The common approach to the analysis of natural texts implies that semantic analysis should following the stage of parsing. However, medical texts are known as very complicated and written in a very specific language. Traditional parsers are demonstrating relatively small productivity here. In this article, we are demonstrating an opposite approach: ontology-based entailing of words in combination with simple shallow parsing rules. It allows us to increase UAS metrics from 0.82 for SpaCy to 0.834 for our approach.

Chomsky Was (Almost) Right: Ontology-Based Parsing of Texts of a Narrow Domain

Medical records contain a textual description of such important information as patients’ complaints, diseases progress and therapy. An extraction of this information could allow starting with processing information stored in medical databases. In this article we introduce a short description of a medical ontology storing information on patients’ complaints. We also describe an algorithm that uses this ontology for extraction of claims from texts of medical records. The algorithm combines both syntactic properties, and peculiarities, of a text and connections between diseases’ properties and their values. The algorithm corrects syntactical mistakes according to the hierarchical information from the ontology. The resulting algorithm was proved on 3000 clinical records of Department of Neurosurgery of FEFU.

Tatiana Aleksandrovna Gorbach

Papers

Formalization of Medical Records Using an Ontology: Patient Complaints

Chomsky Was (Almost) Right: Ontology-Based Parsing of Texts of a Narrow Domain