Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Effects of the Omnipod 5 Hybrid Closed-Loop System in People With Type 1 Diabetes With Impaired Awareness of Hypoglycemia.

Diabetes care·2026
Same author

Real-World Effectiveness and Safety of Tirzepatide in Type 1 Diabetes and Obesity: Impact on Glycaemia, Weight, and Cardiometabolic Risk Markers.

Diabetes, obesity & metabolism·2026
Same author

Impact of Intrinsic Defects and Tungsten Doping on the Catalytic Properties of Two-Dimensional Cu<sub>2</sub>S.

ACS omega·2026
Same author

Artificial Intelligence for Blood Glucose Level Prediction in Type 1 Diabetes: Methods, Evaluation, and Emerging Advances.

Sensors (Basel, Switzerland)·2026
Same author

Personalised Blood Glucose Time Series Forecasting in Type 1 Diabetes: Deep Collaborative Adversarial Learning.

Journal of personalized medicine·2026
Same author

MGPP-DL: deep learning approach for material graph properties prediction.

Journal of molecular modeling·2026

Related Experiment Video

Updated: Sep 29, 2025

Measuring Glucose Uptake in Drosophila Models of TDP-43 Proteinopathy
07:07

Measuring Glucose Uptake in Drosophila Models of TDP-43 Proteinopathy

Published on: August 3, 2021

2.9K

Signal fragmentation based feature vector generation in a model agnostic framework with application to glucose

Heydar Khadem1, Hoda Nemat1, Jackie Elliott2

  • 1Department of Electronic and Electrical Engineering, University of Sheffield, UK.

Talanta
|March 20, 2022
PubMed
Summary

This study introduces a novel feature vector generation method using signal fragmentation for improved glucose quantification from spectroscopy. The approach enhances accuracy across near-infrared (NIR) and mid-infrared (MIR) signals, aiding transparent interpretation.

Keywords:
Glucose quantificationMachine learningMid-infrared spectroscopyNear-infrared spectroscopySHAP

More Related Videos

Bergmeyer Glucose Quantification for Microbiological Samples
07:23

Bergmeyer Glucose Quantification for Microbiological Samples

Published on: January 17, 2025

647
Highly Sensitive and Rapid Fluorescence Detection with a Portable FRET Analyzer
08:27

Highly Sensitive and Rapid Fluorescence Detection with a Portable FRET Analyzer

Published on: October 1, 2016

9.2K

Related Experiment Videos

Last Updated: Sep 29, 2025

Measuring Glucose Uptake in Drosophila Models of TDP-43 Proteinopathy
07:07

Measuring Glucose Uptake in Drosophila Models of TDP-43 Proteinopathy

Published on: August 3, 2021

2.9K
Bergmeyer Glucose Quantification for Microbiological Samples
07:23

Bergmeyer Glucose Quantification for Microbiological Samples

Published on: January 17, 2025

647
Highly Sensitive and Rapid Fluorescence Detection with a Portable FRET Analyzer
08:27

Highly Sensitive and Rapid Fluorescence Detection with a Portable FRET Analyzer

Published on: October 1, 2016

9.2K

Area of Science:

  • Analytical Chemistry
  • Spectroscopy
  • Biomedical Engineering

Background:

  • Accurate glucose quantification is crucial for diabetes management.
  • Absorption spectroscopy offers a non-invasive method for glucose monitoring.
  • Current spectroscopic methods face challenges in signal complexity and interpretation.

Purpose of the Study:

  • To develop an enhanced feature vector generation technique for improved glucose quantification using absorption spectroscopy.
  • To evaluate the proposed method's performance with near-infrared (NIR) and mid-infrared (MIR) spectra.
  • To ensure model transparency and interpretability through SHapley additive exPlanations (SHAP).

Main Methods:

  • Signal fragmentation: Spectra are dissected into optimal fragments.
  • Base-learner analysis: Individual fragments are used to estimate glucose concentration.
  • Feature vector generation: Estimates from fragments are stacked into a comprehensive feature vector.
  • Meta-learner analysis: A final glucose concentration is predicted from the feature vector.
  • Model interpretation: SHapley additive exPlanations (SHAP) are applied for outcome transparency.

Main Results:

  • The proposed feature vector generation method significantly enhances glucose quantification accuracy.
  • The approach demonstrates robust performance with both NIR and MIR spectral data, individually and in combination.
  • The method is compatible with standard spectroscopic preprocessing techniques.
  • SHAP analysis provides clear insights into the quantification process, promoting model interpretability.

Conclusions:

  • The developed feature vector generator improves glucose quantification in absorption spectroscopy.
  • The method offers a reliable and interpretable approach for analyzing spectroscopic data.
  • This technique holds promise for advancing non-invasive glucose monitoring technologies.