Produção de grafos de conhecimento e estratificação de riscos em saúde mental a partir de dicionários de dados semânticos

Abstract

Care teams use mental health risk stratification to determine patients' risk level. The data collected originates from closed or open questions, which generate structured and unstructured data. However, there is still a need to organize and prepare the data to answer different questions about the severity of the stratified signs and symptoms. Ontologies have been used to organize and describe data in order to facilitate the extraction of information and knowledge. They disambiguate terms and preserve the semantics d of the patient data bysemantically annotating the data with metadata. This paper examines how data on mental health risk stratification can be prepared in order to advance data analysis and the definition of levels of mental health care. To this end, an ontology of risk stratification is proposed and its use to annotate the semantics of qualitative and quantitative data with a Semantic Data Dictionary (SDD). SDD generates knowledge graphs from ontologies and some metadata templates. Graphs can be exploited to filter data and verify hypotheses. The graphs generated also facilitate the visualization of information and the discovery of relationships between, for example, diseases, signs and symptoms, serving as a tool to improve the stratification process, in addition to helping todefine levels of patient care.