Keywords: Tumors (Post-Treatment), MR-Guided Radiotherapy
Motivation: To improve RT CTV definition for GBM by leveraging multi-modality bio-clinical-imaging features to uncover hidden patterns of progression.
Goal(s): To develop a cross-modality deep learning model that enables precise, personalized RT planning that predicts and targets GBM tumor spread.
Approach: Using a cross-modality deep learning model, we combined bio-clinical features with a diffusion-weighted, white matter path-length map. This model modified a 3D-SwinUNETR network with cross-attention to bio-clinical data and image features.
Results: The cross-modality model outperformed imaging-only models, achieving higher Dice and Tversky scores, improved coverage of tumor spread, with minimal exposure to normal brain.
Impact: The cross-modality approach uniquely combines bio-clinical and imaging data, capturing complex paths of tumor spread for tailored RT planning. This fusion improves prediction accuracy, minimizes healthy tissue exposure, and enhances personalized care, promising improved outcomes and extended patient survival.
How to access this content:
For one year after publication, abstracts and videos are only open to registrants of this annual meeting. Registrants should use their existing login information. Non-registrant access can be purchased via the ISMRM E-Library.
After one year, current ISMRM & ISMRT members get free access to both the abstracts and videos. Non-members and non-registrants must purchase access via the ISMRM E-Library.
After two years, the meeting proceedings (abstracts) are opened to the public and require no login information. Videos remain behind password for access by members, registrants and E-Library customers.
Keywords