Machine-learning-based diabetes classification method using blood flow oscillations and Pearson correlation analysis of feature importance - abstract

Aug 5, 2026

Upload by Liam

Diabetes is a global health challenge marked by high morbidity and mortality, and conventional diagnosis often relies on invasive blood sampling that increases infection risk and patient stress. This work develops a noninvasive, fast approach using diffuse speckle contrast analysis (DSCA) to obtain blood-flow signals combined with machine learning for diabetes classification. Rat blood-flow oscillation data are analyzed to support effective classification, and blood-flow reactivity tests enable rapid measurement. Feature importance is further examined using Pearson correlation to quantify each signal’s contribution, providing a basis for hemodynamic-based diagnostic development.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages12

Reading Time30 min

AccessFree to download

1. Introduction

Diabetes diagnosis challenges and need for noninvasive methods

Noninvasive monitoring approaches and limitations

Applying AI algorithms for diabetes monitoring

2. Method overview

DSCA-based blood-flow signal acquisition

Machine-learning classification pipeline

Feature importance and Pearson correlation analysis

PAPER • OPEN ACCESS Machine-learning-based diabetes classification method using blood flow oscillations and Pearson correlation analysis of feature importance To cite this article: Hanbeen Jung et al 2024 Mach. Learn. : Sci. Technol. 5 045024 View the article online for updates and enhancements. You may also like -Geometric neural operators (gnps) for data-driven deep learning in non-euclidean settings B Quackenbush and P J Atzberger -A prediction rigidity formalism for low-cost uncertainties in trained neural networks Filippo Bigi, Sanggyu Chong, Michele Ceriotti et al. -Designing the next generation of polymers with machine learning and physics-based models Alex K Chew, Mohammad Atif Faiz Afzal, Anand Chandrasekaran et al. This content was downloaded from IP address [114.71.101.112](114.71.101.112) on 20/12/2024 at 10:37 Mach. Learn.: Sci. Technol. 5 (2024) 045024 [https://doi.org/10.1088/2632-2153/ad861d](https://doi.org/10.1088/2632-2153/ad861d) OPEN ACCESS RECEIVED 9 May 2024 REVISED 19 September 2024 ACCEPTED FOR PUBLICATION 11 October 2024 PUBLISHED 25 October 2024 Original Content from this work may be used under the terms of the Creative Commons Attribution 4 .0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. PAPER Machine-learning-based diabetes classification method using blood flow oscillations and Pearson correlation analysis of feature importance Hanbeen Jung􀁂, Chaebeom Yeo􀁂, Eunsil Jang􀁂, Yeonhee Chang􀁂 and Cheol Song∗􀁂 Department of Robotics and Mechatronics Engineering, Daegu Gyeongbuk Institute of Science and Technology, Daegu 42988, Republic of Korea ∗ Author to whom any correspondence should be addressed. [E-mail:](E-mail: csong@dgist.ac.kr)[ csong@dgist.ac.kr](E-mail: csong@dgist.ac.kr) Keywords: machine learning, feature importance, diffuse speckle contrast analysis, blood flow oscillations, diabetes diagnosis, Pearson correlation Abstract Diabetes is a global health issue affecting millions of people and is related to high morbidity and mortality rates. Current diagnostic methods are primarily invasive, involving blood sampling, which can lead to infection and increased patient stress. As a result, there is a growing need for noninvasive diabetes diagnostic methods that are both accurate and fast. High measurement accuracy and fast measurement time are essential for effective noninvasive diabetes diagnosis; these can be achieved using diffuse speckle contrast analysis (DSCA) systems and artificial intelligence algorithms. In this study, we use a machine learning algorithm to analyze rat blood flow signals measured using a DSCA system with simple operation, easy fabrication, and fast measurement for helping diagnose diabetes. The results confirmed that the machine learning algorithm for analyzing blood flow oscillation data shows good potential for diabetes classification. Furthermore, analyzing the blood flow reactivity test revealed that blood flow signals can be quickly measured for diabetes classification. Finally, we evaluated the influence of each blood flow oscillation data on diabetes classification through feature importance and Pearson correlation analysis. The results of this study should provide a basis for the future development of hemodynamic-based disease diagnostic methods. 1. Introduction Diabetes is a serious global health issue with high morbidity and mortality, and its incidence is increasing. Diabetes is characterized by hyperglycemia, and it can cause various complications such as diabetic neuropathy, diabetic kidney disease, and cardiovascular disease (CVD) [1–5] . CVD is the leading cause of death in diabetics, accounting for 44% of people with type 1 diabetes and 52% of people with type 2 diabetes [6] . Diabetics have a threefold increase in cardiovascular mortality compared to nondiabetics, and younger diabetics have a

FAQs

Why is noninvasive diabetes diagnosis needed?

Conventional diagnosis commonly uses invasive blood sampling, which can cause infection risk and additional patient stress. Noninvasive methods aim to be accurate while reducing discomfort and enabling faster measurement.

How does the study collect data for diabetes classification?

Blood-flow signals are measured using a diffuse speckle contrast analysis (DSCA) system, then used as input to a machine learning algorithm. The study also evaluates blood-flow reactivity tests for quick measurement.

What role do Pearson correlation and feature importance play?

The study analyzes how each blood-flow oscillation feature influences diabetes classification by computing feature importance and using Pearson correlation. This identifies the contribution of individual oscillation data to model performance.

Aug 5, 2026

Upload by Liam