What is NYUAD XNAT?

XNAT (eXtensible Neuroimaging Archive Toolkit) is an open-source platform for managing neuroimaging research data. At NYU Abu Dhabi, we use XNAT to store, organize, process, and share MRI data from our research projects.

What is XNAT?

XNAT is a web-based data management platform specifically designed for neuroimaging research. It provides:

  • Centralized Data Storage: Secure, organized storage for MRI scans and related data

  • Data Organization: Hierarchical structure (Projects → Subjects → Sessions → Scans) that matches research workflows

  • Automated Processing: Integrated pipelines for data conversion, quality control, and preprocessing

  • Collaboration Tools: Easy sharing and access control for research teams

  • Web Interface: Access your data from anywhere through a browser

Why Do We Use XNAT at NYUAD?

XNAT helps our research community by:

1. Streamlining Data Management

  • Automatically receives DICOM files directly from MRI scanners

  • Organizes data in a consistent structure across all projects

  • Eliminates the need for manual file organization

2. Enabling Automated Processing

  • Runs preprocessing pipelines (like fMRIPrep, MRIQC) directly on stored data

  • Converts data between formats (DICOM → NIfTI → BIDS) automatically

  • Processes data on our HPC cluster without manual file transfers

3. Facilitating Collaboration

  • Multiple researchers can access the same data securely

  • Project-based access control ensures data privacy

  • Shared resources and configurations across team members

4. Ensuring Data Quality

  • Built-in quality control pipelines (MRIQC, ARI-validator)

  • Validation checks ensure data meets analysis requirements

  • Quality reports help identify issues early

5. Supporting Reproducible Research

  • Standardized data formats (BIDS) enable reproducible analyses

  • Processing history is tracked and documented

  • Easy access to both raw and processed data

How Do We Use XNAT at NYUAD?

Our typical workflow follows these steps:

1. Data Collection

  • MRI scanners send DICOM files directly to XNAT

  • Data is automatically organized by project, subject, and session

  • Files are stored securely in the XNAT archive

2. Data Conversion

  • Convert DICOM files to NIfTI format using the dcm2niix pipeline

  • Organize data into BIDS format using the dcm2bids pipeline

  • BIDS format enables use of standard analysis pipelines

3. Quality Control

  • Run mriqc to assess data quality automatically

  • Review quality reports to identify problematic scans

  • Validate BIDS datasets using project-specific validators

4. Preprocessing

  • Run preprocessing pipelines like fMRIPrep for functional MRI data

  • Process diffusion data with tractoflow

  • All processing happens on our HPC cluster automatically

5. Data Access

  • Download processed data through the web interface

  • Use the Desktop Client for bulk downloads

  • Access data programmatically using Python or MATLAB scripts

6. Analysis

  • Download analysis-ready data to your local machine or analysis server

  • Use standard neuroimaging tools (FSL, FreeSurfer, etc.) with the processed data

  • Results can be uploaded back to XNAT for sharing

Key Features of NYUAD XNAT

Access and Security - Google OAuth authentication for secure login - Project-based access control - API tokens for programmatic access

Available Pipelines - Data Conversion: dcm2niix, dcm2bids, dcm2hcp - Quality Control: mriqc, ari-validator - Preprocessing: fMRIPrep, tractoflow, HCP Pipeline

Data Formats Supported - DICOM (raw scanner data) - NIfTI (standard analysis format) - BIDS (Brain Imaging Data Structure) - HCP (Human Connectome Project format)

Download Methods - Web browser interface - XNAT Desktop Client - Python scripts - MATLAB scripts

Getting Started

New to XNAT?

  1. Set up your account: See Accessing XNAT for registration and login instructions

  2. Learn about data formats: Read Understanding Data Structure to understand MRI data organization

  3. Explore available tools: Check XNAT Pipelines Overview to see what pipelines are available

  4. Upload your first data: Follow Data Upload Procedures to upload data to XNAT

Already have an account?

Access Information

For more information about XNAT in general, visit the official XNAT documentation.

Next Steps