UNet3D TensorFlow checkpoint trained with AMP
Model Overview
A convolutional neural network for 3D image segmentation.
Model Architecture
3D-UNet was first introduced by Olaf Ronneberger, Philip Fischer, and Thomas Brox in the paper: 3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation. In this repository we host a 3D-UNet version adapted by Fabian Isensee et al. to brain tumor segmentation. 3D-UNet allows for seamless segmentation of 3D volumes, with high accuracy and performance, and can be adapted to solve many different segmentation problems.
The following figure shows the construction of the 3D-UNet model and its different components. 3D-UNet is composed of a contractive and an expanding path, that aims at building a bottleneck in its centermost part through a combination of convolution and pooling operations. After this bottleneck, the image is reconstructed through a combination of convolutions and upsampling. Skip connections are added with the goal of helping the backward flow of gradients in order to improve the training.
Training
This model was trained using script available on NGC and in GitHub repo.
Dataset
The following datasets were used to train this model:
- BraTS19 - Dataset of clinically-acquired pre-operative multimodal MRI scans of glioblastoma (GBM/HGG) and lower grade glioma (LGG), with pathologically confirmed diagnosis and available OS.
Performance
Performance numbers for this model are available in NGC.
References
License
This model was trained using open-source software available in Deep Learning Examples repository. For terms of use, please refer to the license of the script and the datasets the model was derived from.