Glucose Concentration Measurement Using Near-Infrared Spectroscopy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In near-infrared spectroscopy, conventional methods like curve fitting and classical least squares struggle to accurately quantify glucose concentration due to broad component spectra without clear absorption peaks, leading to low reproducibility and high measurement errors, especially for minor components like glucose in biological tissues.
Innovation Solution
A method involving the calculation of a concentration index using specific absorption signals from water, glucose, and fat components, with the use of imaginary spectra and error correction to improve reproducibility and accuracy in glucose concentration measurement, employing near-infrared light and spectroscopic analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional curve fitting method or CLS method is used for glucose concentration quantification in near-infrared spectroscopy, then the measurement can be performed using standard spectroscopic analysis, but the measurement precision and reliability are poor due to broad component spectra without clear absorption peaks and small spectral changes of glucose compared to other components
Solution Approach 1:
The patent segments the complex spectral analysis into distinct components by identifying and analyzing specific absorption peak regions separately. The glucose concentration is determined by focusing on specific wavelength regions (1600-1700 nm and 1900-2000 nm) where glucose exhibits characteristic absorption, rather than analyzing the entire spectrum. This segmentation allows the measurement system to isolate glucose spectral features from interfering components, thereby improving measurement precision despite the broad nature of near-infrared spectra.
2Measurement precision
If the number of components is accurately estimated in CLS method, then accurate quantification can be performed, but the complexity of the measurement process increases due to the need for accurate estimation of multiple parameters including number of components, device errors, and disturbance factors
Solution Approach 1:
The patent extracts and removes disturbance factors from the spectral analysis process. Specifically, it identifies and eliminates the influence of water absorption peaks and other interfering components by focusing analysis on wavelength regions where glucose exhibits characteristic absorption patterns. The method extracts pure glucose spectral information by subtracting or ignoring contributions from other components, thereby simplifying the analysis while maintaining quantification accuracy without requiring complex estimation of all possible components.
3Measurement precision
If multivariate analysis with calibration curve is used for quantitative analysis, then glucose concentration can be quantified using experimental data, but the reproducibility is poor due to broad component spectra and small spectral changes of glucose in biological tissues
Solution Approach 1:
The patent performs preliminary identification and characterization of glucose absorption peak regions before conducting the actual concentration measurement. By pre-establishing the specific wavelength ranges (1600-1700 nm and 1900-2000 nm) where glucose exhibits characteristic absorption, the method creates a reliable foundation for subsequent measurements. This preliminary action ensures that the measurement system is optimized for glucose detection from the outset, improving both accuracy and reproducibility by consistently targeting the same characteristic spectral regions across multiple measurements.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enables high-reproducibility and high-accuracy glucose concentration measurement by mitigating disturbance factors and separating glucose spectra from baseline variations, effectively addressing the limitations of existing methods in near-infrared spectroscopy.
Implementation Method 1
a glucose concentration in an intercellular fluid as an alternative characteristic of a blood glucose level of a living organism is quantified using diffusely reflected light or transmitted light of near-infrared light emitted onto a biological tissue
Implementation Method 2
For the purpose of quantitative analysis in near-infrared spectroscopy, multivariate methods such as principal component regression analysis and PLS regression analysis are frequently used
Data Source
AI summary
A method for quantifying a glucose concentration is a method for quantifying a glucose concentration in which near-infrared light is emitted onto a living organism and a glucose concentration in a biological tissue is measured using a signal obtained by receiving diffusely reflected light or transmitted light from the biological tissue. A concentration calculation step calculates a concentration index of a glucose component by using at least a spectrum of a water component, a spectrum of a glucose component, and a spectrum of a fat component to synthesize a difference spectrum between a measurement spectrum at a time of measurement of a glucose concentration and a spectrum serving as a reference obtained previous to the measurement spectrum. A glucose concentration calculation step calculates a glucose concentration in the living organism using the calculated concentration index.