• slider image 1

About the EDDA Group

image 2

EDDA Group

Tanja Bekhuis, PhD, MS, MLIS, AHIP, leads the Evidence in Documents, Discovery, and Analysis (EDDA) Group, Department of Biomedical Informatics (DBMI), University of Pittsburgh School of Medicine.

Current members

  • Eugene Tseytlin, MS, computer science, systems developer, DBMI
  • Ashleigh Faith, MA, public history; taxonomist, SAE International; adjunct faculty, School of Information Sciences
  • Saja Al-alawneh, MS, computer science, NLM/NIDCR doctoral fellow, DBMI
  • Faina Linkov, PhD, MPH, Associate Professor and epidemiologist, Magee-Womens Research Institute, University of Pittsburgh School of Medicine

Past members of the EDDA Group

  • Kevin Mitchell, MS, systems developer
  • John Frazier, DMD, MSPH, postdoctoral trainee
  • Corey Stein, MS, graduate student researcher


image 2

Screening Nonrandomized Studies for Inclusion in Systematic Reviews of Evidence

(NIH/NLM 5R00LM010943). Dr. Tanja Bekhuis (Principal Investigator) is the recipient of a career development award from the US National Library of Medicine (NLM), National Institutes of Health (NIH). To carry out this research, she often collaborates with Dr. Dina Demner-Fushman, Staff Scientist at the Lister Hill National Center for Biomedical Communications, NLM.

NIH Summary

Translation of biomedical research into practice depends in part on the production of quality systematic reviews that synthesize available evidence. Unfortunately, about 20% of reviews are never completed. Of those that reach fruition, the average time to completion may be 2.4 years, with a reported maximum of 9 years. A major bottleneck occurs when teammates screen studies. In the first step, they independently identify provisionally eligible studies by reading the same set of perhaps thousands of titles and abstracts. To date, researchers have used supervised machine learning (ML) methods in an attempt to automate identification of eligible randomized controlled trials (RCTs). However, finding nonrandomized (NR) studies for inclusion in systematic reviews has yet to be addressed. This is an important problem because RCTs may be unlikely or even unethical for some research questions. The proposed research is significant because it will help support translation of biomedical research to improve human health. Moreover, developing procedures to identify NR studies is essential for the expeditious translation of a very large body of research.


EDDA Study Designs and Publications Terminology

This terminology is available in the NCBO BioPortal http://bioportal.bioontology.org/ontologies/EDDA. Tanja Bekhuis (PI); Eugene Tseytlin and Kevin Mitchell (developers), Ashleigh Faith (taxonomist). For details regarding the rationale for its development, see Bekhuis, Demner‑Fushman, Crowley, JMLA, 2013. To optimize its usefulness for other researchers, we are updating the terminology and will release a new version in the BioPortal.

Improve Identification of Topically Relevant Studies for Provisional Inclusion in a Systematic Review

This is a continuation of research reported in Bekhuis et al., PloS One, 2014. Currently, we are focusing on extracting information from excerpts of full-text scientific articles to improve automatic classification. Additionally, we are working on methods to visualize and label clustered studies.


The EDDA Group is building a prototype to view latent themes in a set of documents for various kinds of users. We envision a Web-based, interactive Lens to assist expert searchers, such as librarians and informationists, clinical researchers, and methodologists.

Finding and Visualizing Studies of Potential Prognostic Biomarkers

We are designing filters (search strategies) to improve retrieval of studies about potential prognostic biomarkers for oral squamous cell carcinoma—the most common form of cancer affecting the oral cavity.

Oral health translational research

Dr. Bekhuis often collaborates with Dr. Heiko Spallek and a group of researchers at the University of Pittsburgh School of Dental Medicine and the American Dental Association. We are interested in promoting implementation and adherence to recommendations in clinical practice guidelines to improve patient care. Currently, we are focused on methods to persuade dentists to adopt a simple, noninvasive procedure to prevent or treat incipient caries (tooth decay), a common and chronic disease. If left untreated, it is the source of significant pain and loss of teeth. Treating tooth decay in the US—excluding costs for impaired quality of life and impact on morbidity—costs about $60 billion annually. Moreover, poor oral health is a risk factor for serious systemic problems, such as diabetes and cardiovascular disease. In 2013, we published a “highly accessed” article in BMC Oral Health titled “Are dentists interested in the oral systemic disease connection? A qualitative study of an online community of 450 practitioners.” This article was also featured in the British Dental Journal, 2013. Natural language processing and pattern matching enabled this research.

New Center for Informatics in Oral Health Translational Research

Dr. Heiko Spallek, Executive Director, Dr. Bernard Costello, Director of Translational Research, and Dr. Tanja Bekhuis, Director of Translational Research Methods, are leading development of a new Center for Informatics in Oral Health Translational Research, in the School of Dental Medicine. The overarching mission of the new Center is to support research and education aimed at improving delivery of dental care and patient outcomes, as well as treatment of oral and maxillofacial conditions, particularly those related to systemic health. This Center significantly expands the mission of the former Center for Dental Informatics.