Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

BraTS-Path Challenge: Assessing Heterogeneous Histopathologic Brain Tumor Sub-regions

Spyridon Bakas, Siddhesh P. Thakur, Shahriar Faghani, Mana Moassefi, Ujjwal Baid, Verena Chung, Sarthak Pati, Shubham Innani, Bhakti Baheti, Jake Albrecht, Alexandros Karargyris, Hasan Kassem, MacLean P. Nasrallah, Jared T. Ahrendsen, Valeria Barresi, Maria A. Gubbiotti, Giselle Y. López, Calixto-Hope G. Lucas, Michael L. Miller, Lee A. D. Cooper, Jason T. Huse, William R. Bell

TL;DR

The BraTS-Path paper addresses the need to automatically identify heterogeneous histologic sub-regions within glioblastoma on digitized tissue, moving beyond prior radiology-focused BraTS benchmarks. It presents a multicenter, expert-annotated whole-slide image dataset with nine distinct sub-regions and a patch-level multiclass classification task, along with a standardized evaluation pipeline using MedPerf and GaNDLF metrics. The work establishes a community benchmark, enabling rigorous comparison of AI approaches for histopathology, and highlights potential clinical impact in tumor diagnosis, grading, and treatment planning. By capturing GBM heterogeneity across sites and providing ground-truth annotations, BraTS-Path aims to accelerate translation of AI tools into pathology workflows and improve patient care through more objective, scalable histopathologic assessment.

Abstract

Glioblastoma is the most common primary adult brain tumor, with a grim prognosis - median survival of 12-18 months following treatment, and 4 months otherwise. Glioblastoma is widely infiltrative in the cerebral hemispheres and well-defined by heterogeneous molecular and micro-environmental histopathologic profiles, which pose a major obstacle in treatment. Correctly diagnosing these tumors and assessing their heterogeneity is crucial for choosing the precise treatment and potentially enhancing patient survival rates. In the gold-standard histopathology-based approach to tumor diagnosis, detecting various morpho-pathological features of distinct histology throughout digitized tissue sections is crucial. Such "features" include the presence of cellular tumor, geographic necrosis, pseudopalisading necrosis, areas abundant in microvascular proliferation, infiltration into the cortex, wide extension in subcortical white matter, leptomeningeal infiltration, regions dense with macrophages, and the presence of perivascular or scattered lymphocytes. With these features in mind and building upon the main aim of the BraTS Cluster of Challenges https://www.synapse.org/brats2024, the goal of the BraTS-Path challenge is to provide a systematically prepared comprehensive dataset and a benchmarking environment to develop and fairly compare deep-learning models capable of identifying tumor sub-regions of distinct histologic profile. These models aim to further our understanding of the disease and assist in the diagnosis and grading of conditions in a consistent manner.

BraTS-Path Challenge: Assessing Heterogeneous Histopathologic Brain Tumor Sub-regions

TL;DR

The BraTS-Path paper addresses the need to automatically identify heterogeneous histologic sub-regions within glioblastoma on digitized tissue, moving beyond prior radiology-focused BraTS benchmarks. It presents a multicenter, expert-annotated whole-slide image dataset with nine distinct sub-regions and a patch-level multiclass classification task, along with a standardized evaluation pipeline using MedPerf and GaNDLF metrics. The work establishes a community benchmark, enabling rigorous comparison of AI approaches for histopathology, and highlights potential clinical impact in tumor diagnosis, grading, and treatment planning. By capturing GBM heterogeneity across sites and providing ground-truth annotations, BraTS-Path aims to accelerate translation of AI tools into pathology workflows and improve patient care through more objective, scalable histopathologic assessment.

Abstract

Glioblastoma is the most common primary adult brain tumor, with a grim prognosis - median survival of 12-18 months following treatment, and 4 months otherwise. Glioblastoma is widely infiltrative in the cerebral hemispheres and well-defined by heterogeneous molecular and micro-environmental histopathologic profiles, which pose a major obstacle in treatment. Correctly diagnosing these tumors and assessing their heterogeneity is crucial for choosing the precise treatment and potentially enhancing patient survival rates. In the gold-standard histopathology-based approach to tumor diagnosis, detecting various morpho-pathological features of distinct histology throughout digitized tissue sections is crucial. Such "features" include the presence of cellular tumor, geographic necrosis, pseudopalisading necrosis, areas abundant in microvascular proliferation, infiltration into the cortex, wide extension in subcortical white matter, leptomeningeal infiltration, regions dense with macrophages, and the presence of perivascular or scattered lymphocytes. With these features in mind and building upon the main aim of the BraTS Cluster of Challenges https://www.synapse.org/brats2024, the goal of the BraTS-Path challenge is to provide a systematically prepared comprehensive dataset and a benchmarking environment to develop and fairly compare deep-learning models capable of identifying tumor sub-regions of distinct histologic profile. These models aim to further our understanding of the disease and assist in the diagnosis and grading of conditions in a consistent manner.
Paper Structure (20 sections, 5 equations, 5 figures, 1 table)

This paper contains 20 sections, 5 equations, 5 figures, 1 table.

Table of Contents

  1. Introduction
  2. Materials & Methods
  3. Data Description
  4. Participating sites
  5. Annotation Protocol
  6. Data Processing
  7. Defining Distinct Sub-regions
  8. Cellular Tumor (CT):
  9. Pseudopalisading Necrosis (PN):
  10. Microvascular Proliferation (MP):
  11. Penetration into White Matter (WM):
  12. Leptomeningeal Infiltration (LI):
  13. Tumor Necrosis (TN):
  14. Regions of Dense Macrophages (DM):
  15. Presence of Lymphocytes (PL):
  16. ...and 5 more sections

Figures (5)

  • Figure 1: GBM Pathology sub-regions considered from one of the WSI in the BraTS-Path 2024 Challenge. The image considers various regions of GBM including Pseudopalasading Necrosis (Red), Microvascular Proliferation (Green), Necrosis (Blue), Infiltration into the cortex (Yellow), Cellular Tumor (Black), and Penetration into white matter (Sky blue). Other regions such as the Presence of Lymphocytes (PL), Regions Dense with Macrophages (DM), and Leptomeningeal Infiltration (LI) are not present in the given slide and are not annotated but are available through other slides.
  • Figure 2: Sample region of interest of selected tissue for Cellular Tumor
  • Figure 3: Sample regions of interest of selected tissues. (a)Microvascular Proliferation, (b) Pseudopalisading Necrosis, (c) White Matter, (d) Leptomeningeal Infiltration.
  • Figure 4: Sample regions of interest of selected tissues. (a) Tumor Necrosis, (b) Regions with Dense Macrophages, (c) Presence of Lymphocytes, (d) Infiltrating Cortex.
  • Figure 5: Pie chart showing the percentage distribution of values of the number of patches by region of interest included in the challenge.