Glucose Signal Extraction from NIR Spectra Using EMSC Regression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Glucose monitoring using Near-Infrared (NIR) spectroscopy is challenging due to the low absorption values of glucose compared to other body constituents, leading to distorted glucose information and noise components in the data.
Innovation Solution
A method involving noise removal from NIR data using Savitzky-Golay and Fourier domain filtering, followed by extracting the glucose signal using Extended Multiplicative Scatter Correction (EMSC) regression and data whitening, and removing temporal drift components to enhance glucose monitoring accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If NIR spectroscopy is used for glucose monitoring, then non-invasive measurement is achieved, but glucose absorption signal is overwhelmed by other body constituents
Solution Approach 1:
The patent segments the complex NIR absorption spectrum into individual contributions from different body constituents (water, fat, protein, glucose) using multivariate analysis. By decomposing the overall absorption signal A = ε1C1d + ε2C2d + ... + εnCnd into separate component signals, the method isolates the glucose absorption component from overwhelming background signals, enabling precise glucose measurement while maintaining non-invasive operation.
Solution Approach 2:
The patent utilizes the unique molecular extinction coefficients (ε) of glucose at specific NIR wavelengths as distinguishing parameters. By selecting wavelengths where glucose has characteristic absorption peaks and applying mathematical transformations to the absorption data, the method enhances the glucose signal relative to other constituents, resolving the contradiction between non-invasive measurement and measurement precision.
2Measurement precision
If conventional NIR spectroscopy is used, then glucose monitoring is attempted, but glucose information is distorted by noise components
Solution Approach 1:
The patent extracts the glucose-specific signal from the noisy NIR spectrum by applying multivariate regression analysis. The method separates the glucose absorption component from noise and interference by using reference spectra and mathematical modeling, effectively taking out the useful glucose information while eliminating distortion from noise components.
Solution Approach 2:
The patent introduces mathematical models and reference spectra as intermediary tools to mediate between the raw noisy NIR signal and the glucose concentration. These intermediaries (calibration models, pure component spectra) act as filters that translate the distorted spectral data into accurate glucose measurements, reducing the impact of noise.
3Object-affected harmful factors
If simple filtering is applied to remove noise, then noise reduction is achieved, but temporal drift components remain in the glucose signal
Solution Approach 1:
The patent applies preliminary signal processing steps including smoothing filters and baseline correction before glucose signal extraction. By performing these preparatory actions on the raw NIR spectrum, the method reduces noise and drift components in advance, making the subsequent glucose quantification more accurate and reliable.
Solution Approach 2:
The patent employs dynamic signal processing that adapts to temporal variations in the NIR spectrum. By using moving window techniques and time-dependent calibration, the method accounts for drift components that change over time, maintaining measurement precision despite temporal variations in the biological system and instrumentation.
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 effectively isolates the glucose signal from other body composition components, reducing noise and temporal drift, thereby improving the accuracy of glucose monitoring and prediction.
Implementation Method 1
NIR waves are generated to pass through the skin and a spectrum indicating absorption of the NIR waves by the blood underneath the skin is used in determining the glucose level
Implementation Method 2
The absorption of the NIR waves is defined based on BEER-Lambert law: A=εCd
Implementation Method 3
The removing the noise comprises removing the noise from the NIR data using at least one of an Savitzky-Golay (SG) filter and a Fourier domain filtering
Implementation Method 4
The removing the noise comprises removing the noise from the NIR data using at least one of an Savitzky-Golay (SG) filter and a Fourier domain filtering
Implementation Method 5
The extracting the glucose signal comprises extracting the glucose signal based on an Extended Multiplicative Scatter Correction (EMSC) regression and data whitening
Implementation Method 6
The data whitening is performed to obtain an orthogonal component of a pure spectrum of a body composition including a glucose
Data Source
AI summary
A method of extracting glucose feature, a method for monitoring glucose using near-infrared (NIR) spectroscopy and a glucose monitoring device are provided. The method comprising: removing noise from a near-infrared (NIR) data; extracting a glucose signal from the NIR data; and removing temporal drift components from the extracted glucose signal.