H-KaaS: uma arquitetura de referência baseada em conhecimento como serviço para e-saúde

Abstract

Facing the need to improve access to knowledge and the establishment of means for sharing and organizing data in the health domain, this research describes a reference architecture based on the paradigm Knowledge as a Service (KaaS), which can be used to assist in clinical diagnosis and decision. The proposed reference architecture, called H-KaaS, offers centralized access to services based on ontologies, knowledge discovery in databases, machine learning, and other means of representing knowledge and reasoning. For this, a detailed description of the architecture is presented, providing implementation examples, and highlighting its main components and communication interfaces. In order to validate the work, two case studies in knowledge-based services were performed, instantiating each service according to the proposed reference architecture and using different means of knowledge representation and extraction. Thus, the development of this research contributed to the creation of a new reference architecture in the health domain able to manage multiple data sources and knowledge models, facilitating its usage and sharing, thus helping to advance the state of the art of health informatics and applications of artificial intelligence and its knowledge-based systems.