Derivation and validation of a machine learning-driven score to predict the diagnostic yield of endomyocardial biopsy

To develop and validate a machine-learning-based score predicting the likelihood of diagnostic endomyocardial biopsy (EMB) using non-invasive data.

EMB yielded a definitive diagnosis in 19.9% of cases, primarily amyloidosis (50%).
Right ventricular late gadolinium enhancement (LGE) was the strongest predictor of diagnostic yield.
The predictive score ranged from 0-100 and demonstrated excellent discrimination with AUC values of 0.92 and 0.91 in cross-validation and testing set, respectively.

The machine-learning-based score may serve as a non-invasive tool to aid in decision-making regarding EMB in clinical practice.

The datasets analyzed are not publicly available due to ethical and privacy restrictions.
The underlying code for the study is not publicly available.

This study presents a promising machine-learning approach to enhance the diagnostic efficacy of EMB in heart failure patients.

Npj Digital Medicine

by Christian Basile, Christian L. Polte, Piero Gentile, Entela Bollano, Araz Rawshani, Anders Oldfors, Charlotta Ljungman, Sven-Erik Bartfay, Pia Dahlberg, Clara Hjalmarsson, Marie Björkenstam, Elena Gualini, Antonio Cannatá, Patrizia Pedrotti, Andrea Garascia, Gianluigi Savarese, Aldo Pietro Maggioni, Kristjan Karason, Emanuele Bobbio
February 9, 2026