File Exchange

image thumbnail


version 0.5.1 (10.1 MB) by Qianqian Fang
A fast and portable NIfTI-1/2 reader and NIfTI-to-JNIfTI converter


Updated 14 Jul 2020

GitHub view license on GitHub

JNIfTI Toolbox - Fast and portable NIfTI-1/2 reader and NIfTI-to-JNIfTI converter
Version: 0.5 (Ascendence)

[Update 10/11/2019] JNIfTI is now available on Fedora/NeuroFedora, install using
sudo dnf install octave-jnifti

== Overview ==
This is a fully functional NIfTI-1/2 reader/writer that supports both MATLAB and GNU Octave, and is capable of reading/writing both non-compressed and compressed NIfTI files (.nii, .nii.gz) as well as two-part Analyze7.5/NIfTI files (.hdr/.img and .hdr.gz/.img.gz).

More importantly, this is a toolbox that converts NIfTI data to its JSON-based replacement, JNIfTI (.jnii for text-based and .bnii for binary-based), defined by the JNIfTI specification ( JNIfTI is a much more flexible, human-readable and extensible file format compared to the more rigid and opaque NIfTI format, making the data much easier to manipulate and share.

== Installation ==
The JNIfTI toolbox includes a stand-alone NIfTI-1/2 parser that works on both MATLAB and GNU Octave without needing additional components. To just reading and writing the un-compressed NIfTI and Analyze7.5 files (.nii, .hdr/.img), one only needs to run addpath('/path/to/jnifti'). For MATLAB, JNIfTI toolbox utilizes memmapfile-based disk-reading, making it very fast. For Octave, memmapfile is currently not implemented, so, a full reading is required.

The JNIfTI toolbox is also capable of reading/writing gzip-compressed NIfTI and Analyze7.5 files (.nii.gz, .hdr.gz, .img.gz). This feature is supported in MATLAB directly without needing another toolbox (MATLAB must be in the JVM-enabled mode).

To process gzip-compressed NIfTI/Analyze files in Octave and MATLAB with -nojvm, one need to install the open-source JSONLab and ZMat toolboxes, both supporting MATLAB and Octave. They can be downloaded at


To save NIfTI-1/2 data as JNIfTI files, one needs to install JSONLab. The JNIfTI data format supports internal compression (as oppose to external compression such as *.gz files). To create or read compressed JNIfTI files in Octave, one must install the ZMat toolbox, as listed above.

== Usage ==

=== nii2jnii - To convert a NIfTI-1/2 file to a JNIfTI file or data structure ===

nii=nii2jnii('test.nii', 'nii'); % read a .nii file as a nii structure
nii=nii2jnii('test.nii.gz'); % read a .nii.gz file as a jnii structure
nii2jnii('test.nii.gz', 'newdata.jnii') ;% read a .nii.gz file and convert to a text-JNIfTI file
nii2jnii('test.nii.gz', 'newdata.bnii','compression','zlib'); % read a .nii.gz file and convert to a binary-JNIfTI file with compression

=== loadnifti - To read a NIfTI-1/2 (.nii or .nii.gz) file (alias to nii2jnii) ===
# Example:

nii=loadnifti('test.nii.gz'); % read a .nii.gz file as a jnii structure
nii=loadnifti('test.nii', 'nii'); % read a .nii file as a nii structure

=== savenifti - To write an image as NIfTI-1/2 (.nii or .nii.gz) file ===

savenifti(img,'test.nii.gz'); % save an array img to a compressed nifti file
savenifti(img, 'test.nii', 'nifti2'); % save an array img to a nifti-2 file file
savenifti(img, 'test.nii', header); % save an array img with an existing header

=== loadjnifti - To read a JNIfTI (.jnii or .bnii) file ===
# Example:

savejnifti(jnii, 'magic10.bnii','Compression','gzip');

=== savejnifti - To write a JNIfTI structure into a file (.jnii or .bnii) ===
# Example:

jnii=jnifticreate(uint8(magic(10)),'Name','10x10 magic matrix');
savejnifti(jnii, 'magic10.jnii');
savejnifti(jnii, 'magic10_debug.bnii','Compression','gzip');

Cite As

Qianqian Fang (2020). JNIfTI (, GitHub. Retrieved .

Comments and Ratings (2)


I have a problem with savenifti
The tutorial says to save a nifti with an existing header;
>> savenifti(img, 'test.nii', header); % save an array img with an existing header
Is the header the same as the NIFTIHeader field available after using loadnifti?
If so , I get errors as not al fields are numeric values so there is an error on line 52 in savenifti.m:
Or do I miss something?

Qianqian Fang

If you use Fedora 29/30/31, starting from today, you may install the latest version of JNIfTI toolbox directly using the below command

sudo dnf install octave-jnifti

to install the demos/sample data, please install

sudo dnf install jnifti-demos

JNIfTI packages are now officially part of Fedora repository!

MATLAB Release Compatibility
Created with R2016a
Compatible with any release
Platform Compatibility
Windows macOS Linux

Inspired by: ZMat, JSONLab (Development Branch)

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!