Multilabel (BRATS-like) output for single-time point post-operative glioma segmentation model

[gusSCIMOV]

Hi Raidionics Team
I've been working a time ago with many of the Colab notebooks added in "processing backend for performing segmentation and computation of standardized report (RADS)" and it has been quite useful

I just wondering if I might get a BRATs-like multilabel output (Edema, enhancing, necrosis) using just the post-operative scans using the MRI_GBM_Postop_FV_1p or FV_3p, instead of the whole tumor mask since that is an important part of therapy response assessment

Thanks, I really appreciate your help

[andreped]

@dbouget Just FYI, this is related to this Issue thread: #67 (comment)

[dbouget]

Hi @gusSCIMOV,

At the moment, we do not have such annotations at our disposal to train in-house models.
The next iteration of the BraTS challenge will include early post-operative segmentation tasks, and will provide the desired class-annotations (i.e., necrosis, enhancing, and edema).

We'll start including such models in Raidionics after the MICCAI conference, scheduled early October 2024, so stay tuned!