The National Library of Medicine’s (NLM) Medical Text Indexer is being used as one of the baselines for the international BioASQ challenge. The Medical Text Indexer (MTI), a system for producing indexing recommendations, assists in the indexing process at NLM. The BioASQ challenge is a series of challenges on biomedical semantic indexing and question answering, with the aim of advancing the state of the art accessibility for researchers and clinicians to biomedical text. The MTI indexing results are providing one of the baselines used in the “Large-scale online biomedical semantic indexing” part of the challenge, which is designed to parallel the human indexing currently being done at NLM. Alan R. Aronson, PhD, Principal Investigator for the MTI project, also will be delivering an invited talk on Indexing The Biomedical Literature In A Time Of Increased Demand And Limited Resources at the BioASQ Workshop in September. Dr. Aronson is a Principal Investigator at the Lister Hill National Center for Biomedical Communications, an Intramural Research Division of the National Library of Medicine.
Archive for the ‘Informatics’ Category
NLM to Participate with Partners in “An Epidemiology of Information: New Methods for Interpreting Disease and Data”
The National Library of Medicine (NLM) has announced its next initiative as part of its ongoing partnership with the National Endowment for the Humanities (NEH). Working with NEH’s Office of Digital Humanities, the National Science Foundation, and Virginia Polytechnic Institute and State University (Virginia Tech), the NLM will be a part of An Epidemiology of Information: New Methods for Interpreting Disease and Data, an interdisciplinary symposium exploring new methods for large-scale data analysis of epidemic disease.
Scheduled to take place at the Virginia Tech Research Center in Arlington, VA, on October 17, 2013, from 8:30 AM to 5:00 PM, “An Epidemiology of Information” will be a unique public forum through which policy makers, public health experts, and scholars can address pressing questions about how new methods of analyzing large-scale datasets can inform research and policy approaches to epidemic disease. Panelists will consider what these new methods suggest for contemporary infodemiology and epidemic intelligence, as well as the implications of data mining as a disease surveillance mechanism, and how new forms of reporting and public health surveillance affect public health policy. The symposium will also explore how these new methods can inform research on the 1918 influenza pandemic, and help to answer lingering questions about the spread of the disease, its pathogenicity, the unusual mortality rates, or the effectiveness of public health responses.
Featured speakers will include Dr. Jeffery Taubenberger, Chief, Viral Pathogenesis and Evolution Section, National Institute of Allergy and Infectious Diseases (NIAID), and Dr. David Morens, Senior Advisor to the Director, NIAID, whose research in data analysis and historical epidemiology has influenced the approaches being adopted and adapted by digital humanities scholars working in the history of medicine. “An Epidemiology of Information” is made possible in part from support received by Virginia Tech through the international Digging into Data Challenge competition sponsored by NEH. Funding for Virginia Tech’s Canadian partner, the Center for E-Health Initiatives of the University of Toronto, comes from the Social Science and Humanities Research Council of Canada. The symposium is free and open to the public, but registration is required.
The National Library of Medicine has announced that Extensible Markup Language (XML) data from the IndexCat™ database is now available for free download. Released with a Document Type Definition (DTD) that allows researchers to validate the data, this new XML release includes the digitized content of more than 3.7 million bibliographic items from the printed, 61-volume Index-Catalogue of the Library of the Surgeon-General’s Office, originally published from 1880 to 1961. The XML describes items spanning five centuries, including millions of journal and newspaper articles, obituaries, and letters; hundreds of thousands of monographs and dissertations; and thousands of portraits. Together, these items cover a wide range of subjects such as the basic sciences, scientific research, civilian and military medicine, public health, and hospital administration.
The NLM release of the Index-Catalogue in XML format opens this key resource in the history of medicine and science to new uses and users. It is one of the monuments of the Library’s longstanding, systematic indexing of the medical literature, an effort which William Henry Welch (1850-1934), the great pathologist and bibliophile, considered to be “America’s greatest contribution to medical knowledge.” This indexing, begun by John Shaw Billings in the nineteenth century at the Library of the Surgeon-General’s Office, United States Army (known today as the NLM), eventually created two distinct products: the Index-Catalogue of the Library of the Surgeon-General’s Office, United States Army, and the Index Medicus, forerunner of MEDLINE®, and now the largest component of PubMed.®
Released alongside the IndexCatalogue XML are an integrated XML file and associated DTD for two collections developed from the electronic database of A Catalogue of Incipits of Mediaeval Scientific Writings in Latin (rev.), by Lynn Thorndike and Pearl Kibre (eTK), and the updated and expanded version of Scientific and Medical Writings in Old and Middle English: An Electronic Reference (eVK2), edited by Linda Ehrsam Voigts and Patricia Deery Kurtz. Also available via the online IndexCat, these resources encompass over 42,000 records of incipits, or the beginning words of a medieval manuscript or early printed book, covering various medical and scientific writings on topics as diverse as astronomy, astrology, geometry, agriculture, household skills, book production, occult science, natural science, and mathematics, as these disciplines and others were largely intermingled in the medieval period of European history. The NLM release of these resources in XML format joins many other freely downloadable resources, including the XML for MEDLINE®/PubMed® data, which includes over 22 million references to biomedical and life sciences journal articles back to 1946, and, for some journals, much earlier.
