Abstract: Magnetic resonance imaging (MRI) scans play a vital role in characterizing anatomical properties and cellular density in brain tumor. In this study, we develop MRI-based radiomics signature for predicting four molecular subgroups of medulloblastoma (MB). Experimental results demonstrated that MRI-based radiomics features were predictive in differentiating molecular subtypes in medulloblastoma.

Learning Objective 1: Develop a MRI-based method for noninavsiely predicting four molecular subgroups of medulloblastoma and to investigate their association with computational MR imaging features.


Zichen Wang, Stanford University
Mu Zhou (Presenter)
Stanford University

Eric Tranvinh, Stanford University
Bryan Lanzman, Stanford University
Katie Shpanskaya, Stanford University
Tina Poussaint, Harvard University
Michael Iv, Stanford University
Kristen Yeom, Stanford University
Olivier Gevaert, Stanford University

Presentation Materials: