Data structures and I/O for volumetric brain imaging with a focus on fMRI. This is the successor to neuroim with improved S4 classes, sparse/dense 3D–4D representations, and fast resampling/filtering.

Website: https://bbuchsbaum.github.io/neuroim2/ CRAN: https://cran.r-project.org/package=neuroim2

Key Features

• NIfTI / AFNI support — read and write .nii, .nii.gz, and AFNI .BRIK/.HEAD files
• S4 class hierarchy — NeuroVol (3D volumes) and NeuroVec (4D time-series) with dense, sparse, memory-mapped, and clustered variants
• Coordinate systems — full affine-aware transforms between voxel, grid, and world coordinates
• Searchlights & ROIs — spherical, cuboid, and random searchlight iterators plus flexible ROI construction
• Spatial filtering — Gaussian blur, guided filter, bilateral filter, and connected-component labelling
• Resampling — volume-to-volume resampling with configurable interpolation
• Visualization — slice montages, orthographic views, and overlay plotting via ggplot2

Installation

CRAN

install.packages("neuroim2")

R-universe (daily builds)

install.packages("neuroim2",
  repos = c("https://bbuchsbaum.r-universe.dev", "https://cloud.r-project.org"))

Development version (GitHub)

# install.packages("remotes")
remotes::install_github("bbuchsbaum/neuroim2")

Usage

Read a NIFTI image and perform simple operations:

library(neuroim2)
fname <- system.file("extdata", "global_mask_v4.nii", package="neuroim2")
vol <- read_vol(fname)

v1 <- vol[1,1,1]
vol2 <- vol + vol
all(vol == (vol2 - vol))

## [1] TRUE

Create a 4D image from volumes:

vec <- vec_from_vols(list(vol, vol, vol2))
series1 <- vec[1,1,1,]
length(series1)

## [1] 3

Vignettes

Browse the full set of articles on the pkgdown site:

Getting Started	Analysis Workflows	Advanced
Overview	Analysis Workflows	Image Volumes
Choosing Backends	Slice Visualization	NeuroVector
Coordinate Systems	Cookbook	Regions of Interest
Volumes & Vectors	Smoothing	Clustered NeuroVec
Resampling		Pipelines

Albers theme

This package uses the albersdown theme. Existing vignette theme hooks are replaced so albers.css and local albers.js render consistently on CRAN and GitHub Pages. The defaults are configured via params$family and params$preset (family = red, preset = homage). The pkgdown site uses template: { package: albersdown }.