/brAInzViewer

.nii NifTI image explorer and automated MRI motion-corruption detection.

Primary LanguageJupyter Notebook

BrainScanMotionDetection

An easy way to explore .nii volumes and slices.

  • Use the file selector to quickily switch between .nii files:

  • Use the volume and slice sliders to explore the volumes and slices:

  • Automated analysis of volumes for motion-induced damage in images:

Our built-in analyzer uses advanced A.I. algorithm to estimate the amount of damage induced by moving subjects during the scan.

Model performance (Jan 1, 2020)

Compiled installers

Check our release page https://github.com/airoscar/brAInzViewer/releases

To run/build from source

  • Python 3.6 is recommended to run or build from source
  • Install all required dependencies from the requirements.txt:

On Mac:

`pip install -r Viewer/requirements/base.txt`

On Windows:

`pip install -r Viewer/requirements/windows.txt`
  • Navigate to the Viewer folder:

    cd Viewer

  • Run from source:

    fbs run

  • Freeze code:

    fbs freeze

Sometimes fbs fails to copy some dependencies, you may need to manually copy the missing packages from your environment (ie: environment/lib/python3.6/site-packagers/) to the fbs build folder. The packages may included but not limited to the following:

  • tensorflow
  • tensorflow_core
  • absl
  • astor
  • google

Copy the above packages to Viewer/target/brAInzViewer/ On Mac, you should also copy the above packages to Viewer/target/brAInzViewer.app/Contents/MacOS/

  • Compile installer:

    fbs installer