Predição do Cumprimento da Lei dos 60 Dias no Tratamento de Câncer de Mama: Comparação de Algoritmos de Aprendizado de Máquina
Resumo
O câncer de mama é a neoplasia maligna mais incidente entre mulheres no Brasil, e o cumprimento do prazo legal de 60 dias para início do tratamento no SUS permanece um desafio. Este estudo compara modelos de aprendizado de máquina (Decision Tree, Random Forest, XGBoost e CatBoost) na predição da violação desse prazo, utilizando dados do Registro Hospitalar de Câncer (2013–2024). O CatBoost apresentou melhor desempenho preditivo (AUC-ROC = 0,801; PR-AUC = 0,862). A análise SHAP identificou o intervalo entre diagnóstico e primeira consulta oncológica e o fluxo assistencial capital–interior como principais preditores. Os resultados evidenciam o potencial desses modelos para subsidiar a gestão do cuidado oncológico.Referências
Alves, M., Monteiro, M., Alves, F. and Figueiredo, F. (2022). Determinants of Lack of Access to Treatment for Women Diagnosed with Breast Cancer in Brazil. International Journal of Environmental Research and Public Health, 19(13):7635.
Barros, A., Araújo, J., Murta-Nascimento, C. and Dias, A. (2019). Clinical pathways of breast cancer patients treated in the Federal District, Brazil. Revista de Saúde Pública, 53:14.
Cabral, A., Giatti, L., Casale, C. and Cherchiglia, M. (2019). Social vulnerability and breast cancer: differentials in the interval between diagnosis and treatment of women with different sociodemographic profiles. Ciência & Saúde Coletiva, 24(2):613-622.
Ferreira, N., Schoueri, J., Sorpreso, I., Adami, F. and Figueiredo, F. (2020). Waiting Time between Breast Cancer Diagnosis and Treatment in Brazilian Women: An Analysis of Cases from 1998 to 2012. International Journal of Environmental Research and Public Health, 17(11):4030.
Ginsburg, O., Yip, C., Brooks, A., Cabanes, A., Caleffi, M., Yataco, J., Gyawali, B., McCormack, V., Anderson, M., Mehrota, R. et al. (2020). Breast Cancer Early Detection: A Phased Approach to Implementation. Cancer, 126(10):2379-2393.
Gioia, S., Galdino, R., Brigagão, L., Valadares, A., Secol, F., Miguel, S., Bukowski, A., Krush, L. and Goss, P. (2020). Prediction of Attendance to the "Law of 60 Days" in Breast Cancer Patients using Machine Learning Classifiers. Acta Scientific Cancer Biology, 4(3):16-28.
Li, J. (2024). Area under the ROC Curve has the most consistent evaluation for binary classification. PLoS ONE, 19(12):e0316019.
Martins, L., Chaves, G., Oliveira, J., Souza, L., Neto, P., Carvalho, F., Vasconcelos, G., Dias, M. and Mello, M. (2026). Epidemiological Profile of Cancer Incidence in Brazil and Regions: Estimates for the 2026-2028 Triennium. Revista Brasileira de Cancerologia, 72(2):e-025587.
McElfresh, D., Khandagale, S., Valverde, J., Prasad, V., Ramakrishnan, G., Goldblum, M. and White, C. When Do Neural Nets Outperform Boosted Trees on Tabular Data? (2024). In 37th Conference on Neural Information Processing Systems (NeurIPS 2023) Track on Datasets and Benchmarks. NeurIPS Foundation.
Medeiros, G., Thuler, L. and Bergmann, A. (2021). Determinants of delay from cancer diagnosis to treatment initiation in a cohort of brazilian women with breast cancer. Health and Social Care in the Community, 00:1-10.
Nogueira, M., Atty, A., Tomazelli, J., Jardim, B., Bustamante-Teixeira, M. and Azevedo e Silva, G. Frequency and factors associated with delay in breast cancer treatment in Brazil, according to data from the Oncology Panel, 2019-2020. Epidemiologia e Serviços de Saúde, 32(1):e2022563.
Barros, A., Araújo, J., Murta-Nascimento, C. and Dias, A. (2019). Clinical pathways of breast cancer patients treated in the Federal District, Brazil. Revista de Saúde Pública, 53:14.
Cabral, A., Giatti, L., Casale, C. and Cherchiglia, M. (2019). Social vulnerability and breast cancer: differentials in the interval between diagnosis and treatment of women with different sociodemographic profiles. Ciência & Saúde Coletiva, 24(2):613-622.
Ferreira, N., Schoueri, J., Sorpreso, I., Adami, F. and Figueiredo, F. (2020). Waiting Time between Breast Cancer Diagnosis and Treatment in Brazilian Women: An Analysis of Cases from 1998 to 2012. International Journal of Environmental Research and Public Health, 17(11):4030.
Ginsburg, O., Yip, C., Brooks, A., Cabanes, A., Caleffi, M., Yataco, J., Gyawali, B., McCormack, V., Anderson, M., Mehrota, R. et al. (2020). Breast Cancer Early Detection: A Phased Approach to Implementation. Cancer, 126(10):2379-2393.
Gioia, S., Galdino, R., Brigagão, L., Valadares, A., Secol, F., Miguel, S., Bukowski, A., Krush, L. and Goss, P. (2020). Prediction of Attendance to the "Law of 60 Days" in Breast Cancer Patients using Machine Learning Classifiers. Acta Scientific Cancer Biology, 4(3):16-28.
Li, J. (2024). Area under the ROC Curve has the most consistent evaluation for binary classification. PLoS ONE, 19(12):e0316019.
Martins, L., Chaves, G., Oliveira, J., Souza, L., Neto, P., Carvalho, F., Vasconcelos, G., Dias, M. and Mello, M. (2026). Epidemiological Profile of Cancer Incidence in Brazil and Regions: Estimates for the 2026-2028 Triennium. Revista Brasileira de Cancerologia, 72(2):e-025587.
McElfresh, D., Khandagale, S., Valverde, J., Prasad, V., Ramakrishnan, G., Goldblum, M. and White, C. When Do Neural Nets Outperform Boosted Trees on Tabular Data? (2024). In 37th Conference on Neural Information Processing Systems (NeurIPS 2023) Track on Datasets and Benchmarks. NeurIPS Foundation.
Medeiros, G., Thuler, L. and Bergmann, A. (2021). Determinants of delay from cancer diagnosis to treatment initiation in a cohort of brazilian women with breast cancer. Health and Social Care in the Community, 00:1-10.
Nogueira, M., Atty, A., Tomazelli, J., Jardim, B., Bustamante-Teixeira, M. and Azevedo e Silva, G. Frequency and factors associated with delay in breast cancer treatment in Brazil, according to data from the Oncology Panel, 2019-2020. Epidemiologia e Serviços de Saúde, 32(1):e2022563.
Publicado
01/06/2026
Como Citar
FERRETE, Lívia F.; RAPOSO, Letícia M..
Predição do Cumprimento da Lei dos 60 Dias no Tratamento de Câncer de Mama: Comparação de Algoritmos de Aprendizado de Máquina. In: CONCURSO DE TRABALHOS DE INICIAÇÃO CIENTÍFICA - SIMPÓSIO BRASILEIRO DE COMPUTAÇÃO APLICADA À SAÚDE (SBCAS), 26. , 2026, Ouro Preto/MG.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 13-18.
ISSN 2763-8987.
DOI: https://doi.org/10.5753/sbcas_estendido.2026.20443.
