Skip to contents

Changelog

Source: NEWS.md

ggseg.extra 1.9.9.9016

Post-creation geometry

Smoothing, dilation and vertex reduction are steps you apply to a finished atlas, not settings baked into a build - retuning one should not cost a rebuild.

  • New atlas_dilate() grows or shrinks region geometry, the post-creation counterpart of the snapshot-stage dilation the pipelines applied. Dilate the structures and leave the context alone: a grey brain grown by even a little closes its sulci and flattens into a blob.

  • dilate on the create_*() functions is deprecated in favour of atlas_dilate(). It is still honoured for now, so no atlas changes underfoot.

  • dilate, smoothness, tolerance and smooth_refinements are no longer formals of the create_*() functions: they are caught in ... instead, so they no longer appear in a signature or a help page while a call that still passes one keeps working and says so. Anything else in ... is an error, as an unused argument always was. The notice for the smoothing arguments is the one they already had; only dilate’s is new.

  • atlas_smooth() and atlas_simplify() now do one thing each. atlas_smooth() decides how round an outline is; atlas_simplify() decides how many vertices it costs. keep is gone from atlas_smooth(), and atlas_simplify() is no longer deprecated: it takes labels and exclude like the others.

    Having both in one function meant atlas_smooth() simplified to keep = 0.05 unless told otherwise, so a call that asked only for smoothing quietly threw away 95% of the vertices - and a second call quietly threw away 95% of what the first left. That is how the bundled cortex silhouettes ended up as blobs. Run them in that order, too: simplifying a rounded outline replaces its curves with straight chords, putting the stair-step back.

Minor improvements and fixes

  • Test describe() calls are namespace-qualified. local_mocked_bindings(.package = "terra") attaches terra, and terra exports a describe() of its own, which then masked testthat’s - turning later describe() blocks into GDAL calls on filenames that do not exist. The error aborted the file, so its remaining blocks never ran. Which files were hit depended on run order, which is why a helper-level pin did not hold; the call sites are qualified instead.

  • The sagittal context slice is now the thinnest section of cortex in the slab, not the densest. Sagittal is the one view where picking the slice with the most cortex is actively wrong: area peaks at both tangential extremes - the medial wall by the midline, and the lateral surface - where the slice skims along the sheet and returns a solid blob. A true cross-section, the one that reads as a gyrified ribbon, is where the area is lowest. The ends of the slab are trimmed before the search so it cannot fall into the midline gap. Axial and coronal still take the densest slice, where the reasoning does not apply.

  • The anatomical context silhouette is no longer dilated. dilate exists to keep small deep structures from vanishing, but it was applied to every snapshot in the directory, including the grey brain, where a two-pixel dilation closes the sulci and flattens the mantle. Structures are dilated as before. The exclusion is deliberately case-sensitive so a Cerebellar_Cortex_* parcel, which is a structure in its own right in some atlases, still gets dilated.

  • The anatomical context silhouette is now taken from a single slice for every view, never projected through the slab. Projecting it unions every sulcus the slab passes through and fills them in, so the grey brain came out as a smooth blob rather than a gyrified mantle; the deeper the slab, the worse it got. cortex_slice_for_slab() already chose a representative slice - the densest cortex slice within the slab - for axial and coronal views as well as sagittal, but only the sagittal branch used it. Structures are still projected through the slab, so small deep ones continue to show up in every view that passes through them.

  • create_tract_from_volume() no longer sweeps the whole label volume once per tract. It collected each tract’s voxels with which(arr == label), so a 35-tract atlas made 35 full passes over the array; the voxels are now gathered in a single pass and grouped by label, and the voxel-to-world affine is applied after thinning rather than before, so it transforms at most 4000 points per tract instead of every voxel. Output is unchanged.

New features

  • The grey anatomical silhouette behind a 2D view is now taken from the slice in the slab holding the most cortex, rather than from the slab midpoint. The silhouette is context, so it should be chosen for legibility — and a midpoint can land somewhere with almost nothing to draw. A mid-sagittal slab centred on the midline cuts the interhemispheric fissure: cvs_avg35_inMNI152 has 797 cortical voxels at x=128 against 6703 four voxels away, so the outline came out in fragments while the structures in front of it were fine. Widening the slab could not fix it, since only the structures project through the slab. Ties are broken toward the midpoint, so a slab whose slices are equally informative keeps the position it had before.

ggseg.extra 1.9.9.9014

New features

  • project_volume_anatomical() now warns when protect_cortex removes a label entirely. The guard blocks the atlas from overwriting the cortical ribbon and cerebral white matter, which is right for deep grey structures but erases a white-matter atlas from the only tissue it describes — and it did so silently. ggsegJHU’s ICBM-DTI-81 atlas shipped for a release with its right superior longitudinal fasciculus missing outright, with nothing in the build output to show it. The warning names the lost labels and points at protect_cortex = FALSE.

