Nilearn — Machine Learning for NeuroImaging

Nilearn — Machine Learning for NeuroImaging

Nilearn is a Python library for machine learning on neuroimaging data (fMRI, MRI). It bridges scikit-learn with nibabel/NifTI data formats, enabling brain decoding, functional connectivity analysis, and biomarker extraction without low-level data wrangling.

Key features

• Mass-univariate GLM for fMRI task analysis (SPM-compatible)
• Functional connectivity matrices and ICA-based parcellations
• 30+ built-in brain atlases (AAL, Destrieux, Schaefer, DiFuMo)
• High-quality 3D/4D brain plot functions (glass brain, stat maps, surfaces)
• Compatible with BIDS-formatted datasets via niworkflows

Quick start

pip install nilearn

from nilearn import datasets, plotting
from nilearn.connectome import ConnectivityMeasure

# Load example resting-state data
data = datasets.fetch_development_fmri(n_subjects=5)
measure = ConnectivityMeasure(kind="correlation")
matrices = measure.fit_transform([img for img in data.func])
plotting.plot_matrix(matrices[0], colorbar=True)
plotting.show()

npx ai-supply add nilearn-neuroimaging-ml

Curated mirror of the open-source Nilearn (BSD-3-Clause). Get it from the source.