This overview of use of and collaborations for the Cancer Data Standards Registry and Repository (caDSR) was initially prepared October 8, 2010 and is updated on an ongoing basis. The following topics are included:
NCI Cancer Data Standards Registry and Repository (caDSR) is a centralized resource for services and web-based information technology tools to meet NCI needs for documenting and sharing unambiguous descriptions of data. These resources are used to develop standards-based IT descriptions for trials and research data shared by NCI and the larger international biomedical community, so data can be effectively and programmatically exchanged, aggregated, analyzed, used and reused for secondary research.
Since 2000, caDSR tools and content have become the primary means by which cancer researchers developing databases and clinical trials data managers describe the data collected in clinical and research studies:
- By NCI and its partners in caBIG®, NIH, other federal agencies, and U.S. and international biomedical and organizations, academia and cancer researchers, such as for documenting and publishing human and computer interpretable representation of standard datasets for cancer registries (NAACCR and SEER) and those used in clinical trials (Cancer Centralized Clinical Database (C3D));
- To document, harmonize access, and publish data descriptions bound to standard value sets biomedical terminologies published by EVS, such as BRIDG 3.0.2;
- For basic, translational, and clinical research, clinical care, epidemiology, public health, administration, and public information.
caDSR data descriptions, software and services are freely available where possible, promoting broader adoption but hindering full tracking of use; restrictions on identifying and surveying users have a similar effect. Like EVS, the caDSR adopters and patterns of use are fairly clear.
NCI/caBIG® Informatics Infrastructure - caDSR
One measure of caDSR usage is the number of data elements recorded by different communities for the purpose of recording the detailed descriptions of data, and enabling sharing of these descriptions among data managers.
caDSR provides the foundational layer for representing the precise meaning and representation of data, and through linkage to EVS terminology and ISO/IEC 11179 standard, the ability to translate the meaning throughout the informatics infrastructure and across various research domains.
Up until 2010, all data, (30K+ "Released" data elements, 44k+ total), and information models (146 including multiple versions of some models) for all systems and applications in caBIG® had to have their meaning coded with controlled terminology, and recorded in caDSR, a repository used by various stakeholder communities. The high number of data element detailed descriptions as caDSR metadata represent the many and varied ways in which similar data are collected and stored in databases throughout the community.
By exposing these details in a common, structured way, caDSR enables automated harmonization when constructing data collection instruments and software heretofore not possible or feasible because the details were hidden away in application programs and paper documentation. In addition to the recording the specific data elements used in applications, the registry also records the reusable semantic components that, when reused, provide the basis for detecting semantic equivalence and potential for data aggregation and common processing. There are 12,383 classes of information defined in caDSR that are reused to create over 30k Common Data Elements (CDEs). For example, the property "Email Address" is reused by 43 or more CDEs to describe data such as "Organization Email Address", "Patient Email Address", "Clinical Trial Participant Email Address", "Investigator Email Address", etc. This fact is recorded in the caDSR so that developers can detect this similarity between disparate data and enable them to discover and potentially combine data from different sources.
Other applications, such as caIntegrator and caB2B, access caDSR via API at runtime to display descriptive information about data on data collection forms or to provide drop down lists for populating fields. Others import CDEs to use to customize data collection forms in local software applications such as a Clinical Data Management System (CDMS), C3D, LimeSurvey and others.
Also worth noting, an open source XML-based version of caDSR was developed by the University of Oxford Computing Lab to support the cancerGrid's clinical trails and other clinical and population studies in the UK. caDSR CDEs were extracted and formed the basis of new content development, reusing many of the NCI's CDEs in UK trials.
Lastly, this XML-based registry was taken by Ohio State University and enhanced to work with caGrid, and provides the basis for annotating services with caDSR CDEs during UML Model design, generating caGrid compatible services for deployment on the grid. OSU has also used this registry, named openMDR, to support its Clinical and Translational Science Award (CTSA) program and another consortium involved in collecting human studies, HSDB described below.
caDSR usage can be measured by the number of communities and applications that have registered their data elements and models in caDSR and are retrieving them via browser downloads and use of caDSR APIs. The caDSR software was designed by Oracle and the Census Bureau in the late 1990's as a centralized repository for ISO/IEC 11179 content, and due to its dependency on an Oracle database and an application server for the Admin Tool and database, is not well suited to installing small cancer centers who are resource-constrained or lack the Oracle skills to get the suite of products installed, customized and working. Instead, the NCI hosts content for those who want to use the infrastructure. Others who want their own registry now also have the cgMDR/openMDR option.
NCI caDSR CDE Browser
The CDE Browser is the starting point for exploring the details of the data elements described and registered in the caDSR. This tool supports browsing, searching, and exporting CDEs in XML or Excel format, within or across end user contexts as follows.
- Search for data elements by NCI Thesaurus Concept, Value Domain Permissible Value, Classifications or simple text searches in the names and definitions of the elements in caDSR.
- Using the Shopping Cart, create a customized set of CDEs for exporting or sharing with other applications in XML or Excel format.
- The CDE Browser publishes its DTD to export/download data elements in XML.
- Access the Tool at https://cdebrowser.nci.nih.gov/CDEBrowser/
USAGE: An average of 1,578 unique users visit the CDE Browser each year, with over 38,869 visits. CTEP has traditionally been the largest NCI user with over 11,500 released and draft CDEs as of October 2019. See Statistical Appendix for more details.
The NCI Cancer Clinical Research (CCR) center has 2,831 selected CDEs that are downloaded and imported into the Cancer Centralized Clinical Database (C3D), with 935,573 instances of CRFs built from these common descriptors and over 150M data points (responses) collected.
Chart showing usage of CDE Browser
Several collections of CDEs have been downloaded in Excel and EXML format from the CDE Browser and pre-packaged for downloading from NCI's Wiki. Pre-packaged downloads include all the CDEs that are in a "released" workflow status, those that are caBIG Standards and CDEs in NACCAR, SEER and BRIDG. Collections are posted to the caDSR Wiki on the caDSR Hosted Data Standards, Downloads, and Transformation Utilities page.
NCI caDSR Form Builder
Form Builder organizes CDEs in forms that replicate the content of Case Report Forms. These forms can include complex form behavior, such as skip patterns. Using the CDE Browser, you can search for CDEs, place them in a shopping cart, and from there insert CDEs as Questions on a Form. As you place the CDEs on the Form, the tool uses the stored metadata to provide default question text and value domain information automatically. If the value domain information is presented as an enumerated list of values, you can perform basic functions to organize the list.
You can place one or more CDEs into one or more sections on the Form, organize them into groups, and save them as Modules that can be copied from one Form to another.
Key capabilities of Form Builder include:
- Define skip patterns between questions based on question responses
- Define repeating groups
- Define default values for questions in repeating groups
- Publish a Form in the caBIG® Context's Form Catalog
- Subscribe to Sentinel Reports that are triggered by changes to CDEs on the Form
- Classify the Form in one or more caDSR Classification Schemes
- Download the Form to Excel
- View and print from a Printer Friendly version of the Form
- Edit, Save or Download the CDE Shopping Cart
- Forms shopping cart for storing collections of form to export to other systems
Access the Tool at https://formbuilder.nci.nih.gov.
USAGE: NCI has been tracking Form Builder usage since January of 2009, though the tool has been available since 2002. On an average day there are 6 curators using Form Builder, with 913 unique visitors over the past year (October 2009 to October 2010), viewing 14,546 pages. There are 407 Protocols defined in caDSR, into which many of the 2,613 CTEP Case Report Forms (CRFs) are grouped, and an additional 264 templates that describe minimum datasets to be collected for various types of cancer organized by disease and type of trial. NCI CTEP uses CDEs and CRFs for reporting trial results for CTEP Sponsored trials, of which there are currently 611 trials.
Commercial vendors Medidata (RAVE), Westat (for CTEP) and Eastern Oncology Centers Group (ECOG) retrieve CRFs using the caDSR API and import them to customize their data collection systems. Refer to the section titled NCI caDSR Database Server/Domain Model and Freestyle APIs for API usage.
ACRIN is involved in registering standard data elements for imaging and using them to create forms in Form Builder.
NCI caDSR Curation Tool
The CDE Curation Tool supports creation and editing of all the primary semantic descriptions for Data Elements used by the community in data repositories or application software. It is intended for use by Context Administrators. This tool's features facilitate the use of the CBIIT Enterprise Vocabulary Services (EVS) to create administered item names and definitions, helping to ensure ISO/IEC 11179 compliance and also use of caDSR naming conventions.
Public users can browse for CDEs, reusable ISO/IEC 11179 value domains (VD), data element concepts (DEC), Object Classes and Properties using identifiers, search strings, Classification Schemes or EVS concepts. The tool leverages the ISO 11179 metamodel to "get associated" items.
Access the Tool at https://cdecurate.nci.nih.gov.
USAGE: Since 2005, on an average day 22 unique users use the Curation Tool, with 4,558 unique visits to the site, and over 1.6M pages served. New CDEs can be created using this tool, the Admin Tool or the UML Model Loader. With the detail of caBIG® models no longer required to be registered in caDSR, that combined with curator harmonization activities designed to ensure and increase reuse of existing CDEs, the number of new CDEs year-over-year is trending down.
Chart showing usage CDE Curation Tool
NCI cgMDR Excel Addin, Template and Bulk Loader
The term “cgMDR” actually refers to a group of components designed to work together to help users get large lists of administered components bulk loaded into caDSR. The initial audience for these products was the National Marrow Donor Program (NMDP) but interest in its use has spread beyond that group.
cgMDR actually stands for the “CancerGrid Metadata Registry” and specifically refers to a downloadable localized database based on ISO/IEC 11179 where you can store and administer a personalized set of data elements and their components. This database and its interface were created by the CancerGrid team at Oxford University Computing Lab in the United Kingdom. The NCI GForge project archive can be found on this wiki.
The CBIIT cgMDR installation also includes a group of components, including two add-ins and two Excel 2007 spreadsheets that work in conjunction with cgMDR as well as with other data repositories. These additional features help provide a complete solution to assist you in creating a list of personalized data elements that can then be bulk loaded into caDSR without the need for creating a UML Model or by individual manual curation of each element.
USAGE: National Marrow Donor Program (NMDP), University of Michigan, ACRIN for batch loading CDEs and Value Domains into caDSR
NCI caDSR UML Model Browser
The UML Model Browser supports web browsing and searching data described by UML Models transformed and loaded in the caDSR repository via the UML Loader. This allows users to find administered items that are part of registered UML models for data services on caGrid. The UML Model Browser supports browsing, searching and exporting the classes, attributes and relationships between classes of a UML domain model. Within the framework of a UML Model CDEs are mapped to the UML attribute level. Search results display the Package Names, Attributes and Java primitive types. The CDEs used for semantic resolution are presented as links to the CDE Browser. CDEs in UML Models can be downloaded from the UML Model Browser.
