U-Net is a Convolutional Neural Network (CNN) that was designed for biological image segmentation. In order to preserve finer feature maps, skipping connections are used that complement the data in deeper layers. In this work, the same architecture was adapted for MRI brain scans to predict one modality given another. This was done by sclicing raw volumetric MRI data scanned in two different modalities into 2D images that the network could be trained on. The network was implemented using matConvNet, a MATLAB toolbox for CNNs.
-
Notifications
You must be signed in to change notification settings - Fork 10
TUM_MLMI_SS16: Convolutional Neural Network using U-Net architecture to predict one modality of a brain MRI scan from another modality.
License
clamesc/Machine-Learning-in-Medical-Imaging--U-Net
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
About
TUM_MLMI_SS16: Convolutional Neural Network using U-Net architecture to predict one modality of a brain MRI scan from another modality.
Topics
Resources
License
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published