Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/adalca/neurite
Neural networks toolbox focused on medical image analysis
https://github.com/adalca/neurite
brain convolutional-neural-networks keras medical-image-computing medical-imaging python tensorflow
Last synced: about 2 months ago
JSON representation
Neural networks toolbox focused on medical image analysis
- Host: GitHub
- URL: https://github.com/adalca/neurite
- Owner: adalca
- License: apache-2.0
- Created: 2017-03-10T01:25:06.000Z (over 7 years ago)
- Default Branch: dev
- Last Pushed: 2024-07-16T20:13:49.000Z (2 months ago)
- Last Synced: 2024-07-17T00:22:02.101Z (2 months ago)
- Topics: brain, convolutional-neural-networks, keras, medical-image-computing, medical-imaging, python, tensorflow
- Language: Python
- Homepage:
- Size: 800 KB
- Stars: 325
- Watchers: 22
- Forks: 66
- Open Issues: 20
-
Metadata Files:
- Readme: README.md
- License: LICENSE.md
- Citation: citations.bib
Awesome Lists containing this project
- awesome-biological-image-analysis - Neurite - Neural networks toolbox focused on medical image analysis. (Other)
README
# Neurite
A neural networks toolbox with a focus on medical image analysis in tensorflow/keras for now.
## Install
To use the Neurite library, either clone this repository and install the requirements listed in `setup.py` or install directly with pip.
```
pip install neurite
```
## Main tools
- [layers](neurite/tf/layers.py): various network layers, sparse operations (e.g. `SpatiallySparse_Dense`), and `LocallyConnected3D` currently not included in `keras`
- [utils](neurite/tf/utils/utils.py): various utilities, including `interpn`: N-D gridded interpolation, and several nonlinearities
- [model](neurite/tf/utils/model.py): `stack_models`: keras model stacking
- [vae](neurite/tf/utils/vae.py): tools for analyzing (V)AE style models
- [seg](neurite/tf/utils/seg.py): segmentation tools
- [models](neurite/tf/models.py): flexible models (many parameters to play with) particularly useful in medical image analysis, such as UNet/hourglass model, convolutional encoders and decoders
- [generators](neurite/tf/generators.py): generators for medical image volumes and various combinations of volumes, segmentation, categorical and other output
- [callbacks](neurite/tf/callbacks.py): a set of callbacks for `keras` training to help with understanding your fit, such as Dice measurements and volume-segmentation overlaps
- [dataproc](neurite/py/dataproc.py): a set of tools for processing medical imaging data for preparation for training/testing
- [metrics](neurite/tf/metrics.py): metrics (most of which can be used as loss functions), such as Dice or weighted categorical crossentropy
- [plot](neurite/py/plot.py): plotting tools, mostly for debugging models
## Papers:
If you use this code, please cite:
**Anatomical Priors in Convolutional Networks for Unsupervised Biomedical Segmentation**
[Adrian V. Dalca](http://adalca.mit.edu), [John Guttag](https://people.csail.mit.edu/guttag/), [Mert R. Sabuncu](http://sabuncu.engineering.cornell.edu/)
CVPR 2018.
[ [PDF](http://www.mit.edu/~adalca/files/papers/cvpr2018_priors.pdf) | [arxiv](http://arxiv.org/abs/1903.03148) | [bibtex](citations.bib) ]
If you are using any of the sparse/imputation functions, please cite:
**Unsupervised Data Imputation via Variational Inference of Deep Subspaces**
[Adrian V. Dalca](http://adalca.mit.edu), [John Guttag](https://people.csail.mit.edu/guttag/), [Mert R. Sabuncu](http://sabuncu.engineering.cornell.edu/)
Arxiv preprint 2019
[ [arxiv](https://arxiv.org/abs/1903.03503) | [bibtex](citations.bib) ]
## Development:
We welcome contributions; please make sure your code respects `pep8`, except for `E731,W291,W503,W504`, by running:
```pycodestyle --ignore E731,W291,W503,W504 --max-line-length 100 /path/to/neurite```
Please open an [issue](https://github.com/adalca/neurite/issues) [preferred] or contact Adrian Dalca at [email protected] for question related to `neurite`.
## Use/demos:
Parts of `neurite` were used in [VoxelMorph](http://voxelmorph.mit.edu) and [brainstorm](https://github.com/xamyzhao/brainstorm/), which we encourage you to check out!