Access the Tool at http://umlmodelbrowser.nci.nih.gov
USAGE: Since tracking began in March 2006, on an average day six visitors come to the site with 2,765 unique visitors through October 2010. 34,522 web pages were viewed in this four year period. There are 198 models loaded into caDSR describing the data for each of these applications using standard ISO/IEC 11179 descriptions. The descriptions of this data in caDSR ensure that the meaning of the data is unambiguously represented for both human and computer interpretation. The UML Model API is used by caGrid and caB2B to explore registered models programmatically. The caGrid Portal exposes the semantic metadata in its portal, which was accessed by 5,163 unique visitors between October 2009 and October 2010. According to ISO 8000, a Data Quality standard, registration of the details of these data and application models ensures the model owner both owns the data and meets guidelines for Data Quality.
NCI caDSR Sentinel Tool
The caDSR Sentinel Tool was first introduced in 2005 to allow users to create and manage Alert Definitions for the caDSR. Alert Definitions are a set of rules that are periodically evaluated against the caDSR. If the conditions in those rules are met, notification is sent to the user by email, with a hyperlink to a report that specifies the changes that have taken place. A script that kicks off this tool runs nightly, but the reports can also be run through the user interface.
The caDSR Sentinel Tool provides the capability to:
- Monitor all changes to Administered Items including Data Elements, Data Element Concepts and Value Domains and Case Report Forms
- Filter report content by Context, Specific forms or templates, Classification Scheme, Class Scheme Item, Creator and Modifier
- Trigger report generation using Workflow Status, Registration Status and Version
- Set reports to automatically be generated daily, weekly, monthly or on demand
- Create a report distribution list which may optionally include a process URL to send the report in XML format to software for evaluation
Access the Tool at https://cadsrsentinel.nci.nih.gov
USAGE: On an average day, 3 users visit the caDSR Sentinel tool site with over 264 unique visitors between October 2009 and October 2010, and 8,354 page views in that time.
NCI caDSR Database Server, Domain Model and Freestyle APIs
All caDSR content is available through various application programming and web service interfaces.
caDSR API allows users to access caDSR content by using a web browser to navigate the caDSR domain model and returns results in HTML or XML. A caDSR Java API provides a set of methods that can be used to retrieve content as XML documents. According to NCI statistics from Wusage, from October 2009 to October 2010, 628 unique visitors came to the site and accessed over 25M documents. The number of pages is slightly inflated sue to the design of the caDSR Domain Model, as multiple pages between 6-20 pages must be served in order to get one logical document. Our estimate is that over 1.2M logical documents were retrieved.
Access the HTTP caDSR Domain Class Browser at http://cadsrapi.nci.nih.gov/cadsrapi40/
Freestyle API is another caDSR interface that provides access to content both via a web browser and application programming interface. From October 2009 to October 2010, 328 unique visitors came to the Freestyle API site and accessed over 9.6k documents. The Freestyle API uses the caDSR Domain API to simplify object retrieval.
Access the Tool at http://freestyle.nci.nih.gov
USAGE: The caDSR API usage information in the statistics appendix lists the top URLs that access the caDSR API. AGNIS uses the API to access Forms.
Illustration listing AGNIS information on using caDSR
From: 'Accessing AGNIS Metadata using the CaDSR'
NCI caDSR Semantic Integration Workbench (SIW) and UML Loader
The Semantic Integration Workbench (SIW) a tool that assists users in adding consistent metadata to a UML model represented as an XMI file, or verifying consistency with existing caDSR content by tagging a domain model with matching concepts from the NCI Thesaurus, or attaching existing caDSR CDEs or Value Domains to attributes in the model. These annotations ensure reuse of CDEs or other metadata elements that have been previously recorded in caDSR. The UML Loader transforms the file and decomposes it into ISO/IEC 11179 descriptive metadata.
Access and use the Tool at http://cadsrsiw.nci.nih.gov (Java WebStart application)
USAGE: As of July 2011, there are 7 new models loaded in 2011, 2 in the queue to be loaded and 2 on hold. 198 caBIG® and NCI UML models representing all versions of these applications' data have been transformed into ISO/IEC 11179 descriptive metadata in caDSR. These models use 26,360 CDEs. Many of the CDEs are reused across models. Thus, for example, if the CDEs were evenly distributed across all models, 133 CDEs would be used per model. However when searching for CDEs in each model, one finds that BRIDG v2.1 has 1,669 CDEs, caAERS v1.1 (Adverse Events Reporting System) has 463, caBIO 4.3 has 465, caTissue Suite 2.0 has 997, caNanoLab has 524 and C3PR v2.0 (Patient Registry) has 180. This includes several versions of BRIDG, caAERS and C3PR. See the Appendix for complete list of models with CDEs in caDSR.
NCI caCORE Training
The caDSR is part of an infrastructure that supports the semantic description of data. While it is based on ISO/IEC standards for internal representation of data elements, and makes it possible to interpret and compare meaning of data across different repositories and domains, it does not guarantee that the intended end users will immediately understand how to use this standard to describe their data, so training has been developed.
The caCORE Training wiki includes information about each course and self-paced workbooks, as well as a curriculum based on end user roles. This training was established in 2007 and has been utilized in training over 1400 caDSR end users. The primary courses are:
- Course 1000: Introduction to caCORE and caDSR
- Course 1010: Introduction to Metadata in the caDSR
- Course 1020: Using the CDE Browser
- Course 1025: Using the UML Model Browser (Created June 2010)
- Course 1030: Introduction to EVS
- Course 1040: Creating Well-formed Metadata and Metadata Business Rules
- Course 1045: Introduction to the Curation Tool (Created June 2010)
The CDE Browser is the primary public access for caDSR content. The course on using the CDE Browser is for advanced users to teach them how to leverage the advanced search and compare features. It has been completed by over 290 end users. According to web usage statistics, the CDE Browser site an average of 500 unique visitors each month.
Chart showing usage of caCORE Training
USAGE: in addition to direct enrollments in caCORE and caDSR training, caDSR training materials are used in the AGNIS training materials.
Illustration showing AGNIS message promoting usage of caDSR
From: 'Accessing AGNIS Metadata using the CaDSR'
caDSR Usage in NCI and caBIG®
National Cancer Institute Cooperative Groups
Project Sponsor: Dr. Jeff Abrams
Project Manager: —
The NCI Cancer Therapy Evaluation Program (CTEP) mandated that variables used in NCI-sponsored trials be registered in the caDSR several years ago. Using a series of unified workflow processes, all new and amended trials go through a review process that verifies the elements are added to the registry both as single elements, and in forms to represent a full clinical trial. There are more than 9600 data elements arranged programmatically in more than 2500 clinical trial-centric forms in the registry to represent the following Cooperative Groups:
- American College of Surgeons Oncology Group (ACOSOG)
- Cancer and Leukemia Group B (CALGB)
- Eastern Cooperative Oncology Group (ECOG)
- Gynecologic Oncology Group (GOG)
- European Organisation for Research and Treatment of Cancer (EORTC)
- National Cancer Institute of Canada (NCIC)
- National Surgical Adjuvant Breast and Bowel Project (NSABP)
- Radiation Therapy Oncology Group (RTOG)
- Southwest Oncology Group (SWOG)
An example of the content housed for each Cooperative Group is illustrated by ECOG content. ECOG is one of the largest clinical cancer research organizations in the United States, and conducts clinical trials in all types of adult cancers. Currently, ECOG has more than 90 active clinical trials in all types of adult malignancies. Annual accrual is 6,000 patients, with more than 20,000 patients in follow-up. These clinical trials use the caDSR to describe case report forms for sharing programmatically and human-understandable detailed descriptions of the data to be collected in clinical trials. This information is pulled electronically to use in customizing local software as well as publicly available for browsing or downloading descriptions via the caDSR CDE Browser.
Examples of ECOG clinical trials that have changed or improved cancer treatment methods include:
- A study that establishes an effective chemotherapy regimen in acute promyelocytic leukemia
- A study that demonstrated that a less-toxic regimen had similar results to the standard treatment of metastatic nonsmall cell lung cancer
- A study that demonstrated that an intensive bone marrow transplant regimen for patients with metastatic breast cancer who responded to initial standard chemotherapy did not improve time to disease progression or lifetime survival
- A study that establishes an effective treatment program involving chemotherapy and radiation therapy for early lung cancer
Cancer Centralized Clinical Database (C3D)
Project Sponsor: —
Project Manager: Christo Andonyadis (CBIIT), Dianne Reeves (CBIIT)
C3D is a clinical trials management system, for which Case Report Forms (CRFs) based on data elements (CDEs) are used to collect data. The CRFs and CDEs are described in caDSR and represented as computable metadata that is shared electronically to support the clinical center information systems required to collect data consistently and accurately across trials. This increases data accuracy, reduces clinician training and increases the possibility of aggregating data across studies.
C3D has been in operation since 2002, first in the NCI intramural research program (Center for Cancer Research (CCR)) and then adopted broadly across a growing number of groups. There are currently more than 315 trials from C3D represented in caDSR; reuse of content across trials and groups supports the need for interoperability, and ultimately the need to aggregate clinical trial data across studies and groups for safety and product reporting. In addition to CCR, C3D provides CDMS support for over 20 cancer centers (see Appendix B), for which trial CRFs are described by the same CDEs enabling the potential for data aggregation across independent groups.
USAGE: These statistics from August 2010 represent usage of caDSR content in the C3D application used by CCR to support these cancer centers.
CDEs used in C3D studies: 3,491
Individual data points (Responses): 150,341,554
Lab CRFs from caDSR download CDEs: 604,128
Data points from batch loaded CRFs: 124,480,823
These figures include the NCCCP RQRS and Breast Cancer Datamart (BCDM). For comparison, for October, 2010 the figures show 935,573 CRFs, an increase of 25,502.
NCI Cancer Therapy Evaluation Program (CTEP)
Project Sponsor: Dr. Jeff Abrams
Project Manager: —
The NCI Cancer Therapy Evaluation Program coordinates the largest, publicly funded oncology clinical trials organization in the world. With over 900 active trials enrolling annually 30,000 study participants, nearly 400 grants and cooperative agreements, and about 100 investigational new drugs (INDs), CTEP's staff of physicians, scientists, pharmacists, nurses and regulatory specialists, assisted by government contractors, works diligently to assure the safe, efficient and ethical conduct of this complex research enterprise. CTEP-sponsored research spans phase 1-3 trials in all cancers and treatment modalities – chemotherapy, immunotherapy, radiation and surgery. All CTEP-reported trials are represented by data elements housed in the caDSR, with Case Report Forms to created to represent #National Cancer Institute Cooperative Groups trials and other trials.
USAGE: CTEP currently has 2,613 CRFs in caDSR with over 9,607 released and draft CDEs. See the #NCI Form Builder description above for CTEP usage details.
