Automatic processing of diabetic patients' hospital documentation

Abstract

The paper presents a rule-based information extraction (IE) system for Polish medical texts. We select the most important information from diabetic patients' records. Most data being processed are free-form texts, only a part is in table form. The work has three goals: to test classical IE methods on texts in Polish, to create relational database containing the extracted data, and to prepare annotated data for further IE experiments.

Slavonic information extraction and partial parsing

“…A system similar in scope is SProUT (Becker et al, 2002), whose shallow parsing language allows to write regular grammars over HPSG-style (Pollard and Sag, 1994) typed feature structures and which includes the operation of unification. Preliminary work on adapting SProUT to the processing of Baltic and Slavonic languages is presented in (Drożdżyński et al, 2003), with much subsequent work devoted to the processing of Polish, especially, in the area of Information Extraction from medical texts Piskorski, 2004a;Piskorski, 2004b;Mykowiecka et al, 2005a;Mykowiecka et al, 2005b;Marciniak and Mykowiecka, 2007).…”