Feature Description
We would like to train and create a new sub-module to obtain from a T1 weighted image all 4 substructures of the Brain Stem (Pons, MidBrain, Medule, SCP)
...
Motivation
The volume of the Pons can be very important to understand the state of certain degenerative diseases.
We are using this module https://surfer.nmr.mgh.harvard.edu/fswiki/BrainstemSubstructures from FreeSurfer to calculate the volumes of each region. We believe that FastSurfer could include a module with this substructures and deliver results Faster.
We propose to generate a Dataset including this regions using the same Open Source Datasets that where used to train FastSurfer, and generate a dataset using the FreeSurfer Module.
The next step would be to create a new Deep Learning model to detect all these 4 different substructures using FastSurferCNN and deliver the module to the community.
We're open to discusion and to new ideas.
Feature Description
We would like to train and create a new sub-module to obtain from a T1 weighted image all 4 substructures of the Brain Stem (Pons, MidBrain, Medule, SCP)
...
Motivation
The volume of the Pons can be very important to understand the state of certain degenerative diseases.
We are using this module https://surfer.nmr.mgh.harvard.edu/fswiki/BrainstemSubstructures from FreeSurfer to calculate the volumes of each region. We believe that FastSurfer could include a module with this substructures and deliver results Faster.
We propose to generate a Dataset including this regions using the same Open Source Datasets that where used to train FastSurfer, and generate a dataset using the FreeSurfer Module.
The next step would be to create a new Deep Learning model to detect all these 4 different substructures using FastSurferCNN and deliver the module to the community.
We're open to discusion and to new ideas.