An automatic clinical document importance estimator for an existing electronic patient record — Architecture and implementation

The goal of the OPTIM project is to optimize the graphical user interface of an electronic health record (EHR) by predicting clinical documents' relevance and provide a ranked list of relevant documents for the given user at a certain time. This paper describes the architecture of the relevance assignment and ranking prototype and some implementation issues. The prototype's design is based on two components: OPTIM Core, with logical representation, estimation server's integration and the webservice layer, and the OPTIM WebUI, with the user interface for presenting the results. The prototype was tested in integration with an EHR using a simulated environment. The results were encouraging but yet they revealed a certain lack of security (confidentiality). It has now the capacity of rating 10 documents per second. Nonetheless, the integration of features such as rating clinical relevance based on mathematical models can be included in existing EHR potentially improving their usability.