Clinical Report: A Multiphasic CT-Based Decision Support Model for Pre-Treatment Differentiation of Pediatric Hepatoblastoma and Focal Nodular Hyperplasia

Overview

This study developed and validated a multiphasic CT-based decision support model to differentiate pediatric hepatoblastoma (HB) from focal nodular hyperplasia (FNH). The model demonstrated high diagnostic performance, with an AUC of 0.980.

Background

Hepatoblastoma (HB) is the most common primary malignant liver tumor in children, requiring aggressive management, while focal nodular hyperplasia (FNH) is a benign lesion typically managed conservatively. Accurate differentiation between these two conditions is crucial to avoid unnecessary interventions and ensure appropriate treatment pathways.

Data Highlights

Key Findings

• The DL+QEF model achieved an AUC of 0.980 in the temporal test cohort.
• DL+QEF demonstrated 94.9% sensitivity for HB and 92.9% specificity for FNH.
• Junior radiologist accuracy improved from 88.4% to 95.0% with model assistance.
• In atypical cases, DL+QEF outperformed junior radiologists (90.2% vs. 68.3%).
• The model integrates deep learning predictions with quantitative enhancement features.

Clinical Implications

The DL+QEF model may assist radiologists in accurately differentiating between HB and FNH, particularly in atypical cases.

Conclusion

The multiphasic CT-based DL+QEF model shows diagnostic performance for differentiating pediatric HB from FNH.

Related Resources & Content

1. European Radiology, 2024 -- Models for Distinguishing Benign and Malignant Liver Lesions Using Multiparametric Dual-Energy Non-Contrast CT
2. European Radiology, 2024 -- Utilizing Radiomics for Distinguishing Between Wilms Tumor and Adrenal Neuroblastoma in Pediatric Patients
3. npj Digital Medicine, 2025 -- Deep learning model for assessing survival benefits in hepatocellular carcinoma patients undergoing intra-arterial therapies based on proliferative subtype
4. The ASCO Post, 2022 -- Study Evaluates Effectiveness of Existing Risk Stratification System for Hepatoblastoma
5. CAP Approved, 2025 -- Liver.Hepatoblastoma_5.0.0.0.REL_CAPCP
6. Imaging of Pediatric Liver Tumors: A COG Diagnostic Imaging Committee/SPR Oncology Committee White Paper - PMC
7. Gadoxetic acid-enhanced MRI in differentiating focal nodular hyperplasia from hepatocellular adenoma in children | British Journal of Radiology | Oxford Academic
8. CAP Approved Liver.Hepatoblastoma_5.0.0.0.REL_CAPCP
9. Imaging of Pediatric Liver Tumors: A COG Diagnostic Imaging Committee/SPR Oncology Committee White Paper - PMC
10. Gadoxetic acid-enhanced MRI in differentiating focal nodular hyperplasia from hepatocellular adenoma in children | British Journal of Radiology | Oxford Academic