Rxnat: An Open-Source R Package for XNAT-Based Repositories.

Area of Science:

• Neuroimaging
• Bioinformatics
• Data Science

Background:

• The Extensible Neuroimaging Archive Toolkit (XNAT) is a widely adopted platform for managing neuroimaging data.
• Key public neuroimaging data repositories like NITRC, ConnectomeDB (Human Connectome Project), and XNAT Central utilize XNAT.
• There is a growing need for tools that facilitate data access for researchers using specific programming languages.

Purpose of the Study:

• To introduce Rxnat, an open-source R package for interacting with XNAT-based repositories.
• To provide R users with capabilities similar to existing Python tools (PyXNAT, XNATpy) for XNAT data access.
• To serve as a unified interface for the R community to access diverse, publicly available neuroimaging datasets.

Main Methods:

• Development of an open-source R package named Rxnat.
• Implementation of functionalities to query multiple XNAT repositories.
• Inclusion of capabilities to download specific subsets or all images from repositories.
• Demonstration using a neuroimaging data normalization task across NITRC and HCP repositories.

Main Results:

• Rxnat enables seamless querying and downloading of neuroimaging data from XNAT repositories using R.
• The package facilitates data access for R users, bridging the gap with existing Python-based solutions.
• Successful illustration of Rxnat's utility in a practical neuroimaging data processing workflow.

Conclusions:

• Rxnat provides a crucial tool for R-based neuroimaging research, enhancing accessibility to public data.
• The package acts as a lingua franca, enabling R analysts to interface with multiple XNAT-based repositories.
• Rxnat has the potential to significantly streamline neuroimaging data analysis workflows for the R community.