Bug fixes

  • write_lut() no longer truncates label names to 29 characters. FreeSurfer parses its colour tables on whitespace and its own FreeSurferColorLUT.txt carries names up to 47 characters, so the cap corrupted every long label – and silently merged any two that were identical up to the cut. Julich-Brain’s Ch_123_(Basal_Forebrain)_right came back as Ch_123_Basal_Forebrain_righ, losing the hemisphere suffix and with it the hemisphere.

  • clean_region_name() now strips the same hemisphere affixes that detect_hemi() recognises, including the L_/R_ prefix convention and _left/_right suffixes. Where the two disagreed the hemisphere ended up in region as well as hemi: R_Fx gave hemi "right" but region "r fx", so the two halves of one structure looked like two unrelated structures to anything grouping on region – including the new ggseg.formats::atlas_view_select(). A name that is only a hemisphere word is left alone rather than stripped to nothing.

ggseg.extra 1.9.9.9013

New features

  • atlas_smooth() gains method. The default "close" is the existing morphological closing, which rounds outlines but fills any hole narrower than the smoothing distance – on a thin cortical ribbon that erases the sulci. "chaikin", "ksmooth" and "spline" come from smoothr::smooth() and move vertices rather than dilating the shape, so enclosed holes stay open. Use "close" for solid shapes such as tract tubes and one of the others where the geometry has holes worth keeping.
  • atlas_smooth(smoothness =) is now a 0–1 strength shared by every method, mapped internally onto each method’s native parameter, so the same value means a comparable amount of smoothing whichever one is chosen. Values outside 0–1 are an error carrying the conversion rule, rather than being silently reinterpreted: divide an old method = "close" distance by 5.

Bug fixes

  • 2D atlas geometry no longer loses its holes. coords2sf() grouped coordinates by (.subid, .id), but .subid is the ring index within a polygon (1 = exterior, >1 = holes) and .id the polygon index, so every ring became its own solid polygon and every enclosed hole was filled. A thin cortical ribbon came out as a blob: the sagittal fsaverage slice measured 20,117 against a true area of 17,481, with 44 of its 137 rings surviving. Rings are now assembled as exterior-plus-holes, and the area matches exactly. This affects every atlas built through get_contours() – cortical, subcortical, wholebrain and tract – so rebuilt atlases will differ from ones built before this fix.

  • Tract 2D projections are no longer drawn in the wrong plane. streamlines_to_volume() built each tube in the template’s native voxel layout and relied on RNifti::orientation(nii) <- "RAS" to convert it. For .mgz templates that assignment was a silent no-op — read_volume() returns a bare array carrying no affine, so orientation() reports "RAS" whatever the layout — while the anatomical reference was converted, being read through read_volume(reorient = TRUE). With an LIA template such as FreeSurfer’s aseg.mgz, voxel axes 2 and 3 ended up swapped between the two, so axial projections were drawn over coronal anatomy and sagittal ones appeared rotated 90°. The volume is now reoriented explicitly.

  • Tract 2D projections are no longer offset from the anatomical reference. center_meshes() translates centerlines to the origin for 3D display, and the snapshot step rasterised those translated coordinates as though they were world RAS, shifting every tract relative to the anatomy. The translation is now recorded and undone before rasterising; 3D output is unchanged.

  • Sagittal cortex reference slices now come from their own slab rather than a fixed plane, matching what axial and coronal already did. A sagittal slab at an explicit position previously had its silhouette drawn somewhere else. An explicit cortex_x still wins, and hemisphere-named views fall back to their lateral positions when the slab holds no cortex to choose from.

