ABSTRACT
Objective To identify the frontiers and hotspots in studies on lung cancer therapy by text mining. Methods PubMed-covered papers on lung cancer therapy were retrieved from 2013 to 2014 . The terms with a broa-der meaning were excluded by mapping the UMLS concept with MetaMap and by limiting their semantic types. A LDA model was established to identify the topics. Results The LDA model could identify the frontiers and hotspots in studies on lung cancer therapy from 2013 to 2014 . Conclusion Frontiers and hotspots in studies on lung cancer therapy can be identified by analyzing the topics and reading the related literature, which can thus provide reference for related medical researchers and managers.
ABSTRACT
OBJECTIVE: For the effective retrieval of clinical information, the elaborate indexing is essential. Two major types of indexing are the human indexing and the automatic or machine indexing. Human indexing shows higher quality but is time consuming, labor-intensive and inconsistent in term assignment activity. METHODS: Using the Unified Medical Language System (UMLS) MetaMap program, we mapped the free text from the diagnosis section of radiology reports into UMLS concepts. To improve the precision of UMLS concept indexing by MetaMap, we evaluated the UMLS subset mapping and semantic type filtering methods, determining the best combination for improved precision. RESULTS: After calculating the candidates from subset combinations, we obtained more enhanced results by semantic-type filtering. CONCLUSION: The results may be improved for the complete automation of indexing process.
Subject(s)
Humans , Abstracting and Indexing , Automation , Diagnosis , Semantics , Unified Medical Language SystemABSTRACT
The UMLS is a long-term NLM research and development effort designed to facilitate the retrieval and integration of information from multiple machine-readable biomedical information sources. It is one of the most comprehensive medical terminology systems. The UMLS is to make available to develop systems such as information retrieval, decision-making support system, natural language processing, and voice recognition in medical fields because it can be easily integrated with other systems or databases. It is now support about 15 languages like French, German, Finnish, and Spanish. The goal of this paper is to estimate the availabilities and usefulness that when we will apply the electronic medical records system and then support to Korean medical vocabulary integration. We tested mapping between chief complaint extracted from the discharge records in Seoul National University Hospital and the UMLS sign or symptom concepts term. Among 35% of chief complaint of the SNUH were Conceptually matched with the UMLS Sign or Symptom concepts. Most of 58% these terms were such that diagnosis, operation names, clinical laboratory test not Sign or Symptom terms. The rest terms of 7% were not found in the UMLS or these terms that different application of used concepts. While the UMLS terms are so specific and diverse about patients sign or symptom, we used simplified and usually based on the anatomical region records. Through this study, we exposed underlying of flaws currently in use clinical terms and analyzed the differences of expression patterns in clinical records between the two vocabulary system, we propose the UMLS application of Korean medical vocabulary translation with appropriateness and Interoperability of the UMLS for development the Korean medical technologies.