Visualization of e-Health Research Topics and Current Trends Using Social Network Analysis.

BACKGROUND: E-health has been grown rapidly with significant impact on quality and safety of healthcare. However, there is a large gap between the postulated and empirically demonstrated benefits of e-health technologies and a need for a clearer mapping of its conceptual domains. Therefore, this study aimed to critically review the main research topics and trends of international e-health through social network analysis. MATERIALS AND METHODS: Medical subject heading terms were used to retrieve 3,023 research articles published from 1979 through 2014 in the PubMed database. We extracted n-grams from the corpus using a text analysis program, generated co-occurrence networks, and then analyzed and visualized the networks using Pajek software. The hub and authority measures identified the most important research topics in e-health. Newly emerging topics by 4-year period units were identified as research trends. RESULTS: The most important research topics in e-health are personal health records (PHR), health information technology, primary care, mobile health, clinical decision support systems (CDSS), and so on. The eight groups obtained through ego network analysis can be divided into four semantically different areas, as follows: information technology, infrastructure, services, and subjects. Also, four historical trends in e-health research are identified: the first focusing on e-health and telemedicine; the second, PHR and monitoring; the third, CDSS and alert; and the fourth, mobile health and health literacy. CONCLUSIONS: This study promotes a systematic understanding of e-health by identifying topic networks, thereby contributing to the future direction of e-health research and education.

Structuralizing biomedical abstracts with discriminative linguistic features.

OBJECTIVE: Nearly 75% of the abstracts in MEDLINE papers present in an unstructured format. This study aims to automate the reformatting of unstructured abstracts into the Introduction, Methods, Results, and Discussion (IMRAD) format. The quality of this reformatting relies on the features used in sentence classification. Therefore, we explored the most effective linguistic features in MEDLINE papers. METHODS: We constructed a feature set consisting of bag of words, linguistic features, grammatical features, and structural features. In order to evaluate the effectiveness, which is the capability of the sentence classification with the features, three datasets from PubMed Central Open Access Subset were selected and constructed: (1) structured abstract (SA) for training, (2) unstructured RCT abstract (UA-1) and (3) unstructured general abstract (UA-2). F-score and accuracy were used to measure the effectiveness on IMRAD section level and the overall classification. RESULTS: Adding linguistic features improves the classification of the abstract sentence from 1.2% to 35.8% in terms of accuracy in three abstract datasets. The highest accuracies achieved were 91.7% in SA, 86.3% in UA-1, and 77.9% in UA-2. Linguistic features (dimensions=15) had fewer dimensions than bag-of-words (dimensions= 1541). All representative linguistic features (n-gram and verb phrase, and noun phrase) for each section are identified in our system (available at http://abstract.bike.re.kr). CONCLUSION: Linguistic features can be used to effectively classify sentence with low computation burden in MEDLINE abstract.

Clinical Data Element Ontology for Unified Indexing and Retrieval of Data Elements across Multiple Metadata Registries.

OBJECTIVES: Classification of data elements (DEs), which is used in clinical documents is challenging, even in across ISO/IEC 11179 compliant clinical metadata registries (MDRs) due to no existence of reliable standard for identifying DEs. We suggest the Clinical Data Element Ontology (CDEO) for unified indexing and retrieval of DEs across MDRs. METHODS: The CDEO was developed through harmonization of existing clinical document models and empirical analysis of MDRs. For specific classification as using data element concept (DEC), The Simple Knowledge Organization System was chosen to represent and organize the DECs. Six basic requirements also were set that the CDEO must meet, including indexing target to be a DEC, organizing DECs using their semantic relationships. For evaluation of the CDEO, three indexers mapped 400 DECs to more than 1 CDEO term in order to determine whether the CDEO produces a consistent index to a given DEC. The level of agreement among the indexers was determined by calculating the intraclass correlation coefficient (ICC). RESULTS: We developed CDEO with 578 concepts. Through two application use-case scenarios, usability of the CDEO is evaluated and it fully met all of the considered requirements. The ICC among the three indexers was estimated to be 0.59 (95% confidence interval, 0.52-0.66). CONCLUSIONS: The CDEO organizes DECs originating from different MDRs into a single unified conceptual structure. It enables highly selective search and retrieval of relevant DEs from multiple MDRs for clinical documentation and clinical research data aggregation.

[A social network analysis of research topics in Korean nursing science].

PURPOSE: This study was done to explore the knowledge structure of Korean Nursing Science. METHODS: The main variables were key words from the research papers that were presented in the Journal of Korean Academy of Nursing and journals of the seven branches of the Korean Academy of Nursing. English titles and abstracts of the papers (n=5,936) published from 1995 through 2009 were included. Noun phrases were extracted from the corpora using an in-house program (BiKE Text Analyzer), and their co-occurrence networks were generated via a cosine similarity measure, and then the networks were analyzed and visualized using Pajek, a Social Network Analysis program. RESULTS: With the hub and authority measures, the most important research topics in Korean Nursing Science were identified. Newly emerging topics by three-year period units were observed as research trends. CONCLUSION: This study provides a systematic overview on the knowledge structure of Korean Nursing Science. The Social Network Analysis for this study will be useful for identifying the knowledge structure in Nursing Science.

Knowledge structure of korean medical informatics: a social network analysis of articles in journal and proceedings.

OBJECTIVES: This study aimed at exploring the knowledge structure of Korean medical informatics. METHODS: We utilized the keywords, as the main variables, of the research papers that were presented in the journal and symposia of the Korean Society of Medical Informatics, and we used, as cases, the English titles and abstracts of the papers (n = 915) published from 1995 through 2008. N-grams (bigram to 5-gram) were extracted from the corpora using the BiKE Text Analyzer, and their cooccurrence networks were generated via a cosine correlation coefficient, and then the networks were analyzed and visualized using Pajek. RESULTS: With the hub and authority measures, the most important research topics in Korean medical informatics were identified. Newly emerging topics by three-year period units were observed as research trends. CONCLUSIONS: This study provides a systematic overview on the knowledge structure of Korean medical informatics.