Multicenter development and validation of machine-learning risk models to predict procedural complete revascularization and in-hospital heart failure in STEMI patients treated with primary PCI - Takeaways - MDSpire

Multicenter development and validation of machine-learning risk models to predict procedural complete revascularization and in-hospital heart failure in STEMI patients treated with primary PCI

  • By

  • Yumin Lin

  • Yufeng Qin

  • Kangkang Ou

  • Jichong Zhu

  • Bizhi Liao

  • May 13, 2026

  • 0 min

Share

  • 1

    Machine learning models were developed to predict in-hospital heart failure and complete revascularization in STEMI patients undergoing primary PCI.

  • 2

    The study utilized a multicenter cohort of STEMI patients, with 734 in the training group and 352 in the independent validation cohort.

  • 3

    CatBoost was identified as the most effective model for predicting in-hospital heart failure and procedural complete revascularization, showing high accuracy and strong calibration.

  • 4

    Key predictors for in-hospital heart failure included LAD involvement, age, symptom duration, and various clinical indicators.

  • 5

    The models demonstrated strong discrimination and clinical applicability, enhancing decision-making in catheterization laboratories for STEMI patients.

Original Source(s)

Multicenter development and validation of machine-learning risk models to predict procedural complete revascularization and in-hospital heart failure in STEMI patients treated with primary PCI

  • by Yumin Lin, Yufeng Qin, Kangkang Ou, Jichong Zhu, Bizhi Liao

  • May 13, 2026

Related Content

FDA Approves Baxdrostat for Resistant Hypertension

Phase 3 findings support the oral therapy in adults with persistent blood pressure elevation despite multidrug treatment.

Clinical outcomes of tricuspid transcatheter edge-to-edge repair in patients with tricuspid regurgitation

Corvad as a bridge to LVAD implantation in end-stage heart disease complicated by cardiogenic shock: a three-case series