Potential Confounders in the Analysis of Brazilian Adolescent's Health: A Combination of Machine Learning and Graph Theory

Autor(es)

Ambriola Oku, Amanda Yumi ^[1] ; Zimeo Morais, Guilherme Augusto ^[2] ; Arantes Bueno, Ana Paula ^[1] ; Fujita, Andre ^[3] ; Sato, Joao Ricardo ^[1]

Afiliação do(s) autor(es)

^[1] Univ Fed ABC, Ctr Math Comp & Cognit, BR-09210580 Santo Andre, SP - Brazil

^[2] Big Data Hosp Israelita Albert Einstein, BR-05652900 Sao Paulo - Brazil

^[3] Univ Sao Paulo, Inst Math & Stat, BR-05508090 Sao Paulo - Brazil

• Tipo de documento: Artigo Científico
• Fonte: INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH; v. 17, n. 1 JAN 2020.
• Citações Web of Science: 0

Resumo

The prevalence of health problems during childhood and adolescence is high in developing countries such as Brazil. Social inequality, violence, and malnutrition have strong impact on youth health. To better understand these issues we propose to combine machine-learning methods and graph analysis to build predictive networks applied to the Brazilian National Student Health Survey (PenSE 2015) data, a large dataset that consists of questionnaires filled by the students. By using a combination of gradient boosting machines and centrality hub metric, it was possible to identify potential confounders to be considered when conducting association analyses among variables. The variables were ranked according to their hub centrality to predict the other variables from a directed weighted-graph perspective. The top five ranked confounder variables were ``gender{''}, ``oral health care{''}, ``intended education level{''}, and two variables associated with nutrition habits-{''}eat while watching TV{''} and ``never eat fast-food{''}. In conclusion, although causal effects cannot be inferred from the data, we believe that the proposed approach might be a useful tool to obtain novel insights on the association between variables and to identify general factors related to health conditions. (AU)

• Processo FAPESP: 19/17907-5 - Inferências de conexões intercerebrais com utilização de espectroscopia de infravermelho próximo: investigação de interações adulto-criança
• Beneficiário: Amanda Yumi Ambriola Oku
• Linha de fomento: Bolsas no Brasil - Doutorado Direto
• Processo FAPESP: 18/21934-5 - Estatística de redes: teoria, métodos e aplicações
• Beneficiário: André Fujita
• Linha de fomento: Auxílio à Pesquisa - Temático
• Processo FAPESP: 16/19376-9 - Integridade estrutural e funcional do circuito de Papez e sua relação com a simtomatologia na Esclerose Lateral Amiotrófica
• Beneficiário: Ana Paula Arantes de Andrade Bueno
• Linha de fomento: Bolsas no Brasil - Doutorado
• Processo FAPESP: 18/04654-9 - Séries temporais, ondaletas e dados de alta dimensão
• Beneficiário: Pedro Alberto Morettin
• Linha de fomento: Auxílio à Pesquisa - Temático