Nilearn

Nilearn is an open-source software project that makes brain imaging analysis approachable and versatile. Designed for researchers and practitioners working with neuroimaging data, it provides statistical and machine-learning tools for analyzing brain volumes and cortical surfaces, with an emphasis on clear, practical guidance and an open community.

The project supports a wide range of workflows, including functional MRI (fMRI) modeling, decoding and prediction from brain images, functional connectivity and connectome analysis, brain parcellation, and manipulation of imaging data. Users can extract signals from regions, build brain networks, apply general linear models, and work with methods such as classification, searchlight analysis, independent component analysis, and dictionary learning. Nilearn also includes tools for masking, smoothing, resampling, and thresholding images, alongside plotting features for visualizing volumes, surfaces, statistical maps, and connectivity patterns.

Its documentation offers quickstarts, tutorials, examples, a user guide, and API references, covering both introductory analyses and advanced workflows. Nilearn also provides dataset-fetching utilities and resources for building and scaling neuroimaging machine-learning pipelines.