostics and therapeutic effect. Therefore, timely analysis of these chemical concentrations has special meaning for modern health care. Invasive measurement is unfavorable when compared with non-invasive methods. Hence, Near-infrared spectroscopy receives global attention as a principal non-invasive diagnostic means.As a matter of fact, many key common issues severely constrain the practical application of this technique. This dissertation discusses the measurement conditions, which are primary factors affecting spectral measurement of human body composition, as a breakthrough to this problem. Measurement conditions affecting measuring accuracy in near-infrared spectral measurement are to be systematically discussed in details.In this dissertation, starting with theoretical analysis, we forward the methodology concerning body composition measurement conditions with spectroscopy methods as well as systematic description of its essential factors. Experiments are done to examine the influence of various measurement conditions on spectral measurement. It is concluded that reproducing measurement conditions is prerequisite to enhancing stability and repeatability in spectral measurement. A discussion of the necessity and importance of measurement conditions is also given.To eliminate or reduce the influence of measurement conditions on near-infrared spectral measurement of human body composition, measurement reproducing theory and smart system towards human body properties are put forwards originally. Image processing, and contact pressure measurement methods are employed to determine the position of probe in the system. Servo relocation device is utilized to relocate the probe position. Accuracy analysis and experimental results reveal that application of measurement conditions reproducing system can significantly reducing spectral difference arising from probe position and contact pressure. The stability and repeatability for spectra measurement can be improved accordingly. For the optimum initial contact pressure in the above measurement condition reproducing system, a contact pressure finite element monte carlo model is forwarded and constructed for the first time, which is based on the formation characteristics of contact pressure between optic fiber probe and the probing part as well as bio-mechanical properties of skin and subcutaneous tissue. Simulation results of the model are combined with experimental outcomes and serve as a guide to experimental design strategy to determine optimum initial contact pressure. Finally, non-invasive blood glucose concentration measurement system by near-infrared spectroscopy is set up according to effectiveness analysis of measuring conditions reproducing system. Clinical experiments are also conducted. Our study reveals: the application of measurement conditions reproducing system can reduce or even eliminate the measuring errors brought by measurement condition variation.With the application of measurement condition reproducing system, near-infrared spectroscopy technique proves to be feasible in blood glucose sensing after the NIR spectra undergo certain mathematical pretreatment. Pre-clinical experimental results demonstrate the great clinical prospect for near-infrared spectroscopy application in non-invasive human body composition analysis technique. Theory and experimental results disclosed by this dissertation laid solid basis for the successful application of this method...