Pediatric brain tumor classification using digital pathology and deep learning: Evaluation of SOTA methods on a multi-center Swedish cohort.

in Brain pathology (Zurich, Switzerland) by Iulian Emil Tampu, Per Nyman, Christoforos Spyretos, Ida Blystad, Alia Shamikh, Gabriela Prochazka, Teresita Díaz de Ståhl, Johanna Sandgren, Peter Lundberg, Neda Haj-Hosseini

TLDR

  • This study implements two weakly supervised MIL approaches to classify pediatric brain tumors in WSIs from a multi-center Swedish cohort, achieving fair generalizability and classification performance.
  • The results show the potential of state-of-the-art computational pathology methods in diagnosing pediatric brain tumors at different hierarchical levels.
  • Future studies could explore the application of these methods to other types of cancer and additional machine learning techniques.

Abstract

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.

Overview

  • The study aims to implement two weakly supervised multiple-instance learning (MIL) approaches to classify pediatric brain tumors in hematoxylin and eosin whole slide images (WSIs) from a multi-center Swedish cohort.
  • FSIs from 540 subjects (age 8.5 ± 4.9 years) diagnosed with brain tumors were gathered from six Swedish university hospitals.
  • The primary objective is to evaluate 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.

Comparative Analysis & Findings

  • 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.
  • The drop in performance from in-site to out-of-site testing was similar across feature extractors.

Implications and Future Directions

  • The study shows 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.
  • Future studies could explore the application of these methods to other types of cancer and additional machine learning techniques.
  • The development of novel attention mechanisms and feature extractors could further improve the performance of these methods in diagnosing pediatric brain tumors.
Read Full Article