Society for Imaging Informatics in Medicine (SIIM) 2011 Abstract
Title: New tools for integrating clinical images into research studies
Background:
The Informatics for Integrating Biology and the Bedside (i2b2, available at http://www.i2b2.org)
software suite enables the repurposing of healthcare data for clinical research [1]. The i2b2 software
provides clinical investigators with the tools necessary to integrate medical record and clinical research
data in the genomics age, a software suite to construct and integrate the modern clinical research
chart (CRC). The i2b2 open-source software is designed to support two general use cases within the
mission of repurposing electronic health records for research. One is to enable queries that can
browse through all the enterprise data to find cohorts of patients that are of interest for further
research while preserving patient privacy. A second is to use the data provided by the medical record
to do a “deep dive” into the phenotype of a set of patients (possibly identified from the first use case)
in support of genomic, outcomes, and environmental studies. While the i2b2 core software enables
such access to a vast majority of medical information, it does not support medical images. Yet many
current and future clinical research studies rely on medical images for quantitative metrics for
diagnosis, prognosis and treatment response. Images are used to quantify disease burden such as
tumor volume, inflammation, hemorrhage, and infarction. Serial imaging is used to quantify the
outcome of interventions, such as changes in tumor size or loss of brain tissue. Importantly, medical
images are providing an ever increasing number of sensitive diagnostic approaches to disease, such as
the use of susceptibility weighted and diffusion tensor MR imaging in the evaluation of brain trauma.
Here we describe a software plug-in (a “cell” within the i2b2 “hive”) that allows clinical images to be
retrieved from a hospital’s clinical Picture Archiving and Communication System (PACS). The i2b2 plugin utilizes a widely available, open-source imaging informatics platform (XNAT [2], available at
http://www.xnat.org) in order to manage the images obtained from the PACS.
The images that become available through XNAT can be navigated through a user interface and
accessed by external applications via web services. Using an application programming interface (API),
research tools have been adapted to compute and extract novel information from the clinical images
using XNAT. These tools include Freesurfer (developed at the Athinoula A. Martinos Center for
Biomedical Imaging at the Massachusetts General Hospital), which allows brain MRIs to be
automatically segmented into anatomic regions, and 3DSlicer (developed by the National Alliance for
Medical Image Computing), which allows non-rigid image registration models to be computed from
conventional MR images. In addition, any medical imaging software that conforms to the DICOM
standard can be interfaced with the XNAT research image repository using the XNAT Gateway
application as a secure bridge
Evaluation:
Ordinarily, i2b2 works by copying all the relevant patient clinical data into a research database called
the Clinical Research Chart (CRC). This works well for the non-imaging data, but the approach is not
optimal for images due to the complexities inherent in the storage, indexing, and retrieval processes.
Furthermore, obtaining images from a clinical PACS can be a complex affair. Leadership at four of the
primary Harvard Medical School-affiliated academic healthcare centers (Brigham and Women’s
Hospital, Massachusetts General Hospital, Children’s Hospital Boston, and Beth Israel Deaconess
Medical Center) were appropriately concerned that retrieval of medical images from the clinical PACS
for research purposes might interfere with and hence compromise their clinical mission. Indeed, a tool
that might pull tens of thousands of images from the clinical PACS could cause a serious clinical
performance impediment. Therefore, a contract was put into place with the software that guaranteed
only a governed number of images would be pulled at a specific rate at specific (off-peak) times.
Furthermore, audit trails and security protocols were added to comply with the Health Insurance
Portability and Accountability Act (HIPAA), where standard DICOM interfaces fail to offer appropriate
assurances.
In addition to ensuring an appropriate interface to clinical systems, the software should be usable by
clinician scientists working in a hospital environment, to best promote its adoption and widespread
use. For most in our target user community, a streamlined workflow, expressed clearly, simply, and
with adequate feedback is required. Our software user interface (UI) is intuitive, guiding a user through
querying for studies by multiple criteria, managing query results, requesting clinical data and
monitoring retrieval status, previewing retrieved data, and transferring data to XNAT. The presentation
layer is designed to also permit flexible navigation among workflow steps, and to be compatible with
existing i2b2 UI paradigms.
Although the software does not need the i2b2 CRC to function, the system is much more effective
when patient queries utilizing the CRC can direct the download of imaging studies for specific
phenotypes. The imaging studies can then be specified and transferred to XNAT, where they are
analyzed and annotated. These annotations can be returned to the CRC for further querying and
analysis with the rest of the phenotypic and genotypic data.
