From Adult Renal Masses to Pediatric Wilms Tumors: Evaluating Cross-Domain Generalization of Kidney Lesion Segmentation on MRI

Sarah de Boer; Myrthe A. D. Buser; Hartmut Häntze; Jawed Nawabi; Lisa C. Adams; Alida F. W. van der Steeg; Marry M. van den Heuvel-Eibrink; Annemieke S. Littooij

cs.CVOct 16, 2026

From Adult Renal Masses to Pediatric Wilms Tumors: Evaluating Cross-Domain Generalization of Kidney Lesion Segmentation on MRI

Authors: Sarah de Boer, Myrthe A. D. Buser, Hartmut Häntze, Jawed Nawabi, Lisa C. Adams, Alida F. W. van der Steeg, Marry M. van den Heuvel-Eibrink, Annemieke S. Littooij,

Organizations: Department of Medical Imaging, Radboudumc, Nijmegen, The Netherlands · Department of Radiology, Charité - Universitätsmedizin Berlin, Berlin, Germany · Department of Neuroradiology, Charité - Universitätsmedizin Berlin, Berlin, Germany · Department of Diagnostic and Interventional Radiology, Klinikum rechts der Isar, TUM University Hospital, Technical University of Munich, Munich, Germany · Princess Máxima Center for Pediatric Oncology, Utrecht, The Netherlands · Wilhelmina Children’s Hospital, Division of Childhealth, Utrecht, The Netherlands · Department of Radiology and Nuclear Medicine, University Medical Center Utrecht, Utrecht, The Netherlands · National Center for Tumor Diseases West, Essen, Germany · Institute of Artificial Intelligence in Medicine, University Hospital Essen, Essen, Germany · Department of Diagnostic and Interventional Radiology and Neuroradiology, University Hospital Essen, Essen, Germany

Abstract

Deep learning segmentation models for pediatric oncology are often trained on small, single-center cohorts, whereas much larger datasets exist for adult diseases, such as kidney cancer. Whether models trained on adult renal lesions generalize to pediatric Wilms tumors remains unclear. We trained a segmentation model on single-sequence T1- or T2-weighted adult kidney MRI and evaluated it on a held-out adult test set and a pediatric Wilms tumor cohort, where it was compared with a pediatric-specific nnU-Net. Segmentation performance was quantified using the Dice similarity coefficient (DSC) and 95th-percentile Hausdorff distance. On the adult test set, the model achieved a median DSC of 0.86 for renal lesions. On the pediatric cohort, the adult-trained model achieved a median DSC of 0.85 compared with 0.90 for the pediatric-specific model. Despite differences in age, anatomy, and tumor type, the adult-trained model showed promising performance on pediatric Wilms tumors, with only a modest difference in median DSC compared with the pediatric-specific model. However, non-inferiority could not be shown, highlighting the value of pediatric training data. In future work, a fine-tuning strategy could be explored.

Figures & tables

Figure 1: (a) Semi-supervised model training: MRSegmentator was used to localize both kidneys. A temporary nnU-Net was trained on manually annotated crops and used to generate pseudo-labels for previously unlabeled kidneys. Manual and automatic annotations were then merged to train the final model. (b) Tumor examples from the adult and pediatric cohorts.

	Adult cohort	Pediatric cohort
Institution	Charité – Universitätsmedizin Berlin	Princess Máxima Center for Pediatric Oncology
Use in this study	Training and in-domain test	External test (zero-shot)
Patients (sequences)	689 (1794)	45 (45)
Training	551 (1435)	–
Test	138 (359)	45 (45)
Age range	38–74 \ts years	7–109 \ts months

Table 1: Overview of the adult and pediatric cohorts. Renal mass volume is reported as the median and interquartile range.

Figure 2: Case-level DSC comparison between the adult- and pediatric-trained models on the Wilms tumor cohort.

Figure 3: Subgroup analysis comparing the performance (in DSC) per patient age group for the adult (left) and pediatric (right) models.

Medical Image Segmentation

Medical image segmentation delineates organs, lesions, and other structures in scans to support measurement, diagnosis, treatment planning, and research.

