search

nipreps / mriqc

Automated Quality Control and visual reports for Quality Assessment of structural (T1w, T2w) and functional MRI of the brain
http://mriqc.readthedocs.io
Apache License 2.0
300 stars 132 forks source link

ENH: Drop FSL BET to estimate the "outskin" (head) mask #1105

Closed oesteban closed 1 year ago

oesteban commented 1 year ago

This PR contributes toward #1032 by eliminating the need for FSL BET. This implementation requires Dipy for the nonlinear means denoising.

BTW - the nonlinear means denoising could be leveraged to calculate the sigma of the snr calculations more precisely, obtaining much better estimates.

By using an implementation closer to the gradient-based of (Mortamet, 2009), there is no more need for BET.