Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
November 9, 2022PeerJOpen Access

The XGBoost model demonstrated the highest predictive value for coronary heart disease in young and middle-aged patients, achieving an AUC of 0.940 and an F1 score of 0.887.

View Full Paper
Ask AI
Bookmark
Share

Why the study?

The study was conducted to identify coronary heart disease risk factors in young and middle-aged individuals and develop a tailored risk prediction model.

Do machine learning models accurately predict coronary heart disease risk in young and middle-aged patients?

Population

553 young and middle-aged patients undergoing coronary angiography at a tertiary hospital in Anhui Province

Comparison

Coronary heart disease (n = 201) vs non-coronary heart disease (n = 352)

Design

Retrospective cohort study

Key result

The XGBoost model demonstrated the highest predictive value for coronary heart disease in young and middle-aged patients, achieving an AUC of 0.940 and an F1 score of 0.887.

Authors

JCJiaoyu CaoLZLixiang ZhangSMShaolin Mao

Discussion

Loading...

Member takes

Overview

May aid early CHD risk stratification in younger adults; leaves open need for prospective validation before clinical adoption.

Study Design

Type

Cohort (n=553)

Multicenter

No

Structured PICO

Do machine learning models accurately predict coronary heart disease risk in young and middle-aged patients?

P
Population
553 young and middle-aged patients who underwent coronary angiography at a tertiary hospital in China between 2017 and 2020.
E
Exposure
Machine learning risk prediction models (XGBoost, random forest, BP neural network) and logistic regression
C
Comparator
Comparison between different prediction models
O
Outcome
Model prediction efficiency (Area Under the Curve [AUC] and F1 score)surrogate

Main Result

Absolute Event Rate: 0.94% vs 0.829%

An XGBoost machine learning model can highly accurately predict the risk of coronary heart disease in young and middle-aged individuals, potentially aiding in clinical screening.

Cite This Study

Cao et al. (2022) conducted a cohort in Coronary heart disease (n=553). XGBoost model vs. Logistic regression, BP neural network, and random forest models was evaluated on Area under the curve (AUC) for predicting coronary heart disease. The XGBoost model demonstrated the highest predictive value for coronary heart disease in young and middle-aged patients, achieving an AUC of 0.940 and an F1 score of 0.887.

synapsesocial.com/papers/6a484db0a567c8cbc92f7d98https://doi.org/10.7717/peerj.14078
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Coronary Heart Disease Prediction Using Machine Learning Algorithms2026
  2. 2Using machine learning algorithms to identify chronic heart disease: National Health and Nutrition Examination Survey 2011–20182023 · 4 citations
  3. 3Use of machine learning to identify risk factors for coronary artery disease2023 · 33 citations
  4. 4Development and validation of a machine learning model for on-site prediction of coronary heart disease in high-risk adults using clinical data2025 · 1 citations
  5. 5Development and validation of a machine learning-based model for assessing coronary artery disease risk in postmenopausal women: a dual-center retrospective study.2026