Analysis of Textual Features in the Automation of Medical Regulation

• Kauan Vaz do Nascimento UFPI
• Raimundo Santos Moura UFPI

DOI: https://doi.org/10.5753/sbcas.2024.2140

Abstract

In Brazil, private health plans face many financial problem, due to increasing demand for medical services, and fraud or abuse in the use of services. The adoption of prior authorization was important, but the maintenance of specialized teams for this task is still expensive, generating the need to automate this process. In this work, we investigate eight Machine Learning (ML) models (classical and deep learning) with the MIMIC-CXR data translated into Portuguese. The results show 95% accuracy with an RNN, in addition to identify important features using the LinearSVC model.

References

Goldberger, A., Amaral, L., Glass, L., Hausdorff, J., Ivanov, P., Mark, R., Mietus, J., Moody, G., Peng, C., and Stanley, H. (2000). Physiobank, physiotoolkit, and physionet: Components of a new research resource for complex physiologic signals. Circulation, 101(23).

Graham, B., Bond, R., Quinn, M., and Mulvenna, M. (2018). Using data mining to predict hospital admissions from the emergency department. IEEE Access, 6:10458–10469.

Johnson, A., Lungren, M., Peng, Y., Lu, Z., Mark, R., Berkowitz, S., and Horng, S. (2019a). MIMIC-CXR-JPG chest radiographs with structured labels (version 2.0.0). PhysioNet. DOI: 10.13026/8360-t248.

Johnson, A., Pollard, T., Mark, R., Berkowitz, S., and Horng, S. (2019b). MIMIC-CXR database (version 2.0.0). PhysioNet. DOI: 10.13026/C2JT1Q.

Johnson, A. E., Pollard, T. J., Berkowitz, S. J., Greenbaum, N. R., Lungren, M. P., Deng, C.-y., Mark, R. G., and Horng, S. (2019c). MIMIC-CXR, a de-identified publicly available database of chest radiographs with free-text reports. Scientific data, 6(1):317.

Johnson, A. E., Pollard, T. J., Greenbaum, N. R., Lungren, M. P., Deng, C.-y., Peng, Y., Lu, Z., Mark, R. G., Berkowitz, S. J., and Horng, S. (2019d). MIMIC-CXR-JPG, a large publicly available database of labeled chest radiographs. arXiv preprint arXiv:1901.07042.

Magalhães Junior, G. V., Vieira, J. P. A., Santos, R. L. S., Barbosa, J. L. N., Santos Neto, P., and Moura, R. S. (2019). A study of the influence of textual features in learning medical prior authorization. In 2019 IEEE 32nd International Symposium on Computer-Based Medical Systems (CBMS).

Roquette, B. P., Nagano, H., Marujo, E. C., and Maiorano, A. C. (2020). Prediction of admission in pediatric emergency department with deep neural networks and triage textual data. Neural Networks, 126:170–177.

Wang, B., Li, W., Bradlow, A., Bazuaye, E., and Chan, A. T. (2023). Improving triaging from primary care into secondary care using heterogeneous data-driven hybrid machine learning. Decision Support Systems, 166:113899.