Título: Thyroid Syndrome Detection using Machine Learning Algorithms: A Comparative Analysis
Autores: Caio M. V . Cavalcante, Vinicius A. Almeida, Marcos Barros, Nathalee Lima, and Rosana C. B. Rego
Resumo: A thyroid syndrome necessitates early and proper diagnosis to facilitate adequate treatment. However, subjectivity in analyzing test results poses a challenge. In this work, we explored and analyzed the potential of machine learning algorithms. These algorithms include decision trees, random forest, logistic regression, naive Bayes, XGBoost, LightGBM, and a stacking ensemble model. The goal was to classify the euthyroid syndrome, which is a medical condition impacting the thyroid gland, by utilizing attributes obtained from blood tests. These attributes encompass thyroxine, thyroid stimulating hormone, free thyroxine index, total thyroxine, and triiodothyronine. The findings indicate the efficacy of employing these algorithms in accurately classifying the syndrome and providing diagnostic support
Palavras-chave: Thyroid syndrome, Machine learning, Classification, Healthcare
Páginas: 7
Código DOI: 10.21528/CBIC2023-088
Artigo em pdf: CBIC_2023_paper088.pdf
Arquivo BibTeX: CBIC_2023_088.bib