BIG - Current Projects
Below is a listing of BIGs current projects.
Profiles Research Networking Software is an NIH-funded open source tool to speed the process of finding researchers with specific areas of expertise for collaboration and professional networking. Profiles RNS imports and analyzes "white pages" information, publications, and other data sources to create and maintain a complete searchable library of web-based electronic CV's. Built-in network analysis and data visualization tools allow administrators to generate research portfolios of their institution, discover connections between parts of their organization, and understand what factors influence collaboration.
The CTSA Biobank Consortium is automating biobank operations through the development of a custom prototype online software application commonly referred to as SLED (Sample Location and Enhancement Distribution). This is an internally developed, automated IT system which provides for sample request management (request, approval, distribution, tracking, and reporting), while allowing the data and samples to remain with and under the control of the primary investigator.
Many dental schools, including UT Health, use the Axium Electronic Dental Record as their primary data source for teaching and student education. The Consortia of Oral Health Related Informatics (COHRI), comprised of members from many of the leading dental schools in US, would like to be able to use the data from the Axium Instances for research and quality improvement purposes. However, because Consortium members maintain their own Axium Instances each member wanting access to another member's data must individually broker reciprocal access agreements, author custom scripts and provide in-house support to the common membership. The Consortia recognizes the current process is cumbersome costly and time consuming and adds no intrinsic value to the research being undertaken.
The enterprise Master Patient Index (eMPI) project was created as a means to reconcile duplicate patient data within and across institution databases. For a given patient, multiple records may exist within a single database and across databases which creates a scattered patient history and an incomplete history if viewing a single record. The number of duplicate records may be at least 15% of several million, making manually reconciling records an impossible task. A consolidated eMPI will help standardize patient records, reduce the potential for errors in patient identification, and increase opportunities for improved patient treatment and care. Equally important, consolidated patient records provide a comprehensive source of patient information for researchers performing clinical and retrospective studies.
Collaboration in cyberspace often requires that various aspects of the identities of relying parties, both human and non-person entities, be explicitly known and trusted. This is indispensible when authentication, authorization, digital signatures, information integrity, individual accountability and privacy must dynamically occur across institutional boundaries and among numerous individuals and systems.
Shibboleth infrastructure allows users to authenticate at their local institution using their local credentials to securely access the centrally stored resources of a groups participating members.
The goal of the Natural Language Processing (NLP) research project, led by Dr. Jorge Herskovic, is to design an automated process by which the free text in clinical notes can be effectively interpreted for the benefit of future discoveries, developing research hypothesis, and determining comparative effectiveness of specific treatments.
The NLP project is funded through a National Institute of Health (NIH) Challenge Grant (part of the American Recovery and Reinvestment Act, ARRA). The main focus of this project is transforming clinical records into concept streams, delivering a process which enables quick and privacy-preserving comparative effectiveness research (CER). BIG’s NLP project for CER focuses on three crucial categories: Disorders, Medications, and Procedures.
The NLP Project comprises five separate initiatives:
- Gold Standard Development
- Clinical Text Concept Extraction
- Clinical Documentation Summarization
- Biomedical Literature Summarization
- Biomedical Literature Indexing