PROSISY: PRospective Stroke Identification SYstem Based on Cognitive Radio Theory and Machine Learning

• Autor: Nieto-Chaupis, Huber
• Formato: artículo
• Fecha de Publicación: 2020
• Institución: Universidad Autónoma del Perú
• Repositorio: AUTONOMA-Institucional
• Lenguaje: inglés
• OAI Identifier: oai:repositorio.autonoma.edu.pe:20.500.13067/1706
• Enlace del recurso: https://hdl.handle.net/20.500.13067/1706; https://doi.org/10.1109/CBMS49503.2020.00026
• Nivel de acceso: acceso restringido
• Materia: Sugar; Pollution measurement; Sensors; Cognitive radio; Task analysis; Surveillance; Internet; https://purl.org/pe-repo/ocde/ford#2.02.04

In this paper, a surveillance system expected to run in the prospective technology called Internet of Bio-Nano Things is presented. For this end the theory of Cognitive Radio as well as the Machine Learning criteria based on the hypothesis of Tom Mitchell are employed. In addition the Feynman's propagator model is also used. Essentially this paper focuses on the events where diabetes patients might have initialized a stroke event, so that the necessity to make the best decision is critic in order to guarantee a fast recover in the short term. Therefore this paper is focused on the following clinic variables: (i) cardiac pulse, (ii) blood pressure, (iii) glucose, and (iv) cholesterol. When all these variables are fully interconnected among them the full response might very encouraging in those cases where critic and non-critic patients might to anticipate unexpected events against their wellness in the shortest times in comparison with current systems.