NCI Division of Cancer Prevention (DCP)*
Project Sponsor: Dr. Leslie Ford
Project Manager: —
The Division of Cancer Prevention (DCP) is the primary unit of the National Cancer Institute devoted to cancer prevention research. DCP provides funding and administrative support to clinical and laboratory researchers, multidisciplinary teams, and collaborative, research-based networks. DCP also requires that all trials sponsored by the organization use variables that are included in the caDSR. Over the past decade nearly 1,300 data elements with 58 protocol forms and templates have been developed within the caDSR. DCP has also authored a significant number of the data standards vetted and approved by the caBIG® community over the past few years, including standards for the capture and reporting of:
- Address components
- Person name
- Person height
- Person weight
These data standards are widely reused across the NCI clinical research enterprise.
USAGE: DCP has 1,317 CDEs in caDSR. These are organized into categories such as Adverse Events Reporting, Protocol Deviation Notification, and types of conditions such as the 201 CDEs to collect for Barretts Esophagus, or the 152 CDEs for Bladder Cancer.
By using the caDSR and ISO/IEC 11179 to harmonize the descriptions of data to be collected in these studies, and designating the same CDEs to collect on different forms for the same data across these conditions, caDSR enables DCP to simplify data collection and training, as well as be assured of being able to aggregate data and perform comparative analysis across studies.
NCI Center for Cancer Research (CCR)
Project Sponsor: Caryn Steakley, Elizabeth Ness, Alison Wise
Project Manager: —
The Center for Cancer Research is the NCI's intramural research program. The Center (CCR) is home to more than 250 scientists and clinicians working in intramural research at NCI. CCR is organized into over 50 branches and laboratories, each one grouping scientists with complementary interests. CCR investigators are basic, clinical, and translational scientists who work together to advance our knowledge of cancer and AIDS and to develop new therapies against these diseases. CCR investigators collaborate with scientists at the more than 20 other Institutes and Centers of the National Institutes of Health (NIH), as well as with extramural scientists in academia and industry (as further discussed inthe Appendix). CCR uses the NCI's C3D for study and trial data collection.
USAGE: As all C3D studies are described using CDEs from caDSR, so are all CCR studies. The CDEs are available electronically as an Excel download from a web browser-based UI, the CDE Browser, and then used to create forms in C3D. See Cancer Centralized Clinical Database (C3D) for details.
Specialized Programs of Research Excellence (SPOREs)
Project Sponsor: —
Project Manager: —
Specialized Programs of Research Excellence (SPOREs) are funded through specialized center grants (P50s) that promote interdisciplinary research and move basic research findings from the laboratory to clinical settings, involving both cancer patients and populations at risk of cancer.
The outcome of interdisciplinary research is a bidirectional approach to translational research, moving laboratory discoveries to clinical settings or clinical observations to the laboratory environment. Laboratory and clinical scientists share the common goal of bringing novel ideas to clinical care settings that have the potential to reduce cancer incidence and mortality as well as improve survival and the quality of life. Since 2003 SPOREs studies predominantly for head and neck and lung cancer have been added to the caDSR. There are studies and CRFs for two Iloprost trials in the registry, as well as studies for the University of Colorado (GO studies for advanced lung cancer; Principal Investigator is Dr. Paul Bunn).
USAGE: As of October 2010, the SPORE program has registered 29 CRFs in caDSR from 704 CDEs. Of special note is the fact that the University of Colorado now has refined its protocol templates to the point that builds of new studies require essentially no creation of new elements; existing content can be rapidly used to build forms, increasing the interoperability and ability to aggregate collected data.
The Lung SPORE at Emory University Winship Cancer Institute has designated 91 CDEs to collect for Lung Cancer Patients.
The National Marrow Donor Program® (NMDP) and our Be The Match FoundationSM Project
Sponsor: Dr. Roy Jones, MD Anderson; Dr. Doug Rizzo, Medical College of Wisconsin
Project Manager: Robinette Alley
The National Marrow Donor Program is a member of the Center for International Blood and Marrow Transplant Research. NMDP has located in Minneapolis and Milwaukee, with teams of data element curators who have been trained by preceptor NCI CBIIT staff. NMDP curators are using the caDSR Form Builder and Microsoft Excel add-in along with manual curation tools to register all the data collection forms and make them available via the caDSR web-based tools and programming interfaces. This organization began with a requirement to add nearly 100 CRFs to the registry; at the present time nearly all of the high volume forms have been completed, with CRFs designed for much more rarely seen malignancies slated for future curation.
Currently this group is evaluating the applicability of caAERS, an application designed by NCI CBIIT to document and report adverse events, for transplant research. They are also actively exploring the harmonization of their variables with those identified in the NCI Case Report Form Harmonization and Standardization initiative. The resulting set of variables for use in Transplant reporting will be vetted and evaluated as standards for the larger Transplant community. These activities have the strong support of Dr. Douglas Rizzo in Milwaukee, and Dr. Roy Jones at the MD Anderson Cancer Center. Dr. Jones is also a member of the caBIG® CTMS Steering Committee, and remains a strong proponent of the development of core data standards for clinical trial reporting.
USAGE: This group has identified for reuse or curated 1,535 of a planned 14,000 CDEs selected to use in their hematopoietic stem cell transplant (HSCT) program. NMDP utilizes AGNIS, a network based electronic forms system based on FormsNet to collect donor information. To harmomize the data across these 98 forms and render the CDEs programmatically accessible via caDSR APIs, NMDP has chosen caDSR to record their reusable data elements in ISO/IEC 11179 format, allowing them to develop sharable forms in caDSR Form Builder to build the collections of CDEs to match their existing varied and unharmonized FormsNet forms. The use of caDSR and careful planning by the NMDP curation team is allowing NMDP to harmonize fields across all donor program data collection forms and efforts (caBIG® 2010 Annual Meeting Poster 82).
Program Manager: Juli Klemm
caIntegrator is a web-based software package that allows researchers to set up custom, caBIG®-compatible web portals to conduct integrative research, without requiring programming experience. These portals bring together heterogeneous clinical, microarray and medical imaging data to enrich multidisciplinary research.
Using caIntegrator, researchers can execute, save and share queries to identify and collect many types of data, combining clinical information with genetic and genomic data to enable multidimensional analysis. caIntegrator uses caGrid analytical services such as GenePattern and BioConductor to perform analysis on the integrated study data, including clinical survival data.
caIntegrator leverages the Cancer Data Standards Registry and Repository (caDSR) to map experimental data to well-defined datatypes and utilizes caGrid and Java client APIs to access data from caBIG® applications such as caArray and the National Biomedical Imaging Archive (NBIA). caIntegrator is also integrated with caBIO to perform queries on genes and pathways. NCI hosts an online version of the caIntegrator application.
Usage: The caIntegrator has 362 CDEs decribed in caDSR, enabling users of the application to better understand the kind of data that is available in caIntegrator without having to look at the caIntegrator code. For example, the UML Model Browser can be used to see that there are 70 classes in the database, and that the field for "Class Comparison Analysis Adjustment Type MultipleComparisonAdjustmentType" with public ID and version 2529267v1.0, with only 3 possible values in this field: "fdr", "fwer" and "none". The caDSR metadata provides the meaning of this data field, the datatype, representation and the meaning of each of the three values.
This information is then available to be accessed programmatically when designing the application, or at run time for the end user. An example of the details that are available for every CDE can be seen in these search results.
Project Manager: Rakesh, Washington State University
cancer Bench-to-Beside (caB2B) is an open-source query tool that permits translational research scientists to search and combine data from virtually any caGrid data service.
The caB2B suite is composed of three core components: the Web Application, the Client Application and the Administrative Module. The caB2B Web Application provides query templates that leverage semantic metadata from caDSR to allow easy search and retrieval of microarray data (from caArray), imaging data (from the National Biomedical Imaging Archive (NBIA)), specimen data (from caTissue) and nanoparticle data (from caNanoLab) across the grid. Searches can be performed on selected locations using either form-based or keyword searches and data can be exported in the CSV format.
Using model metadata stored in caDSR to detect relationships between data classes, the caB2B Client Application is a thick Java application that enables advanced end users to create and execute queries across caGrid data services. The query component consists of a diagrammatic view that allows the user to create a directed acyclic graph of the query that is to be executed and also helps the user to connect two or more classes to be searched. Users can save the query and data returned may be saved in the form of a 'virtual experiment.' These data can be visualized using various graphical components.
The Administrative Module provides a graphical user interface for customizing a particular instance of caB2B. For example, an administrator can select models and service instances that may be queried, curate paths between classes in models, create user-friendly categories using two or more classes from a model, and create inter-model joins.
USAGE: caB2B utilizes caDSR CDEs and UML Model association information to support semantic queries and data aggregation across semantically and syntactically similar data on the grid.
The Cancer Genome Atlas (TCGA)
Letter: Dr. Kenna Shaw, Director, TCGA
TCGA teams work iteratively with NCI curator staff to identify disease-centric data element collections that use existing standards and extend standard content with TCGA domain expert definitions. Currently the TCGA project uses more than 1000 CDEs registered in the caDSR. This collaboration with the NCI caDSR team began in the spring of 2010, and continues to represent a major program commitment of caDSR resources to the TCGA program. The result has been a rich collection of elements that is heavily reused by other members of the caDSR extended community. Vocabulary needs are managed rapidly to allow the creation of new elements to proceed on an expedited timeline.
USAGE: The Cancer Genome Atlas has 636 CDEs registered in caDSR.
Imaging (In Vivo Imaging Workgroup, DICOM, ACRIN)
Project Sponsor: —
Project Manger: —
Imaging content is central to the oncology community and research. To this end, and based on early work with Cancer Imaging Program (CIP), the addition of data elements and models to the caDSR began in 2004. Early efforts to move legacy databases and applications quickly transitioned into the creation of new applications and intersections with the Imaging standards community. Content to reflect the Digital Imaging and Communications in Medicine or DICOM standard has been added to the caDSR using manual caCORE tools, and working collaboratively and iteratively with the imaging community.
A variety of applications as well as imaging variables used in clinical trials has raised the full set of imaging content in the registry to nearly 1400 variables. These variables are all annotated and referenced in a manner to maintain their links to the underlying DICOM standard. This task is facilitated by members of the American College of Radiology Imaging Network or ACRIN, who work within caBIG® and with curators to reinforce the need for registration of imaging standards.
C3D is one area in which the benefits of working with Imaging colleagues is most apparent. To the clinical research content required by a clinical trial we have added imaging content in caDSR, required by the NCI sponsors, resulting in a set of content that reuses clinical trials elements while extending out the set with most commonly ACRIN standards. This approach allows the rapid build of imaging clinical trials reporting their data through NCI mechanisms.
