| Abstract |
Evidence-based medical guidelines should be regularly updated, so that they can serve medical practice using latest medical research evidence. A usual approach to detecting new evidences is to use a set of terms which appear in a guideline statement and create queries over a bio-medical search engine such as PubMed with a ranking over a selected subset of terms to search for relevant evidences. However, the sizes of found relevant evidences are usually very large (i.e. over a few hundreds, even thousands). Namely the precision of the searches is low. That would make medical professionals quite difficult to find which evidences are really interesting and useful. In my presentation, we propose an approach to detecting new evidences for evidence-based medical guideline update with journal filtering, so that the evidences appear in those top journals are preferred. We will report our experiments with journal filtering and show that this new approach can indeed gain a much higher precision (69.73% difference) with a slightly lower recall (14.29% difference) for detecting new evidences. |