Neuroinformatics Research Group

Projects

XNAT

Open-Source Software Platform

XNAT is an open-source software platform developed to support imaging informatics research. The platform offers permission-controlled storage of imaging and clinical assessment data, as well as support for containerized data processing to support reproducible science. XNAT has been used in hundreds of research projects and clinical trials at institutions around the world.

Development on XNAT began in 2001 at Washington University in the Buckner Lab, which is now located at Harvard University. The first developers were hired in 2001-02 with funding from the Howard Hughes Medical Institute, and the first release followed soon after. XNAT joined the Biomedical Informatics Research Network in 2005. In 2008 the project received independent funding under an NIBIB R01 grant.

With ongoing funding from NIBIB and NCI, many more releases and advances in core XNAT technology have followed through the years, with institutions and teams of outside developers contributing to the extended suite of XNAT plugins and integrations.

XNAT is currently funded by NIH grant R01 EB009352.

XNAT allows the combination of imaging and non-imaging data in one research repository.
XNAT allows the combination of imaging and non-imaging data in one research repository.

Human Connectome Project

Mapping the Brain Across the Human Lifespan

The Human Connectome Project (HCP) encompasses over 20 funded studies to map the brain across the human lifespan and in a range of neurological and psychiatric disorders. The initial HCP Young Adult study imaged more than 1100 young using a cutting edge scanner and set of structural and functional sequences. The resulting data set was shared openly with the research community through the XNAT-based ConnectomeDB platform. Since the end of the original HCP grant, the HCP operations in the NRG has been supported by the Connectome Coordinating Facility (CCF), enabling 19 other disease and aging related studies. Ongoing study of the aging brain continues with the Aging Adult Brain Connectome (AABC)

The Human Connectome Project uses XNAT imaging and processing to help map the human brain.
The Human Connectome Project uses XNAT imaging and processing to help map the human brain.

OASIS

Open Access Series of Imaging Studies

OASIS is a project aimed at making neuroimaging data sets of the brain freely available to the scientific community. By compiling and freely distributing neuroimaging data sets, we hope to facilitate future discoveries in basic and clinical neuroscience.

To date, three OASIS datasets have been released:

  • OASIS-1: Cross-sectional MRI Data in Young, Middle Aged, Nondemented and Demented Older Adults (416 subjects, 434 image sessions)
  • OASIS-2: Longitudinal MRI Data in Nondemented and Demented Older Adults (150 subjects, 373 image sessions)
  • OASIS-3: Longitudinal Neuroimaging, Clinical, and Cognitive Dataset for Normal Aging and Alzheimer’s Disease (1098 subjects, 3776 MR and PET image sessions)

OASIS is funded by multiple NIH grants as described in the OASIS Data Use Agreement.

For more information, email Pamela LaMontagne or visit the OASIS website.

The Open Access Series of Imaging Studies (OASIS) makes neuroimaging data sets freely available to the scientific community.

MIR Research Imaging Repository

Mallinckrodt Institute of Radiology – Research Image Repository (MIRRIR) is a datastore consisting of radiological images and reports used by Washington University investigators to conduct translational research. This includes development for computational algorithms for disease prediction, diagnosis and treatment guidance. MIRRIR is designed to hold all radiology exams obtained in BJC HealthCare hospitals and the associated reports and observations generated by MIR radiologists.

MIRRIR includes a suite of integrated applications that together orchestrate a secure workflow to retrieve data from clinical systems, anonymize the data, and deliver it to authorized users. To date, anonymized data from more than 200,000 subjects and 300,000 image sessions and clinical assessments has been made available to authorized researchers.

Aggregation of large-scale clinical data like this has already yielded promising returns, when combined with machine learning and image analysis pipelines.

The MIR Research Imaging Repository is a vast image data store, using XNAT as its backbone.

For more information, email Woonchan Cho.

Translational Imaging Portal

Subhead

The goal of translational medicine is to speed the application of analytic methods developed in the lab into clinical practice. The Translational Imaging Portal (TIP) is the result of a collaboration between the CIRC and radiologists at Barnes-Jewish Hospital. What started as a pilot program and proof-of-concept in 2015 is now a continuing part of surgical planning at BJC HealthCare.

The TIP process sounds simple. Pre-surgical patients are scanned and those scans are sent to a highly secured instance of XNAT where radiology fellows perform rapid image processing and send the results back to the surgical planning theater. Examples of the kind of processing include mapping of functional brain areas in brain tumor and epilepsy patients and analysis of regional brain volumes in aging populations. To date, more than 4,000 patient exams have been analyzed on the TIP platform.

TIP is a collaboration between CIRC, the NIL-RC, the Department of Neurosurgery at Washington University School of Medicine, and BJC HealthCare’s 1Rad team.

For more information, email Pamela LaMontagne.

The Translational Imaging Portal applies the latest brain mapping research to patients' presurgical planning in real time.
The Translational Imaging Portal applies the latest brain mapping research to patients’ presurgical planning in real time.

Integrative Imaging Informatics for Cancer Research Center

The Integrative Imaging Informatics for Cancer Research Center (I3CR) focuses on expanding the open source XNAT informatics platform to better support computational workflows in cancer imaging. I3CR has also developed knowledge management tools to better track data processing and analysis, including tools for orchestrating and tracking container-based computing pipelines.

Artificial intelligence (AI) and other computationally intensive methods have the potential to revolutionize cancer imaging research and patient care. The broad adoption of these technologies depends on the availability development of imaging informatics tools to assist users in managing massive data sets, generating well-curated annotations, and accessing scalable computing resources. I3CR’s current development focus standards-based clinical interfaces, cohort discovery services with natural language processing support, and extension of XNAT container-based computing service to support high performance computing and cloud computing environments. Similar efforts are underway to support imaging of animal models under the Preclinical Imaging with XNAT Informatics (PIXI) project in partnership with Kooresh Shoghi.

Through the I3CR, XNAT is incorporating imaging-based machine learning tools into cancer research workflows

I3CR and PIXI are funded by NIH grant. U24 CA258483 and U24 CA253531.

For more information, email Daniel Marcus.

More in Marcus Lab