Filter Atlas by ROI Attributes

Subsets an atlas object to include only ROIs matching specified criteria. Uses non-standard evaluation (NSE) similar to dplyr::filter().

Usage

filter_atlas(x, ..., .dots = NULL)

# S3 method for class 'atlas'
filter_atlas(x, ..., .dots = NULL)

Arguments

x

An atlas object

...

Filter expressions using column names from roi_metadata(x). Multiple conditions are combined with AND.

.dots

A list of quosures for standard evaluation (advanced use)

Value

A new volume atlas object of the same class containing only the matching ROIs. The returned atlas has updated ids, labels, hemi, orig_labels, cmap, network (if present), and roi_metadata fields. The underlying volume data is also subset.

Details

The filter operation creates a new atlas containing only the specified ROIs. For volume atlases, voxels belonging to excluded ROIs are set to zero. ROI IDs are preserved (not renumbered) to maintain consistency with the original atlas labeling. Surface-atlas subsetting is not implemented; use get_roi() to extract labelled surface regions.

See also

roi_metadata for viewing available filter columns, roi_attributes for listing available attributes, get_roi for extracting ROI data

Examples

if (FALSE) { # \dontrun{
# Filter Schaefer atlas to left hemisphere visual network
atlas <- get_schaefer_atlas(parcels = "200", networks = "7")
left_vis <- filter_atlas(atlas, hemi == "left", network == "Vis")

# Filter ASEG to left hemisphere structures
aseg <- get_aseg_atlas()
left_aseg <- filter_atlas(aseg, hemi == "left")

# Filter by label pattern
hippo <- filter_atlas(aseg, grepl("Hippocampus", label))
} # }