Philipp Daumke

Subword-based Semantic Retrieval of Clinical and Bibliographic Documents

OBJECTIVES The increasing amount of electronically available documents in bibliographic databases and the clinical documentation requires user-friendly techniques for content retrieval. METHODS A domain-specific approach on semantic text indexing for document retrieval is presented. It is based on a subword thesaurus and maps the content of texts in different European languages to a common interlingual representation, which supports the search across multilingual document collections. RESULTS Three use cases are presented where the semantic retrieval method has been implemented: a bibliographic database, a department EHR system, and a consumer-oriented Web portal. CONCLUSIONS It could be shown that a semantic indexing and retrieval approach, the performance of which had already been empirically assessed in prior studies, proved useful in different prototypical and routine scenarios and was well accepted by several user groups.

The Clinical Data Intelligence Project

This article is about a new project that combines clinical data intelligence and smart data. It provides an introduction to the “Klinische Datenintelligenz” (KDI) project which is founded by the Federal Ministry for Economic Affairs and Energy (BMWi); we transfer research and development results (R&D) of the analysis of data which are generated in the clinical routine in specific medical domain. We present the project structure and goals, how patient care should be improved, and the joint efforts of data and knowledge engineering, information extraction (from textual and other unstructured data), statistical machine learning, decision support, and their integration into special use cases moving towards individualised medicine. In particular, we describe some details of our medical use cases and cooperation with two major German university hospitals.

Biomedical information retrieval across languages

This work presents a new dictionary-based approach to biomedical cross-language information retrieval (CLIR) that addresses many of the general and domain-specific challenges in current CLIR research. Our method is based on a multilingual lexicon that was generated partly manually and partly automatically, and currently covers six European languages. It contains morphologically meaningful word fragments, termed subwords. Using subwords instead of entire words significantly reduces the number of lexical entries necessary to sufficiently cover a specific language and domain. Mediation between queries and documents is based on these subwords as well as on lists of word-n-grams that are generated from large monolingual corpora and constitute possible translation units. The translations are then sent to a standard Internet search engine. This process makes our approach an effective tool for searching the biomedical content of the World Wide Web in different languages. We evaluate this approach using the OHSUMED corpus, a large medical document collection, within a cross-language retrieval setting.

Cross-language mining for acronyms and their completions from the web

We propose a method that aligns biomedical acronyms and their long-form definitions across different languages. We use a freely available search and extraction tool by which abbreviations, together with their fully expanded forms, are massively mined from the Web. In a subsequent step, language-specific variants, synonyms, and translations of the extracted acronym definitions are normalized by referring to a language-independent, shared semantic interlingua.

Intelligent Healthcare Applications

Currently three client systems are provided by Medico: the stationary clinical radiological workplace which consists of a semantic annotation prototype to be used as a semantic reporting tool for the radiologists in-daily routine; the semantic search prototype; and a mobile system. The semantic annotation tool is designed to enable the physician to validate and complete the automatically generated semantic annotations, whereas the semantic search prototype is to be used to ensure a diagnosis by searching for similar cases in medical databases. The mobile application (Usage of mobile applications for medical purposes might be restricted in several countries.) runs on the iPad and connects to the Semantic Server of Medico, which enables the radiologist to quickly view and manipulate the semantic annotations anywhere, e.g., at the bedside, using touch screen gestures and speech.

MorphoSaurus in ImageCLEF 2006: the effect of subwords on biomedical IR

In the 2006 ImageCLEF Medical Image Retrieval task we evaluate the effects of deep morphological analysis for mono-and cross-lingual document retrieval in the biomedical domain. The morphological analysis is based on the MorphoSaurus system in which subwords are introduced as morphologically meaningful word units. Subwords are organized in language specific lexica that were partly manually and partly automatically generated and currently cover six European languages. They are linked together in a multilingual thesaurus. The use of subwords instead of full words significantly reduces the number of lexical entries that are needed to sufficiently cover a specific language and domain. A further benefit of the approach is its independence from the underlying retrieval system. We combined MorphoSaurus with the open-source search engine Lucene and achieved precision gains of up to 25% over the baseline for a monolingual setting and promising results in a multilingual scenario.

