Exploring Adult Glioma through Magnetic Resonance Imaging: A Global Review of Publicly Available Datasets#

Authors#

Meryem Abbad Andaloussi, Raphael Maser, Frank Hertel, François Lamoline, Andreas Husch

Abstract#

Publicly available health data are essential for digital health research, from machine learning models to data-driven mathematical models. Glioma is the most common malignant intracranial tumor, and magnetic resonance imaging (MRI) is a widely used modality for its diagnosis and treatment. However, the availability and quality of public datasets for Glioma MRI are not well known. In this review, we searched for public datasets for Glioma MRI using Google Dataset Search, The Cancer Imaging Archive (TCIA) and Synapse. We found 26 datasets published between 2005 and 2023, containing 52542 images from 5333 patients. We analyzed the characteristics of these datasets, such as the origin, size, format, annotation, and accessibility. We also examined the distribution of tumor types, grades, and stages among the datasets. We discuss how the WHO classification evolution affects research and highlights research questions that can be addressed using these datasets. Finally, we discussed research questions that some datasets may be capable of addressing, such us tumor evolution through malignant transformation, MRI normalisation, and tumor segmentation. Within the analysed datasets, only one follows the latest WHO classification with histopathological and molecular proof. This review aims to offer a comprehensive overview of the currently available public datasets for Glioma MRI, underscoring the challenges and opportunities for future research endeavors.