Dženan Zukić

Use MRIQC or MRQy to establish classical image quality metrics

2

Classical image quality metrics can be additional information, in addition to outputs of our trained neural network models. MRIQC: https://github.com/poldracklab/mriqc MRQy: https://github.com/ccipd/MRQy

Supersedes and closes #52.

A search for `image->Allocate();` has 149 hits while `image->FillBuffer(` has 64 hits. The instances of `Allocate` not followed by a `FillBuffer` should be examined, and probably turned into [image->Allocate(true);](https://github.com/InsightSoftwareConsortium/ITKSphinxExamples/commit/fc491a42151e2dd27f3e98b821d0c2fd35b55b2b#diff-a388a43bc1ceb769184a4daea4a146c73cbe1c5aeaf7003d1c698f51f95f0ad6R113) calls.

Outputting pixel values it tests should probably not be done for images bigger than maybe 100 pixels, as it swamps the log file making failures hard to find.

Requirement introduced in #109.

It would be good if this supported 2D color images, for those who prefer Slicer over FAIR's online demo.

This piece of code: https://github.com/Project-MONAI/SlicerMONAIViz/blob/27c6fa84d373b9be30b01251c59ea28cfc1bb370/MONAIViz/MONAIViz.py#L440-L445 could be refactored to use this: https://github.com/Project-MONAI/SlicerMONAIViz/blob/27c6fa84d373b9be30b01251c59ea28cfc1bb370/MONAIViz/MONAIVizLib/utils.py#L46 That should make it more reliable and future proof.

See https://discourse.itk.org/t/how-to-use-itkmontage-to-stitch-segmentation-label-images/6611.

Expose test's #if as command line parameter

2

If the else part works, this [#if](https://github.com/InsightSoftwareConsortium/ITKColorNormalization/blob/805e28e2bdfdfbf0bf562e10ca2596243f056782/test/itkStructurePreservingColorNormalizationFilterTest.cxx#L117) should be exposed as a command line parameter and invoked as a new [test case](https://github.com/InsightSoftwareConsortium/ITKColorNormalization/blob/805e28e2bdfdfbf0bf562e10ca2596243f056782/test/CMakeLists.txt#L9).