TCGA-Brain glioma

Citation Author(s):
Submitted by:
Ju Liu
Last updated:
Tue, 12/13/2022 - 05:33
0 ratings - Please login to submit your rating.


This is just a preliminary collation of the relevant TCGA datasets collated and used in our methodology. We will continue to upload the full dataset later for your reference and use. We hope to make a small contribution to the study of automatic 3D MRI classification of gliomas and the problem of domain adaptation on medical images.


We select the source and target domain datasets for the Glioma Grading experiment from the BraTS18 dataset and the TCGA-Brain dataset, respectively, with the number of specific categories shown in Table I. Since WHO further classifies glioma into different grades based on their characteristics: grade II, grade III and grade IV. Gliomas are treated by doctors with surgery, radiotherapy or chemotherapy alone or in combination, depending on the grade diagnosed. Due to the scarcity of data, the complex structure of medical images and the lack of medical expertise, mining the visual structure for the grading features of gliomas is a enormous challenge. Therefore, we attempt to complete this fine-grained grading experiment using the existing UDA method and the proposed STDA method. The implementation detail is identical to that of the Glioma Binary Classification experiment.


My focus is on applying the meta-learning technique to the challenge of diagnosing rare diseases. To advance my research, I need access to this data set.
The sole reason I need this data collection is for research.

Submitted by Kuljeet Singh on Sat, 05/13/2023 - 00:13