  • The anatomical reference for tract atlases now includes the brainstem, cerebellar cortex and deep grey structures alongside the cortical ribbon, so tracts descending out of the cerebrum are drawn against anatomy instead of empty space. White matter is excluded, cerebral and cerebellar alike: tracts run through it, and filling it would bury them. Excluding cerebellar white matter also keeps the cerebellum a foliated shell like the cerebral ribbon instead of a solid mass that merges with the occipital lobe in sagittal views.

  • create_tract_from_volume() and create_tract_from_tractography() now honour atlas_name for the atlas object itself, not only for the output directory. The finished atlas previously took the name derived from its tract labels (e.g. "tracts"), ignoring what the caller asked for.

ggseg.extra 1.9.9.9012

New features

  • prepare_subcortical_mni152() embeds a subcortical parcellation supplied in fixed FSL-MNI152 space into a FreeSurfer subject’s aseg (via the known mni152.register.dat transform), replacing the lumped aseg structures the parcels subdivide and returning a merged volume plus a matching colour table ready for create_subcortical_from_volume(). It is the fixed-registration counterpart to prepare_subcortical_anatomical(), which instead computes an mri_coreg registration to cvs_avg35_inMNI152; use this one when the atlas already lives in a standard MNI152 template and should keep fsaverage5 context. aseg_subcortical_labels() returns the lumped subcortical structure ids a finer parcellation typically subdivides.

  • create_tract_from_volume() builds a white-matter tract atlas from a volumetric tract label map (one integer label per tract) rather than from streamlines: each tract’s voxel cloud is reduced to a principal-curve centerline and handed to create_tract_from_tractography(), which builds the 3D tubes and 2D projection. This suits probabilistic tract atlases distributed as NIfTI label volumes (e.g. AtlasTrack). Adds princurve to Suggests.

Bug fixes

  • create_tract_from_tractography() no longer errors when input_tracts is an in-memory list of coordinate matrices: the setup log interpolated the matrices into a {.path} inline style, which cli cannot format as file paths.

ggseg.extra 1.9.9.9011

Minor improvements and fixes

  • Workflows written by use_atlas_github_actions() now run on pull requests. pkgdown previously had no pull request trigger at all, so a broken reference index or vignette could only fail after merging; code-quality filtered pull requests to those targeting main, so a stacked pull request skipped linting entirely. Neither publishes from a pull request: the shared workflows guard the pkgdown deploy on github.event_name, and the coverage badge and README commits on github.ref.

ggseg.extra 1.9.9.9010

New features

  • use_atlas_github_actions() adds the shared ggsegverse GitHub Actions workflows to a package, in the style of usethis::use_github_action(). atlas_github_actions() lists what is available. Run it on a freshly scaffolded atlas package, or on an existing one to replace hand-maintained workflows with the shared set.
  • setup_atlas_repo() gains github_actions, TRUE by default.

Minor improvements and fixes

  • Atlas packages now receive workflows as short caller stubs for the reusable workflows in ggsegverse/.github, rather than inline copies. The scaffold previously shipped 229 lines of inline workflow, none of which called the shared workflows, so every new package started out with CI that had already drifted.
  • setup_atlas_repo() no longer copies the atlas template’s .github/ directory. That directory is the template’s own infrastructure — its smoke-test workflow and the scripts driving it — and had been leaking into every generated package.
  • The offline fallback now produces the same workflows as the downloaded template, since they no longer come from the template at all. Previously the fallback silently omitted CI.

ggseg.extra 1.9.9.9009

Bug fixes

  • The bundled atlas template no longer generates a broken package. Template placeholders were spelled {GGSEG} / {REPO}, which R parses as brace blocks, so air format rewrote .{GGSEG} into a separate expression and split library({REPO}) across lines. Generated packages had an accessor that returned nothing and a test suite that referenced an undefined object.
  • data-raw/create-atlas.R in the scaffold now calls the current API. It previously used read_freesurfer_lut() (never exported), color_lut, input_gifti, input_cifti, and a tract input_volume argument, none of which exist, and omitted the required source / desc arguments to create_cortical_from_neuromaps(). Every uncommented section would have errored.