The release also coincides with the NLM’s participation in “Shared Horizons: Data, Biomedicine, and the Digital Humanities,” an interdisciplinary symposium exploring the intersection of digital humanities and biomedicine, being held April 10-12, 2013, in partnership with the National Endowment for the Humanities’ Office of Digital Humanities, Maryland Institute for Technology in the Humanities at the University of Maryland, and Research Councils UK. Shared Horizons will create opportunities for disciplinary cross-fertilization through a mix of formal and informal presentations, combined with breakout sessions designed to promote a rich exchange of ideas about how large-scale quantitative methods can lead to new understandings of human culture. Bringing together researchers from the digital humanities and bioinformatics communities, the symposium will explore ways in which these two communities might fruitfully collaborate on projects that bridge the humanities and medicine around the topics of sequence alignment and network analysis, two modes of analysis that intersect with “big data.” All Shared Horizons sessions will be live-streamed with a monitored back channel for the public to post/tweet comments. Recordings of all talks will also be posted to the Shared Horizons website, with the ability to comment pre- and post-event.
Research funded by the National Library of Medicine provides new insight into why patients stop taking drugs that lower their cholesterol, and what happens when patients try those drugs, known as statins, a second time. Researchers found that more than 90% of patients who stopped taking statins because of an adverse reaction could tolerate the medication when tried again. The study is published in the April 2, 2013, issue of the Annals of Internal Medicine.
NLM grantee Alexander Turchin MD, MS, of Brigham and Women’s Hospital, a teaching affiliate of Harvard Medical School, notes that statins are commonly stopped even though their benefits are well documented. He and colleagues wanted to better understand why statins are discontinued and whether adverse reactions play a role. They conducted a retrospective study, analyzing clinical data in an electronic medical record (EMR) system. Researchers examined structured data as well as the narrative electronic notes of health providers. Those notes frequently are the only place in an EMR where adverse reactions to medications are documented. Using the NLM grant, researchers developed natural language processing software and scoured more than 5 million notes, on more than 107,000 patients, recorded over nearly a decade. The software generated data on a scale that could not have been done manually. Researchers say the next step would be to conduct a clinical trial to determine if outcomes are improved when statins are tried again, after an adverse event.
The National Library of Medicine, part of the National Institutes of Health, conducts and funds research in biomedical informatics, which involves applying computers and communications technology to the field of health. This research was supported by NLM’s Division of Extramural Programs grant RC1-LM010460. This was an NIH Challenge Grant, supported by NLM with funds from the American Recovery and Reinvestment Act. For additional information, visit the Brigham and Women’s Hospital News Release.
The National Library of Medicine Division of Extramural Programs has announced the 2013 NLM Informatics Lecture series showcasing NLM-funded research in biomedical informatics. The program kicks off March 6 when Chunhua Weng, PhD, presents “Bridging the Semantic Gap Between Research Eligibility Criteria and Clinical Data: Methods and Issues.” Dr. Weng is the Florence Irving Assistant Professor of Biomedical Informatics at Columbia University. Her research centers on developing human-computer collaborative approaches to help clinical researchers make the best use of health information technology. She currently is focusing on problems that include interactive query formulation to assist clinical researchers in interrogating large clinical databases.
On June 5, Graciela Gonzalez, PhD, of Arizona State University, will present “Mining Social Network Postings for Mentions of Potential Adverse Drug Reactions.” The final 2013 lecture, November 13, will be presented by Timothy Cardozo, MD, PhD, of the New York University School of Medicine. He will discuss “A Chemical Biology Network for Personalized Medicine.” The lectures will be held 2-3pm (Eastern) in Balcony A of the Natcher Building (Building 45) on the NIH campus. All three talks will be archived at http://videocast.nih.gov/.
New Course Announced for Spring 2013 at the National Library of Medicine: A Librarian’s Guide to NCBI!
