Kemal Hakan Gülkesen1, Reinhold Haux1. 1. Peter L. Reichertz Institute for Medical Informatics, TU Braunschweig and Hannover Medical School, Braunschweig, Germany.
Abstract
OBJECTIVES: To identify major research subjects and trends in medical informatics research based on the current set of core medical informatics journals. METHODS: Analyzing journals in the Web of Science (WoS) medical informatics category together with related categories from the years 2013 to 2017 by using a smart local moving algorithm as a clustering method for identifying the core set of journals. Text mining analysis with binary counting of abstracts from these journals published in the years 2006 to 2017 for identifying major research subjects. Building clusters based on these terms for the complete time period as well as for the periods 2006-2008, 2009-2011, 2012-2014, and 2015-2017 for identifying trends. RESULTS: The identified cluster includes 17 core medical informatics journals. By text mining of these journals, 224,992 different terms in 14,414 articles were identified covering 550 specific key terms. Based on these key terms five clusters were identified: "Biomedical Data Analysis," "Clinical Informatics," "EHR and Knowledge Representation," "Mobile Health," and "Organizational Aspects of Health Information Systems." No shifts in the clusters were observed between the first two 3-year periods. In the third period, some terms like "mobile phone," "mobile apps," and "message" appear. Also, in the third period, a "Clinical Informatics" cluster appears and persists in the fourth period. In the fourth period, a rearrangement of clusters was observed. CONCLUSIONS: Beside classical subjects of medical informatics on organizing, representing, and analyzing data, we observed new developments in the context of mobile health and clinical informatics. These subjects tended to grow over the past years, and we can expect this trend to continue. Georg Thieme Verlag KG Stuttgart · New York.
OBJECTIVES: To identify major research subjects and trends in medical informatics research based on the current set of core medical informatics journals. METHODS: Analyzing journals in the Web of Science (WoS) medical informatics category together with related categories from the years 2013 to 2017 by using a smart local moving algorithm as a clustering method for identifying the core set of journals. Text mining analysis with binary counting of abstracts from these journals published in the years 2006 to 2017 for identifying major research subjects. Building clusters based on these terms for the complete time period as well as for the periods 2006-2008, 2009-2011, 2012-2014, and 2015-2017 for identifying trends. RESULTS: The identified cluster includes 17 core medical informatics journals. By text mining of these journals, 224,992 different terms in 14,414 articles were identified covering 550 specific key terms. Based on these key terms five clusters were identified: "Biomedical Data Analysis," "Clinical Informatics," "EHR and Knowledge Representation," "Mobile Health," and "Organizational Aspects of Health Information Systems." No shifts in the clusters were observed between the first two 3-year periods. In the third period, some terms like "mobile phone," "mobile apps," and "message" appear. Also, in the third period, a "Clinical Informatics" cluster appears and persists in the fourth period. In the fourth period, a rearrangement of clusters was observed. CONCLUSIONS: Beside classical subjects of medical informatics on organizing, representing, and analyzing data, we observed new developments in the context of mobile health and clinical informatics. These subjects tended to grow over the past years, and we can expect this trend to continue. Georg Thieme Verlag KG Stuttgart · New York.