Synapse
⌘+K
Synapse
PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
January 8, 2026Scientific ReportsOpen Access

XGBoost based machine learning prediction model for major adverse cardiovascular events after PCI in STEMI patients

View Full Paper
Ask AI
Bookmark
Share

Key result

The XGBoost model demonstrated superior predictive capability for MACE in STEMI patients post-PCI with an AUC of 0.71 on the test set.

Authors

NZNing ZhangJWJing WangCSChen Shen

Discussion

Loading...

Member takes

Overview

Machine learning predicts major adverse cardiovascular events in STEMI patients after PCI, suggesting improved risk assessment methods.

Key Points

  • This research aims to develop a predictive model for major adverse cardiovascular events (MACE) in STEMI patients after percutaneous coronary intervention (PCI).
  • Analyzed clinical data from 1,011 STEMI patients post-PCI.
  • Collected 37 variables including demographics, hemodynamic data, and lab results.
  • Employed six machine learning algorithms to identify key predictors for MACE.
  • Utilized XGBoost for final model construction and performance evaluation.
  • Applied Shapley additive explanations (SHAP) and partial dependence plots (PDP) for model interpretation.
  • The XGBoost model showed the best performance with an AUC of 0.81 on the training set and 0.71 on the test set.
  • Key predictors included LCX occlusion, KILLIP classification, lymphocyte count, and LVEF among others.
  • SHAP analysis indicated that higher levels of lymphocytes and male sex were negatively associated with MACE.
  • PDP demonstrated interactions between AST, LCX, and lymphocyte count affecting MACE risk.

Study Design

Type

Observational (n=1,011)

Multicenter

Yes

PICO

P
Population
ST-segment elevation myocardial infarction (STEMI) (n=1,011)
I
Intervention / Comparator
XGBoost model vs Conventional clinical indicators, TIMI and PAMI risk scores
O
Primary Outcome
Major Adverse Cardiovascular Events (MACE) — null (null), p=<0.05

Main Result

Effect estimate: null (95% CI null)

p-value: p=<0.05

Limitations

  • Retrospective design may introduce bias.
  • Single-center study may limit generalizability.

Cite This Study

Zhang et al. (2026) conducted an observational in ST-segment elevation myocardial infarction (STEMI) (n=1,011). XGBoost model vs. Conventional clinical indicators, TIMI and PAMI risk scores was evaluated on Major Adverse Cardiovascular Events (MACE) (null, 95% CI null, p=<0.05). The XGBoost model demonstrated superior predictive capability for MACE in STEMI patients post-PCI with an AUC of 0.71 on the test set.

synapsesocial.com/papers/69608779fa51ca23bb9809a0https://doi.org/10.1038/s41598-025-34441-1
View Full Paper
Ask AI
Bookmark
Share