Minor improvements and fixes

  • Template placeholders are now bare identifiers (ATLASNAME, PKGNAME, YEARNUM) so template sources parse as valid R and cannot be rewritten by R tooling. setup_atlas_repo() still substitutes the legacy brace-wrapped spelling, so templates published before this change continue to work.
  • Added air.toml excluding inst/templates/ and inst/rstudio/templates/ from formatting.
  • The scaffold now includes an atlas_smooth() post-processing step and no longer passes the deprecated tolerance argument.
  • Generated packages set Config/Needs/website: ggsegverse/ggseg.docs and use the ggseg.docs pkgdown template instead of an inlined bslib theme, and no longer set LazyData without a data/ directory.
  • Generated packages now pass R CMD check cleanly. The template declared License: CC0 while shipping an MIT-style LICENSE file, which produced a NOTE in every new package; it now declares MIT + file LICENSE, matching both the bundled file and the other ggseg atlas packages.
  • Generated packages ship a .lintr excluding data-raw/, so the deliberately commented-out scaffold no longer trips commented_code_linter.

ggseg.extra 1.9.9.9008

Minor improvements and fixes

  • Tract 2D projections now reuse the same centerline as the 3D tube (built with the configured n_points / centerline_method) instead of recomputing a different 50-point mean centerline, so the two representations agree.
  • read_tractography() reads .tck files without the previous quadratic slow-down on large bundles.
  • create_tract_from_tractography() validates tube_segments (integer >= 3), and degenerate leading tract segments no longer yield NaN tube vertices.
  • Cerebellar trilinear resampling skips non-finite deformation coordinates rather than raising an error, and a deep nucleus that cannot be converted to polygons is now reported instead of silently dropped.
  • create_wholebrain_from_volume() splits cortex at the correct (1-based) midline voxel; the left/right boundary was previously off by one voxel.
  • Reading a FreeSurfer LUT warns about malformed lines; write_lut() validates its input; read_lut()/read_dpv() fail with clear messages on malformed headers, and read_dpv() handles zero-face surfaces.
  • CIFTI files with more than one label map warn that only the first is used.
  • Annotation files whose colour table already defines an “unknown” region no longer produce a duplicate unknown label.
  • Contour extraction reports a clear error when no region yields a contour instead of a cryptic downstream failure.
  • Subcortical mesh tessellation honors the requested verbosity inside parallel workers.
  • Reading a subcortical surface surfaces the underlying conversion error when the fallback reader is unavailable, and the interactive atlas preview warns on 3D render failures instead of failing silently.
  • setup_atlas_repo() no longer rewrites files under a template’s .git directory and reports failed file copies/renames.

ggseg.extra 1.9.9.9007

Bug fixes

  • create_cerebellar_from_volume() no longer errors on float-typed parcellation volumes. Sampling labels at the SUIT surface forced an integer vapply() template, so a double array — e.g. the SUIT volume written by transform_mni_to_suit() — aborted the build. The volume is now coerced to integer before sampling.
  • read_tractography() reads .trk files whose header track count is 0. The TrackVis format uses 0 to mean “count not recorded, read to end of file”; the reader previously trusted the count and returned no streamlines, yielding an empty atlas. It now reads to the end of the file and treats a positive count as an upper bound.
  • create_cortical_from_neuromaps() / read_neuromaps_volume() no longer abort with 'breaks' are not unique when a continuous map has tied values (common for thresholded maps or maps with many zeros). Duplicate quantile breaks are collapsed and the bin count is reduced with a warning; an all-medial-wall hemisphere now errors with a clear message.
  • Tract atlases built from in-memory streamline matrices (e.g. create_tract_from_tractography(input_tracts = list(cst = matrix(...)))) now detect voxel- versus RAS-space correctly. Detection previously flattened the matrices and always assumed RAS, misplacing voxel-space tracts.
  • create_tract_from_tractography() now reads the voxel-to-world affine from FreeSurfer .mgz headers correctly, and warns instead of silently falling back to an approximate origin-centering heuristic when a template’s affine cannot be read.
  • create_subcortical_from_volume() derives a default atlas_name of aseg (not aseg.nii) from a .nii.gz input.
  • mri_info is now called with a shell-quoted volume path, so cerebellar deep-nuclei meshing works for volume paths that contain spaces.
  • Contour extraction (subcortical and tract 2D geometry) no longer crashes on empty or all-NA region rasters, keeps the valid contours when only some region geometries are empty, and reports a clear error when no region yields any contour instead of a cryptic dplyr failure.
  • Verbosity and boolean options parse spelled-out strings consistently: GGSEG_EXTRA_VERBOSE=false now silences output, and string values such as "yes" or "1" are honored across the explicit, option, and environment-variable channels.

