Skip to content
Neural networks toolbox focused on medical image analysis
Branch: master
Clone or download
Type Name Latest commit message Commit time
Failed to load latest commit information.
neuron adding compose; cleanup of keras compatibility Aug 15, 2019
.gitattributes 🎉 Added .gitattributes & .gitignore files Mar 10, 2017
.gitignore adding .nfs to gitignore Jun 14, 2017 Create Sep 23, 2018 adding imputation Mar 11, 2019
bibtex.txt adding imputation Mar 11, 2019


A Neural networks toolbox with a focus on medical image analysis in tensorflow/keras for now.
Note: this is under ongoing development

Main tools

  • layers: various network layers, including a rich SpatialTransformer layer for N-D (dense and affine) spatial transforms, a vector integration layer VecInt, sparse operations (e.g. SpatiallySparse_Dense), and LocallyConnected3D currently not included in keras
  • utils: various utilities, including interpn: N-D gridded interpolation, transform: warp images, integrate_vec: vector field integration, stack_models: keras model stacking
  • models: flexible models (many parameters to play with) particularly useful in medical image analysis, such as UNet/hourglass model, convolutional encoders and decoders
  • generators: generators for medical image volumes and various combinations of volumes, segmentation, categorical and other output
  • callbacks: a set of callbacks for keras training to help with understanding your fit, such as Dice measurements and volume-segmentation overlaps
  • dataproc: a set of tools for processing medical imaging data for preparation for training/testing
  • metrics: metrics (most of which can be used as loss functions), such as Dice or weighted categorical crossentropy
  • vae_tools: tools for analyzing (V)AE style models
  • plot: plotting tools, mostly for debugging models


  • tensorflow, keras and all of their requirements (e.g. hyp5)
  • numpy, scipy, tqdm
  • pytools lib


If you use this code, please cite:

Anatomical Priors in Convolutional Networks for Unsupervised Biomedical Segmentation
Adrian V. Dalca, John Guttag, Mert R. Sabuncu
CVPR 2018.
[ PDF | arxiv | bibtex ]

If you are using any of the sparse/imputation functions, please cite:

Unsupervised Data Imputation via Variational Inference of Deep Subspaces
Adrian V. Dalca, John Guttag, Mert R. Sabuncu
Arxiv preprint 2019
[ arxiv | bibtex ]


Please open an issue [preferred] or contact Adrian Dalca at for question related to neuron.


Parts of neuron were used in VoxelMorph and brainstorm, which we encourage you to check out!

You can’t perform that action at this time.