Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
April 30, 2026Frontiers in Cardiovascular MedicineOpen Access

The stacked ensemble model achieved an AUC of 0.977 and accuracy of 0.942 in internal validation, and an AUC of 0.929 and accuracy of 0.885 in external validation.

View Full Paper
Ask AI
Bookmark
Share

Why the study?

The study aimed to develop and externally validate a coronary heart disease risk model from routine clinical indicators and identify key predictors.

Can a machine learning model based on routine clinical indicators accurately predict coronary heart disease risk?

Population

Framingham Heart Study cohort (n = 4,240) and external hospital cohort (n = 200)

Design

Retrospective cohort model development and validation study

Authors

HXHui XiongXCXiang CaoHXHan Xiao

Discussion

Loading...

Member takes

Overview

May support routine-data CHD risk stratification; leaves open prospective validation before clinical use.

Structured PICO

Can a machine learning model based on routine clinical indicators accurately predict coronary heart disease risk?

P
Population
Framingham Heart Study cohort (n=4,240) for training and internal validation, and a retrospective hospital cohort (n=200; 2024-2025) for external validation.
I
Intervention
Stacked ensemble machine learning model (gradient boosting, random forest, XGBoost with logistic-regression meta-learner) using ten routine clinical variables.
O
Outcome
Model performance measured by AUC, accuracy, precision, recall, and F1 for coronary heart disease risk prediction.

A machine learning model using routine clinical indicators demonstrated strong discrimination for predicting coronary heart disease risk, generalizing well to an external cohort.

Cite This Study

Xiong et al. (2026) studied this question.

synapsesocial.com/papers/6a703bba26a7f98052dcbc98https://doi.org/10.3389/fcvm.2026.1821221
View Full Paper
Ask AI
Bookmark
Share