Advancing precision risk stratification in adult diffuse gliomas through DSC-MRI-based habitat analysis

Area of Science:

• Neuroimaging
• Oncology
• Radiomics

Background:

• Diffuse gliomas present significant molecular and spatial heterogeneity.
• Accurate identification of aggressive phenotypes is crucial for risk stratification and treatment planning.

Purpose of the Study:

• To evaluate habitat-based radiomics models using dynamic susceptibility contrast MRI (DSC-MRI) and conventional MRI.
• To identify aggressive diffuse glioma phenotypes linked to integrated histologic-molecular risk.

Main Methods:

• Retrospective analysis of 197 adult diffuse glioma patients.
• Habitat identification using K-means clustering on multiparametric MRI data.
• Radiomic feature extraction and selection, followed by random forest classification for risk prediction.

Main Results:

• Habitat-based models significantly outperformed whole-region analyses (AUC 0.949 vs. 0.931).
• Hemodynamic features (CBF + MTT) from DSC-MRI showed superior predictive accuracy over conventional MRI sequences (T1C+T2FLAIR).
• The optimal model utilized only hemodynamic features from combined habitats, indicating their discriminative power.

Conclusions:

• DSC-MRI-based habitat analysis effectively resolves perfusion heterogeneity, outperforming conventional imaging.
• Hemodynamic features are promising for predicting the integrated histologic-molecular risk status of adult diffuse gliomas.
• This approach may serve as a valuable preoperative imaging biomarker.