ONTODRC rompendo o silêncio da doença renal crônica

• Autor(a) principal: Gomes, Cecília Neta Alves Pegado
• Data de Publicação: 2018
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da UFPB
• Texto Completo: https://repositorio.ufpb.br/jspui/handle/123456789/13001
• topic: Doença renal crônica; Inteligência artificial; Ontologia biomédica; Chronic kidney disease; Artificial intelligence; Biomedical ontology; CNPQ::CIENCIAS DA SAUDE::SAUDE COLETIVA

Chronic Kidney Disease (CKD) is a public health problem that affects nearly 10% of the global adult population, with a mortality rate of 15% per year. This disease is classified in five progressive stages (1>2>3>4>5), and it is shocking to know that 70% of the patients that start Kidney Replacement Therapy (KRT), only in the last stage, do not know they were previously infected by it. It is clear that those patients in the early stages of CKD are underdiagnosed and deprived of health policies to optimize diagnosis and to warn them of the need for KRT and of the catastrophic cardiovascular outcomes that are the main cause of death in this segment. The barriers identified for kidney care were the factors related to the knowledge and attitude of physicians and patients, besides the geography. In this context, there is a need to develop well-designed strategies to guide decision-making and enhance the care for patients with CKD. In this, work, we aimed to develop an Artificial Intelligence tool called ontology, with a view to optimizing the early diagnosis of Chronic Kidney Disease in Primary Health Care. For this purpose, after acquiring knowledge about the disease domain, we drew up the global guideline for the assistance of the patient with kidney disease in the form of a Rule-Based System that was subsequently implemented manually following the model 101 for construction of ontologies in the Protégé software, an ontology editor, having as its reasoning element the Hermit, for the necessary inferences. Accordingly, the constructed ontology called ONTODRC is characterized as a Clinical Decision Support System that has been validated computationally and in real HULW cases. After validating ONTODRC, we applied it in a sample of 185 primary care physicians in the town of João Pessoa, in two moments. In the first one, we applied a questionnaire to assess the pre and post knowledge on ONTODRC; and, in the second, we measured the perception of ease and usefulness of the tool with the technology acceptance model (TAM). In order to assess the effect of the intervention in the Knowledge assessment, we used the McNemar’s Test; and, to check the reliability of the TAM constructs, we used the Cronbach’s Alpha indicator (C.A.). In the results, we noted that the constructed ontology has an approximate capacity to respond to 90% of the surveyed requirements and also has the capacity to bring knowledge to the users. In turn, these users have considered the tool as useful and easy in their daily lives. We can conclude that the ONTODRC has the capacity to optimize the early diagnosis of CKD, in order to provide human, economic and environmental benefits.