Bongchu Shim

Finger tissue model and blood perfused skin tissue phantom

Measurements of optical properties of fingernail and underlying tissues using OCT are presented. Review and measurements of Raman spectra of tissue and phantom compounds were done. Updating of modeling algorithm for scattering coefficient calculation on the basis of integrating sphere measurements accounting for particle size-distribution was also done. The adequate fingernail and underlying tissue optical model at 830 nm was evaluation. Tissue phantoms potentially suitable for calibration of Raman instrumentation for glucose sensing were designed and tested on the basis of epoxy resin, TiO2-nanoparticles and micron-sized silica particles with the capillary net-work.

The high-quality spectral fingerprint of glucose captured by Raman spectroscopy in non-invasive glucose measurement

Optical spectroscopy has the inherent advantage of directly measuring glucose levels by means of the unique nature of a spectral fingerprint for the target analyte, glucose, which is determined in the form of a calibration spectrum (a product of multivariate analysis method). The literature is replete with claims of successful in-vivo glucose measurements in terms of Clark error grid analysis. However, all of them fail to demonstrate an in vivo glucose-specific calibration spectrum. In this study we use Raman spectroscopy to capture a high quality spectral fingerprint of the glucose molecule noninvasively and demonstrate calculated correlation coefficients between a pure glucose spectrum and calibration spectra at over 0.8.

An investigation of the effect of in-vivo interferences on Raman glucose measurements

Raman spectroscopy is a promising technology for noninvasive blood glucose monitoring because of its good selectivity for the glucose molecule. The low sensitivity of the Raman signal however, makes it difficult to quantify the concentration of glucose directly from the Raman spectra. To solve this, statistical methods such as PCA (principle component analysis) and PLS (partial least square) are traditionally used. These statistical methods general work very well and give highly accurate results, provided there is no interference. In the in-vivo case however, there are many interferences such as the inhomogeneity of tissue, physiological changes, and denaturation of the tissue by the light source. This study investigates the affect of in-vivo interferences on Raman glucose measurements. In this study, a high throughput dispersive Raman system was constructed with an 830nm multimode laser, a multiple conductor optical fiber bundle, and a back-illuminated CCD spectrometer. A simply phantom was devised, which was comprised of a plastic cuvette fitted with a human fingernail window and glucose doped human serum used as the sample. To test the inhomogeneity of tissue samples, different sites of the phantom were exposed to the laser. In the case of denaturation, tests were conducted under two laser power densities: low (3.7mW/mm2) and high density (110mW/mm2). To simulate the physiological change, gelatin phantoms of varied concentration were investigated. The results of the study indicate that the dominant interferers for Raman in-vivo glucose measurements are the inhomogeneity of the tissue and the denaturation by the laser power density. The next phase for this study will be the design of a high SNR Raman system which affords a low power density laser sample illumination as well as larger volumetric illumination to mitigate the effects of tissue inhomogeneity.

SIMULATION FOR THE EFFECT OF VARIOUS COMPONENTS IN THE Raman SPECTRA

Raman spectra measured in the finger are a combination of backscattered signals induced from incident light. They contain Raman scattering, intrinsic tissue fluorescence and noise. Our goal of this study is to find perturbing components which are able to prohibit an accurate prediction of target materials and propose a new terminology for the SNR of regression coefficient vector. We assumed that Raman spectra are consist of fluorescent background spectrum (F), Raman spectrum of glucose (Rglucose), Raman spectrum of the skin (Rskin), random noise (RN) and unknown spectrum from other components (Runknown). And we used partial least squares regression (PLSR) to evaluate the main perturbing components under the condition that is various combinations of each signal. In simulation, Rskin is the main component to cause the error between reference and predicted glucose concentration. F is also able to affect accuracy at the low signal to noise ratio (SNR) of glucose signal. To minimize the perturbing effects, enhancement of Rglucose is more important than any other things.