Related Experiment Video
Updated: Feb 27, 2026
05:45
Quantitative Immunohistochemistry of the Cellular Microenvironment in Patient Glioblastoma Resections
Published on: July 31, 2017
10.2K
Relationship between Glioblastoma Heterogeneity and Survival Time: An MR Imaging Texture Analysis
1From the School of Biomedical Engineering (Y.L., X.P.X., L.L.Y., X.Z., H.B.L.), Fourth Military Medical University, Xi'an, Shaanxi, China.
AJNR. American Journal of Neuroradiology
|July 1, 2017
Summary
Glioblastoma heterogeneity assessed via MR imaging textures can predict patient survival. Texture analysis, particularly regional variations, shows promise in stratifying glioblastoma patients for improved outcomes.
Area of Science:
- Radiology
- Oncology
- Medical Imaging
Background:
- Glioblastoma exhibits significant heterogeneity, leading to variable patient prognoses.
- Accurate prognostic markers are crucial for tailoring glioblastoma treatment strategies.
Purpose of the Study:
- To investigate the utility of MR imaging texture analysis in quantifying glioblastoma heterogeneity.
- To evaluate the impact of identified textural features on patient survival outcomes.
Main Methods:
- Retrospective analysis of 133 primary glioblastoma patients from The Cancer Genome Atlas.
- Extraction of texture features (co-occurrence matrix, run-length matrix, histogram) from postcontrast T1-weighted MRI.
- Support vector machine classification and recursive feature elimination for feature selection and survival stratification.
Main Results:
- Combined texture features demonstrated superior performance in differentiating survival groups.
- An optimal subset of 43 features was identified, with run-length matrix features showing high importance.
- Regional texture features, particularly those emphasizing high gray-levels, were highly ranked.
Conclusions:
- MR imaging texture analysis effectively captures glioblastoma heterogeneity.
- Local and regional textural features are significant predictors of survival in glioblastoma patients.
