Paper Info
Title
Visualizing unstructured patient data for assessing diagnostic and therapeutic history.
Abstract
Having access to relevant patient data is crucial for clinical decision making. The data is often documented in unstructured texts and collected in the electronic health record. In this paper, we evaluate an approach to visualize information extracted from clinical documents by means of tag cloud. Tag clouds will be generated using a bag of word approach and by exploiting part of speech tags. For a real word data set comprising radiological reports, pathological reports and surgical operation reports, tag clouds are generated and a questionnaire-based study is conducted as evaluation. Feedback from the physicians shows that the tag cloud visualization is an effective and rapid approach to represent relevant parts of unstructured patient data. To handle the different medical narratives, we have summarized several possible improvements according to the user feedback and evaluation results.
Year
DOI
Venue
2014
10.3233/978-1-61499-432-9-1158
Studies in Health Technology and Informatics
Keywords
Field
DocType
Visualization,EHR,Text Mining,Tag Cloud
Data science,Data mining,World Wide Web,Medicine
Conference
Volume
ISSN
Citations 
205
0926-9630
0
PageRank 
References 
Authors
0.34
0
2
Name
Order
Citations
PageRank
Yihan Deng182.83
Kerstin Denecke214023.57