Skip to contents

Locate vertices belonging to specific cortical ROIs

Source: R/utils.R
locate_roi.Rd

This function identifies which vertices in a cortical surface mesh correspond to one or more specified regions of interest (ROIs), based on the 36 regions of the Desikan-Killiany atlas distributed with FreeSurfer (aparc.annot). If no ROIs are specified, it returns a full lookup table of all ROIs with vertex counts, proportions, and lobe assignments.

Usage

locate_roi(rois = NULL, n_verts = 163842, hemi = c("lh", "rh"), verbose = TRUE)

Arguments

rois

Optional character vector of ROI names to locate (e.g., c("insula", "precentral")). If NULL (default), the function returns a full ROI lookup table.

n_verts

Integer. Total number of surface vertices to consider (default: 163842 "fsaverage").

hemi

String: "lh" or "rh". Used to point to the hemisphere-specific aparc annotation file.

verbose

Logical (default: TRUE).

Value

  • If rois is NULL, returns a data.frame with:

    • roi_id: ROI numeric ID

    • roi_label: ROI name

    • vw_count: Number of vertices in the ROI

    • vw_prop: Proportion of total vertices in the ROI

    • roi_lobe: Anatomical lobe classification

    • lobe_count: Total number of vertices in the lobe

    • lobe_prop: Proportion of vertices in the lobe

  • If rois is provided, returns a logical vector of length n_verts, where TRUE indicates the vertex belongs to one of the selected ROIs.

Examples

# Get the full ROI lookup table
roi_table <- locate_roi()

# Get a logical vector of vertices in the insula
insula_mask <- locate_roi(rois = "insula")
#>  * selected 5229 vertices (3.2%)