Keywords: Epilepsy, Radiomics, multi-task learning
Motivation: Prediction of IDH mutation status and epilepsy occurrence are important to glioma patients, but the correlation between these two issues has not been fully explored.
Goal(s): To develop collaborative radiomics models to identify IDH mutation status and epilepsy in patients with glioma II-IV, using both shared and task-specific features.
Approach: Radiomic features were extracted from multiparametric MRI and ROIs were segmented using a pre-trained deep learning model. We combined classic LASSO with multi-task LASSO to select task-specific and shared features respectively.
Results: The proposed models achieved better performance with fewer features than classic LASSO models on IDH status and epilepsy identification.
Impact: The collaborative radiomics models exploited the underlying connections between IDH status and epilepsy identification tasks with fewer features, improved the model performance, and can potentially been used to find valuable generalized biomarkers for multiple clinical problems.
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