Machine learning-based prediction of carotid intima–media thickness progression - a three-year prospective cohort study

Aug 5, 2026

Upload by Eden

Machine learning models are developed and validated to predict carotid intima–media thickness (CIMT) progression using routine clinical biomarkers. Using a three-year prospective cohort of 904 participants with three consecutive annual CIMT measurements, outcomes are defined by CIMT thickening or nonthickening. Seven algorithms are compared with discrimination, calibration (Platt scaling), and decision curve analysis. Elastic net shows the best performance and supports three-tier risk stratification with improved calibration and clinical net benefit across thresholds.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages17

Reading Time43 min

AccessFree to download

Background

CIMT as a surrogate marker for atherosclerosis

Limitations in current CIMT clinical application

Methods

Three-year prospective cohort and CIMT measurement protocol

Outcome definition for progression

Machine learning models and evaluation metrics

Calibration and clinical utility assessment

Results

Model comparison and key predictors

Calibration improvements after Platt scaling

Decision curve analysis and risk stratification

Subgroup findings

Conclusion

Clinical value of machine learning for CIMT progression prediction

OPEN ACCESS EDITED BY Taminul Islam, Southern Illinois University Carbondale, United States REVIEWED BY Gang Ye, Sichuan Agricultural University, China Qiaoqiao Xu, Third Affiliated Hospital of Anhui Medical University, China *CORRESPONDENCE Jiangang Wang [395896584@qq.com](395896584@qq.com)[ ](395896584@qq.com)Wei-Dong Li [liweidong98@tmu.edu.cn](liweidong98@tmu.edu.cn) †These authors have contributed equally to this work RECEIVED 14 March 2025 ACCEPTED 28 May 2025 PUBLISHED 12 June 2025 CITATION Zhou A, Chen K, Wei Y, Ye Q, Xiao Y, Shi R, Wang J and Li W-D (2025) Machine learning-based prediction of carotid intima– media thickness progression: a three-year prospective cohort study. Front. Med. 12:1593662 . doi: 10.3389/fmed.2025.1593662 COPYRIGHT © 2025 Zhou, Chen, Wei, Ye, Xiao, Shi, Wang and Li. 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 12 June 2025 DOI 10.3389/fmed.2025.1593662 Machine learning-based prediction of carotid intima– media thickness progression: a three-year prospective cohort study An Zhou 1†, Kui Chen 2,3†, Yonghui Wei 1†, Qu Ye 1,4†, Yuanming Xiao 2†, Rong Shi 1, Jiangang Wang 2* and Wei-Dong Li 1* 1 Department of Genetics, College of Basic Medical Sciences, Tianjin Medical University, Tianjin, China, 2 Health Management Medical Center, Third Xiangya Hospital, Central South University, Changsha, China, 3State Key Laboratory of Ultrasound in Medicine and Engineering, College of Biomedical Engineering, Chongqing Medical University, Chongqing, China, 4 Department of Clinical Laboratory, Peking University First Hospital, Beijing, China Background: Early detection of subclinical atherosclerosis progression is crucial for preventing atherosclerotic cardiovascular disease (ASCVD) . Carotid intima– media thickness (CIMT) is a recognized surrogate marker for atherosclerosis, but accurate prediction of its progression remains challenging. This study aimed to develop and validate machine learning models for predicting CIMT progression via routine clinical biomarkers. Methods: In this three-year prospective cohort study, we analyzed data from 904 participants from the Third Xiangya Hospital of Central South University Health Examination Cohort who underwent three consecutive annual CIMT measurements. The participants were categorized into CIMT thickening and nonthickening groups on the basis of a final CIMT ≥1.0 mm or an increase ≥0.1 mm across consecutive measurements. We evaluated seven machine learning algorithms: logistic regression, random forest, XGBoost, support vector machine (SVM), elastic net, decision tree, and neural network. Model performance was assessed through discrimination (AUC, sensitivity, specificity) and calibration metrics, with Platt scaling applied to optimize probability estimates. Clinical utility was evaluated through decision curve analysis. Results: Compared with the more complex algorithms, the elastic net model demonstrated superior performance (AUC 0.754) . Baseline CIMT, absolute monocyte count, sex, age, and LDL-C were identified as the most influential predictors. After Platt scaling, the calibration improved significantly across all the models. Decision curve analysis revealed a positive net benefit across a wide threshold range (0 .01–0. 5) . On the basis of calibrated probabilities, we developed a three-tier risk stratification framework that identified distinct groups with progressively higher event rates: medium-risk (13 .9%), high-risk (50 .0%), and very-high-risk (60 .0%) . Subgroup a

FAQs

What is the main goal of this study?

To develop and validate machine learning models that predict CIMT progression using routine clinical biomarkers in a three-year prospective cohort.

How is CIMT progression defined in the study?

Participants are grouped as CIMT thickening or nonthickening based on a final CIMT threshold (≥1.0 mm) or an increase of at least 0.1 mm across consecutive measurements.

Which model performed best and what predictors were most influential?

The elastic net model shows superior performance (AUC 0.754), and baseline CIMT, absolute monocyte count, sex, age, and LDL-C are identified as the most influential predictors.

Aug 5, 2026

Upload by Eden