A transformer-based survival model for prediction of all-cause mortality in patients with heart failure: a multi-cohort study - Takeaways - MDSpire

A transformer-based survival model for prediction of all-cause mortality in patients with heart failure: a multi-cohort study

  • By

  • Shishir Rao

  • Nouman Ahmed

  • Gholamreza Salimi-Khorshidi

  • Christopher Yau

  • Huimin Su

  • Nathalie Conrad

  • Folkert W. Asselbergs

  • Mark Woodward

  • Rod Jackson

  • John GF Cleland

  • Kazem Rahimi

  • January 8, 2026

  • 0 min

Share

  • 1

    Heart failure (HF) has a variable prognosis, making medium-term outcome prediction crucial for effective patient management.

  • 2

    Current risk assessment models for HF often rely on resource-intensive tests and overlook important comorbidities affecting mortality.

  • 3

    The Transformer-based Risk assessment survival model (TRisk) utilizes electronic health records to improve mortality prediction in HF patients.

  • 4

    TRisk demonstrated superior discrimination (C-index: 0.845) compared to the MAGGIC-EHR model (C-index: 0.728) in predicting all-cause mortality.

  • 5

    The TRisk model effectively captures the complete patient journey, incorporating diverse medical data types for enhanced prognostication.

Original Source(s)

A transformer-based survival model for prediction of all-cause mortality in patients with heart failure: a multi-cohort study

  • by Shishir Rao, Nouman Ahmed, Gholamreza Salimi-Khorshidi, Christopher Yau, Huimin Su, Nathalie Conrad, Folkert W. Asselbergs, Mark Woodward, Rod Jackson, John GF Cleland, Kazem Rahimi

  • January 8, 2026

Related Content

Thoracoscopically assisted resection of left ventricular myxoma: a case report

Transcatheter closure of PDA: how the pathway changed from classical fluoroscopy approach

Validation of 2 Syncope Risk Scores and Comparison With Physician Risk Estimation