Quantitative radiomics and morphometry analysis of brain sub-regions reveals neurodevelopmental alterations in children with Tetralogy of Fallot

BACKGROUND: Tetralogy of Fallot (TOF) is one of the most common forms of cyanotic congenital heart disease. Although surgical advances have markedly improved survival rates, children with TOF remain at increased risk for neurodevelopmental impairments. However, the underlying neural mechanisms contributing to these deficits remain unclear, and reliable early biomarkers for neurodevelopmental risk are still lacking. METHODS: This study enrolled 60 children with TOF and 61 healthy controls who underwent high-resolution three-dimensional T1-weighted structural MRI (sMRI) and neurodevelopmental assessment. Radiomics and morphometric features were extracted from 68 brain regions using Pyradiomics and FreeSurfer. Least absolute shrinkage and selection operator (LASSO) regression and support vector machine (SVM) models were used to identify significant alterations. To ensure robustness and avoid overfitting, both employing 10-fold cross-validation for parameter tuning and internal validation. Associations between these features and neurodevelopmental scales and clinical indicators were also analyzed. RESULTS: Children with TOF showed significant alterations in brain regions involved in verbal, motor, visual, and executive functions. The radiomics-morphometry fusion model outperformed single-feature models, achieving an The area under the receiver operating characteristic curve(AUC) of 0.99 (training) and 0.77 (testing). Significant correlations were observed between altered features and neurodevelopmental and clinical metrics. CONCLUSION: This study developed a radiomics-morphometry fusion model that detects TOF-related brain alterations from both macro- and microstructural perspectives. These radiomics and morphometric features may serve as imaging biomarkers for early neurodevelopmental intervention, offering insights into potential therapeutic targets.