ggseg.extra 1.9.9.9005

Bug fixes

  • read_volume() now reorients FreeSurfer .mgz volumes to RAS+, matching its long-standing behaviour for NIfTI inputs. Previously only niftiImage objects were reoriented, so .mgz volumes (e.g. FreeSurfer’s LIA-oriented aseg.mgz) reached the RAS+-assuming projection code still in LIA order. Subcortical atlases built directly from a .mgz therefore came out left-right flipped in axial views, top-bottom flipped in coronal, and 90-degrees rotated in sagittal; atlases built from reoriented .nii.gz volumes (and tract atlases, whose geometry is already in scanner RAS) were unaffected. Volumes whose header carries no valid RAS information fall back to native voxel order.

Subcortical atlas builder helpers

New thin compositions of the existing ggseg.formats atlas ops and the volume reader, distilled from the repeated boilerplate in the ggsegFreeSurfer subcortical build scripts:

  • subcortical_slabs() builds a slab table from the bounding box of a set of labels, reading the volume in the same frame the builder uses so coronal/axial/sagittal slabs can’t be pointed at the wrong slices.
  • aseg_context() collapses the standard post-processing chain (punch cortical white matter, strip the structures aseg doesn’t draw, demote everything outside focus to grey context, drop empty views) into one call. The focus set is subtracted from the context set with exact, case-sensitive matching, so a region is never swallowed by a context entry that is a substring of its name (e.g. Thalamus vs hypothalamus). aseg_hidden_labels() returns the default stripped set.
  • lut_add() / lut_combine() append and merge FreeSurfer-style colour tables (validating with is_lut() and warning on index clashes) for atlases that add custom prefixed labels.
  • create_subcortical_from_volume() gained two opt-in arguments: slabs now also accepts a subcortical_slabs() list spec (e.g. slabs = list(labels = 801:810, coronal = 3)), and a new context argument runs aseg_context() on the finished 2D atlas (e.g. context = list(focus = "Hippocampus")). Both thread through create_wholebrain_from_volume()’s subcortical_opts.

sf smoothing moves out of atlas creation

Pipeline-time simplification of 2D sf geometry was the most common reason to re-run an otherwise expensive create_*() pipeline (10+ minutes for volumetric atlases). All create_*() functions now return raw, unsmoothed sf polygons. The (cheap) atlas_smooth() post-processing step is the single place where simplification level is decided, so you can iterate freely on a cached atlas:

atlas <- create_cortical_from_annotation(...) |>
  atlas_smooth(keep = 0.2, exclude = "cortex_")
  • 3D mesh smoothing (tessellation, FreeSurfer mris_smooth, decimation) is unchanged.
  • tolerance, smoothness and smooth_refinements on every create_*() function are now soft-deprecated. Supplying any of them emits a lifecycle::deprecate_warn() and the value is otherwise ignored.
  • atlas_smooth() gained labels / exclude regex arguments so the brain-outline geometry can stay crisp while everything else is simplified.
  • atlas_smooth() also gained a smoothness argument that applies a positive-then-negative sf::st_buffer() (morphological closing) after vertex simplification. This restores the rounded sulcal curves the old pipeline smoothness argument produced, but as a per-region, post-atlas operation. Pass keep = NULL to skip vertex reduction and only round off voxel-edge stair-steps.
  • The “large atlas” warning text now points at atlas_smooth() instead of the deprecated tolerance argument.
  • create_wholebrain_from_volume() no longer injects a default smooth_refinements = 2L into the cerebellar sub-pipeline.

API naming and argument consistency pass

A pass over the atlas-builder family (create_cortical_from_*(), create_subcortical_from_volume(), create_cerebellar_from_*(), create_wholebrain_from_volume(), create_tract_from_tractography()) and the LUT helpers to make the API “as similar as possible” across atlas types. Old names/arguments keep working with a lifecycle::deprecate_warn().