OpenMDR – Ohio State University
Poster: Rakesh Dhaval, Claixto Melean, David Ervin, Philip Payne, PhD
USAGE: OpenMDR adopted ISO/IEC 11179 as the basis for describing data elements in caGrid services. openMDR allows model and application owners to search for reusable content from caDSR using Enterprise Architect and the MDR generator to create caGrid services with semantics based on CDEs (caBIG® 2010 Annual Meeting Poster 112).
DECAMAP - Data Element Curation & Mapping Platform
Project Sponsor: Dr. Chris Chute, Dr. Robert Freimuth, Dr. Jyotishman Pathak
"DECAMAP is a tool that allows PGRN researchers to harmonize their local data dictionaries to existing metadata and terminology standards such as the caDSR (Cancer Data Standards Registry and Repository), NCIT (NCI Thesaurus) and SNOMED-CT (Systematized Nomenclature of Medicine-Clinical Terms). This tool can be used to search/browse metadata related to different studies, create new study dictionaries and its related metadata, export metadata in excel format and assist investigators with dbGAP submission. DECAMAP is being created by the PHONT group in the PGRN network with the first version to be released July 2011. The overarching aim of our PGRN Ontology Network Resource (PHONT) is to enable codification of standardized phenotype definitions and relationships, in coordination with other established government-funded efforts. Because the standardized representation of phenotypes depends on standards, our first two aims focus on making vocabularies and ontologies (computationally formal vocabularies) about clinical and physiologic concepts accessible and usable to the PGRN community."
The focus of this research is on matching to the ISO 11179 DEC and Value Domain Permissible Values in caDSR to compare/string match values across different resources. The UI allows item by item searching. The user creates their own data dictionary and then uses that dictionary content to search in caDSR, NCIt, SNOMED and SDTM. An expert curator reviews the submitted data dictionary entries to find the best match. Based on sample studies, 69% matched to caDSR, 49% mapped to NCIt, lesser to SNOMED and SDTM.
eMerge - Data Element Curation & Mapping Platform
Project Sponsor: Dr. Chris Chute, http://www.ncbi.nlm.nih.gov/pubmed?term=%22Pathak%20J%22%5BAuthor%5D Dr. Dan Masys, Wang J , Kashyap S , Basford M , Li R , Dr. Jyotishman Pathak
Abstract: Systematic study of clinical phenotypes is important for a better understanding of the genetic basis of human diseases and more effective gene-based disease management. A key aspect in facilitating such studies requires standardized representation of the phenotype data using common data elements (CDEs) and controlled biomedical vocabularies. In this study, the authors analyzed how a limited subset of phenotypic data is amenable to common definition and standardized collection, as well as how their adoption in large-scale epidemiological and genome-wide studies can significantly facilitate cross-study analysis.
Conclusions: This study emphasizes the requirement for standardized representation of clinical research data using existing metadata and terminology resources and provides simple techniques and software for data element mapping using experiences from the eMERGE Network.
J Am Med Inform Assoc. 2011 Jul-Aug;18(4):376-86. Epub 2011 May 19.
caDSR Usage in NIH
National Heart Lung and Blood Institute (NHLBI)
Project Sponsor: Dr. Jenny Larson, Dr. Douglas Rizzo
Project Manager: —
The National Heart, Lung and Blood Institute (NHLBI) began partnering with the NCI to add data elements to the caDSR in 2005 to reflect the variables used for a Family Blood Pressure Program (FBPP) study.
Working with NCI CBIIT curators, NHLBI achieved reuse of some existing elements, with registration of a significant amount of cardiology-specific content over the ensuing year. This is also the context into which the NMDP elements are being added.
USAGE: There are currently more than 4000 data elements registered in caDSR; a number of these are being used by oncology groups, demonstrating the ability to reuse variables across the research community regardless of the domain of expertise.
National Institute of Dental and Craniofacial Research (NIDCR)
Project Manager: Alice Birnbaum, Director of Biostatistics, Axio Research, LLC
The National Institute of Dental and Craniofacial Research (NIDCR) was one of the first non-cancer adopters of the caDSR as an attractive place for the registration of data elements. Since early 2005, a number of curation teams across the US have been trained by NCI CBIIT staff in the best practices for data element creation and maintenance.
USAGE: To date, more than 2300 data elements have been registered in the caDSR, and 81 Case REport forms, notably with the addition of valuable and unique dental-specific content. This organization is one of the groups driving the development of batch curation and registration functionality for the caDSR; the curators created a set of content that has been loaded using the caDSR Bulk Loader. We continue to capture the need for enhancements and emerging business requirements.
National Institute of Child Health and Development (NICHD)
Project Sponsor: Dr. Steven Hirschfeld
Project Manager: —
The National Institute of Child Health and Development is one of the newest non-oncology users of the caDSR. Beginning in November,2009, a model to represent a Newborn Assessment was annotated and loaded into the caDSR. The resulting 107 data elements were reviewed and released for widespread community review.
USAGE: Currently a small team is being trained by NCI CBIIT staff to act as curators for additional content that is being planned for addition to the caDSR, as of 2011 114 CDEs have been created for use in this research domain, 7 have been reused from other Contexts.
National Institute of Neurological Diseases and Stroke (NINDS)
Project Sponsor: Joanne Odenkirchen, NINDS
Project Manager: Staci Grinnon, Yun Lu, KAI Research
The National Institute of Neurological Diseases and Stroke (NINDS) has developed a set of data elements for use in clinical trials over the past several years. In the past year members of an NINDS CRO, KAI Research, recommended to NINDS that their core data elements may be candidates for addition to the caDSR. Based on their evaluation of the caDSR, a comparison analysis between NINDS content and NCI content was conducted. The result was in excess of 80% of core research variables across the two groups, again demonstrating the potential reuse of content regardless of the area of expertise. A small team of curators has been trained by NCI CBIIT, and curation of an identified core set of NINDS variables was slated to begin in August, 2010. After the core set is loaded, next steps will be determined.
USAGE: NINDS has reused and created 310+ CDEs found on 21 NINDS CRFs (123 reused CDEs) used to collect for neurological clinical research data.
caDSR External Adoption
NCI caDSR provides open access to all its content via publicly accessible browserw and representatives have spoken at a number of external conferences on the caDSR approach to data management and services as a means for sharing data. This has drawn the attention of a number of external groups and international parties looking to find data descriptions for clinical trial and research datasets as highlighted below.
Biopathology Center, National's Childrens Hosptial
Project Sponsor: Dave Billiter
Cooperative Group Specimen Banks are using caDSR content via caDSR API in their Group Banking Committee tooling to inspect and harmonize across different specimen banks. They use the caDSR API and the UML Model Browser and CDE Browser to identify CDEs that need to be harmonized.
The Toolkit provides standard measures related to complex diseases, phenotypic traits and environmental exposures. Use of PhenX measures facilitates combining data from a variety of studies, and makes it easy for investigators to expand a study design beyond the primary research focus. The data elements use to create these standard measures are based on 660 CDEs from caDSR.
Toolkit Users can:
- Search or Browse the Toolkit to review and select PhenX measures
- Request a Data Dictionary and Data Collection Worksheet for selected PhenX measures
Project Sponsor: The Antibody Society
Guidelines and Information About Therapy Experiments (GIATE) is a data standard for recording data about therapy experiments. It is being promoted by the Antibody Society, as a data standard for antibody therapy experiments but it has become apparent that the scope of materials used in antibodies as therapeutic agents, means that the data standard can equally be used to describe a wide range of therapeutics.
GIATE categorises the information about therapy experiments into three classes, that is, information about the target, therapy agent and models used to test the therapy. Nested within these classes are additional classes describing different aspects and properties of the parent class. Therefore, a logical representation of GIATE would be a tree, with three main branches, which stems into smaller branches and leaves. The contents of the GIATE tree are downloaded from the caDSR, each concept contains a reference to the caDSR data element. http://www.genscript.com/giate-viewer/
American Heart Association (AHA) and American College of Cardiology Foundation (ACCF)
Project Sponsor: AHA, Duke
"The American College of Cardiology Foundation (ACCF) and the American Heart Association (AHA) support their members' goalto improve the prevention and care of cardiovascular diseases through professional education, research, and development of guidelines and standards and by fostering policy that supports optimal patient outcomes. The ACCF and AHA recognize the importance of the use of clinical data standards for patient management, assessment of outcomes, and conduct of research, and the importance of defining the processes and outcomes of clinical care, whether in randomized trials, observational studies, registries, or quality-improvement initiatives." The AHA and ACCF have chosen the caDSR define and disseminate clinical data standard s- sets of standardized data elements and corresponding definitions - to collect data relevant to cardiovascular conditions. http://circ.ahajournals.org/content/124/1/103.extract
Winthrop P. Rockefeller Cancer Institute at the University of Arkansas for Medical Sciences
Project Sponsor: Umit Topaloglu, PhD
The institute has decided to implement standard operating procedures (SOPs) requiring that all investigator led trials use CDEs from caDSR to conduct their research. The first project created 12 CRFs in the program's Clinical Data Management System (CDMS), using 72 CDEs, 38 of which were already in caDSR (50% reused). They created the remaining 33 CDEs using the Curation Tool. They have installed openMDR (OSU caDSR ISO/IEC 11179 compatible registry that can pull CDEs from caDSR for a local repository). The program plans to use this for ongoing collaborative research studies (caBIG® 2010 annual meeting Poster 80).
Union of Light Ion Centers in Europe (ULICE)
Project Sponsor: EU, WP Lead Dr. Bleddyn Jones, Manchester University, Oxford Computing Laboratory, Oxford UK
Informatics PIs: Professor Jim Davies, DPhil. Steve Harris, PhD
Project: WP7 Common database and grid infrastructures for improving and catalysing access to RI for the broad European community
ULICE is a 4-year project set up by 20 leading European research organizations, including 2 leading European industrial partners (Siemens and IBA), to respond to the need for greater access to hadron-therapy facilities for particle therapy research. Project coordinator is the Italian Research Infrastructure Facility CNAO (Milan).
WP7 is concerned in part with the compilation of a sufficient body of evidence to support future evidence-based evaluations of CPT – identifying new, preferred treatment protocols for specific cancers and ensuring that patients receive the best treatment.
By establishing a registry using key shared terminology and caDSR-type data elements to collect data, WP7 will collect information on every patient treated. However, such a registry will deliver the evidence only if the data gathered is comparable or interoperable. It is necessary to be able to reliably combine data recorded in different contexts, in different clinics, and in different health care systems.
Given the variety of (changing) practices and situations, it is unrealistic to expect that a single quality standard might be imposed for data collection (Bray and Parkin 2009) (Parkin and Bray 2009), so the registry has been set up to allow all data collected in the registry to be properly described, with detailed explanations of protocol for each clinical observation, and detailed definitions for each of the possible values recorded. These detailed explanations of the data will also be collected using caDSR-style common data elements. The ULICE project is in Year 1 of a 48 month project; it is aligning data elements, wherever possible, with caDSR. Currently 50% of the data elements to describe ULICE studies are caDSR based CDEs, as are 25% of the Baseline CDEs and 70% of the Followup CDEs.