Discussion:
By adhering to the DICOM standards for query and retrieval of images we believe our work will have
broad applicability at many research hospitals. The Harvard hospitals represent a microcosm in which
equipment from different PACS vendors, different policies on-line, near-line, and off-line storage, and
different radiology IT infrastructures can all be accommodated early in the design and implementation
process. The result should be a system that can be deployed with minimal changes at other
institutions.
The National Center for Research Resources (NCRR) funding enabled the development of this software
plug-in for i2b2 that allows clinical images to be retrieved from a hospital’s clinical Picture Archiving
and Communication System (PACS). The i2b2 plug-in utilizes the widely available, open-source imaging
informatics platform XNAT, supported by the Biomedical Informatics Research Network (BIRN), in
order to manage the images obtained from the PACS. Many National Institute of Heath ClinicalTranslational Science Center awarded entities use the i2b2 software suite to enable the repurposing of
healthcare data for clinical research, and we expect the ability to add clinical images from the hospital
PACS to broadly serve the community.
Conclusion:
The imaging plug-in for i2b2 will become available shortly as an open source addition to the software
platform. Development is supported by both NCRR (1 UL1 RR 025758-01) and the BIRN (1 U24 RR
025736-01) funding.
References:
[1] Murphy, S. N., G. Weber, et al. (2010). "Serving the enterprise and beyond with informatics for
integrating biology and the bedside (i2b2)." J Am Med Inform Assoc 17(2): 124-130.
[2] Marcus, D. S., T. R. Olsen, et al. (2007). "The Extensible Neuroimaging Archive Toolkit: an
informatics platform for managing, exploring, and sharing neuroimaging data."
Neuroinformatics 5(1): 11-34.
Primary Author:
First Name: Shawn
Middle Name/Initial: N
Last Name: Murphy
Degree: MD, PH.D.
Title: Medical Director, Partners Healthcare Reseach Computing
Department: Information Systems
Organization: Partners HealthCare
Address: One Constitution Center, 261
City: Charlestown
State/Province: MA
Zip Code: 02129
Phone: 617-643-5876
Fax: 617-643-5280
Email: snmurphy@partners.org
SIIM Member? Yes
Secondary Authors:
No
Author 2:
Author 3:
Author 4:
First: Daniel
First: Christopher
First: Timothy
Middle:
Middle: D.
Middle:
Last: Marcus
Last: Herrick
Last: Olsen
Degree: Ph.D.
Degree: MBA
Degree: MCS
Title: Director, Neuroinformatics Research Group
Title: Senior Technical Architect, Title: Lead Developer XNAT,
Information Systems
Neuroinformatics Research
Group
Organization: Partners
Organization: Washington
University, St. Louis, St. Louis,
MO
HealthCare System, Inc.,
Charlestown, MA
Organization: Washington
University St. Louis, St. Louis, MO
Author 5:
Author 6:
Author 7:
First: Randy
First: Steven
First: Yanbing
Middle: L.
Middle: D.
Middle:
Last: Gollub
Last: Pieper
Last: Wang
Degree: MD, Ph.D.
Degree: Ph.D.
Degree: MSc
Title: Associate Director of
Psychiatric Neuroimaging
Research and Director of the
Center for Biomedical Imaging
Title: CEO
Title: Software Developer,
Information Systems
Organization: Isomics, Inc.,
Cambridge, MA
Organization: Massachusetts
General Hospital, Boston, MA
Organization: Partners
HealthCare System, Inc.,
Charlestown, MA
Author 8:
Author 9:
Author 10:
First: Wendy
First: Darren
First: Jesse
Middle:
Middle:
Middle:
Last: Plesniak
Last: Sack
Last: Wei
Degree: Ph.D.
Degree: BS, RT®
Degree: MD
Title: Research Fellow
Title: Corporate Manager,
Enterprise Medical Imaging
Title: Director, Radiology
Informatics
Organization: Brigham and
Women’s Hospital, Boston, MA Organization: Partners
Organization: Beth Israel
HealthCare System, Inc., Boston, Deaconess Medical Center,
MA
Boston, MA
Author 11:
Author 12:
First: Katherine
First: Paul
Middle: P.
Middle:
Last: Andriole
Last: Lamonica
Degree: Ph.D.
Degree: BA
Title: Assoc Professor of
Radiology
Title: Sr. Pacs Systems Analyst,
Neuroimaging Center
Organization: Brigham and
Organization:
Women’s Hospital, Boston, MA Children’s Hospital, Boston, MA