Documenting and predicting topic changes in Computers in Biology and Medicine: A bibliometric keyword analysis from 1990 to 2017. (2018)

• Record Type: Journal Article
• Authors: Faust, Oliver
• Is Part Of: Informatics in medicine unlocked. Volume 11(2018)
• Page Start: 15
• Page End: 27
• Publication Date: 2018
• Subjects: Research trends -- Topic analysis -- Topic detection and tracking -- Text mining -- Computers in biology and medicine; Medical informatics -- Periodicals; 610.285
• DOI: 10.1016/j.imu.2018.03.002
• Languages: English
• ISSNs: 2352-9148

Abstract

Abstract: The Computers in Biology and Medicine (CBM) journal promotes the use of computing machinery in the fields of bioscience and medicine. Since the first volume in 1970, the importance of computers in these fields has grown dramatically, this is evident in the diversification of topics and an increase in the publication rate. In this study, we quantify both change and diversification of topics covered in. This is done by analysing the author supplied keywords, since they were electronically captured in 1990. The analysis starts by selecting 40 keywords, related to Medical (M) (7), Data (D) (10), Feature (F) (17) and (AI) (6) methods. Automated keyword clustering shows the statistical connection between the selected keywords. We found that the three most popular topics in CBM are: Support Vector Machine (SVM), Electroencephalography (EEG) and IMAGE PROCESSING. In a separate analysis step, we bagged the selected keywords into sequential one year time slices and calculated the normalized appearance. The results were visualised with graphs that indicate the CBM topic changes. These graphs show that there was a transition from Artificial Neural Network (ANN) to SVM. In 2006 SVM replaced ANN as the most important AI algorithm. Our investigation helps the editorial board to manage and embrace topic change. Furthermore, our analysis is interesting for the general reader, as the results can help them to adjust their research directions.