Winthrop P. Rockefeller Cancer Institute, Cancer Control Core at the University of Arkansas for Medical Sciences; Breast Mammography
Project Sponsor: Umit Topaloglu, PhD
USAGE: The cancer institute is using caDSR CDEs and LimeSurvey, an open source tool, to recreate questionnaires for breast cancer mammography. The center recreates or reuses the survey CDEs in caDSR, then imports the CDEs into LimeSurvey. This ensures that the data are collected with high quality and common, standard data descriptions across studies. As a result of doing this work, the existing survey data was analyzed, corrected and standardized and can be reused for other studies (caBIG® 2010 Annual Meeting poster 81).
Oncology Patient Enrolment Network (OPEN) - Sponsor-CTEP/NCI
Project Manager: Mike Montello (Project officer)
Bioinformatics: Ravi Rajaram
The OPEN System has adopted the caCORE Application Programming Interface (API) and the Form Builder application, in addition to integrating and adapting the Common Security Model (CSM) components, as part of its architecture. OPEN uses the caDSR API to download the case report form (CRF) metadata from the caDSR. In addition, OPEN has adapted the CSM in such a way that the authorization feature of the CSM is used while authenticating the user against the National Cancer Institute – Cancer Therapy Evaluation Program (NCI CTEP) Identity and Access Management (IAM) system. The CSM user provisioning tool (UPT) is being used to provide instance and attribute level security.
Duke Translational Medicine Institute
Bioinformatics Sponsors: Dr. Jeffrey Ferranti, Patricia Gunter
M. Nahm, PhD, Associate Director, DTMI Biomedical Informatics
H. Shang, Director of Business Information Services, Duke Health Technology Solutions
Rob Califf, M.D., Vice Chancellor of Clinical Research, Duke University Health Systems
Asif Ahmad, CIO, Duke Health Technology Solutions
Dwight Smith, Director, Information Technology Application Development, Duke Health Technology Solutions
Dr. Ferranti at DTMI recognized a business problem in the lack of standardized clinical terminology across the enterprise, preventing meaningful sharing and reuse of the data – semantic interoperability.
Following NCI's lead for using controlled terminology as the basis for metadata descriptions (caDSR), the DIMI solution is to build a metadata registry that would become a foundation for future data management and data quality efforts. Currently Duke is considering caDSR open source as a possible solution versus building anew.
Silos of operational databases and data entry applications exist in Patient registration, Patient billing, Clinical systems – surgical, emergency, labs, and Disease registries in over 150 clinics with over 3 million HL7 transactions per day. To promote consistent semantics, DTMI would leverage controlled terminologies – LOINC, SNOMED-CT, ICD-9, CPT, and others, and the ISO/IEC 11179 standard for metadata registries and HL7 messaging standards. Like many institutions, DTMI data was collected on forms. The current repository collected these forms as a blob of data. DTMI used its 11179 registry to create structured descriptions of the data on these forms and used the metadata descriptions as a means by which to extract and transform the data into a data warehouse. The major difference between the NCI's caDSR and Duke's metadata registry is the terminology basis. NCI has used the NCI Thesaurus as the basis for describing data; DTMI is using concepts from the HL7 RIM as the basis for the metadata descriptions.
Taiwan Cancer Registry
ISO/IEC 11179 has been used to describe the cancer data set for the Taiwanese cancer registry. A paper entitled "Annotating Taiwan Cancer Registry to caDSR for International Interoperability" in "Future Wireless Networks and Information Systems", 2012 explains that "It is very difficult to exchange and integrate the data among different cancer institutions, in order to discover useful biomedical information for the cancer community. We developed a cancer Biomedical Informatics GridTM (caBIGTM) Silver level compliant cancer registry database system, called the gridTCR (grid of the Taiwan Cancer Registry), which integrates the demographic data, clinical history, pathology data, and clinical outcome data including treatment, recurrence and vital status from cancer registry databases in Taiwan. In this manuscript, we will be developing the common data elements using vocabulary standards, ontology and semantic modeling methodology. The Taiwan Cancer Common Data Element Project (TCCDEP) developed 40 data elements to annotate the cancer registry data collected. The aim of this project is to creat a core set of data elements for annotating the cancer registry data and achieve the interoperability over the caBIG community. We describe the process required to develop the model, the caDSR CDEs, and the results of the modeling effort. We address difficulties we encountered and modifications for solution. Currently, the Taiwan cancer registry CDEs are released and available in CDE browser for reusing. Furthermore, we will extend our CDEs to daily clinical practice and trials, along with how the methods were used to fully implemented in hospitals and cancer research centers in Taiwan."
Children's Oncology Group (COG) and National Childhood Cancers Foundation
John Deardurff, Nationwide Childrens Hospital
The COG and National Childhoold Cancers Foundation are using caDSR APIs to retrieve CDEs for development of studies using common data descriptors
University of Miami Ontology Modeling
Yevs R. Jean-Mary
Using UML models registered in caDSR and related NCIt concepts, the University of Miami uses semantic web technologies including OWL to develop executable queries against NCIt concepts that can return instances of caGrid Data , (caBIG® 2010 annual meeting poster 87).
Center for International Blood and Marrow Transplant Research (CIBMTR)
Kirt Schapner, Robinette Aley, Barb Kramer, Dr. Douglass Rizzo
This project groups similar data elements using ISO/IEC 11179 structures.
The Center for International Blood and Marrow Transplant Research (CIBMTR) is collecting data on 45 forms ove 5,000 data points. The program uses clustering techniques to analyze the parts of CDEs (Data Element Concept, Object Class and Property and Value Domain concept associations) to discover similar data points. These will be used to help create an HSCT-specific database model in the future (caBIG® 2010 annual meeting poster 95).
The Human Studies Database (HSDB)
UCSF (Leader): Ida Sim (Project Leader), Simona Carini, Rob Wynden
The project goals are to:
- Enhance national clinical and translational research capability: As a computable database with detailed scientific information about past and ongoing human studies, HSDB will be a critical infrastructure for large-scale analysis and reuse of human studies data.
- Ensure standardized computable descriptions of human studies through use of ontologies and controlled clinical vocabularies. The semantic foundation for HSDB is the Ontology of Clinical Research, with mappings to and integration with other relevant models and vocabularies (for example, BRIDG, SNOMED). By adopting ISO/IEC 11179 as the framework for representing the meaning of data, this semantic foundation is independent of the particular technology or architecture used for implementing data sharing.
- Provide a set of tools for institutions to contribute their human studies data to HSDB and to share this data over caGrid. We are focusing first on federating descriptions of study designs before federating study results. Definition and reuse of openMDR and caDSR CDEs will help to ensure consistency in the data collected across these study designs.
Current project members are primarily from Clinical and Translational Science Award (CTSA) institutions, the HSDB project is open and other participating institutions are:
Duke University: Meredith Nahm, Swati Chakraborty
Columbia University: Suzanne Bakken
Johns Hopkins: Harold P. Lehmann
Mayo Clinic: Chris Chute
The Rockefeller University: Ed Barbour, Shamim A. Mollah, Knut M. Wittkowski
Stanford: Samson Tu
UC Davis: Davera Gabriel, Hien Nguyen
University of Colorado: Jessica Bondy
University of Manchester, UK: Alan Rector
UT Health Science Center San Antonio: Brad H. Pollock
UT Southwestern: Herbert K. Hagler, Richard H. Scheuermann
University of Washington: Jim Brinkley, Todd Detweiller
Washington University St. Louis: Rakesh Nagarajan, Jahangheer Shaik
This consortium has adopted the caDSR-compatible openMDR as the basis for registering semantic descriptions of data in the HSDB database.
Illustration of usage by The Human Studies Database (HSDB)
Towards Unambiguous formal descriptions of cancer therapy experiments
Poster (and a paper reference TBD), authors: Alejandra González-Beltrán, PhD et al, Wolfson Institute of Biomedical Research and Department of Computer Science and Cancer Institute, University College London
caDSR CDEs and new CDEs based on ISO/IEC 11179 structure were developed for the Guidelines for Information about Therapy Experiments (GIATE). The CDEs are annotated with concepts from the NCI Thesaurus. Though GAITE was developed independently of NCI Thesaurus, by creating the structured CDEs, using OWL and following a semantic web and linked data approach, an unambiguous and formal description of GIATE was developed from the CDE structures and used to compare GAITE to the ontologies in NCI Thesaurus. "This matching will facilitate interoperability between GIATE compliant knowledge bases and caBIG® services." (caBIG® 2010 annual meeting poster 78)
Ontology based queries for caGrid Infrastructure
Poster, authors: Alejandra González-Beltrán, PhD, Ben Tagger, Eng.D., Anthony Finkelstein, B.Eng, M.Sc., Ph.D., C.Eng, FIET, CTP, FBCS et al, Wolfson Institute of Biomedical Research and Department of Computer Science and Cancer Institute, University College London
Using OWL as a formal language for representing knowledge, this team was able to leverage caDSR mappings between UML Models and NCI Thesaurus concepts and ontology which servs as a conceptual unified view of the data services. This project developed a service for caGrid that provides methods to take caDSR registered models and turn them into OWL. The project has developed a query process in light of OWL2 profiles (caBIG® Annual Meeting poster 79).
Information Management Services, Inc. (IMS) BIS, NAACCR and DCCPS/caSEER program
The program rovides NIH, pharmaceutical, academic and other research organizations with biomedical computing support including:
- Web and Application Development
- caBIG® tool adoption and enhancement support
- Repository Management Systems
- RISMA Compliant Computer Center Hosting
- Biomedical Research Dta Management and Study Research Support
- Statistical Analysis and Reporting
IMS supports the development of CDEs for caDSR for the Person Age standard and the NAACCR CDEs. The program developed data models for seven data services for DCCPS/caSEER Program for sharing through the grid and has registered the CDEs in the caDSR. The program's databases contain over 6 million records that can be searched and summarized to produce tables, graphs, and geographic maps for caSEER. Applications developed by IM are designed to extract and publish specimen data using caBIG® CDEs and the caBIG® Common Business Model.
The CTSA Health Ontology Mapper
This application allows users to take local data and map it to ontology codes using caDSR and LexEVS. The application queries against the NCI LexEVS api server, as well as caDSR and caGRID. Further information can be found on the Health Onology Mapper website.
A Growable Network Information System (AGNIS)
MDACC: Dr. Roy Jones, Charles Martinez
AGNIS, or A Growable Network Information System, began as an idea to transmit hematopoietic stem cell transplant data between organizations but developed into a standards-based communications model. AGNIS was originally intended to be built on an existing messaging system, However, analysis of the National Institutes of Health (NIH) caBIG® project produced a strong case for using caBIG® tools, as caBIG® is becoming a recognized leader in creating standards for grid computing and data definition. The mission to connect the cancer community to accelerate research discoveries and improve patient outcomes fits the purpose of AGNIS, which is the implementation of clinical data exchange across the HSCT community to decrease the time it takes for patient follow-up data to be available for research.