Anatomical-context coregistration helpers

  • New coregister_volume() wraps mri_coreg to align an atlas volume to a FreeSurfer subject’s T1 grid (default cvs_avg35_inMNI152), returning a reusable LTA file.
  • New project_volume_anatomical() resamples each atlas label with trilinear interpolation onto the target aparc+aseg grid, takes the argmax across labels, and produces a merged volume that combines the source aparc+aseg (anatomical brain-outline context) with the user’s atlas labels. It returns a list(volume, lut, id_offset): the merged volume and a colour table aligned to it (FreeSurfer names for the surviving aparc+aseg context labels plus the user’s labels at their shifted ids), so the context regions render with names aseg_context() recognises.
  • New prepare_subcortical_anatomical() chains both in a single call, returning the same list(volume, lut, id_offset). create_subcortical_from_volume() now accepts that list directly as input_volume, unpacking the matching colour table for you (an explicit input_lut still wins).
  • Removes the ~50 lines of per-atlas boilerplate previously hand-rolled in ggsegShen and ggsegHO build scripts.
  • id_offset parameter (default 200L) shifts every input label ID in the merged volume so they don’t collide with FreeSurfer aparc+aseg IDs (e.g. an atlas where 11 means “Putamen” while FS uses 11 for “Caudate”). The returned colour table is shifted in lock-step.
  • protect_cortex parameter (default TRUE) keeps aparc+aseg cortex voxels (labels 1000-2999) and cortical white matter (2, 41) intact even when the user’s argmax wins above threshold — this preserves the brain-outline geometry that the subcortical pipeline draws as anatomical context.
  • The per-voxel argmax streams the running winner across labels instead of materialising an n_voxels x n_labels probability matrix, so projecting a many-region atlas (e.g. Shen-268 onto a 256^3 grid) no longer needs tens of gigabytes of memory.

ggseg.extra 1.9.9.9004

Template-based atlas repo scaffolding

  • setup_atlas_repo() now downloads the atlas template from ggsegverse/ggseg-atlas-template instead of bundling template files inside the package. This makes the template a single source of truth that can be updated independently.
  • The generated scaffold includes all modern atlas repo conventions: Quarto README, Bootstrap 5 pkgdown config, code-quality workflow, render-readme workflow, update-codemeta workflow, and AI agent instructions.
  • data-raw/create-atlas.R now scaffolds all pipeline methods (cortical, subcortical, cerebellar, tract, wholebrain) with commented sections.
  • Falls back to a bundled minimal template when offline.
  • Rich CLI messaging throughout the scaffolding process.

ggseg.extra 1.9.9.9003

Large-atlas warning

  • warn_if_large_atlas() now scales its threshold with region count via per_region = 50 (threshold = max(max_vertices, per_region * n_regions)). Prevents spurious warnings for high-resolution parcellations (e.g. Kong 1000-parcel) where keep_shapes = TRUE sets a ~40 vertices/region floor.
  • Fixed the follow-up hint which previously suggested raising tolerance to reduce vertices; lower values simplify more aggressively.

Deep cerebellar nuclei support

  • create_cerebellar_from_volume() now detects deep cerebellar nuclei (Dentate, Interposed, Fastigial) that have volume voxels but no SUIT surface vertices. These are tessellated as individual 3D meshes with proper tkRAS-to-MNI coordinate transform, and rendered as smoothed coronal projection sf geometries in a separate “nuclei” view.
  • Orphaned surface parcels (e.g. buckner17 17Networks_14) that are too small for any SUIT vertex to land on are now rescued by assigning the nearest surface vertex, keeping them on the flatmap.
  • Voxel neighbor fill radius expanded from 1 to 3 (configurable) to better capture small regions during volume-to-surface sampling.
  • Restored colour auto-fill in read_suit_parcellation() and read_neuromaps_volume().

Cerebellar atlas type and SUIT flatmap pipeline

New “cerebellar” atlas type added across the ggseg ecosystem (ggseg.formats, ggseg, ggseg3d). Cerebellar atlases use SUIT flatmap sf polygons for 2D rendering and per-region meshes for 3D (like subcortical).

Three creation pipelines:

Supporting functions:

Boundary triangle splitting

Boundary triangles (where vertices belong to different atlas regions) are now split into sub-polygons along edge midpoints instead of being assigned wholesale to a single region. This eliminates the sawtooth artifacts at region borders that resulted from the triangular mesh geometry.

  • 2-region boundaries: triangle is split at the midpoints of the two cross-boundary edges — the majority region gets a quadrilateral, the minority region gets a triangle.
  • 3-region boundaries: triangle is divided into three quadrilaterals meeting at the centroid.
  • Default tolerance increased from 0.5 to 1 — the smoother borders tolerate higher simplification without visible degradation.

