Impact of deep learning on CT-based organ-at-risk delineation for flank irradiation in paediatric renal tumours: a SIOP-RTSG radiotherapy committee study

Mianyong Ding; Matteo Maspero; Semi Harrabi; Emmanuel Jouglar; Sabina Vennarini; Timothy Spencer; Britta Weber; Henriette Magelssen; Karen Van Beek; Remus Stoica; Simonetta Saldi; Tom Boterberg; Patrick Melchior; Marry M van den Heuvel-Eibrink; Geert O Janssens

doi:10.1016/j.ctro.2025.101051

Impact of deep learning on CT-based organ-at-risk delineation for flank irradiation in paediatric renal tumours: a SIOP-RTSG radiotherapy committee study

Clin Transl Radiat Oncol. 2025 Sep 19:56:101051. doi: 10.1016/j.ctro.2025.101051. eCollection 2026 Jan.

Authors

Mianyong Ding^{1

2}, Matteo Maspero^{2

3}, Semi Harrabi⁴, Emmanuel Jouglar⁵, Sabina Vennarini⁶, Timothy Spencer⁷, Britta Weber⁸, Henriette Magelssen⁹, Karen Van Beek¹⁰, Remus Stoica¹¹, Simonetta Saldi¹², Tom Boterberg¹³, Patrick Melchior¹⁴, Marry M van den Heuvel-Eibrink^{1

15}, Geert O Janssens^{1

2}

Affiliations

¹ Princess Máxima Centre for Paediatric Oncology, Utrecht, Netherlands (the).
² Department of Radiation Oncology, Imaging and Cancer Division, University Medical Centre Utrecht, Utrecht, Netherlands (the).
³ Computational Imaging Group for MR Diagnostics & Therapy, Centre for Image Sciences, University Medical Centre Utrecht, Utrecht, Netherlands (the).
⁴ University Hospital Heidelberg, Heidelberg Ion Beam Therapy Centre (HIT), Dept. of Radiation Oncology, Heidelberg, Germany.
⁵ Department of Radiation Oncology, Institut Curie, PSL University, Paris, France.
⁶ Paediatric Radiotherapy Unit, IRCCS Fondazione Istituto Nazionale Tumori-Milan, Italy.
⁷ Bristol Cancer Institute, Bristol, UK.
⁸ Danish Centre for Particle Therapy, Aarhus, Denmark.
⁹ Department of Oncology, Oslo University Hospital, Oslo, Norway.
¹⁰ Department of Radiation Oncology, University Hospital Leuven, Belgium.
¹¹ Emergency Clinical Hospital for Children "Marie Skłodowska Curie", Bucharest, Romania.
¹² Hospital Santa Maria della Misericordia, Perugia, Italy.
¹³ Department of Radiation Oncology, Ghent University Hospital, Ghent, Belgium.
¹⁴ Department of Radiation Oncology, University of Saarland, Homburg, Germany.
¹⁵ Wilhelmina Children's Hospital-Division of CHILD HEALTH, University Medical Centre Utrecht, University of Utrecht, Utrecht, Netherlands (the).

Abstract

Background and purpose: Integrating deep learning (DL) for auto-contouring has significantly improved organ-at-risk (OAR) delineation in adult radiotherapy. However, its application in paediatric radiotherapy remains limited. This study evaluates DL-based auto-contouring of OARs, followed by manual revisions, for paediatric flank irradiation, focusing on delineation time, accuracy, and inter-observer variability (IOV).

Materials and methods: Twelve paediatric radiation oncologists from nine countries affiliated with the SIOP Renal Tumour Study Group participated in a two-day workshop. Participants were randomly divided into two groups: one performed manual delineation first, followed by DL-based revision, while the other group performed in reverse order. Eight thoracoabdominal OARs were delineated on non-contrast CTs of renal tumour patients (ages 1-6). DL-based contours were generated using a model for paediatric abdominal cases. Delineation time was recorded, accuracy and IOV were assessed using the Dice similarity coefficient (DSC), 95th percentile Hausdorff distance, mean surface distance against a STAPLE consensus (threshold = 0.95), and an expert reference.

Results: In total, 122 manual delineations and 254 DL-based revisions were collected. DL-based auto-contouring reduced delineation time by 59 %, from 25.5 to 10.2 min. The mean DSC of all eight OARs improved from 0.91 to 0.97 using STAPLE reference and from 0.89 to 0.93 using expert reference. The pancreas exhibited the largest gain, with mean DSC increases ranging from 0.18 to 0.25. Delineation accuracy was significantly improved for seven OARs (p < 0.05), while IOV significantly decreased for the pancreas and heart in both references (p < 0.05).

Conclusion: Manually revising DL-based auto-contouring reduces delineation time, enhances accuracy, and reduces inter-observer variability in paediatric CT-based OAR delineation.

Keywords: Artificial intelligence; Auto-contouring; Flank irradiation; Inter-observer variability; Organs-at-risk; Wilms tumors.