An Interpretable Radiomics Model Based on Pituitary MRI to Predict Growth Hormone Deficiency in Short-statured Children: A Multicenter Study.
Authors
Affiliations (6)
Affiliations (6)
- Department of MRI, the First Affiliated Hospital of Henan University of Chinese Medicine, Zhengzhou, China (F.S., Q.X., J.S., Y.Y., J.S., L.Z.); First Clinical Medical College of Henan University of Chinese Medicine, Zhengzhou, China (F.S., Q.X., J.S., Y.Y., J.S.).
- Department of Pediatrics, the First Affiliated Hospital of Henan University of Traditional Chinese Medicine College of Pediatrics, Henan University of Traditional Chinese Medicine, Zhengzhou, China (X.R.).
- Department of Radiology, Children's Hospital Affiliated of Zhengzhou University, Henan Children's Hospital Zhengzhou Children's Hospital, Zhengzhou, China (S.S.).
- Department of Radiology, the Affiliated Children's Hospital of Xiangya School of Medicine, Central South University (Hunan Children's Hospital), Changsha, China (K.J., F.L.).
- Department of Radiology, Gold Coast University Hospital, Southport, Australia (J.Z.).
- Department of MRI, the First Affiliated Hospital of Henan University of Chinese Medicine, Zhengzhou, China (F.S., Q.X., J.S., Y.Y., J.S., L.Z.). Electronic address: [email protected].
Abstract
To develop and validate an interpretable radiomics model based on pituitary MRI to predict growth hormone deficiency (GHD) in children with short stature. This retrospective multicenter study enrolled 202 children (105 GHD, 97 idiopathic short stature [ISS]) as an internal cohort (7:3 ratio for training/testing cohorts) from institution I, and 138 children (61 GHD, 77 ISS) from institution II and institution III as an external validation cohort. Radiomics features were selected by SelectKBest and least absolute shrinkage and selection operator (LASSO), subsequently used to construct six machine learning models. Diagnostic performance of model was evaluated by the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and calibration curves. The interpretability of the model was assessed using Shapley additive explanations (SHAP). A total of 17 radiomics features were selected. Among all classifiers, support vector machine (SVM)-based radiomics model exhibited the highest diagnostic performance, with AUCs of 0.877 (95% CI: 0.813, 0.928), 0.878 (95% CI: 0.786, 0.951), and 0.885 (95% CI: 0.833, 0.937) in training, testing, and external validation cohorts, respectively. The SVM-integrated clinical-radiomics model yielded comparable efficacy, with AUCs of 0.874 (95% CI: 0.812, 0.928), 0.878 (95% CI: 0.786, 0.952), and 0.889 (95% CI: 0.830, 0.939) across the same cohorts. Both radiomics-based models significantly outperformed the clinical model (all p<0.001), while no statistically significant difference was observed between the radiomics and clinical-radiomics models (all p>0.05). The SHAP analysis identified three key radiomics features with significant differences between GHD and ISS groups (all p<0.001). The interpretable radiomics-driven SVM model effectively predicts GH levels, providing a clinically viable, non-invasive alternative to GH stimulation test in children with short stature.