Diagnosis of patients with blood cell count for COVID-19: An explainable artificial intelligence approach

Autores

  • Kaike Wesley Reis Universidade Federal da Bahia
  • Karla Patricia Oliveira-Esquerre Universidade Federal da Bahia

Palavras-chave:

Artificial Intelligence, Diagnosis, Blood Cell Count

Resumo

Objective: Present an explainable artificial intelligence (AI) approach for COVID-19 diagnosis with blood cell count. Methods: Five AI algorithms were evaluated: Logistic Regression, Random Forest, Support Vector Machine, Gradient Boosting and eXtreme Gradient Boosting. A Bayesian optimization with 5-Fold cross-validation was used to hyper-parameters tuning. The model selection evaluated three results: cross validation performance, test set prediction performance and a backtest: performance on identifying patients negative for COVID-19, but positive for others respiratory pathologies. Shapley Additive explanations (SHAP) was used to explain the chosen model. Results: A Random Forest model was obtained with 77.7% F1-Score (IC95%:57.1;92.3), 85.9% AUC (IC95%:73.7;95.9), 74.4% Sensitivity (IC95%:50.0;92.1) and 97.5% Specificity (IC95%:93.6;100.0). The main features were leukocytes, platelets and eosinophils. Conclusion: The research highlights the importance of model interpretability, demonstrating blood cell count as a possibility for COVID-19 diagnosis. The methodological structure developed, using TRIPOD’s guidelines, can be extrapolated to other pathologies.

Downloads

Não há dados estatísticos.

Biografia do Autor

Kaike Wesley Reis, Universidade Federal da Bahia

Graduando de Engenharia de Controle e Automação de Processos do atual Departamento de Química na escola Politécnica da Universidade Federal da Bahia. Área de aprendizagem de máquinas.

Karla Patricia Oliveira-Esquerre, Universidade Federal da Bahia

Professora Associada do Departamento de Engenharia Química, Universidade Federal da Bahia, Salvador, Bahia, Brasil.

Downloads

Publicado

10-06-2021

Como Citar

Reis, K. W., & Oliveira-Esquerre, K. P. (2021). Diagnosis of patients with blood cell count for COVID-19: An explainable artificial intelligence approach. Journal of Health Informatics, 13(2). Recuperado de https://jhi.sbis.org.br/index.php/jhi-sbis/article/view/779

Edição

Seção

Artigo Original

Artigos Semelhantes

1 2 3 4 5 6 7 8 > >> 

Você também pode iniciar uma pesquisa avançada por similaridade para este artigo.