Bug fixes

  • ensure_fs_compatible_nifti() no longer errors when the NIfTI header cannot be read (e.g. .mgz files or nonexistent paths). It now falls through gracefully and lets downstream FreeSurfer commands handle the file.

ggseg.extra 2.0.1

Cortical pipeline: mesh projection

The cortical atlas pipeline now projects inflated mesh triangles directly to 2D polygons via orthographic projection, replacing the screenshot-based contour extraction from v2.0.0.

  • Much faster — atlas creation completes in ~5 seconds instead of minutes.
  • Cleaner geometry — no pixel staircase artifacts from rasterisation.
  • Fewer dependencies — no FreeSurfer rendering, ImageMagick, or Chrome needed for 2D geometry (FreeSurfer is still required to read annotation files).
  • Better small-region visibility — boundary faces are assigned to the smallest neighbouring region so tiny parcels are not swallowed by their neighbours.
  • Smooth region borders — boundary triangles (vertices in different regions) are split along edge midpoints so each region gets a clean polygon slice, eliminating the sawtooth artifacts from whole-triangle assignment.

Breaking changes

  • Removed method, snapshot_dim, smoothness, and steps parameters from all create_cortical_from_*() functions. The pipeline always reads data and projects to 2D in one pass — no step-based control needed.
  • Changed default tolerance from 0.5 to 1 — the triangle-splitting approach produces smoother borders that tolerate higher simplification.

Lighter dependency footprint

  • Moved chromote, htmlwidgets, magick, smoothr, terra, RNifti, and freesurfer from Imports to Suggests. Users who only need the cortical pipeline no longer need these packages installed. They are checked at runtime and requested when needed (subcortical, tract, and volumetric pipelines).

New internals

  • Added R/mesh-projection.R with the full geometric projection algorithm: orthonormal view basis computation, backface culling, per-face label assignment, and triangle-to-polygon union via sf.

ggseg.extra 2.0.0

  • Major rewrite of atlas creation pipelines with modular step-based architecture
  • Added GIFTI (.label.gii) and CIFTI (.dlabel.nii) annotation support
  • Added neuromaps surface and volume annotation pipelines
  • Added whole-brain atlas creation from volumetric parcellations
  • Added white-matter tract atlas creation from tractography files
  • Added three-level verbosity control (silent/standard/debug)
  • Deprecated ggseg_atlas_repos(), install_ggseg_atlas(), and install_ggseg_atlas_all() in favour of ‘ggseg.hub’
  • Moved convert_legacy_brain_atlas() to ‘ggseg.formats’ (re-exported)
  • Removed rgdal, purrr, reticulate, and tidyr dependencies
  • Replaced reticulate/kaleido snapshots with chromote
  • Protected all parallel operations against multicore fork crashes
  • Removed dead FreeSurfer wrapper functions
  • Fixed read_ctab for multi-word labels
  • Fixed subcortical label classification in whole-brain pipeline

ggseg.extra 1.6

1.6.0

  • Removed rgdal dependency, replaced with sf/terra (#49, #59)
  • Fixed r-universe API calls (JSON array format change)
  • Fixed vignette build issues with conditional evaluation for suggested packages
  • Replaced reticulate/kaleido with webshot2 for plotly screenshots
  • Updated system setup vignette with new requirements
  • Added documentation for parallel processing and progress bars
  • Added note about freesurfer dev version requirement
  • Updated CITATION to use bibentry()
  • Updated pkgdown site with ggseg brand styling
  • Fixed mris_label2annot example documentation

ggseg.extra 1.5

1.5.33.003

  • small bug fix that prevented calls to FreeSurfer
  • Possibility to initiate new atlas project from the RStudio Project GUI

ggseg.extra 1.5.33

  • removes purrr dependency
  • used ggseg r-universe as install repo for install functions

ggseg.extra 1.5.32

  • non-standard columns in 3d atlas are retained in 2d atlas
  • Freesurfer annotation file custom S3 class implemented
  • progressbar for region snapshots

ggseg.extra 1.5.3

  • Added pipeline functions for:
    • creating ggseg3d-atlas from annotation files
    • creating ggseg3d-atlas from volumetric files
    • creating ggseg-atlas from cortical ggseg3d-atlas
    • creating ggseg-atlas from volumetric files
  • Added a NEWS.md file to track changes to the package.