Note

This page is a reference documentation. It only explains the function signature, and not how to use it. Please refer to the user guide for the big picture.

nilearn.glm.cluster_level_inference

nilearn.glm.cluster_level_inference(stat_img, mask_img=None, threshold=3.0, alpha=0.05, verbose=0)

Report the proportion of active voxels for all clusters defined by the input threshold.

This implements the method described in Rosenblatt et al.[1].

Parameters:

stat_img3D Niimg-like object or SurfaceImage with a single sample.

statistical image (presumably in z scale)

mask_imgNiimg-like object, or SurfaceImage or None, default=None

mask image

thresholdNon-negative float, int, or list of non-negative float or int, default=3.0

Cluster-forming threshold in z-scale.

alphafloat or list, default=0.05

Level of control on the true positive rate, aka true discovery proportion.

verbosebool or int, default=0

Verbosity level (0 or False means no message).

Returns:

proportion_true_discoveries_imgNifti1Image or SurfaceImage

The statistical map that gives the true positive.

References

[1]

Jonathan D. Rosenblatt, Livio Finos, Wouter D. Weeda, Aldo Solari, and Jelle J. Goeman. All-resolutions inference for brain imaging. NeuroImage, 181:786–796, November 2018. doi:10.1016/j.neuroimage.2018.07.060.

Examples using nilearn.glm.cluster_level_inference

Example of surface-based first-level analysis

Example of surface-based first-level analysis

Second-level fMRI model: true positive proportion in clusters

Second-level fMRI model: true positive proportion in clusters