[ENH] Implement pure-Python ICA with mixture modeling

Brainhack-Donostia / ica-aroma-org

Repository to prepare the ICA-AROMA tutorial for Brainhack Donostia.

Apache License 2.0

1 stars 3 forks source link

[ENH] Implement pure-Python ICA with mixture modeling #51

Closed tsalo closed 4 years ago

tsalo commented 4 years ago

Closes #2.

TODO:

[x] Incorporate license and attribution info for nipy's GGM.
[x] Convert pseudocode to real code (this mostly means supporting images in addition to arrays and getting computefeats2 working).
[ ] Compare mixture modeling results to MELODIC.
[ ] Add tests.
[x] Integrate changes into workflow.

Changes proposed in this pull request:

Add GGM class from nipy for mixture modeling.
Use tedana for PCA + ICA.

eurunuela commented 4 years ago

As discussed with @CesarCaballeroGaudes yesterday, the mixture modeling is not necessary but it helps with the accuracy.

The ICA-AROMA paper states:

To increase accuracy of the spatial features, IC spatial maps are first thresholded using a Gamma/Gaussian alternative testing approach (p N 0.5). This approach is automated in FSL's MELODIC.

So, we could try running the workflow without the mixture modeling first and see if the results look good. If we see the mixture modeling is critical for accuracy, we can open a new PR with it.

tsalo commented 4 years ago

To be honest, I'm not confident that either the features or the classifier in AROMA are robust enough to handle inputs that don't exactly match the original method. From a conceptual standpoint, unthresholded maps may be fine, but from a practical one, I'm not so sure.

tsalo commented 4 years ago

I have https://github.com/Brainhack-Donostia/aroma/pull/22 open now, so I'm going to close this.