Radiomics-Based Classification of Severe Pulmonary Regurgitation in Patients with Repaired Tetralogy of Fallot by Cardiac Magnetic Resonance

Cardiac magnetic resonance (CMR) radiomics is a novel technique for advanced cardiac image phenotyping. This study aimed to develop a radiomics-based classification model by CMR images and thereby to identify rTOF patients with severe pulmonary regurgitation (PR). In our results, the radiomics-based classification model can successfully identify rTOF patients with PR ≥25% by routine CMR cine images. The extracted features underlined the RV intracardiac flow alteration due to PR and the potential right ventricular remodeling in rTOF patients with severe PR. This radiomics-based information may be helpful in determining the appropriate timing for pulmonary valve replacement in the future.