AGNIS works by acting as a translator between two centers. When organizations attempt to communicate their data to one another, they must first massage the data into a format that the other organization can recognize. The translation mechanism is actually standardized common data elements (CDEs) curated and stored on the NIH's publicly accessible caDSR. Due to the reusable nature of elements in caDSR, everyone benefits as more terms are created.
USAGE: The AGNIS network uses the CDE Browser, Curation Tool and caDSR APIs to create and access standard CDEs.
cancerGrid, cgMDR UK
Project Sponsors: Peter McCallum, Jim Davies, Steve Harris
cancergrid is an initiative involving scientists at the Universities of Oxford and Cambridge, working together to reduce the cost of clinical research, and to increase its value through effective data sharing. It is building upon the success of a four-year project funded by the UK Medical Research Council, to address a wider range of scientific goals, with support from Microsoft Research.
The cancergrid team have developed standards and tools for the automatic production of the systems needed to support clinical studies and translational research. The vision:
- the researcher creates a model of their study or dataset, based upon standard templates, using a simple study designer tool
- the software artifacts - forms and services - needed to run the study, or interact with the dataset, are then produced automatically from the model
Along with a meta-model (or model template) for clinical studies, based upon the CONSORT statement, and work on the classification and registration of clinical trials, the project has produced an ISO11179-compliant metadata registry, a semantically-aware trials design/ management system, and a toolkit for clinical data transformation and integration.
These models and software applications have been tested through initial deployment on a small number of clinical studies in Oxford and Cambridge. The metadata registry software has been adopted for use within the US caBIG initiative, and is being evaluated by a number of organisations within the UK. Development continues at Oxford, with clinical collaboration in Cambridge and London, and technical collaboration with the caBIG team in the US.
USAGE: cgMDR was built to be compatible with caDSR, including extensions to ISO 11179 to reference concepts as the grounding for semantics. They have developed an import that can take caDSR CDEs downloaded into an XML file and import then into local cgMDR installations. They have included the "Standard" CDEs in their repository as well as including BRIDG CDEs in their registry download so that local installations may more easily reuse/harmonize with NCI standards as much as possible.
CDC, PHIN Reportable Conditions and Common Core Data Elements
Project Sponsors: Michael Pray, Sundak Gunesan, Catherine Staes
Public health data elements and value sets described in over 90 CSTE position statements which are the foundation building blocks of Public Health Case Reporting (PHCR), ELR and Case Notification. A project is underway collaboratively distribute these data elements and value sets using the existing tools and applications at the National Cancer Institute through the caDSR CDE browser and CDC Vocabulary Server (VADS).
Related Software Engineering Research Highlights
"Quality evaluation of cancer study Common Data Elements using the UMLS Semantic Network"
Authors: Guoqian Jiang, Harold R. Solbrig, Christopher G. Chute
This paper discusses of UMLS SemNet to discover disjointness of semantic network types in caSDR Object Class(OC) and Property(Prop) as a QA mechanism for validating the OC and property in the Data Element Concept. This could be the basis for new collaborative curation platform, these kinds of rules could be incorporated as OWL reasoning for automated validation.
It also discusses using UMLS SemNet as an alternate classification scheme for browsing content in caDSR.
Appendix A - Statistics
Below are a few statistics on some key caDSR resources.
Data Element Descriptions
44,682 Community CDEs across all Contexts
- 24,749 with Workflow Status of "Released"
- 26,939 used in caBIG®
- 10,462 used by CTEP
- 2,272 in CCR implementations
- 3,491 used in C3D studies
- 628 by the Specialized Program of Research Excellence (SPOREs Program)
- 1,373 used by DCP
- 129 used by EDRN
- 288 used by Cancer Imaging Program (CIP)
- 4,007 used by National Heart, Lung and Blood Institute (NHLBI), including National Marrow Donor Program
- 2,333 used by National Institute of Dental and Craniofacial Research (NIDCR)
- 114 used by National Institute of Child Health and Development (NICHD)
- 310 used by National Institute of Neurological Disorders and Stroke (NINDS)
- 1,112 used by Population Sciences and Cancer Control (PS&CC)
As seen in the first chart below, the trend in total numbers of CDEs has been increasing since the project's inception in 2002, with 100 to thousands of draft new CDEs added with each new UML Model and with each new adopter's project (NCI, NIH, External).
While the total number of elements in the registry is increasing, the focus in 2010 has been on harmonization and releasing CDEs that are used, vetted, or both by the community and on identifying and retiring unused content as depicted in the second table below. Retiring unused content will make it easier for new users to find good reusable content.
Chart showing trends in total numbers of CDEs
Chart showing effect of harmonization and retirement of unused content
Case Report Forms and Surveys
3,328 Forms created by the Community
- 2,923 CRFs in CTEP used in 611 trials
- 152 in CCR used in 322 Active Trials
- 102 in NHLBI
- 99 in caBIG®
- 92 in NIDCR
- 40 in DCP
- 29 in SPOREs
- 910,071 CRFs created in C3D from caDSR CDEs
- 1 in NICHD
- 1 in NHINDS
- 1 in REDCap (Demography)
144 Models registered by the Community
- 130 with a Released Status
- NICHD (1)
- Newborn Examinations
- NCI Population Sciences and Cancer Control (5)
- NHIS2-005, caSEER, HINTS2005, SNP500Cancer, GridEnabled Measures
Chart showing UML Models loaded by year
caDSR Tools and Browser
The statistics below only cover the tools most commonly used by the community.
CDE Browser (for the month of June 2011)
544 unique visitors to the site
Average number of unique visitors per day:
Unique visitors past 3 months: Apr 550, May 554, June 544
The top 10 non-NIH domains accessing the CDE Browser were:
- itsg.sdc-moses.com, portale-x.east.saic.com, vpn.nmdp.org, 220.127.116.11 (Nationwide Children S Hospital), 18.104.22.168 (Medical College of Wisconsin), host226.kai-research.com, chi-portal.chi.ohio-state.edu, mail.scenpro.com, 65-113-146-98.dia.static.qwest.net, c-67-99-175-226.roswellpark.org
Curation Tool (as of Feb 2011)
129 unique users visited the site this month.
Curation Tool was linked to by 119 distinct web sites.
Unique visitors past 3 months: Nov 108, Dec 111, Jan 115
(Compared with 2010 Nov 539, Dec 430, Jan 490 --)
Trend Year/Year - 19% Decrease, Average of 117 unique visitors for past 3 months (144 same period in 2009-10).
The top non-NIH Domains using Curation Tool were:
- verizon.net, nmdp.org, saic.net, comcastbusiness.net, direcpc.net, scenpro.com
Form Builder (as of Feb 2011)
101 unique visitors to Form Builder this month.
Form Builder was linked to by 31 distinct web sites.
Unique Visitors the past 3 months: Nov 83, Dec 75, Jan 101
Trend Year/Year - 12% Decrease, Average of 86 unique visitors for past 3 months (97 same period in 2009-10).
For the month, the top 5 non-NIH domains using Form Builder were:
- nmdp.org, Comcast.net, saic.com, verizon.net, direcpc.com
UML Model Browser (as of Feb 2011)
104 unique users visited the site this month.
UML Model Browser was linked to by Not Available distinct web sites.
Unique visitors past 3 months: Nov 99, Dec 96, Jan 104
Trend Year/Years - 22% Decrease, Average 99 unique visitors for past 3 months (138 same period 2009-10)
For the month, the top 5 non-NIH domains using the site were:
- cox.net, duke.edu, nmdp.org, scenpro.com, uams.edu
caDSR API services (as of June 2011)
161 unique users visited the caDSR API site this month.
caDSR API was linked to by 17 distinct web sites.
The top NIH referring URL was the Freestyle API, see statistics below.
Unique visitors past 3 months: Nov 531, Dec 163, and Jan 63
The top 10 non-NIH domains accessing caDSR API were:
- 22.214.171.124 (training Grid), vpn.nmdp.org, hdcob50.mdsol.com, www3.ctsu.org, ip-16-117.wustl.edu, cache1a.mayo.edu, 99-29-162-87.lightspeed.miamfl.sbcglobal.net, newt.westat.com, 91.eabccf.client.atlantech.net
Freestyle services (as of Feb 2011)
52 unique users visited the site.
Freestyle was linked to by 12 distinct web sites.
Unique visitors past 3 months: Nov 33, Dec 33, Jan 43
Trend – Freestyle was incorporated in SIW for searching for existing caDSR content when annotating models. It is currently used by caIntegrator.
Average 43 unique visitors per month.
For the month, the top 5 non-NIH domains using the site were:
- UNKNOWN (4,849 accesses), saic.com, cox.net, gatech.edu, as12448.com
Appendix B - Support for Clinical Trials
Cancer Centers Supported by caDSR CDE development for clinical trials
Abramson Cancer Center of the University of Pennsylvania
Albert Einstein Medical Center
Arizona Cancer Center
UCI / Chao Family Comprehensive Cancer Center
Massachusetts General Hospital/Dana Farber
Georgetown / Lombardi Medical Center
Johns Hopkins Medical Institute
MD Anderson Cancer Network
National Cancer Institute Center for Cancer Research
National Cancer Institute Cancer Diagnosis Program
Oregon Health & Science University
University of Colorado Health Sciences Center
St. Joseph Hospital of Orange
St. Joseph Hospital / Candler
UCSF Carol Franc Buck Breast Care Center
University of Arkansas Medical Sciences
University of Minnesota
University of Nebraska Medical Center
University of Pennsylvania Health Sciences
University of Washington Medical Center
Virginia Commonwealth University
Appendix C - Models
Models with CDEs registered in caDSR, listed alphabetically and by Version within NCI program/Context owner:
Some of these models are retired. Model details are accessible via the UML Model Browser.
