PuSH - Publication Server of Helmholtz Zentrum München

Piffer, A.* ; Buchner, J.A. ; Gennari, A.G.* ; Grehten, P.* ; Sirin, S.* ; Ross, E.* ; Ezhov, I.V.* ; Rosier, M. ; Peeken, J.C.* ; Piraud, M. ; Menze, B.* ; Crane, C.A.* ; Jakab, A.* ; Kofler, F.

Enhancing efficiency in paediatric brain tumour segmentation using a pathologically diverse single-center clinical dataset.

Neurooncol. Adv. 8:vdag024 (2026)
Publ. Version/Full Text Postprint DOI PMC
Open Access Gold
Creative Commons Lizenzvertrag
BackgroundBrain tumours are the most common solid malignancies in children, encompassing diverse histological, molecular subtypes and imaging features and outcomes. Paediatric brain tumours (PBTs), including high- and low-grade gliomas (HGG, LGG), medulloblastomas (MB), ependymomas, and rarer forms, pose diagnostic and therapeutic challenges. Deep learning (DL)-based segmentation offers promising tools for tumour delineation, yet its performance across heterogeneous PBT subtypes and MRI protocols remains uncertain.MethodsA retrospective single-centre cohort of 174 paediatric patients with HGG, LGG, medulloblastomas (MB), ependymomas, and other rarer subtypes was used. MRI sequences included T1, T1 post-contrast (T1-C), T2, and FLAIR. Manual annotations were provided for four tumour subregions: whole tumour (WT), T2-hyperintensity (T2H), enhancing tumour (ET), and cystic component (CC). A 3D nnU-Net model was trained and tested (121/53 split), with segmentation performance assessed using the Dice similarity coefficient (DSC) and compared against intra- and inter-rater variability.ResultsThe model achieved robust performance for WT and T2H (mean DSC: 0.85), comparable to human annotator variability (mean DSC: 0.86). ET segmentation was moderately accurate (mean DSC: 0.75), while CC performance was poor. Segmentation accuracy varied by tumour type, MRI sequence combination, and location. Notably, T1, T1-C, and T2 alone produced results nearly equivalent to the full protocol.
Altmetric
Additional Metrics?
[➜Log in]
Edit extra informations Login
Publication type Article: Journal article
Document type Scientific Article
Keywords Segmentation ; Sørensen–dice Coefficient ; Radiomics ; Magnetic Resonance Imaging ; Similarity (geometry) ; Neuroradiology ; Tumour Heterogeneity; Management
ISSN (print) / ISBN 2632-2498
e-ISSN 2632-2498
Journal Neuro-oncology advances
Quellenangaben Volume: 8, Issue: 1, Pages: , Article Number: vdag024 Supplement: ,
Publisher Oxford University Press
Publishing Place Great Clarendon St, Oxford Ox2 6dp, England
Reviewing status Peer reviewed
Institute(s) Helmholtz Artifical Intelligence Cooperation Unit (HAICU)
Institute of Radiation Medicine (IRM)
Grants Anna Valentina Lioba Eleonora Claire Javid Mamasani and the Gemeinntzige Hertie Stiftung
Forschungszentrum fr das Kind Universitt Zrich
Prof. Max Cloetta Foundation
Swiss National Science Foundation
EMDO Foundation and Vontobel Foundation