MRIQC - New release 0.9.7!

We are very happy to announce MRIQC version 0.9.7, just a few steps away from 1.0 and becoming a beta software.

This release accompanies the pre-print We would love your feedback.

We have also released OpenNeuro (, where you’ll find MRIQC installed and ready for use online!

The CHANGES of MRIQC from 0.9.6 are mostly adaptations and bugfixes required for the revision of the pre-print. So we have worked fundamentally on the classifier:

  • [ENH] Clip outliers in FD and SPIKES group plots (#593)
  • [ENH] Second revision of the classifier (#555):
    • Set matplotlib plugin to agg in docker image
    • Migrate scalings to sklearn pipelining system
    • Add Satra’s feature selection for RFC (with thanks to S. Ghosh for his suggestion)
    • Make model selection compatible with sklearn Pipeline
    • Multiclass classification
    • Add feature selection filter based on Sites prediction (requires pinning to development sklearn-0.19)
    • Add RobustLeavePGroupsOut, replace RobustGridSearchCV with the standard GridSearchCV of sklearn.
    • Choice between RepeatedStratifiedKFold and RobustLeavePGroupsOut in mriqc_clf
    • Write cross-validation results to an .npz file.
  • [ENH] First revision of the classifier (#553):
    • Add the possibility of changing the scorer function.
    • Unifize labels for raters in data tables (to rater_1)
    • Add the possibility of setting a custom decision threshold
    • Write the probabilities in the prediction file
    • Revised mriqc_clf processing flow
    • Revised labels file for ds030.
    • Add IQMs for ABIDE and DS030 calculated with MRIQC 0.9.6.
  • ANNOUNCEMENT: Dropped support for Python<=3.4

Version 0.9.6 supposed a great change for MRIQC, and you should expect the features to change a bit. We have implemented a mechanism to keep track of these changes, so from now on we will indicate in the changelog when some metrics might have been modified. For a detailed list of the changes of that version, please visit

Finally, I would like to comment on the latest feature in MRIQC: the Web API. Unless you opt-out (which is really easy using the argument --no-sub when running the participant analysis level), the quality metrics extracted from your datasets will be annonimized (if there is some remaining identifiers in the subject name or the session name) and uploaded to a public repository hosted by the NIMH -

We will use all those crowdsourced measures to improve MRIQC, so please contribute!.

Thanks a lot. Happy QC’ing