Pediatric Personalized Deep Learning Models for Segmentation of Hepatoblastoma at CT and MRI.
AI interpretation is pending for this paper.
Open original publication →What the AI sees
Not AI summarized yet.
Research significance
Pending deeper interpretation.
Source abstract
Purpose To evaluate the generalizability of adult-trained models for hepatoblastoma segmentation to pediatric patients and to develop two deep learning (DL) models, M P C T and M p M R I , specifically trained on pediatric contrast-enhanced CT and T2-weighted MRI scans, respectively. Materials and Methods Imaging data from the multicenter Children's Oncology Group AHEP0731 trial (NCT00980460; May 2008-July 2018) were analyzed. DL models employing the three-dimensional U-Net architecture were trained using DCT-Train and DMRI-Train. These models were evaluated on DCT-Val and DMRI-Val using the Dice similarity coefficient (DSC), and model segmentations were compared with manual segmentations from three annotators (R1, R2, and R3), their consensus (Rc), and adult-trained model ( M A C T ) segmentations. Volume percentage error analysis was performed to evaluate segmentation precision. Results A total of 104 participants (mean age ± SD, 28.2 months ± 30.5; 64 male; DCT-Train = 56, DCT-Val = 48) were included in the CT dataset and 123 (31.5 months ± 38.4; 87 male; DMRI-Train = 50, DMRI-Val = 73) in the MRI dataset. M P C T achieved good agreement with consensus segmentation (DSC = 0.86 [95% CI: 0.80, 0.91]) and exhibited higher agreement than M A C T with R1 (0.83 vs 0.55), R2 (0.85 vs 0.55), R3 (0.84 vs 0.54), and Rc (0.86 vs 0.55) segmentations. Volume percentage error analysis revealed that M P C T achieved segmentation results on par with or better than those of a novice annotator (R3) in high-precision scenarios. M P M R I also achieved a DSC of 0.86, demonstrating good agreement with Rc. Conclusion The pediatric-trained DL-based models outperformed adult-trained models for accurate segmentation of pediatric hepatoblastoma. Keywords: Pediatrics, Deep Learning, Liver, MR-Imaging, Abdomen/GI, Algorithm Development ClinicalTrials.gov NCT00980460 Supplemental material is available for this article. © The Author(s) 2026. Published by the Radiological Society of North America under a CC BY 4.0 license.