Data processing from mHealth patient data acquisition related to extracting structured data from EH records

Abstract

Application of mobile devices in healthcare is a pervasive way how to asses patient health status. Adding just another data source to overwhelmed physician requires technologies for their effective processing. Despite the fact that the problem of extracting clinical information from free-text health reports for computerized applications or for decision support systems is a lot discussed issue, it is still complicated and unresolved question. In general Natural language processing (NLP) systems are implemented to solve the task. However NLP system works only for relatively narrow clinical domains because the format of the language used in the reports is not standardized and the reports vary depending on the domain. We describe methodology suggested for extracting structured data from EHR focusing especially on EHR in Slovak language. Further, we discuss problems concerning a task of extracting required data from free-text health reports. In the conclusion we present test results and possible implementations. © Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2013.