Launch Cluster Explorer Shiny App — cluster

Interactive explorer linking a cluster summary table, parcel brain map, and design-aware signal plots. Clusters are formed volumetrically from stat_map; visualization is parcel-level on surfatlas.

Usage

cluster_explorer(
  data_source = NULL,
  atlas = NULL,
  stat_map = NULL,
  surfatlas = NULL,
  sample_table = NULL,
  design = NULL,
  threshold = 3,
  min_cluster_size = 20,
  connectivity = c("26-connect", "18-connect", "6-connect"),
  tail = c("two_sided", "positive", "negative"),
  series_fun = NULL,
  overlay_space = NULL,
  overlay_density = NULL,
  overlay_resolution = NULL,
  overlay_fun = c("avg", "nn", "mode"),
  overlay_sampling = c("midpoint", "normal_line", "thickness"),
  prefetch = TRUE,
  prefetch_max_clusters = 200,
  prefetch_max_voxels = 1e+05,
  palette = "vik",
  selection_engine = c("cluster", "parcel", "sphere", "custom"),
  parcel_ids = NULL,
  sphere_centers = NULL,
  sphere_radius = 6,
  sphere_units = c("mm", "voxels"),
  sphere_combine = c("separate", "union"),
  selection_provider = NULL,
  analysis_plugins = NULL,
  default_analysis_plugin = "none"
)

Arguments

data_source: A sample-wise data source supporting neuroim2::series(data_source, i) where i is voxel coordinates (n x 3 matrix) or indices. Rows of the returned matrix correspond to samples/design rows.
atlas: A volumetric atlas object used for parcel annotation. When atlas and stat_map dimensions differ, the atlas is automatically resampled to stat_map space (nearest-neighbor labels) before cluster annotation.
stat_map: A NeuroVol statistic image used for thresholding and clustering.
surfatlas: A surface atlas object used by plot_brain(). If NULL, the function attempts to infer a surface atlas from a compatible `atlas` input (for example Schaefer or Glasser). If inference is not possible, and non-demo inputs are otherwise present, input validation fails.
sample_table: Optional data frame with one row per sample. If NULL, a default table with .sample_index is created.
design: Optional design table (one row per sample). When provided, it is column-bound to sample_table.
threshold: Numeric threshold used for cluster formation.
min_cluster_size: Minimum number of voxels required to keep a cluster.
connectivity: Connectivity passed to neuroim2::conn_comp().
tail: Clustering mode: "positive", "negative", or "two_sided".
series_fun: Optional function override for extracting voxel-wise sample series. Must accept (data_source, i) and return a matrix-like object with one row per sample.
overlay_space: Optional surface space override used to fetch white/pial meshes for volumetric cluster projection. Defaults to surfatlas$surface_space.
overlay_density: Optional TemplateFlow density override for overlay surface loading.
overlay_resolution: Optional TemplateFlow resolution override for overlay surface loading.
overlay_fun: Reduction used by neurosurf::vol_to_surf().
overlay_sampling: Sampling strategy for neurosurf::vol_to_surf().
prefetch: Logical default for eager cluster signal prefetch.
prefetch_max_clusters: Default max cluster count allowed for prefetch.
prefetch_max_voxels: Default max total cluster voxels allowed for prefetch.
palette: Continuous palette passed to plot_brain().
selection_engine: Selection backend. "cluster" uses the built-in connected-component workflow. "custom" delegates data assembly to selection_provider.
parcel_ids: Optional parcel ids used when selection_engine = "parcel". Default NULL uses all parcels.
sphere_centers: Optional sphere centers used when selection_engine = "sphere". Provide numeric vector length 3 or matrix/data.frame with 3 columns.
sphere_radius: Sphere radius used when selection_engine = "sphere".
sphere_units: Units for sphere_centers and sphere_radius: "mm" (world coordinates) or "voxels" (grid coordinates).
sphere_combine: How to combine multiple spheres: "separate" (one region per sphere) or "union".
selection_provider: Optional function used when selection_engine = "custom". Must return a list compatible with build_cluster_explorer_data() output.
analysis_plugins: Optional list of analysis plugin definitions. Plugins can transform extracted time-series/design data before plotting.
default_analysis_plugin: Optional default plugin id. Defaults to "none".

Value

A shiny.appobj.

Details

Calling cluster_explorer() with no arguments launches a synthetic demo dataset so the UI can be explored immediately.