On Building and Evaluating a Medical Records Exploration Interface Using Text Mining Techniques

Abstract

Medical records contain many terms that are difficult to process. Our aim in this study is to allow visual exploration of the information in medical databases where texts present a large number of syntactic variations and abbreviations by using an interface that facilitates content identification, navigation, and information retrieval. We propose the use of multi-term tag clouds as content representation tools and as assistants for browsing and querying tasks. The tag cloud generation is achieved by using a novelty mathematical method that allows related terms to remain grouped together within the tags. To evaluate this proposal, we have carried out a survey over a spanish database with 24,481 records. For this purpose, 23 expert users in the medical field were tasked to test the interface and answer some questions in order to evaluate the generated tag clouds properties. In addition, we obtained a precision of 0.990, a recall of 0.870, and a F1-score of 0.904 in the evaluation of the tag cloud as an information retrieval tool. The main contribution of this approach is that we automatically generate a visual interface over the text capable of capturing the semantics of the information and facilitating access to medical records, obtaining a high degree of satisfaction in the evaluation survey.

Keywords

• knowledge representation
• electronic health records
• health information systems
• content identification
• visual interface