David Eichmann

Cross-language information retrieval with the UMLS metathesaurus

We investigate an automatic method for Cross Language Information Retrieval (CLIR) that utilizes the multilingual UMLS Metathesaurus to translate Spanish and French natural language queries into English. Two experiments are presented using OHSUMED, a subset of MEDLINE. Both experiments examine retrieval effectiveness of the translated queries. However, in the second experiment, the query translation procedure is augmented with digram based vocabulary normalization procedures. In this comparative study of retrieval effectiveness the measures used are: 11-point-average precision score (11-AvgP); average interpolated precision at recall of 0.1; and noninterpolated (i.e., exact) precision after 10 retrieved documents. Our results indicate that for Spanish the UMLS Metathesaurus based CLIR method appears equivalent to multilingual dictionary based approaches investigated in the current literature French yields less favorable results and our analysis suggests that linguistic differences may have caused the performance differences.

Detecting Wikipedia vandalism with active learning and statistical language models

This paper proposes an active learning approach using language model statistics to detect Wikipedia vandalism. Wikipedia is a popular and influential collaborative information system. The collaborative nature of authoring, as well as the high visibility of its content, have exposed Wikipedia articles to vandalism. Vandalism is defined as malicious editing intended to compromise the integrity of the content of articles. Extensive manual efforts are being made to combat vandalism and an automated approach to alleviate the laborious process is needed. This paper builds statistical language models, constructing distributions of words from the revision history of Wikipedia articles. As vandalism often involves the use of unexpected words to draw attention, the fitness (or lack thereof) of a new edit when compared with language models built from previous versions may well indicate that an edit is a vandalism instance. In addition, the paper adopts an active learning model to solve the problem of noisy and incomplete labeling of Wikipedia vandalism. The Wikipedia domain with its revision histories offers a novel context in which to explore the potential of language models in characterizing author intention. As the experimental results presented in the paper demonstrate, these models hold promise for vandalism detection.

Synergies and Distinctions between Computational Disciplines in Biomedical Research: Perspective from the Clinical and Translational Science Award Programs

Clinical and translational research increasingly requires computation. Projects may involve multiple computationally oriented groups including information technology (IT) professionals, computer scientists, and biomedical informaticians. However, many biomedical researchers are not aware of the distinctions among these complementary groups, leading to confusion, delays, and suboptimal results. Although written from the perspective of Clinical and Translational Science Award (CTSA) programs within academic medical centers, this article addresses issues that extend beyond clinical and translational research. The authors describe the complementary but distinct roles of operational IT, research IT, computer science, and biomedical informatics using a clinical data warehouse as a running example. In general, IT professionals focus on technology. The authors distinguish between two types of IT groups within academic medical centers: central or administrative IT (supporting the administrative computing needs of large organizations) and research IT (supporting the computing needs of researchers). Computer scientists focus on general issues of computation such as designing faster computers or more efficient algorithms, rather than specific applications. In contrast, informaticians are concerned with data, information, and knowledge. Biomedical informaticians draw on a variety of tools, including but not limited to computers, to solve information problems in health care and biomedicine. The paper concludes with recommendations regarding administrative structures that can help to maximize the benefit of computation to biomedical research within academic health centers.

siSPOTR: a tool for designing highly specific and potent siRNAs for human and mouse

RNA interference (RNAi) serves as a powerful and widely used gene silencing tool for basic biological research and is being developed as a therapeutic avenue to suppress disease-causing genes. However, the specificity and safety of RNAi strategies remains under scrutiny because small inhibitory RNAs (siRNAs) induce off-target silencing. Currently, the tools available for designing siRNAs are biased toward efficacy as opposed to specificity. Prior work from our laboratory and others’ supports the potential to design highly specific siRNAs by limiting the promiscuity of their seed sequences (positions 2–8 of the small RNA), the primary determinant of off-targeting. Here, a bioinformatic approach to predict off-targeting potentials was established using publically available siRNA data from more than 50 microarray experiments. With this, we developed a specificity-focused siRNA design algorithm and accompanying online tool which, upon validation, identifies candidate sequences with minimal off-targeting potentials and potent silencing capacities. This tool offers researchers unique functionality and output compared with currently available siRNA design programs. Furthermore, this approach can greatly improve genome-wide RNAi libraries and, most notably, provides the only broadly applicable means to limit off-targeting from RNAi expression vectors.

