Machine Learning–Driven Risk Prediction Model in Transthyretin Amyloid Cardiomyopathy - Summary - MDSpire
Coming Soon: Introducing MDSpire News. Learn more
Conexiant’s news site is now MDSpire News. Learn more

A Machine Learning Model for Predicting Risk in Transthyretin Amyloid Cardiomyopathy

  • By

  • Giovanni Baj

  • Nicola Ciocca

  • Pooya Mohammadi Kazaj

  • Xuan Ma

  • Annina A. Studer Bruengger

  • Simon F. Stämpfli

  • Niklas F. Ehl

  • Sarah Hugelshofer

  • Otmar Pfister

  • Joëlle Lehmann

  • Christoph Ryffel

  • Lukas Hunziker

  • Michael Poledniczek

  • Andreas Kammerlander

  • George C. M. Siontis

  • Stephan Windecker

  • Moritz J. Hundertmark

  • Christian Nitsche

  • Isaac Shiri

  • Christoph Gräni

  • October 1, 2026

Share

Objective:

To develop and validate an AI-based model predicting all-cause mortality or heart-failure hospitalization in patients with transthyretin cardiac amyloidosis (ATTR-CM).

Approach:
  • Study population: Patients with confirmed ATTR-CM were prospectively recruited from six Swiss tertiary centers and the Medical University of Vienna.
  • Outcome and data: The primary outcome was all-cause mortality or hospitalization for heart failure. Baseline clinical, laboratory, imaging, and genetic data were collected.
  • Model development and validation: A random survival forest was evaluated using internal-external cross-validation, successively reserving each center as a test set. Training included MissForest imputation, LASSO feature selection, and hyperparameter optimization.
Interpretation:

The provided context describes the study rationale and methods but does not report model performance or outcome results.

Limitations:
  • The supplied article context ends before reporting validation metrics or other results.

Original Source(s)

Machine Learning–Driven Risk Prediction Model in Transthyretin Amyloid Cardiomyopathy

  • by Giovanni Baj, Nicola Ciocca, Pooya Mohammadi Kazaj, Xuan Ma, Annina A. Studer Bruengger, Simon F. Stämpfli, Niklas F. Ehl, Sarah Hugelshofer, Otmar Pfister, Joëlle Lehmann, Christoph Ryffel, Lukas Hunziker, Michael Poledniczek, Andreas Kammerlander, George C. M. Siontis, Stephan Windecker, Moritz J. Hundertmark, Christian Nitsche, Isaac Shiri, Christoph Gräni

  • October 1, 2026

Related Content

Representation of the LGBTQIA+ Patient Community in Cardiovascular Disease Clinical Guidelines

Hybrid data balancing and ensemble learning with SHAP-based explainability to predict cardiovascular disease using a unified dataset

AI-Assisted Carotid Ultrasound for Nonexpert Use in Primary Care Cardiovascular Prevention: User-Centered Iterative Development Study