See all 768 papers

Overcoming data scarcity through multi-center federated learning for organs-at-risk segmentation in pediatric upper abdominal radiotherapy
May 7, 2026Mianyong Ding, Maximilian Knoll, Semi Harrabi +7Medical Image Segmentation Oncology
Department of Radiation Oncology, Imaging and Cancer Division, University Medical Centre Utrecht, Utrecht, The Netherlands · University Hospital Heidelberg, Heidelberg Ion Beam Therapy Centre (HIT), Department of Radiation Oncology, Heidelberg, Germany · Princess Máxima Centre for Pediatric Oncology, Utrecht, The Netherlands +3
Benchmarking Foundation Models for Renal Lesion Stratification in CT
May 8, 2026Hartmut Häntze, Sarah de Boer, Myrthe Buser +7Medical Imaging Foundation Models Medical Image Classification
Charité - Universitätsmedizin Berlin, Department of Radiology, Berlin, Germany · Radboudumc, Diagnostic Image Analysis Group, Nijmegen, The Netherlands · Klinikum rechts der Isar, TUM University Hospital, Technical University of Munich, Munich, Germany +2
VIDS-Seg: Towards Reliable Uncertainty Quantification in Pediatric Cardiac Ultrasound Segmentation
Aug 11, 2026Paul Fischer, Ece OzkanUltrasound Image Segmentation Uncertainty Quantification
Department of Biomedical Engineering, University of Basel, Basel, Switzerland

May 7, 2026J/K move · Enter open · S save

Overcoming data scarcity through multi-center federated learning for organs-at-risk segmentation in pediatric upper abdominal radiotherapy

Mianyong Ding, Maximilian Knoll, Semi Harrabi, Martine van Grotel+6

Department of Radiation Oncology, Imaging and Cancer Division, University Medical Centre Utrecht, Utrecht, The Netherlands · University Hospital Heidelberg, Heidelberg Ion Beam Therapy Centre (HIT), Department of Radiation Oncology, Heidelberg, Germany · Princess Máxima Centre for Pediatric Oncology, Utrecht, The Netherlands+3

Medical Image Segmentation Oncology Federated Learning

Open paper AskPDF arXiv

Deep learning-based organs/structures-at-risk(OARs) auto-contouring models can improve radiotherapy workflows, but models trained on adult data often underperform in pediatric patients. Developing robust pediatric-specific models is hindered by data scarcity and fragmentation across centers. Federated learning (FL) enables privacy-preserving collaborative training without the need for data sharing. We evaluated the feasibility and performance of FL for developing pediatric-specific OAR segmentation models across two European medical centers. Computed tomography (CT) images from pediatric patients from Utrecht and Heidelberg with a renal tumor or abdominal neuroblastoma were retrospectively collected and locally processed. An nnU-Net-based framework segmented 19 OARs using local and FL schemes. FL was implemented with secure weight exchange on a cloud storage across institutional firewalls. Performance was assessed using the Dice similarity coefficient (DSC), 95th percentile Hausdorff distance, and mean surface distance. Robustness to patient orientation, false-positive segmentation of surgically removed kidneys, and failure cases were identified. A total of 310 postoperative CTs from 272 patients (105 renal tumors, 167 neuroblastomas) were included. Local models performed well on their respective center data but showed significantly reduced cross-center performance for four to seven of the nine evaluated OARs (DSC). In contrast, the FL model matched local performance for at least seven of nine OARs and achieved the best cross-center results across three metrics, with DSC gains of 0.003-0.007 over local models. FL also maintained stable performance across patient orientations and reduced false-positive kidney segmentations. Real-world FL improves cross-center robustness of CT-based OAR segmentation models in pediatric upper abdominal tumors.

Abstract

Medical Image Segmentation

Explore similar work

Overcoming data scarcity through multi-center federated learning for organs-at-risk segmentation in pediatric upper abdominal radiotherapy

Benchmarking Foundation Models for Renal Lesion Stratification in CT

VIDS-Seg: Towards Reliable Uncertainty Quantification in Pediatric Cardiac Ultrasound Segmentation