A data-driven machine learning algorithm to predict the effectiveness of inulin intervention against type II diabetes

Aug 5, 2026

Upload by Levi

A data-driven machine learning framework was developed to estimate which type 2 diabetes mellitus (T2DM) patients may benefit from inulin intervention. LASSO regression selected six baseline patient factors from 758 individuals, and an XGBoost model was trained and tested using predictive metrics including accuracy, specificity, positive predictive value, and negative predictive value. Model interpretation relied on SHAP value ranking, highlighting HbA1c and glucose-related variables, with performance evaluated via ROC, calibration, and decision curves. Results support use of the XGBoost-SHAP approach to connect nutrition intervention with individualized treatment decisions.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages8

Reading Time20 min

AccessFree to download

Introduction

Inulin and nutritional therapy in T2DM

Methods

Feature selection with LASSO

Prediction modeling with XGBoost

Model evaluation and interpretation

Results

HbA1c response after inulin intervention

Key prognostic factors from LASSO

SHAP variable importance

Model performance on training and testing sets

Conclusion

XGBoost-SHAP framework for individualized prediction

OPEN ACCESS EDITED BY Elizabethe Esteves, Universidade Federal dos Vales do Jequitinhonha e Mucuri, Brazil REVIEWED BY Kulvinder Kochar Kaur, Kulvinder Kaur Centre For Human Reproduction, India Andrea Deledda, Azienda Ospedaliero-Universitaria Cagliari, Italy *CORRESPONDENCE Hualin Wang [wanghualin313@163.com](wanghualin313@163.com)[ ](wanghualin313@163.com)Ralf Weiskirchen [rweiskirchen@ukaachen.de](rweiskirchen@ukaachen.de)[ ](rweiskirchen@ukaachen.de)RECEIVED 31 October 2024 ACCEPTED 18 December 2024 PUBLISHED 07 January 2025 CITATION Yang S, Weiskirchen R, Zheng W, Hu X, Zou A, Liu Z and Wang H (2025) A data-driven machine learning algorithm to predict the effectiveness of inulin intervention against type II diabetes. Front. Nutr. 11:1520779. doi: 10.3389/fnut.2024.1520779 COPYRIGHT © 2025 Yang, Weiskirchen, Zheng, Hu, Zou, Liu and Wang. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. TYPE Original Research PUBLISHED 07 January 2025 DOI 10.3389/fnut.2024.1520779 A data-driven machine learning algorithm to predict the effectiveness of inulin intervention against type II diabetes Shuheng Yang 1, Ralf Weiskirchen 2*, Wenjing Zheng 1, Xiangxu Hu 1, Aibiao Zou3, Zhiguo Liu 1 and Hualin Wang 1* 1School of Life Science and Technology, Wuhan Polytechnic University, Wuhan, China, 2 Institute of Molecular Pathobiochemistry, Experimental Gene Therapy and Clinical Chemistry (IFMPEGKC), RWTH University Hospital, Aachen, Germany, 3 Research Center of Medical Nutrition Therapy, Cross-strait Tsinghua Research Institute, Xiamen, China Introduction: The incidence of type 2 diabetes mellitus (T2DM) has increased in recent years. Alongside traditional pharmacological treatments, nutritional therapy has emerged as a crucial aspect of T2DM management. Inulin, afructan-type soluble fiber that promotes the growth of probiotic species like Bifidobacterium and Lactobacillus, is commonly used in nutritional interventions for T2DM. However, it remains unclear which type of T2DM patients are suitable for inulin intervention. The aim of this study was to predict the effectiveness of inulin treatment for T2DM using a machine learning model. Methods: Original data were obtained from a previous study. After screening T2DM patients, feature election was conducted using LASSO regression, and a machine learning model was developed using XGBoost. The model’s performance was evaluated based on accuracy, specificity, positive predictive value, negative predictive value and further analyzed using receiver operating curves, calibration curves, and decision curves. Results: Out of the 758 T2DM patients included, 477 had their glycated hemoglobin (HbA1c) levels reduced to less than 6. 5% after inulin intervention, resulting in an incidence rate of 62.93% . LASSO regression identified six key factors in patients prior to inulin treatment. The SHAP values for interpretation ranked the characteristic variables in descending order of importance: HbA1c, difference between fasting and 2 h-postprandial glucose levels, fasting blood glucose, high-density lipoprotein, age, and body mass index. The XGBoost prediction model demonstrated a training set accuracy of 0. 819, specificity of 0.913, positive predictive value of 0. 818, and negative predictive value of 0.820. The testing set showed an accuracy of 0.709, specificity of 0.909, positive predictive value of 0.705, and negative predictive value of 0.710. Conclusion: The XGBoost-SHAP framework for predicting the impact of inulin intervention in T2DM treatment proves to be effec

FAQs

What problem does the study address about inulin in type 2 diabetes?

It targets uncertainty about which T2DM patients are suitable for inulin intervention. The study builds a machine-learning model to predict inulin effectiveness for individual patients.

How were predictive features selected in the study?

LASSO regression was used to perform feature election on baseline patient data, resulting in six key factors prior to inulin treatment.

How does the study interpret the XGBoost model’s predictions?

SHAP values were calculated to rank the importance of characteristic variables, helping identify which patient features most influence predicted effectiveness.

Aug 5, 2026

Upload by Levi