Cum sociis natoque penatibus et magnis

Clinical Practice Ontology Automatic Learning from SOAP Reports

David Mendes, Irene Pimenta Rodrigues, Carlos Fernandes Baeta


Extensão da representação de Conhecimento orientado por ontologias Biomédicas.


We show how we implemented an end-to-end process to automatically develop a clinical practice knowledge base acquiring from SOAP notes. With our contribution we intend to overcome the “Knowledge Acquisition Bottleneck” problem by jump-starting the knowledge gathering from the most widely available source of clinical information that are natural language reports. We present the different phases of our process to populate automatically a proposed ontology with clinical assertions extracted from daily routine SOAP notes.

The enriched ontology becomes a reasoning able knowledge base that depicts accurately and realistically the clinical practice represented by the source reports. With this knowledge structure in place and novel state-of-the-art reasoning capabilities, based in consequence driven reasoners, a clinical QA system based in controlled natural language is introduced that reveals breakthrough possibilities regarding the applicability of Artificial Intelligence techniques to the medical field.

in Handbook of Research: Trends in the Diagnosis and Treatment of Chronic Conditions. pp 1-15, 2016.


David Mendes, Irene Pimenta Rodrigues, Carlos Fernandes Baeta.


Knowledge Acquisition Bottleneck; Ontology Learning:; Semantic Web; SOAP.