EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Machine Learning in Predicting Cardiac Events for ESRD Patients: A Framework for Clinical Decision Support

Chien-Wei Chuang, Chung-Kuan Wu, Chao-Hsin Wu, Ben-Chang Shia iD, Mingchih Chen iD

Diagnostics · 22.04.2025 · 15 (9) · 1063

DOI: 10.3390/diagnostics15091063 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Diagnostics, online erschienen am 22.04.2025. Verfasst von Chien-Wei Chuang und 4 weiteren Autor:innen. Beteiligte Einrichtung: Graduate Institute of Business Administration, Fu Jen Catholic University, New Taipei City 242062, Taiwan. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 48 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science and Technology Council, Fu Jen Catholic University, Shin Kong Wu Ho-Su Memorial Hospital Research Foundation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background/Objectives: Patients with end-stage renal disease (ESRD) are at an increased risk of major adverse cardiac events (MACEs), highlighting the need for accurate risk prediction and personalized interventions. This study aims to develop and evaluate machine learning (ML) models to identify key predictive features and enhance clinical decision-making in MACE risk assessment.

Methods

A dataset comprising 84 variables, including patient demographics, laboratory findings, and comorbidities, was analyzed using CatBoost, XGBoost, and LightGBM. Feature selection, cross-validation, and SHAP (SHapley Additive exPlanations) analyses were employed to improve model interpretability and clinical relevance.

Results

CatBoost exhibited the highest predictive performance among the models tested, achieving an AUC of 0.745 (0.605–0.83) with balanced sensitivity and specificity. Key predictors of MACEs included antiplatelet use, the grade of left ventricular hypertrophy, and serum albumin levels. SHAP analysis enhanced the interpretability of model outputs, supporting clinician-led risk stratification.

Conclusions

This study highlights the potential of ML-based predictive modeling to improve MACE risk assessment in patients with ESRD. The findings support the adoption of ML models in clinical workflows by integrating explainable AI methods to enable individualized treatment planning. Future integration with electronic health record systems may facilitate real-time decision-making and enhance patient outcomes.

Schlagwörter

Stichwörter der Autor:innen

ESRDartificial intelligencemachine learningmedical decision-makingrisk factor

Autor:innen und Institutionen

  • Graduate Institute of Business Administration, Fu Jen Catholic University, New Taipei City 242062, Taiwan.
  • Artificial Intelligence Development Center, Fu Jen Catholic University, New Taipei City 242062, Taiwan.
  • Division of Nephrology, Shin Kong Wu Ho-Su Memorial Hospital, Taipei 111045, Taiwan.
  • Dialysis Access Management Center, Shin Kong Wu Ho-Su Memorial Hospital, Taipei 111045, Taiwan.
  • School of Medicine, Fu Jen Catholic University, New Taipei City 242062, Taiwan.

ORCID-Profile: Ben-Chang Shia, Mingchih Chen

Förderung

  • National Science and Technology Council
  • Fu Jen Catholic University
  • Shin Kong Wu Ho-Su Memorial Hospital Research Foundation

Zitieren

APA

Chuang, C. W., Wu, C. K., Wu, C. H., Shia, B. C., & Chen, M. (2025). Machine Learning in Predicting Cardiac Events for ESRD Patients: A Framework for Clinical Decision Support. Diagnostics, 15(9), 1063. https://doi.org/10.3390/diagnostics15091063

Vancouver

Chuang CW, Wu CK, Wu CH, Shia BC, Chen M. Machine Learning in Predicting Cardiac Events for ESRD Patients: A Framework for Clinical Decision Support. Diagnostics. 2025;15(9):1063. doi:10.3390/diagnostics15091063

BibTeX
@article{Chuang2025,
  author = {Chuang, Chien-Wei and Wu, Chung-Kuan and Wu, Chao-Hsin and Shia, Ben-Chang and Chen, Mingchih},
  title = {Machine Learning in Predicting Cardiac Events for ESRD Patients: A Framework for Clinical Decision Support},
  journal = {Diagnostics},
  year = {2025},
  volume = {15},
  number = {9},
  pages = {1063},
  doi = {10.3390/diagnostics15091063}
}
Kostenlos starten →