A Librarian’s Guide to NCBI comprises a pre-course, “Fundamentals in Bioinformatics and Searching,” offered online (asynchronous), during March 2013, and a five-day, in-person course offered at the National Library of Medicine, April 15-19, 2013. The course provides basic knowledge and skills for librarians interested in helping their clientele use online molecular databases and tools from the NLM’s National Center for Biotechnology Information (NCBI). Topics include using the BLAST sequence similarity search and Entrez text search systems to find relevant data. The course describes the various kinds of molecular data available, and explains how these are generated and used in modern biomedical research. Instructors will be NCBI staff and Diane Rein, Ph.D., from the University at Buffalo. The course will be limited to 18 students. More information and an online application will be available in December, 2012.
Applications for the 2013 sessions of the NLM Biomedical Informatics course in Woods Hole, MA, are now being accepted. The submission deadline is January 11, 2013. Selection decisions will be made by mid-February. The costs for attending this course, including travel, housing, and meals, are fully supported by the National Library of Medicine. Sessions are limited to 30 participants. This is a National Library of Medicine fellowship program directed at medical educators, medical librarians, medical administrators, and young faculty, who are not currently knowledgeable but can become agents of change in their institutions.
This week-long survey course is designed to familiarize individuals with the application of computer technologies and information science in biomedicine and health science. Through a combination of lectures and hands-on computer exercises, participants will be introduced to the conceptual and technical components of biomedical informatics. The conceptual components will include principles of database design, human-computer interfaces, medical terminologies and coding systems, medical decision analysis methods, clinical information systems architectures, and methods for measuring costs and benefits in health care systems. The technical components will include use of the Internet for biomedical applications, current and emerging wide area network technologies, use of literature and molecular sequence databases, and systems for telemedicine. Evening workshops will include hands-on project development to give students the opportunity to bring their own expertise to bear, and apply the lessons learned in class to a relevant informatics application.
The course dates for 2013 are:
- Spring Session: May 26 – June 1, 2013
- Fall Session: September 15 – 21, 2013
More information, including course objectives, activity agenda, and list of instructors, is available on the course page. Many PSR Network members have attended over the years, and they can attest to the value of the experience!
The July 2012 International Release of SNOMED CT is available for download. The International Release is updated each year in January and July. The recent download contains SNOMED CT files in both Release Format 1 (RF1) and the new Release Format 2 (RF2) versions. Contents of the download include SNOMED CT terminology, cross maps, subsets, User Guide (content, principles and uses), Technical Implementation Guide (design of applications using SNOMED CT), Technical Reference Guide (file layouts, field sizes, required values, data diagrams), and ICD-9-CM Cross Map. Additionally, updated RF2 to RF1 compatibility tools are available for download from the same Web page. SNOMED CT terminology will be available in Metathesaurus format in UMLS 2012AB Release, in November 2012.
Funding Announcement: NLM Administrative Supplements for Informationist Services in NIH-funded Research Projects
A new funding announcement offering support for informationists to work on NIH-funded research grants was published on the NIH Guide web site today. These supplements provide funds to researchers who have existing research grants from any of the Institutes listed in the announcement (NLM, NCI, NEI, NIA, NIAAA, NIBIB, NIDCD, NIDCR), to pay for adding an informationist to the project. The principal investigator of the grant must apply for this funding, so librarian/informationist colleagues in academic settings might want to identify partners of interest and reach out to them to suggest that they apply, or alert people with whom they already work. An easy way to find potential partners would be to use the NIH RePORTER resource to search by state and funding agency. Applications must be submitted electronically by the deadline of June 5, 2012. The earliest funding start date is September, 2012.
The purposes of the administrative supplement program are (1) to enhance collaborative, multi-disciplinary basic and clinical research by integrating an information specialist, also known as in-context information specialist, into the research team in order to improve the capture, storage, organization, management, integration, presentation, and dissemination of biomedical research data; and (2) to assess and document the value and impact of the informationist’s participation.
The next UMLS Webcast will be on September 28, 2011 at 2:00 pm ET. The topic is “An Update of RxNorm” presented by Dr. Stuart Nelson, Head of the Medical Subject Headings (MeSH®) Section. The Webcast will cover:
- Inclusion of National Drug Codes from RxNorm content providers First DataBank (National Drug Data File Plus) and Micromedex (RED BOOK®)
- New term types for RxTerms functionality
- Current Prescribable Content Subset
The Webcast link is https://webmeeting.nih.gov/rxnorm_update. It will be archived and linked from the UMLS Webcasts Web page.
From MEDLARS Management Section