To support an end to end Question and Answering system to help the clinical practitioners in a cardiovascular healthcare environment, an extended discourse representation structure CIDERS is introduced. This extension of the well-known DRT (Discourse Representation Theory) structures, go beyond single text representation extending them to embrace the general clinical history of a given patient. Introduced is a proposed and developed ontology framework, Ontology for General Clinical Practice, enhancing the currently available state-of-the-art ontologies for medical science and for the cardiovascular specialty, It's shown the scientific and philosophical reasons of its present dual structure with a deeply expressive (SHOIN) terminological base (TBox) and a highly computable (EL++) assertions knowledge base (ABox).

No abstract

We present an end to end Question and Answering system to help the clinical practitioners in a cardiovascular healthcare environment. We introduce our proposed ontology framework, Ontology for General Clinical Practice, that we developed enhancing the currently available state-of-the-art ontologies for medical science and for the cardiovascular specialty, extending upon the OBO Foundry principles. It's shown the scientific and philosophical reasons of its present dual structure with a deeply expressive (SHOIN) terminological base (TBox) and a highly computable (EL++) assertions knowledge base (ABox). The knowledge base is automatically populated by means of a tutored acquisition from clinical reports using Controlled Natural Language rendering an ontology driven question answering system with high recall, precision and F-Measure that competes in its specific sub-domain with the more advanced current NLP systems developed in general for the healthcare domain.

No abstract

We introduce the Ontology for General Clinical Practice (OGCP) for better knowledge representation support in the Clinical Practice domain. We followed the established OBO Foundry principles to leverage the ontological relations that might be present in the ontology axioms we harvest from clinical reports text segments. In accordance to the Ontological Realism principles we expect the reasoning inferred from the ontological relations to render more acceptable consequences then logical relations alone. We enhance the Ontology for General Medical Science (OGMS) with the Computer-Based Patient Record Ontology (CPR) structure and propose knowledge base creation/enhancing automatically extracting from clinical reports written in the, well known to the medical community, SOAP format. Reasoning over the resulting (OGCP) knowledge base with novel parallel algorithms that appeared recently in literature is presented. We finally propose Controlled Natural Language justifications of the inferred knowledge intending to achieve wider acceptance among clinicians. 1 MOTIVATION AND RESEARCH QUESTIONS Originally our research intention was the development of personal CDS 1 tools to help the healthcare professionals in scarce resource countries like most in Africa and Asia. After evaluating the State-of-the-Art presented ahead we found that relevant work is yet to be done in the KR 2 area regarding the Clinical Practice domain. We believe that some developments that have been achieved recently motivate us to incorporate our expertise in NLP 3 into effective ontology population. Our main intention is to be able to automatically produce clinical practice knowledge bases extracting from healthcare reports text. Research Questions. Ontologies in the sub-domain of Clinical Medicine 4 are lacking some thorough study. These can be stated as current problems for the effectiveness of using them as knowledge support for 1 Clinical Decision Support 2 Knowledge Representation 3 Natural Language Processing 4 The study of disease by direct examination of the living patient clinical reasoning. Problems found in current ontologies and enumerated in literature (Hoehndorf et al., 2011) that lead to reasoning hurdles are: • Lack of adequate modularization to achieve the minimum amount of implicit differentiation among primitive concepts.