Skip to content

Latest commit

 

History

History
36 lines (24 loc) · 2.26 KB

README.md

File metadata and controls

36 lines (24 loc) · 2.26 KB
Raw

nobrainer Python API Guides

This set of guides demonstrates basic usage of the nobrainer python API to train and use neural network models for neuroimaging. The nobrainer project was developed under the support of NIH RF1MH121885 and R01EB020470. It is distributed under the Apache 2.0 license.

Accessing the guides

These guides are available in both jupyter notebook versions and as standalone python scripts.

Jupyter notebooks Open In Colab

See the notebooks directory for the guides in jupyter notebook format. These notebooks can be run for free on Google Colab, accessed here.

Standalone python scripts

See the scripts directory for the guides in python format. These notebooks can be run anywhere with a an installation of python >= 3.8.

List of guides

  1. Getting started (notebook, python) Basic nobrainer info and installation
  2. Preparing training data (notebook, python) How to obtain some example neuroimaging data and prepare it for neural network training.
  3. Train brain extraction (notebook, python) Train a standard brain extraction model on the example dataset.
  4. Train brain volume generation (notebook, python) Train a model to generate synthetic brain volumes on the example dataset.
  5. Augmentation of training data (notebook, python) Use augmented training data to train a standard brain extraction model on the example dataset.

Adding to the guide

These guides are maintained in parallel using jupytext. Edits should be made to the python files in the scripts directory, then notebooks can be generated via

jupytext --sync scripts/<python-file>