Search

The high heterogeneity of biomedical vocabulary is a major obstacle for information retrieval in large biomedical collections. Therefore, using biomedical controlled vocabularies is crucial for managing these contents. We investigate the impact of query expansion based on controlled vocabularies to improve the effectiveness of two search engines. Our strategy relies on the enrichment of users' queries with additional terms, directly derived from such vocabularies applied to infectious diseases and chemical patents. We observed that query expansion based on pathogen names resulted in improvements of the top-precision of our first search engine, while the normalization of diseases degraded the top-precision. The expansion of chemical entities, which was performed on the second search engine, positively affected the mean average precision. We have shown that query expansion of some types of biomedical entities has a great potential to improve search effectiveness; therefore a fine-tuning of query expansion strategies could help improving the performances of search engines.

We report on the original implementation of named entity recognition (NER) modules based on an automatic text categorization pipeline, so-called ToxiCat (Toxicogenomic Categorizer), developed to perform biomedical documents classification and prioritization for the previous Biocreative campaign in order to speed up the curation of the Comparative Toxicogenomics Database (CTD). ToxiCat NER modules are a group of components that analyse text for enclosed information. These modules are based on an information retrieval engine for MEDLINE (EAGLi), a gene normalization (GN) service (NormaGene) developed for a previous BioCreative campaign, gene ontology categorizer (GOCat) and finally an entity recognizer for diseases and chemicals. The NER services are publically available as RESTful web services at http://pingu.unige.ch:8080/Toxicat.

Health-related information retrieval is complicated by the variety of nomenclatures available to name entities, since different communities of users will use different ways to name a same entity. We present in this report the development and evaluation of a user-friendly interactive Web application aiming at facilitating health-related patent search. Our tool, called TWINC, relies on a search engine tuned during several patent retrieval competitions, enhanced with intelligent interaction modules, such as chemical query, normalization and expansion. While the functionality of related article search showed promising performances, the ad hoc search results in fairly contrasted results. Nonetheless, TWINC performed well during the PatOlympics competition and was appreciated by intellectual property experts. This result should be balanced by the limited evaluation sample. We can also assume that it can be customized to be applied in corporate search environments to process domain and company-specific vocabularies, including non-English literature and patents reports.

Objective: While the broad use of antibiotics has reached its limits with the apparition of bacterial resistance, it became of major importance to regulate antibiotic prescriptions. In this paper, we present KART, a system to facilitate the creation of clinical guidelines in the context of infectious diseases. Methods: This system is composed of three main modules. The first module aims at facilitating the step of systematic reviews with the use of question-answering techniques, in order to create recommendations by querying a question-answering engine. Then, the second module proposes an approach to semi-automatically normalize the different parameters forming clinical recommendations. Finally, the third module aims to automatically formalize and store the generated recommendations into a knowledge repository. Results: The question-answering module is able to answer about two thirds of the queries correctly. The normalization, however, has had very mixed success upon evaluation of the system by infectious diseases specialists. Conclusion: In conclusion, KART is an innovative and promising system to create clinical guidelines; however, preliminary qualitative and quantitative evaluations show the need for further improvements.

We present a new approach to perform biomedical documents classification and prioritization for the Comparative Toxicogenomics Database (CTD). This approach is motivated by needs such as literature curation, in particular applied to the human health environment domain. The unique integration of chemical, genes/proteins and disease data in the biomedical literature may advance the identification of exposure and disease biomarkers, mechanisms of chemical actions, and the complex aetiologies of chronic diseases. Our approach aims to assist biomedical researchers when searching for relevant articles for CTD. The task is functionally defined as a binary classification task, where selected articles must also be ranked by order of relevance. We design a SVM classifier, which combines three main feature sets: an information retrieval system (EAGLi), a biomedical named-entity recognizer (MeSH term extraction), a gene normalization (GN) service (NormaGene) and an ad-hoc keyword recognizer for diseases and chemicals. The evaluation of the gene identification module was done on BioCreativeIII test data. Disease normalization is achieved with 95% precision and 93% of recall. The evaluation of the classification was done on the corpus provided by BioCreative organizers in 2012. The approach showed promising performance on the test data.

The expansion of biomedical knowledge, reductions in computing costs and spread of IT facilities have led to an explosion of the biomedical electronic data. However, these data are rarely integrated and analysed because of lack of tools. The integration process is complex due to technical and semantic heterogeneity as well as lack of reliability in such distributed system. In addition, for the specific case of biomedical data, privacy is a crucial constraint. This paper presents a pilot system that will be used in the European FP7 DebugIT project to integrate biomedical data from several healthcare centres across Europe.

We report on the design and evaluation of an original system to help assignment ICD (International Classification of Dis-ease) codes to clinical narratives. The task is defined as a multi-class multi-document classification task. We combine a set of machine learning and data-poor methods to generate a single automatic text categorizer, which returns a ranked list of ICD codes. The combined ranking system currently ob-tains a precision of 75% at high ranks and a recall of about 63% for the top twenty returned codes for a theoretical upper bound of about 79% (inter-coder agreement). The performance of the data-poor classifier is weak, whereas the use of tempo-rally-typed contents such as anamnesis and prescription free text sections results in a statistically significant improvement.

PURPOSE: We propose a question-answering (QA) driven generation approach for automatic acquisition of structured rules that can be used in a knowledge authoring tool for antibiotic prescription guidelines management. METHODS: The rule generation is seen as a question-answering problem, where the parameters of the questions are known items of the rule (e.g. an infectious disease, caused by a given bacterium) and answers (e.g. some antibiotics) are obtained by a question-answering engine. RESULTS: When looking for a drug given a pathogen and a disease, top-precision of 0.55 is obtained by the combination of the Boolean engine (PubMed) and the relevance-driven engine (easyIR), which means that for more than half of our evaluation benchmark at least one of the recommended antibiotics was automatically acquired by the rule generation method. CONCLUSION: These results suggest that such an automatic text mining approach could provide a useful tool for guidelines management, by improving knowledge update and discovery.

Khresmoi is a European Integrated Project developing a multilingual multimodal search and access system for medical and health information and documents. It addresses the challenges of searching through huge amounts of medical data, including general medical information available on the internet, as well as radiology data in hospital archives. It is developing novel semantic search and visual search techniques for the medical domain. At the MIE Village of the Future, Khresmoi proposes to have two interactive demonstrations of the system under development, as well as an overview oral presentation and potentially some poster presentations.