Biomedical and Health Informatics Weekly

November 23-27, 2009


NEWS AND ANNOUNCEMENTS

Collaborative Technologies in Health Seminar - MEBI 591D

Monday, November 23, 2009
11:00 - 11:50 a.m., Health Sciences K-523A
Facilitator: Andrea Civan

Topic: Kasia Wilamowska will discuss collaboration and EMRs/open MRs


BHI Lecture Series - MEBI 590

Tuesday, November 24, 2009
12:00 - 12:50 p.m., Health Sciences T-739

Speaker:      Kenric W. Hammond, MD, Joint Clinical Associate Professor,
                    Psychiatry and Behavioral Sciences/Medical Education and
                    Biomedical Informatics, University of Washington;
                    Director, Medical Informatics Post-Doctoral Fellowship and 
                    Staff Psychiatrist, VA Puget Sound Health Care System

Title:   "Assessing Information Value in Computerized Patient
           Care Documentation Systems"


Semantic Web in Biomedicine Seminar - MEBI 591B

Tuesday, November 24, 2009
1:00 - 1:50 p.m., Health Sciences I-132
Facilitator:  Michal Galdzicki

Discussion topic:  Nolan Nichols will lead the discussion based on an article by Daniel Rubin and Ion-Florin Talos presented at Translational AMIA Summit 2008, "Computational neuroanatomy: ontology-based representation of neural components and connectivity."  BMC Bioinformatics 2009.  Link to article pdf.  Link to website.


Upcoming Dissertation Defense - Eithon Cadag

Wednesday, November 25, 2009
9:00 a.m., Location:  Health Sciences, BB 1602

Title:  "Automated learning of protein involvement in pathogenesis
          using integrated queries"

Abstract:  "Methods of weakening and attenuating pathogens' abilities to infect and propagate in a host, and thus allowing the natural immune system to more easily decimate invaders, have gained attention as alternatives to broad-spectrum targeting approaches. The following work describes a technique to identifying proteins involved in virulence by relying on latent information computationally gathered across biological repositories. A lightweight method for data integration is introduced, which links information regarding a protein via a path-based query graph and supports both exploratory and logical queries; data gathered in this way is characterized with experiments on retrieving high-quality annotation data. A system and method of weighting is then applied to query graphs that can serve as input to various statistical classification methods for discrimination, and the combined usage of both data integration and learning methods are leveraged against the problem of generalized and specific virulence function prediction. This approach improves coverage of functional data over a protein, outperforms other recent approaches to identification of virulence factors, is robust to different weighting schemes of varying complexity and is found to generalize well to traditional function prediction."


Upcoming General Exam - Casey Overby

Monday, November 30, 2009
1:00 p.m., Location:  Health Sciences, BB-1404

Title:   "A Model for Incorporating Pharmacogenomic Information Into
           Electronic Medical Records for Drug Therapy Individualization:
           A Microcosm of Personalized Medicine"

Abstract:  With the promise of personalized medicine, there is a push for genomics to be incorporated into clinical practice. Pharmacogenomics (PGx) research (the study of how variations in the human genome affect an individual's response to medications) is producing information that is of particular interest to the pursuit of this goal.  Incorporation of PGx-related information in treatment decisions has the potential to increase safety and effectiveness of drug treatment.  With support from governmental agencies such as the US Food and Drug Administration, these studies are leading to the development of guidelines (or supporting information) for prescribing drugs based on a persons' genotype.  Even so, there are still challenges to incorporating the use of these guidelines for PGx-related decisions into clinical practice.  For example, much of the available information on genetic tests to identify individuals as potential slow and fast metabolizers and nonresponders of particular drugs don't tell the physician how and when to adjust drug doses.  In this research, I will develop a prototype system that builds on a production electronic medical record system.  The prototype will present information useful for making PGx-related treatment decisions.  Towards facilitating the use of PGx information in clinical decision-making, the overarching goal for this research is to answer the question: "What are the characteristics and the value of PGx knowledge in the context of clinical decision support within an electronic medical record?"


MEBI Department Holiday Buffet Luncheon
All faculty, staff, students and alumni are invited

Friday, December 11, 2009
12:00 - 1:30pm

Ivar's Salmon House
401 NE Northlake
Seattle, WA 98105
206-632-0767

Directions: http://ivars.net/index.php?page=locations_salmonhouse


Welcome to
Meliha Yetisgen-Yildiz                                                                                  

Meliha Yetisgen-Yildiz is joining the BHI core faculty as an Assistant Professor starting December 1st.  She received her BS degree in Computer Engineering and Information Science from Bilkent University (Ankara, Turkey) and MS degree on Computer Engineering from Middle East Technical University (Ankara, Turkey).  She received her PhD from the University of Washington with a thesis on Biomedical Informatics in December 2007.  After working as a post-doctoral researcher at UW, she joined Kiha, Inc. as a text mining researcher.  Her current research interests include biomedical text mining and information extraction.  She will be focused on a project working with Peter Tarczy-Hornoch and other leadership of the Northwest Institute for Genetic Medicine (http://www.nwigm.org/) using Natural Language Processing approaches to extract phenotype from the electronic medical record for research purposes (funded by the NWIGM LSDF award).