Related Experiment Video
Updated: Aug 31, 2025
06:32
Evaluation of Biomarkers in Glioma by Immunohistochemistry on Paraffin-Embedded 3D Glioma Neurosphere Cultures
Published on: January 9, 2019
8.0K
Image-based deep learning identifies glioblastoma risk groups with genomic and transcriptomic heterogeneity: a
Jing Yan1, Qiuchang Sun2,3, Xiangliang Tan4
1Department of MRI, The First Affiliated Hospital of Zhengzhou University, Jian she Dong Road 1, Zhengzhou, 450052, Henan province, China.
European Radiology
|August 24, 2022
Summary
A novel deep learning imaging signature (DLIS) from MRI accurately predicts survival in glioblastoma (GBM) patients. This DLIS correlates with key molecular pathways and genetic alterations, aiding personalized treatment strategies.
Area of Science:
- Neuro-oncology
- Radiomics
- Machine Learning
Background:
- Glioblastoma (GBM) is an aggressive brain tumor with poor prognosis.
- Accurate risk stratification and understanding of underlying biology are crucial for effective treatment.
Purpose of the Study:
- To develop and validate a deep learning imaging signature (DLIS) for risk stratification in GBM patients.
- To investigate the biological pathways and genetic alterations associated with the DLIS.
Main Methods:
- Developed a DLIS using multi-parametric MRI on a training set (n=600) and validated across multiple independent datasets.
- Employed a radiogenomics framework integrating imaging, transcriptome, and genome data from 127 patients.
Main Results:
- The DLIS demonstrated significant association with survival (p < 0.001) and served as an independent predictor.
- DLIS correlated with GBM core pathways (P53, RB, RTK) and genetic alterations (del_CDNK2A).
- An integrated nomogram incorporating DLIS showed improved prognostic accuracy.
Conclusions:
- The developed DLIS provides a biologically interpretable tool for predicting GBM patient survival.
- This signature aids in understanding GBM prognosis and guiding individualized treatment approaches.
- DLIS effectively stratifies GBM patients into distinct molecular and risk groups.