Personal health records: a randomized trial of effects on elder medication safety

PURPOSE To examine the impact of a personal health record (PHR) on medication-use safety among older adults. BACKGROUND Online PHRs have potential as tools to manage health information. We know little about how to make PHRs accessible for older adults and what effects this will have. METHODS A PHR was designed and pretested with older adults and tested in a 6-month randomized controlled trial. After completing mailed baseline questionnaires, eligible computer users aged 65 and over were randomized 3:1 to be given access to a PHR (n=802) or serve as a standard care control group (n=273). Follow-up questionnaires measured change from baseline medication use, medication reconciliation behaviors, and medication management problems. RESULTS Older adults were interested in keeping track of their health and medication information. A majority (55.2%) logged into the PHR and used it, but only 16.1% used it frequently. At follow-up, those randomized to the PHR group were significantly less likely to use multiple non-steroidal anti-inflammatory drugs-the most common warning generated by the system (viewed by 23% of participants). Compared with low/non-users, high users reported significantly more changes in medication use and improved medication reconciliation behaviors, and recognized significantly more side effects, but there was no difference in use of inappropriate medications or adherence measures. CONCLUSIONS PHRs can engage older adults for better medication self-management; however, features that motivate continued use will be needed. Longer-term studies of continued users will be required to evaluate the impact of these changes in behavior on patient health outcomes.

Direct2Experts: A pilot national network to demonstrate interoperability among research-networking platforms

Research-networking tools use data-mining and social networking to enable expertise discovery, matchmaking and collaboration, which are important facets of team science and translational research. Several commercial and academic platforms have been built, and many institutions have deployed these products to help their investigators find local collaborators. Recent studies, though, have shown the growing importance of multiuniversity teams in science. Unfortunately, the lack of a standard data-exchange model and resistance of universities to share information about their faculty have presented barriers to forming an institutionally supported national network. This case report describes an initiative, which, in only 6 months, achieved interoperability among seven major research-networking products at 28 universities by taking an approach that focused on addressing institutional concerns and encouraging their participation. With this necessary groundwork in place, the second phase of this effort can begin, which will expand the networks functionality and focus on the end users.

Design lessons for older adult personal health records software from older adults

Online Personal Health Records (PHR) software has the potential to provide older adults with tools to better manage several aspects of their health, including their use of medications. In spite of this potential, we still know little about how to make PHRs accessible for older adults. We also know little about how to design PHRs in a way that will enable older adults to get a valuable return on their time investment in using such systems. In this paper, we present our experience partnering with a group of older adults to obtain design guidelines for the design of a PHR with a focus on medication management. We discuss the outcomes of our design partnership and provide an overview of the design of a web-based PHR we designed based on these outcomes.

Commentary: Unintended consequences of health information technology: A need for biomedical informatics

Professor and Chair Department of Biomedical Informatics, University of Pittsburgh School ofMedicine, Pittsburgh, PAIn all science, error precedes the truth, and it is better it should go first than last.-- Sir Hugh Walpole (English novelist, 1884–1941)Health information technology (HIT) can address important problems in clinical care andbiomedical research. These problems include lack of compliance with clinical practiceguidelines [1], insufficient use of preventative medicine services [2] and numerousimpediments to clinical/translational research [ 3]. However, front-line patient care informationsystems that can influence care may worsen outcomes as well as improve them. Increasingly,there is evidence of significant, unintended and deleterious effects of well-meaning HIT efforts[4]. In this paper we present examples of such deleterious effects and argue that: 1) HIT is atool that can influence health care and biomedicine (for good or ill) and 2) biomedicalinformatics efforts are needed to ensure that HIT fulfills its promise in biomedicine.

Standardized nursing language in the systematized nomenclature of medicine clinical terms: A cross-mapping validation method

Many standardized healthcare languages have been mapped to the Systematized Nomenclature of Medicine Clinical Terms known as SNOMED CT, which was developed by the College of American Pathologists. This study describes a methodology for detecting misassigned concepts from source systems to SNOMED CT and presents the results of applying this methodology to a subset of concepts from two standardized nursing languages, the Nursing Interventions Classification and the Nursing Outcomes Classification. The methodology is based on comparing the knowledge representations of a set of nursing concepts between source systems (nursing languages) and SNOMED CT. If any nursing concept differs in knowledge representation in the target system compared with the source system, editorial misassignment of the concept was declared and recommendations for target system developers were made. In a total of 75 nursing concepts used to test this method, five misassigned concepts(6.6%) were found in SNOMED CT. This method can be used to validate other healthcare languages.

A cluster-based approach to broadcast news

We present an approach to detection and tracking of topics in multilingual broadcast news based upon a dynamic clustering scheme. Our approach derives from a system used to filter Web searches from multiple sources, with extensions for pipelining document clusters, part-of-speech tagging and extraction of named entities for use in an extended similarity measure.