Related Experiment Video
Updated: Jun 23, 2025
07:53
Author Spotlight: Using the Chick Embryo Brain as a Model for In Vivo and Ex Vivo Analyses of Human Glioblastoma Cell Behavior
Published on: May 26, 2023
3.6K
Histological Hyperspectral Glioblastoma Dataset (HistologyHSI-GB)
Samuel Ortega1,2,3, Laura Quintana-Quintana4, Raquel Leon4
1Seafood Industry Department, Norwegian Institute of Food, Fisheries and Aquaculture Research (Nofima), Tromsø, Norway. sortega@iuma.ulpgc.es.
Scientific Data
|June 24, 2024
Summary
This study introduces the HistologyHSI-GB dataset, featuring hyperspectral images of glioblastoma brain tumors. This resource aids in analyzing tissue characteristics for improved disease detection and research.
Area of Science:
- Medical imaging
- Spectroscopy
- Computational pathology
Background:
- Hyperspectral (HS) imaging (HSI) integrates imaging and spectroscopy to analyze morphological and chemical properties.
- HSI offers insights into light-tissue interactions, enabling detection of disease-specific patterns, cells, or biomarkers.
- Accurate disease identification is crucial in fields like neuropathology.
Purpose of the Study:
- To introduce and describe the HistologyHSI-GB dataset for glioblastoma research.
- To provide a valuable resource for developing and validating HSI-based diagnostic tools for brain tumors.
- To facilitate advancements in computational pathology and medical diagnostics.
Main Methods:
- Acquisition of 469 HS images from 13 glioblastoma patients using custom HSI instrumentation.
- Microscopic capture at 20× magnification, covering a spectral range of 400–1000 nm.
- Standard H&E staining and expert histopathological diagnosis with image-level annotations.
Main Results:
- The HistologyHSI-GB dataset comprises 469 annotated HS images of glioblastoma tissue.
- The dataset captures detailed spectral and spatial information crucial for pathological analysis.
- Expert annotations provide ground truth for developing machine learning models.
Conclusions:
- The HistologyHSI-GB dataset is a significant contribution to hyperspectral imaging in neuropathology.
- This dataset will accelerate research in automated glioblastoma detection and characterization.
- HSI technology holds promise for enhancing diagnostic accuracy in cancer research.
