Modelos de Aprendizado de Máquina na Predição de Diabetes Tipo 1 na Gestação usando Dados do Sistema Único de Saúde

• Jorge R. H. Moreira UFJF
• Heder S. Bernardino UFJF
• Helio J. C. Barbosa UFJF / LNCC
• Alex B. Vieira UFJF

DOI: https://doi.org/10.5753/sbcas.2021.16082

Abstract

Pre-existing or developed diabetes mellitus during pregnancy can bring serious health risks to the pregnant woman and the baby, throughout the pregnancy-puerperal cycle. In this sense, predicting the presence of one of the types of diabetes, even before its first symptoms, can generate positive impacts on public health systems. This work generates a classification model, using data from the outpatient production of the Unified Health System, which can predict the presence of one of the types of diabetes (type 1), according to the characteristics and history of monitoring the pregnant patient. The proposed classifier is capable of separating pregnant women in the health system according to their predisposition to the disease, making it possible to generate an alert in the system and direct attention to the monitoring of pregnant women in the context of this health condition. The results obtained were relevant, presenting sensitivity and precision greater than 90%. Thus, it is believed that the proposed model can be another resource for improving the system.

References

Acharjee, S., Ghosh, B., Al-Dhubiab, B., and Nair, A. (2013). Understanding type 1 diabetes: etiology and models. Canadian Journal of Diabetes, 37(4):269–276.

Albuquerque, J. P. D., Vasques, E. P., and Machado, G. R. (2011). Ambivalent implications of health care information systems: A study in the Brazilian public health care system. Revista de Administração de Empresas, 51.

Beeck, A. and Eizirik, D. (2016). Viral infections in type 1 diabetes mellitus why the cells? Nature Reviews Endocrinology, 12:263–273.

Cavalcante, G. L., Moura, M. C. L., Lago, A. F. V., and de Oliveira Lima, J. V. (2020). Perl farmacoepidemiológico de pacientes com diabetes mellitus tipo 1. Research, Society and Development, 9:184953361.

Chaves, L. and Marques, G. (2021). Data Mining Techniques for Early Diagnosis of Diabetes: A Comparative Study. Applied Sciences, 11(5).

Di Nubila, H. B. V. and Buchalla, C. M. (2008). O papel das Classicações da OMS-CID e CIF nas denições de deciência e incapacidade. Revista Brasileira de Epidemiologia, 11(2):324–335.

Feitosa, A. C. R. and íAvila, A. N. (2016). Uso do prontuário eletrônico na assistência pré-natal às portadoras de diabetes na gestação. Revista Brasileira de Ginecologia e Obstetrícia, 38:9 – 19.

IDF (2015). IDF Diabetes Atlas. 7th ed. Brussels: International Diabetes Federation. http://www.diabetesatlas.org, Acessado em: 24/09/2020.

IDF (2020). What is diabetes. International Diabetes Federation.

Javad, M. O. M., Agboola, S. O., Jethwani, K., Zeid, A., and Kamarthi, S. (2019). A Reinforcement Learning-Based Method for Management of Type 1 Diabetes: Exploratory Study. JMIR Diabetes.

Mainenti, G., Campanile, L., Marulli, F., Ricciardi, C., and Valente, A. S. (2020). Machine learning approaches for diabetes classication: Perspectives to articial intelligence methods updating. In 5th International Conference on Internet of Things, Big Data and Security (IoTBDS 2020), pages 533–540.

MGH (2020). Massachusetts General Hospital. Disponível em: https://www.massgeneral.org. Acessado em: 04/09/2020.

Ministério da Saúde (2006). Caderno de atenção básica. diabetes mellitus. n.16. série a. normas e manuais técnicos brasília.

Ministério da Saúde (2012). Departamento de Atenção Básica. Atenção ao Pré-Natal de Baixo Risco. Normas e Manuais Técnicos, Cadernos de Atenção Básica, n° 32. Brasília, DF.

Moulin, C. M., Portella, R. B., Pinheiro, V. S., Oliveira, M. M., Fuks, A. G., Cunha, E. F., and Gomes, M. B. (2003). Prevalência de sobrepeso e obesidade em pacientes com diabetes tipo 1. Arquivos Brasileiros de Endocrinologia & Metabologia, 47.

OPAS (2016). Organização Pan-Americana da Saúde. Ministério da Saúde. Federação Brasileira das Associações de Ginecologia e Obstetrícia. Sociedade Brasileira de Diabetes. Rastreamento e diagnóstico de diabetes mellitus gestacional no Brasil. Brasília.

Rodríguez-Rodríguez, I., Rodríguez, J.-V., Woo, W. L., Wei, B., and Pardo-Quiles, D.-J. (2021). A Comparison of Feature Selection and Forecasting Machine Learning Algorithms for Predicting Glycaemia in Type 1 Diabetes Mellitus. Applied Sciences, 11(4).

Sociedade Brasileira de Diabetes (2019). Diretrizes Brasileiras de Diabetes 2019-2020.

Taser, P. Y. (2021). Application of bagging and boosting approaches using decision tree based algorithms in diabetes risk prediction. Proceedings, 74(1).

UCI (2020). UC Irvine Machine Learning Repository . Disponível em: https://archive.ics.uci.edu/ml/datasets/Early+stage+diabetes+risk+prediction+dataset.

WHO (2019). International Statistical Classification of Diseases and Related Health Problems 10th Revision. World Health Organization.

WHO (2020a). Ficha T´ecnica Diabetes.World Health Organization. www.who.int/newsroom/fact-sheets/detail/diabetes, Acessado em: 25/03/2021.

WHO (2020b). Int. Statistical Classification of Diseases and Related Health Problems (ICD). https://www.who.int/standards/classifications/classification-of-diseases. Acessado em: 24/09/2020.