caBIG® (NCI cancer Biomedical Informatics Grid Data Services)
C3D Connector 1.5
caArray Internal 2.4
CAD Markup 1.0
CAD Order 1.0
caFE Server 2.0
Cancer Molecular Pages 1.0
CAP Cancer Checklists 1.0
Cardiovascular Model 1.0
caTISSUE CAE 1.2
caTISSUE Core 1.0
caTRIP Annotation Engine 1.0
caTRIP Tumor Registry 1.0
CDC NCPHI Proof of Concept 1.0
Center for Epidemiologic Studies Depression Scale (CES-D) 1.0
Chromosomal Segment Overlap Finder Across Samples 1.0
Chromosomal Segment Overlap Finder Across Sources 1.0
Clinical Trials Lab Model 1.0
Clinical Trials Object Data System (CTODS) .53
CoCaNUT (Colon Cancer Knowledge Utility Toolbox) 1.0
Copy Number Analysis Tool 1.0
CTEP Enterprise Services 1.1
CTMS Metadata 1.0
DNAcopy Analytical Service 1.0
DSD (Dynamic Service Deployment) 1.0
EVS Core Grid Analytical Service 1.0
GenePattern Based Copy Number Analytical Service 1.0
Generic Image 1.0
Generic Parameters 1.0
Genomic Identifiers 1.0
Grid-enablement of Protein Information Resource (PIR) 1.1
Grid-enablement of Protein Information Resource (PIR) 1.2
LabKey CPAS Client API 2.1
Lymphoma Enterprise Architecture Data System (LEADS) 1.0
NBIA (National Biomedical Imaging Archive) 5.0
NCI-60 Drug 1.0
NCI-60 SKY 1.0
NCIA_(National Cancer Imaging Archive) Model 3.0
NHLBI (Protein DB)1.0
omnispect (Multispectral Image Unimxing) 1.0
Organism Identification 1.0
Patient Study Calendar 2.6
PCTA (ProteomeCommon Tranche Annotations) 1.0
Reactome Database Sharing 1.0
SVM (Support Vector Machines) 1.0
Transcription Annotation Prioritization and Screening System 1.0
Utah (Federated Utah Research Translational Health e-Repository-(FURTHeR)) 1.0
WCI1116-05 – Winship's Cancer Institute's Protocol 1116-05 1.0
PS&CC (NCI Population Sciences & Cancer Control)
NHIS 2005 1.0
caCORE (NCI Core Infrastructure)
MDR Object Cart 1.0
MicroArray Gene Expression Object Model (Mage-OM) 1.0
Appendix D - ARCHIVE Publications - see Current list of Publications
Publications citing caDSR and CDEs
- Hu, H., Correll, M., Kvecher, L., Osmond, M., Clark, J., Bekhash, A., Schwab, G., Gao, D., Gao, J., Kubatin, V., Shriver, C.D., Hooke, J.A., Maxwell, L.G., Kovatich, A.J., Sheldon, J.G., Liebman, M.N., Mural, R.J.
"DW4TR: A Data Warehouse for Translational Research". Journal of Biomedical Informatics volume 44, issue 6, year 2011, pp. 1004 - 1019
- Hsu, Chien-Yeh, and Chi-Hung Huang. "Annotating Taiwan Cancer Registry with CaDSR for International Interoperability." Future Wireless Networks and Information Systems. By Shin-Bo Chen. Springer Verlag, 2012. 257-62. Print.
- Martínez Costa, C., Menárguez-Tortosa, M., Fernández-Breis, J.T., "Clinical data interoperability based on archetype transformation". Journal of Biomedical Informatics volume 44, issue 5, year 2011, pp. 869 - 880
- James P. McCusker, Joanne Luciano, and Deborah L. McGuinness, "Towards an Ontology for Conceptual Modeling". Tetherless World Constellation Department of Computer Science Rensselaer Polytechnic Institute, http://tw.rpi.edu/media/2011/03/14/da62/cmo.pdf
- Alejandra González-Beltrán, Ben Tagger, Anthony Finkelstein*,* "Federated Ontology-based Queries over Cancer Data". Accepted for publication BMC Bioinformatics
- Weintraub WS, Karlsberg RP, Tcheng JE, Buxton AE, Boris JR, Dove JT, Fonarow GC, Goldberg LR, Heidenreich P, Hendel RC, Jacobs AK, Lewis W, Mirro MJ, Shahian DM. ACCF/AHA 2011 key data elements and definitions of a base cardiovascular vocabulary for electronic health records: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Clinical Data Standards. Circulation. 2011;124:103--123.
- Pathak J, Wang J, Kashyap S, Basford M, Li R, Masys DR, Chute CG "Mapping clinical phenotype data elements to standardized metadata repositories and controlled terminologies: the eMERGE Network experience", JAMIA doi:10.1136/amiajnl-2010-000061, May 2011
- Guoqian Jiang, Harold R. Solbrig, Christopher G. Chute "Quality Evaluation of Cancer Study Common Data Elements Using the UMLS Semantic Network". MIA Summit on Clinical Research Informatics (AMIA CRI 2011) http://proceedings.amia.org. San Francisco, CA, USA, March 11-12, 2011
- Cui Tao*, PhD, Guoqian Jiang*, PhD, Weiqi Wei, MM, Harold R. Solbrig,and Christopher G. Chute MD, DrPH "Towards Semantic-Web Based Representation and Harmonization of Standard Meta-data Models for Clinical Studies". AMIA Summit on Clinical Research Informatics (AMIA CRI 2011) http://proceedings.amia.org. San Francisco, CA, USA. March 11-12, 2011
- Informatics in radiology: an information model of the DICOM standard. Charles E CE Kahn, Curtis P CP Langlotz, David S DS Channin and Daniel L DL Rubin crossref 31(1):295-304 1 Jan 2011 PMID 20980665, doi: 10.1148/rg.311105085
- Krikov S, Price RC, Matney SA, Allen-Brady K, Facelli JC Enabling GeneHunter as a Grid Service. "A Case Study for Implementing Analytical Services in Biomedical Grids".. Methods Inf Med. 2010 Oct 20;49(6).
- Park YR, Kim JH. Achieving interoperability for metadata registries using comparative object modeling. Stud Health Technol Inform. 2010;160(Pt 2):1136-9.
- Amin W, Singh H, Dzubinski LA, Schoen RE, Parwani AV. Design and utilization of the colorectal and pancreatic neoplasm virtual biorepository: "An early detection research network initiative." J Pathol Inform [serial online] 20101:22. Available from: http://www.jpathinformatics.org/text.asp?2010/1/1/22/70831
- McCusker, James P., Joshua A. Phillips, Alejandra González Beltrán, Anthony Finkelstein, Michael Krauthammer, "Semantic web data warehousing for caGrid", BMC Bioinformatics, Vol 10, Supp 10, 2009.
- Frey LJ, Turner, S., et al. (2010 submitted - under review). "Obtaining Desiderata in the Common Terminology Criteria for Adverse Events: A Case Study", JAMIA
- Waqas Amin, Harpreet Singh, Andre K. Pople, Sharon Winters, Rajiv Dhir, Anil V. Parwani, and Michael J. Becich, "A decade of experience in the development and implementation of tissue banking informatics tools for intra and inter-institutional translational research.", J Pathol Inform. 2010; 1: 12. Published online 2010 August 10. doi: 10.4103/2153-3539.68314.
- Joshua Phillips, Alejandra González Beltrán, Anthony Finkelstein, Jyotishman Pathak "Exposing caGrid Data Services as Linked Data" In the proceedings of the [2010 AMIA Summit on Clinical Research Informatics (AMIA CRI 2010)|http://crisummit2010.amia.org/. San Francisco, CA, USA. March 12-13, 2010.
- Alejandra González Beltrán, Anthony Finkelstein, J Max Wilkinson, Jeff Kramer "Domain Concept-Based Queries for Cancer Research Data Sources " In the proceedings of the 22nd IEEE International Symposium on Computer-Based Medical Systems 2009 (CBMS 2009). Special track on HealthGrid Computing - Applications to Biomedical Research and Healthcare
- Sabados, William Thomas, Thesis: Comparing semantic matching results of schema matchers and metadata registry enabled systems. Pub date:2010. Pages:xii, 164 leaves
- McCarthy J.L., Warzel D, Kendall E., Bargmeyer B., Solbrig H, Keck K, and Gey F., "Data Modeling and Harmonization with OWL: Opportunities and Lessons Learned", 5th International Workshop on Semantic Web Enabled Software Engineering, 8 th International Semantic Web Conference Proceedings and Session, 2009, Virginia, USA
- Sambit K Mohanty, Amita T Mistry, Waqas Amin, Anil V Parwani, Andrew K Pople, Linda Schmandt, Sharon B Winters, Erin Milliken, Paula Kim, Nancy B Whelan, Ghada Farhat, Jonathan Melamed, Emanuela Taioli, Rajiv Dhir, Harvey I Pass, and Michael J Becich., "The development and deployment of Common Data Elements for tissue banks for translational research in cancer – An emerging standard based approach for the Mesothelioma Virtual Tissue Bank", BMC Cancer. 2008; 8: 91. Published online 2008 April 8. doi: 10.1186/1471-2407-8-91.
- Thora Jonsdottir, Johann Thorsson, Ebba Thora Hvannberg, Jan-Eric Litton, Helgi Sigurdsson, "The Nordic Common Data Element repository for describing cancer data - International Journal of Metadata, Semantics and Ontologies" . Inderscience 232 Int. J. Metadata, Semantics and Ontologies, Vol. 4, No. 4, 2009
- Davies, J., Gibbons, J., Harris, S,, Warzel, D. "Evolving Health Informatics Semantic Frameworks and Metadata-Driven Architectures", Position Paper in Proceedings and session presented at Microsoft eScience Conference, (December 2008), Indiana, USA
- Frey LJ, Maojo V and Mitchell JA. (2008). "Genome Sequencing: a Complex Path to Personalized Medicine". In Kim S, Tang H, Mardis ER (Eds.), Advances in Genome Sequencing Technology and Algorithms (pp. 53-75). Norwood, MA: Artech House Publishers, Inc.
- Waqas Amin, Anil V Parwani, Linda Schmandt, Sambit K Mohanty, Ghada Farhat, Andrew K Pople, Sharon B Winters, Nancy B Whelan, Althea M Schneider, John T Milnes, Federico A Valdivieso, Michael Feldman, Harvey I P, Rajiv Dhir, Jonathan Melamed and Michael J Becich., "National Mesothelioma Virtual Bank: A standard based biospecimen and clinical data resource to enhance translational research.", BMC Cancer 2008, 8:236doi:10.1186/1471-2407-8-236
- Cryer, M and Frey LJ (2009) "Agent Based Modeling Supporting the Migration of Registry Systems to Grid Based Architectures." Proceedings of AMIA Summit on Translational Bioinformatics, San Francisco, CA.
- Komatsoulis, G.A., Warzel, D.B., Hartel, F.W., Shanbhag, K, Chilukuri, R, Fragoso, G., de Coronado, S, Reeves, D.M., Hadfield, J.B., Ludet, C., and P.A. Covitz (2007) "caCORE version 3: Implementation of an model driven, service-oriented architecture forsemantic interoperability+." +Journal of Biomed Informatics. 2008 February; 41(1): 106--123. Published online 2007 April 2. doi: 10.1016/j.jbi.2007.03.009.
