Automatic uncertainty-based quality controlled T1 mapping and ECV analysis from native and post-contrast cardiac T1 mapping images using Bayesian vision transformer

Tewodros Weldebirhan Arega, Stéphanie Bricq, François Le Grand, Alexis Jacquier, Alain Lalande, Fabrice Mériaudeau

Abstract: Highlights•Automatic uncertainty-based quality control framework for T1 mapping and extracellular volume (ECV) analysis.•Segmentation of cardiac structures from T1 mapping images using a Bayesian Swin transformer-based U-Net.•A novel uncertainty-based quality control (QC) that leverages image-level uncertainty metrics.•The automatically computed myocardial T1 and ECV values can characterize different myocardial tissues.