Pediatric Personalized Deep Learning Models for Segmentation of Hepatoblastoma at CT and MRI. Radiol Imaging Cancer

Modanwal G, Kumar S, Viswanathan V, Morin CE, Rees MA, Squires JH, Tang ER, Katzenstein HM, Towbin AJ, Madabhushi A, Schooler GR. Pediatric Personalized Deep Learning Models for Segmentation of Hepatoblastoma at CT and MRI. Radiol Imaging Cancer. 2026 Mar;8(2):e250041. doi: 10.1148/rycan.250041. PMID: 41718532; PMCID: PMC13036680.

To evaluate the generalizability of adult-trained models for hepatoblastoma segmentation to pediatric patients and to develop two deep learning (DL) models, M CT/P and M MRI/P, 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.