Similarity of Medical Cases in Health Care Using Cosine Similarity and Ontology

The increasing use of digital patient records in hospital saves time and reduces risks of wrong treatments caused by lack of information. Digital patient records also enable efficient spread and transfer of experience gained from diagnosis and treatment of individual patient which is now-a-days mostly manual (speaking with colleagues) and rarely aided by computerized system. Most of the content in patient records is semi-structured textual information. In this paper, we propose a hybrid textual case-based reasoning system promoting experience reuse. This is derived from structured or unstructured patient records, case-based reasoning and similarity measurement based on cosine similarity metric improved by a domain specific ontology and the nearest neighbor method. As a result, hospital staffs can learn not only new cases but also add comments to existing cases and thus enables prototypical cases.