Prediction of Fontan failure and correlates of Fontan-associated liver disease severity using machine learning and radiomic features from multi-parametric abdominal MRI

Models incorporating radiomic features from abdominal MRI in Fontan circulation patients demonstrate moderate diagnostic performance for predicting Fontan failure as well as correlates of FALD severity. These models outperformed models containing only clinical electronic health record data and did not improve with ensembled radiomic and clinical data.