Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
June 17, 2025Reviews in Cardiovascular MedicineOpen Access

The pooled C-index for ML models in the validation sets was 0.77 (95% CI 0.74-0.81) in predicting MACEs post MI, with a sensitivity of 0.78 (95% CI 0.73-0.82) and specificity of 0.85 (95% CI 0.81-0.89).

View Full Paper
Ask AI
Bookmark
Share

Why the study?

Despite the development of machine learning models to predict MACEs following MI, evidence-based proof validating their performance remained lacking.

Do machine learning models accurately predict major adverse cardiovascular events in patients following myocardial infarction?

Comparison

Machine learning models for predicting MACEs

Design

Systematic review and meta-analysis

Authors

YXYi XiangXXXiaoman Xiong

Discussion

Loading...

Member takes

Overview

ML models show good accuracy for post-MI MACE prediction; leaves open need for prospective validation before routine use.

Structured PICO

Do machine learning models accurately predict major adverse cardiovascular events in patients following myocardial infarction?

P
Population
59,392 patients with myocardial infarction (MI) across 28 studies, including a subgroup of patients who underwent percutaneous coronary intervention (PCI).
I
Intervention
Machine learning (ML) models (e.g., logistic regression, random forest, deep learning) for predicting major adverse cardiovascular events (MACEs).
C
Comparator
Traditional clinical risk scoring tools (GRACE, TIMI scores).
O
Outcome
Predictive performance for MACEs, measured by C-index, sensitivity, and specificity in validation sets.composite

Machine learning models demonstrate favorable predictive accuracy for major adverse cardiovascular events following myocardial infarction, generally outperforming traditional scoring tools like GRACE and TIMI.

Limitations

  • Quality of the included studies varies widely with flaws in analytical methods
  • Significant heterogeneity among the models regarding prediction time points, MACE definitions, and number of variables
  • Insufficient sample sizes in some studies leading to increased risk of overfitting
  • Lack of external validation in many studies
  • No reports on clinical implementation effects

Cite This Study

Xiang et al. (2025) studied this question.

synapsesocial.com/papers/6a7e3e28dea41f5fbab7b002https://doi.org/10.31083/rcm37224
View Full Paper
Ask AI
Bookmark
Share