Can Radiomics Identify Severe Right Ventricular Dilation in Patients with Repaired Tetralogy of Fallot by LV Cardiac Magnetic Resonance Images?

Radiomics analysis of cardiac magnetic resonance (CMR) may provide new insights into the quantitative analysis of cardiac imaging by extracting many computational quantitative metrics. This study aimed to develop a radiomics-based classification model by left ventricular (LV) CMR images to identify repaired tetralogy of Fallot (rTOF) patients with severe right ventricular (RV) dilation. In our results, the AUC were 0.81 and 0.91 for LV radiomics and LV radiomics+LV indices models, respectively. The extracted features underlined the LV intracardiac flow alteration due to RV dilation and the potential LV remodeling in rTOF patients with severe RV dilation.