Identification of important predictors, using V-Cramer technique and Permutation feature importance, in predicting Tuberculosis treatment status

• Jacó M. Santos UFAM
• Elson P. Vasques SAMSA
• Adriane F. Valentin SAMSA
• Eliane C. Nogueira SAMSA
• Ericle L. Costa SAMSA
• Juan C. Souza SAMSA
• John K. S. Silva SAMSA
• Elias E. R. Marques SAMSA
• July E. S. Silva SAMSA
• João G. S. Gomes SAMSA

DOI: https://doi.org/10.5753/sbcas.2024.2826

Abstract

In this study, we investigate the use of a fully connected neural network implemented with Keras to create prediction models for the treatment outcomes of tuberculosis patients. The data used were from the Sistema de Informação de Agravos de Notificação (SINAN) of the year 2023, integrated by iTB, a software developed by the Health Department of Manaus-AM. This work is still in progress. So far, we have achieved an accuracy of 92.80% and F1-Score values of 0.90 for cure, 0.89 for abandonment, 0.98 for TB-related death, and 0.99 for drug resistance. Additionally, applying the permutation-importance technique, we obtained an accuracy of 93.31%. Experiments with patient consultation data will still be added to the test data, along with a Cramer's V analysis.

References

Acharya, V., Dhiman, G., Prakasha, K., Bahadur, P., Choraria, A., Prabhu, S., Chadaga, K., Viriyasitavat, W., & Kautish, S. (2022). AI-Assisted Tuberculosis Detection and Classification from Chest X-Rays Using a Deep Learning Normalization-Free Network Model. Hindawi Computational Intelligence and Neuroscience, 2022. DOI: 10.1155/2022/2399428

Bahaweres, R. B., Agustian, F., Hermadi, I., Suroso, A. I., & Arkeman, Y. (2020). Software defect prediction using neural network based smote. International Conference on Electrical Engineering, Computer Science and Informatics (EECSI), 2020-October. DOI: 10.23919/EECSI50503.2020.9251874

Inacio B, A., Henrique Q, L., Arruda, I., Carminati S, P., Caroline C V, A., & Maia M S, C. (2023). View of Predição do desfecho de pacientes com Tuberculose. REVISTA OBSERVATORIO DE LA ECONOMIA LATINOAMERICANA. DOI: 10.55905/oelv21n9-094

Murrugara, L. K. S., Lima, L. V., Pavinati, G., Silva, I. G. P., Monteiro, L. R. S., Gil, N. L. M., & Magnabosco, G. T. (2023). Desfecho do tratamento da tuberculose latente em pessoas vivendo com vírus da imunodeficiência humana. Revista de Saúde Pública do Paraná, 6(1), 1–15. DOI: 10.32811/25954482-2023v6n1.709

Perlaza, C. L., Mosquera, F. E. C., Murillo, L. M. R., Sepulveda, V. B., & Arenas, C. D. C. (2023). Factors of abandonment of tuberculosis treatment in the public health network. Revista de Saude Publica, 57. DOI: 10.11606/S1518-8787.2023057004454

Singh, M., Pujar, G. V., Kumar, S. A., Bhagyalalitha, M., Akshatha, H. S., Abuhaija, B., Alsoud, A. R., Abualigah, L., Beeraka, N. M., & Gandomi, A. H. (2022). Evolution of Machine Learning in Tuberculosis Diagnosis: A Review of Deep Learning-Based Medical Applications. Electronics 2022, Vol. 11, Page 2634, 11(17), 2634. DOI: 10.3390/ELECTRONICS11172634

Tabosa de Oliveira, T., Sampaio, V., & Endo, P. T. (2020). Configuração de hyper-parâmetros de modelos deep learning para auxílio no pós-diagnóstico de Tuberculose. Anais Estendidos do Simpósio Brasileiro de Computação Aplicada à Saúde (SBCAS), 87–92. DOI: 10.5753/SBCAS.2020.11563

Zhang, J., Qiang, Y., & Xu, X. (2021). Detection of Tuberculosis based on Deep Learning based methods You may also like An Ultrasensitive Electrochemical Aptasensor for Thrombin Detection Using MoS 2 Nanoparticles Loaded Iron-Porphyrinic Metal-Organic Framework as Signal Amplifier. J. Phys.: Conf. Ser, 1767, 12004. DOI: 10.1088/1742-6596/1767/1/012004