EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Enhancing Explainable AI Stability with Realistic Synthetic Data for Cardiovascular Risk Prediction

Chang Sun, Michel Dumontier

Studies in Health Technology and Informatics · 21.05.2026

DOI: 10.3233/shti260317 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Studies in Health Technology and Informatics, erschienen am 21.05.2026. Verfasst von Chang Sun und 1 weiteren Autor:innen. Beteiligte Einrichtung: Institute of Data Science, Maastricht University. Hauptthemen laut MEDLINE: Cardiovascular Diseases, Machine Learning. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Data-driven such as machine learning models in rare disease research faces a significant challenge: scarce data cause performance degradation and unstable explanations that undermine clinical trust. This study investigated whether synthetic data augmentation can maintain prediction performance and improve explainability stability for cardiovascular disease risk prediction in Chronic Myeloid Leukemia patients. Using 235 real-world patients’ data, we employed DP-CGANS to generate synthetic records and evaluated their impact across decreasing sample sizes, comparing real-data-only versus real + synthetic data scenarios. Synthetic augmentation showed stable performance while real-data-only models degraded substantially. We observed synthetic data can significantly improve SHAP stability. These findings demonstrate that synthetic data augmentation has potential to address performance degradation and explanation instability in building reliable ML models for diseases with scarce patient data.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Cardiovascular DiseasesMachine LearningHumansPredictive Learning ModelsRisk AssessmentArtificial IntelligencePrediction AlgorithmsLeukemia, Myelogenous, Chronic, BCR-ABL PositiveLarge Language Models

Stichwörter der Autor:innen

Decision supportExplainable AILLMMulti-agentsSynthetic data

Autor:innen und Institutionen

  • Institute of Data Science, Maastricht University.
  • Department of Advanced Computing Sciences, Maastricht University.

Zitieren

APA

Sun, C., & Dumontier, M. (2026). Enhancing Explainable AI Stability with Realistic Synthetic Data for Cardiovascular Risk Prediction. Studies in Health Technology and Informatics. https://doi.org/10.3233/shti260317

Vancouver

Sun C, Dumontier M. Enhancing Explainable AI Stability with Realistic Synthetic Data for Cardiovascular Risk Prediction. Studies in Health Technology and Informatics. 2026. doi:10.3233/shti260317

BibTeX
@article{Sun2026,
  author = {Sun, Chang and Dumontier, Michel},
  title = {Enhancing Explainable AI Stability with Realistic Synthetic Data for Cardiovascular Risk Prediction},
  journal = {Studies in Health Technology and Informatics},
  year = {2026},
  doi = {10.3233/shti260317}
}
Kostenlos starten →