In this study we propose the use of text mining and machine learning methods to predict and detect Surgical Site Infections (SSIs) using textual descriptions of surgeries and post-operative patients’ records, mined from the database of a high complexity University hospital. SSIs are among the most common adverse events experienced by hospitalized patients; preventing such events is fundamental to ensure patients’ safety. Knowledge on SSI occurrence rates may also be useful in preventing future episodes. We analyzed 15,479 surgery descriptions and post-operative records testing different preprocessing strategies and the following machine learning algorithms: Linear SVC, Logistic Regression, Multinomial Naive Bayes, Nearest Centroid, Random Forest, Stochastic Gradient Descent, and Support Vector Classification (SVC). For prediction purposes, the best result was obtained using the Stochastic Gradient Descent method (79.7% ROC-AUC); for detection, Logistic Regression yielded the best performance (80.6% ROC-AUC).
Resumo-Este artigo propõe uma extensão à Ontologia do Orçamento Federal Brasileira, hoje disponível no regime de Dados Abertos, de modo a integrar a tal ontologia conceitos de geolocalização que permitem a tomada de decisão no domínio orçamentário, considerando as características regionais do país. A extensão proposta envolve a transformação dos dados relacionais de localizadores, elemento de dados chave do modelo orçamentário, associando-os ao conceito de geolocalização de Regiões do Brasil, para uso na composição das triplas RDF publicadas como dados abertos. Para avaliar o efeito dessa adição, este trabalho analisa os resultados obtidos pela consulta de localizadores relacionados à educação e saúde, comparando os índices com aqueles da tendência extraída do Índice de Desenvolvimento Humano correspondente, concluindo com uma análise da tendência de destinação regionalizada de recursos.Palavras-chave: ontologia; finanças; orçamento; governo eletrônico; dados abertos.Abstract-This paper proposes an extension to the Ontology of the Brazilian Federal Budget, available today as Open Data, in order to integrate to this ontology geolocation concepts that enable decision making in the budgetary field, considering the regional characteristics of the country. The proposed extension involves the transformation of relational data representing locators which are key elements of the budget data model, linking them to the concept of geo-regions of Brazil, so as to use those locators in the composition of RDF triples published as open data. To evaluate the effect of this addition, this paper analyzes the results obtained by consulting locators related to education and health, comparing the obtained rates with those of the trends extracted from the corresponding Human Development Index, and concluding with a trend analysis regarding the regionalized allocation of budgetary resources.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.