Pediatric brain tumor classification using digital pathology and deep learning: Evaluation of SOTA methods on a multi‐center Swedish cohort
Brain Pathology · 2026
DOI: 10.1111/bpa.70029 ↗Kurz gefasst
Multizentrische Studie in der Zeitschrift Brain Pathology, 2026. Verfasst von Iulian Emil Tampu und 9 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biomedical Engineering, Linköping University, Linköping, Sweden. Hauptthemen laut MEDLINE: Brain Neoplasms, Deep Learning. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 46 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: Linköping University's Cancer Strength Area, ALF Grants, Region Östergötland, Vinnova via Medtech4Health and Analytic Imaging Diagnostics Arena.
Abstract (Originalsprache)
Brain tumors are the most common solid tumors in children and young adults, but the scarcity of large histopathology datasets has limited the application of computational pathology in this group. This study implements two weakly supervised multiple‐instance learning (MIL) approaches on patch features obtained from state‐of‐the‐art histology‐specific foundation models to classify pediatric brain tumors in hematoxylin and eosin whole slide images (WSIs) from a multi‐center Swedish cohort. WSIs from 540 subjects (age 8.5 ± 4.9 years) diagnosed with brain tumors were gathered from the six Swedish university hospitals. Instance (patch)‐level features were obtained from WSIs using three pre‐trained feature extractors: ResNet50, UNI, and CONCH. Instances were aggregated using attention‐based MIL (ABMIL) or clustering‐constrained attention MIL (CLAM) for patient‐level classification. Models were evaluated on three classification tasks based on the hierarchical classification of pediatric brain tumors: tumor category, family, and type. Model generalization was assessed by training on data from two of the centers and testing on data from four other centers. Model interpretability was evaluated through attention mapping. The highest classification performance was achieved using UNI features and ABMIL aggregation, with Matthew's correlation coefficient of 0.76 ± 0.04, 0.63 ± 0.04, and 0.60 ± 0.05 for tumor category, family, and type classification, respectively. When evaluating generalization, models utilizing UNI and CONCH features outperformed those using ResNet50. However, the drop in performance from the in‐site to out‐of‐site testing was similar across feature extractors. These results show the potential of state‐of‐the‐art computational pathology methods in diagnosing pediatric brain tumors at different hierarchical levels with fair generalizability on a multi‐center national dataset.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Biomedical Engineering, Linköping University, Linköping, Sweden.
- Center for Medical Image Science and Visualization, Linköping University, Linköping, Sweden.
- Crown Princess Victoria Children's Hospital and Department of Health, Medicine and Caring Sciences, Linköping University, Linköping, Sweden.
- Department of Radiology and Department of Health, Medicine and Caring Sciences, Linköping University, Linköping, Sweden.
- Department of Oncology-Pathology, Karolinska Institutet, Solna, Sweden.
- Department of Clinical Pathology and Cancer Diagnostics, Karolinska University Hospital, Huddinge, Sweden.
ORCID-Profile: Iulian Emil Tampu
Förderung
- Linköping University's Cancer Strength Area
- ALF Grants, Region Östergötland
- Vinnova via Medtech4Health and Analytic Imaging Diagnostics Arena
- Swedish Childhood Cancer Fund
- Joanna Cocozza's Foundation for Children's Medical Research
Zitieren
APA
Tampu, I. E., Nyman, P., Spyretos, C., Blystad, I., Shamikh, A., Prochazka, G., de Ståhl, T. D., Sandgren, J., Lundberg, P., & Haj-Hosseini, N. (2026). Pediatric brain tumor classification using digital pathology and deep learning: Evaluation of SOTA methods on a multi‐center Swedish cohort. Brain Pathology. https://doi.org/10.1111/bpa.70029
Vancouver
Tampu IE, Nyman P, Spyretos C, Blystad I, Shamikh A, Prochazka G, et al. Pediatric brain tumor classification using digital pathology and deep learning: Evaluation of SOTA methods on a multi‐center Swedish cohort. Brain Pathology. 2026. doi:10.1111/bpa.70029
BibTeX
@article{Tampu2026,
author = {Tampu, Iulian Emil and Nyman, Per and Spyretos, Christoforos and Blystad, Ida and Shamikh, Alia and Prochazka, Gabriela and de Ståhl, Teresita Díaz and Sandgren, Johanna and Lundberg, Peter and Haj-Hosseini, Neda},
title = {Pediatric brain tumor classification using digital pathology and deep learning: Evaluation of SOTA methods on a multi‐center Swedish cohort},
journal = {Brain Pathology},
year = {2026},
doi = {10.1111/bpa.70029}
}
EUVIMED