Semantic Processing of Medical Data

Medical images increase in quality and quantity: More and more detailed image content can be represented on the pixel level, and increasing amounts of medical images are produced in the context of clinical diagnosis. Technological solutions are needed to enhance existing clinical IT solutions helping clinicians to access and use medical images optimally. Within Medico, we developed methods and tools (a) to parse and describe the content of medical images, (b) to extract and annotate the related information from radiology reports, and (c) to provide and manage medical ontologies as a common language for labeling and integrating the various information entities.

Smart Medical Information Technology for Healthcare (SMITH)

Summary Introduction: This article is part of the Focus Theme of Methods of Information in Medicine on the German Medical Informatics Initiative. “Smart Medical Information Technology for Healthcare (SMITH)” is one of four consortia funded by the German Medical Informatics Initiative (MI-I) to create an alliance of universities, university hospitals, research institutions and IT companies. SMITH’s goals are to establish Data Integration Centers (DICs) at each SMITH partner hospital and to implement use cases which demonstrate the usefulness of the approach. Objectives: To give insight into architectural design issues underlying SMITH data integration and to introduce the use cases to be implemented. Governance and Policies: SMITH implements a federated approach as well for its governance structure as for its information system architecture. SMITH has designed a generic concept for its data integration centers. They share identical services and functionalities to take best advantage of the interoperability architectures and of the data use and access process planned. The DICs provide access to the local hospitals’ Electronic Medical Records (EMR). This is based on data trustee and privacy management services. DIC staff will curate and amend EMR data in the Health Data Storage. Methodology and Architectural Framework: To share medical and research data, SMITH’s information system is based on communication and storage standards. We use the Reference Model of the Open Archival Information System and will consistently implement profiles of Integrating the Health Care Enterprise (IHE) and Health Level Seven (HL7) standards. Standard terminologies will be applied. The SMITH Market Place will be used for devising agreements on data access and distribution. 3LGM 2 for enterprise architecture modeling supports a consistent development process. The DIC reference architecture determines the services, applications and the standards-based communication links needed for efficiently supporting the ingesting, data nourishing, trustee, privacy management and data transfer tasks of the SMITH DICs. The reference architecture is adopted at the local sites. Data sharing services and the market place enable interoperability. Use Cases: The methodological use case “Phenotype Pipeline” (PheP) constructs algorithms for annotations and analyses of patient-related phenotypes according to classification rules or statistical models based on structured data. Unstructured textual data will be subject to natural language processing to permit integration into the phenotyping algorithms. The clinical use case “Algorithmic Surveillance of ICU Patients” (ASIC) focusses on patients in Intensive Care Units (ICU) with the acute respiratory distress syndrome (ARDS). A model-based decision-support system will give advice for mechanical ventilation. The clinical use case HELP develops a “hospital-wide electronic medical record-based computerized decision support system to improve outcomes of patients with blood-stream infections” (HELP). ASIC and HELP use the PheP. The clinical benefit of the use cases ASIC and HELP will be demonstrated in a change of care clinical trial based on a step wedge design. Discussion: SMITH’s strength is the modular, reusable IT architecture based on interoperability standards, the integration of the hospitals’ information management departments and the public-private partnership. The project aims at sustainability beyond the first 4-year funding period.

A CLIR interface to a web search engine

Medical document retrieval presents a unique combination of challenges for the design and implementation of retrieval engines. We introduce a method to meet these challenges by implementing a multilingual retrieval interface for biomedical content in the World Wide Web. To this end we developed an automated method for interlingual query construction by which a standard Web search engine is enabled to process non-English queries from the biomedical domain in order to retrieve English documents.