• amyloidosis
• cardiac magnetic resonance
• deep learning
• diagnosis

Background: Cardiac magnetic resonance (CMR) is part of the diagnostic work-up for cardiac amyloidosis (CA). Deep learning (DL) is an application of artificial intelligence that may allow to automatically analyze CMR findings and possibly establish the likelihood of CA.
Methods: Subjects with suspected CA (n=206 patients; n=100, 49% with unexplained left ventricular – LV – hypertrophy; n=106, 51% with blood dyscrasia and suspected light-chain amyloidosis) underwent 1.5 T CMR. Patients were randomized to the training (n=134, 65%), validation (n=30, 15%), and testing subgroups (n=42, 20%). Short axis, 2-chamber, 4-chamber late gadolinium enhancement (LGE) images were evaluated by 3 networks (DL algorithms). Tags “amyloidosis present” or “absent” were attributed when the average probability of CA from the 3 networks was ≥50% or <50%, respectively. The DL strategy was compared to a machine learning (ML) algorithm that reproduced exam reading by an experienced operator, and combined all CMR features deriving from manual extraction: biventricular volumes and function, LGE presence and pattern, early darkening, pericardial and pleural effusion.
Results: The DL algorithm displayed good diagnostic accuracy (88%), with an area under the curve (AUC) of 0.982. The precision (positive predictive value), recall score (sensitivity), and F1 score (a measure of test accuracy) were 83%, 95%, and 89% respectively. A ML algorithm considering all CMR features had a similar diagnostic yield to DL strategy (AUC 0.952 vs. 0.982; p=0.39).
Conclusions: A DL-based approach relying on LGE acquisitions displayed a similar diagnostic performance for CA to the simulation of CMR reading by experienced operators.