- Ashokkumar A. Patel, John R. Gilbertson, Louise C. Showe, Jack W. London, Eric Ross, Michael Ochs, Joseph Carver, Andrea Lazarus, Anil V. Parwani, Rajiv Dhir, J. Robert Beck, Michael Liebman, Fernando U. Garcia, Jeff Prichard, Myra Wilkerson, Ronald B. Herberman, Michael J. Becich, and the PCABC, "A Novel Cross-Disciplinary Multi-Institute Approach to Translational Cancer Research: Lessons Learned from Pennsylvania Cancer Alliance Bioinformatics Consortium (PCABC)", Cancer Inform. 2007; 3: 255--274. Published online 2007 June 8.
- Frey LJ, Maojo V and Mitchell JA. (2007). "Bioinformatics Linkage of Heterogeneous Clinical and Genomic Information in Support of Personalized Medicine." IMIA Yearbook of Medical Informatics, 21, 98 - 105.
- Crowley R, Wright L, Warzel D, Sioutos N, Mohanty S, Komatsoulis G, Chilukuri R, Tobias J, "The CAP cancer protocols - a case study of caCORE based data standards implementation to integrate with the Cancer Biomedical Informatics Grid", BMC Medical Informatics and Decision Making (June 20 2006) 6:25
- Covitz P, Warzel D, Fragoso G, Chilukuri R, Phillips J, "The caCORE Software Development Kit: Streamlining construction of interoperable biomedical information services.", BMC Medical Informatics and Decision Making (2006) 6:2
- Ashokkumar A Patel, John R Gilbertson, Anil V Parwani, Rajiv Dhir, Milton W Datta, Rajnish Gupta, Jules J Berman, Jonathan Melamed, Andre Kajdacsy-Balla, Jan Orenstein, Michael J Becich and the Cooperative Prostate Cancer Tissue Resource (CPCTR), "An informatics model for tissue banks – Lessons learned from the Cooperative Prostate Cancer Tissue Resource", BMC Cancer 2006, 6:120doi:10.1186/1471-2407-6-120
- Ashokkumar A Patel, André Kajdacsy-Balla, Jules J Berman, Maarten Bosland, Milton W Datta, Rajiv Dhir, John Gilbertson, Jonathan Melamed, Jan Orenstein, Kuei-Fang Tai and Michael J Becich, "The development of common data elements for a multi-institute prostate cancer tissue bank: the Cooperative Prostate Cancer Tissue Resource (CPCTR) experience.", BMC Cancer 2005, 5:108 doi:10.1186/1471-2407-5-108
- Warzel D, Edelstein C, Lin C, Winget M, Thornquist M, "Proteomics Knowledge databases: facilitating collaboration and interaction between academia, industry, and federal agencies" in Informatics in Proteomics, Srivastava S, Zhang Z, Ravichandran V, Hanash S, Gangal R, Medjahed D, Lockett S, Crichton D, Fenyo D, Kowalski J, Eng J, Beer D, Hitt B, Taylor & Francis/CRC Press. 2005
- Gao Q, Zhang YL, Xie ZY, Zhang QP, Hu ZZ.. "caCORE: core architecture of bioinformation on cancer research in America" Beijing Da Xue Xue Bao 2006 Apr 18;38(2):218-21. Chinese.
- Winget MD, Baron JA, Spitz MR, Bremmer DE, Warzel D, Kincaid H, Thornquist M, Feng Z, "Development of common data elements: The experience of and recommendations from the early detection research network", International Journal of Medical Informatics, 2003, 70:41-48
- Silva, J., Chute, C., Cancer informatics: essential technologies for clinical trials, 2002. Springer Verlag, New York
Abstracts, Posters and Talks
- Topaloglu U., Jett T., Lane C., Hogan W., Hicks A., Kieber-Emmons T., Hutchins L., Curating All Data Elements in a Clinical Trial with CDEs from caDSR, Poster caBIG® Annual Meeting, September 13-15, 2010, Washington, D.C., U.S.A.
- Alejandra González Beltrán, Ben Tagger, Anthony Finkelstein "Ontology-based queries for the caGrid infrastructure", In "Building a Collaborative Biomedical Network", caBIG® Annual Meeting, September 13-15, 2010, Washington, D.C., U.S.A.
- Alejandra González Beltrán, May Yong, Richard Begent "Towards an unambiguous and formal description of cancer therapy experiments", In "Building a Collaborative Biomedical Network", caBIG® Annual Meeting, September 13-15, 2010, Washington, D.C., U.S.A.
- Warzel D. et al, "caCORE Evolves: Semantics II – Future Innovations", MIT Information Quality Symposium July, 2010
- Warzel, D., Reeves, D., Alley, R., Avdic, D., Mathur, A., Gagne, B., Harris, S and Tsui, A., "Cancer Informatics Research: for Recent advances in semantic interoperability cancer informatics+", caBIG® Architecture and Vocabulary and Common Data Elements Face-To-Face, June 2010, Washington, DC+
- Warzel D, , "Semantics II – Future Innovations", Keynote Talk in ISO JTC1 SC32 WG2 Standards and Semantic Web,13 th Open Forum on Metadata Registries (May 2010), ,Kunming, China.
- Alejandra González Beltrán, Anthony Finkelstein, J Max Wilkinson, Jeff Kramer, "Semantic concept-based queries for ONIX - caGrid case " in "Solving Basic and Clinical Research Challenges in Cancer and Beyond", caBIG® Annual Meeting, July 20-22, 2009, Washington, D.C., U.S.A.
- McCusker, James, Michael Krauthammer, Joshua Phillips, Alejandra González Beltrán_,_ Anthony Finkelstein, "Semantic Web Data Warehousing for caGrid", In "Solving Basic and Clinical Research Challenges in Cancer and Beyond", caBIG® Annual Meeting, July 20-22, 2009, Washington, D.C., U.S.A.
- Warzel D., Crichton, C, Davies J. DPhl, Harris, S. PhD, Tsui A., Hastings S., Avdic D., Mathur A, Ludet, C, Elahi, B, "Recent advances in semantic interoperability for cancer informatics," Joint NCI/NCRI Informatics – caBIG® Conference (2009), London, UK
- Warzel D, "Emerging Semantic Technology Standards, Making Life Easier in the Information World", (June 2009), Keynote Talk in ISO JTC1 SC32 WG2 Standards and Semantic Web,12 th Open Forum on Metadata Registries, Seoul, Korea
- Alejandra González Beltrán, Anthony Finkelstein, Jeff Kramer, J. Max Wilkinson. "ONIX Semantic Query Infrastructure", Poster NCI/NCRI Joint Conference: Biomedical Research Without Borders. Bethesda, USA. 2-3 September 2008
- Frey LJ, Stroup N, Cryer, M., He T, Meystre S, Rowe K, & Hartz A. Enabling Data Sharing across Borders with NAACCR and caBIG®. Poster session at Biomedical Informatics without Borders: Enabling Collaboration to Strengthen Research and Care, Bethesda, USA, September 2008
- Warzel D, "Using Vocabularies on the GRID." Abstract and session presented at NKOS/CENDI Joint Conference New Dimensions In Knowledge Organization Systems, (September 2008), Washington DC, USA
- Warzel D, Ludet, C, Davies J, Harris S, Tsui A, +"cancerGrid cgMDR for decentralized development and registration of caBIG® compatible systems",+ Poster, NCRI/NCI Biomedical Informatics without Borders, September 2008, London, UK
- Alejandra González Beltrán, Anthony Finkelstein, J Max Wilkinson, "Platform Architecture and Requirements Testing (PART2) for ONIX Federated Queries" In "Getting Connected with caBI®, Poster, caBIG® Annual Meeting, June 23-25, 2008, Washington, D.C., U.S.A.
- Warzel D, Reeves D, Ludet, C, Davies J, Harris S, Tsui A, "Forms annotation and registration in caDSR facilitated by cancerGrid cgMDR", Poster, NCRI/NCI Biomedical Informatics without Borders, (September 2008), London, UK
- Kunz, Issac, Lin, Ming-Chin Lin, Frey, Lewis, "Metadata mapping and reuse in caBIG®, The First Summit on Translational Bioinformatics 2008 10--12 March 2008, San Francisco, CA, USA
- Kunz I, Lin, MC, and Frey LJ. (2008). "Population Studies for the Management of Colorectal Cancer using caBIG®. Poster session presented at caBIG® Annual Meeting, September 2008, Washington, D.C.
- Frey LJ, He T, Stroup N, Meystre S, Rowe K, & Hartz A. "A Grid Demonstration Project Combining Colorectal Cancer Data Sets Across Utah to Examine Population-Based Health Issues". Annual Conference of the North American Association of Central Cancer Registries (NAACCR), Inc (2008).
- Warzel D, "Exchanging Components in Health Registries", Poster and Talk 11th Open Forum on Metadata Registries, Metadata Down Under, (May 2008), Sydney, Australia
- Macallum, Peter, Warzel, D, "Achieving International Interoperability: Distributed Metadata Registries" – Poster, Joint NCI/NCRI Informatics - caBIG® Conference (2007), Washington DC, USA
- Lin MC and Frey LJ. "Tooling to Support Reuse of CDEs in caBIG® UML Models". Poster session presented at caBIG® Annual Meeting, 2007, Washington, D.C.
- Curtis T, Gundry K, Komatsoulis G, Warzel D, "Using Metadata For Semantic Interoperability: caCORE and the NCI's Cancer Biomedical Informatics Grid (caBIG®)", Poster and Talk, Data Managers Association (DAMA) 2006 Conference, Denver, Colorado
- Yu Rang Park, MS and Ju Han Kim, M.D., Ph.D."Metadata registry and management system based on ISO 11179 for cancer clinical trials information system", AMIA Annual Symposium, 2006
- Warzel D, "The NCI's cancer Data Standards Repository (caDSR): Harmonization of Ontology, Information Models and Metadata" 9th Open Forum on Metadata Registries Harmonization of Terminology, Ontology and Metadata, 2006, Abstract and Talk, Kobe, Japan
- Warzel D, Chilukuri R, De Coronado S, Fragoso G, "The NCI's caCORE: Leveraging Standards and Business Vocabulary for Semantic Interoperability", Abstract and Talk, Second Annual Semantic Technology Conference 2006 (STC06), San Jose, California
- Frey LJ. (2005). "Exploring Harmonization of UML models for caBIG®." Poster session presented at Advancing Practice, Instruction and Innovation through Informatics (APII), Lake Tahoe, CA.
- Warzel D, Andonyadis C, McCurry C, Chilukuri R, Ishmukhamedov S, Covitz P, "Common Data Element (CDE) Management and Deployment in Clinical Trials", Poster presented at American Medical Informatics Association (AMIA), Annual Symposium, September 2003, Washington DC, USA
- Covitz P, Warzel D, "Metadata Management and Information Modeling", Systemics, Cybernetics and Informatics (SCI 2003), July 2003 Abstract and Talk at The 7th World Multiconference, Orlando, Florida
Help Downloading Files
For help accessing PDF, audio, video, and compressed files on this wiki, go to Help Downloading Files.