Machine learning-based prediction of carotid intima–media thickness progression: a three-year prospective cohort study
BackgroundEarly 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. T...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-06-01
|
| Series: | Frontiers in Medicine |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fmed.2025.1593662/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850160610027765760 |
|---|---|
| author | An Zhou Kui Chen Kui Chen Yonghui Wei Qu Ye Qu Ye Yuanming Xiao Rong Shi Jiangang Wang Wei-Dong Li |
| author_facet | An Zhou Kui Chen Kui Chen Yonghui Wei Qu Ye Qu Ye Yuanming Xiao Rong Shi Jiangang Wang Wei-Dong Li |
| author_sort | An Zhou |
| collection | DOAJ |
| description | BackgroundEarly 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.MethodsIn 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.ResultsCompared 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 analysis revealed better predictive performance in younger participants (<50 years), those with lower baseline CIMT (<0.8 mm), and females.ConclusionMachine learning approaches, particularly the elastic net model, can effectively identify individuals at high risk for CIMT progression via routine clinical biomarkers. The superior performance of simpler models suggests predominantly linear relationships between predictors and CIMT progression. Following appropriate calibration, the model demonstrated strong clinical utility across diverse decision thresholds, supporting a stratified approach to atherosclerosis prevention. |
| format | Article |
| id | doaj-art-bf2ffc2671cd4c6fa216affb7bdf6112 |
| institution | OA Journals |
| issn | 2296-858X |
| language | English |
| publishDate | 2025-06-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Medicine |
| spelling | doaj-art-bf2ffc2671cd4c6fa216affb7bdf61122025-08-20T02:23:06ZengFrontiers Media S.A.Frontiers in Medicine2296-858X2025-06-011210.3389/fmed.2025.15936621593662Machine learning-based prediction of carotid intima–media thickness progression: a three-year prospective cohort studyAn Zhou0Kui Chen1Kui Chen2Yonghui Wei3Qu Ye4Qu Ye5Yuanming Xiao6Rong Shi7Jiangang Wang8Wei-Dong Li9Department of Genetics, College of Basic Medical Sciences, Tianjin Medical University, Tianjin, ChinaHealth Management Medical Center, Third Xiangya Hospital, Central South University, Changsha, ChinaState Key Laboratory of Ultrasound in Medicine and Engineering, College of Biomedical Engineering, Chongqing Medical University, Chongqing, ChinaDepartment of Genetics, College of Basic Medical Sciences, Tianjin Medical University, Tianjin, ChinaDepartment of Genetics, College of Basic Medical Sciences, Tianjin Medical University, Tianjin, ChinaDepartment of Clinical Laboratory, Peking University First Hospital, Beijing, ChinaHealth Management Medical Center, Third Xiangya Hospital, Central South University, Changsha, ChinaDepartment of Genetics, College of Basic Medical Sciences, Tianjin Medical University, Tianjin, ChinaHealth Management Medical Center, Third Xiangya Hospital, Central South University, Changsha, ChinaDepartment of Genetics, College of Basic Medical Sciences, Tianjin Medical University, Tianjin, ChinaBackgroundEarly 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.MethodsIn 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.ResultsCompared 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 analysis revealed better predictive performance in younger participants (<50 years), those with lower baseline CIMT (<0.8 mm), and females.ConclusionMachine learning approaches, particularly the elastic net model, can effectively identify individuals at high risk for CIMT progression via routine clinical biomarkers. The superior performance of simpler models suggests predominantly linear relationships between predictors and CIMT progression. Following appropriate calibration, the model demonstrated strong clinical utility across diverse decision thresholds, supporting a stratified approach to atherosclerosis prevention.https://www.frontiersin.org/articles/10.3389/fmed.2025.1593662/fullcarotid intima–media thickness (CIMT)machine learningatherosclerosis progressionrisk predictioncardiovascular prevention |
| spellingShingle | An Zhou Kui Chen Kui Chen Yonghui Wei Qu Ye Qu Ye Yuanming Xiao Rong Shi Jiangang Wang Wei-Dong Li Machine learning-based prediction of carotid intima–media thickness progression: a three-year prospective cohort study Frontiers in Medicine carotid intima–media thickness (CIMT) machine learning atherosclerosis progression risk prediction cardiovascular prevention |
| title | Machine learning-based prediction of carotid intima–media thickness progression: a three-year prospective cohort study |
| title_full | Machine learning-based prediction of carotid intima–media thickness progression: a three-year prospective cohort study |
| title_fullStr | Machine learning-based prediction of carotid intima–media thickness progression: a three-year prospective cohort study |
| title_full_unstemmed | Machine learning-based prediction of carotid intima–media thickness progression: a three-year prospective cohort study |
| title_short | Machine learning-based prediction of carotid intima–media thickness progression: a three-year prospective cohort study |
| title_sort | machine learning based prediction of carotid intima media thickness progression a three year prospective cohort study |
| topic | carotid intima–media thickness (CIMT) machine learning atherosclerosis progression risk prediction cardiovascular prevention |
| url | https://www.frontiersin.org/articles/10.3389/fmed.2025.1593662/full |
| work_keys_str_mv | AT anzhou machinelearningbasedpredictionofcarotidintimamediathicknessprogressionathreeyearprospectivecohortstudy AT kuichen machinelearningbasedpredictionofcarotidintimamediathicknessprogressionathreeyearprospectivecohortstudy AT kuichen machinelearningbasedpredictionofcarotidintimamediathicknessprogressionathreeyearprospectivecohortstudy AT yonghuiwei machinelearningbasedpredictionofcarotidintimamediathicknessprogressionathreeyearprospectivecohortstudy AT quye machinelearningbasedpredictionofcarotidintimamediathicknessprogressionathreeyearprospectivecohortstudy AT quye machinelearningbasedpredictionofcarotidintimamediathicknessprogressionathreeyearprospectivecohortstudy AT yuanmingxiao machinelearningbasedpredictionofcarotidintimamediathicknessprogressionathreeyearprospectivecohortstudy AT rongshi machinelearningbasedpredictionofcarotidintimamediathicknessprogressionathreeyearprospectivecohortstudy AT jiangangwang machinelearningbasedpredictionofcarotidintimamediathicknessprogressionathreeyearprospectivecohortstudy AT weidongli machinelearningbasedpredictionofcarotidintimamediathicknessprogressionathreeyearprospectivecohortstudy |