Estimating Hospital Bed Occupancy for COVID-19 Treatment Using Temporal Data

• Thiago M. Ventura UFMT
• Raphael S. R. Gomes UFMT
• Gabriel S. G. Pedroso UFMT
• Daniel A. Vecchiato UFMT
• Rebeca L. Rezende UFMT

DOI: https://doi.org/10.5753/sbcas.2025.6909

Abstract

Efficient management of the quantity and types of hospital beds is a critical challenge in public healthcare, as highlighted during the COVID-19 pandemic. This study proposes an approach based on artificial neural networks of the Multi-Layer Perceptron (MLP) type and time series analysis, using data collected in the state of Mato Grosso between 2020 and 2022. The model was trained with hospitalization data from the previous two weeks to predict the demand for clinical and complementary beds, achieving a root mean square error (RMSE) of 45.83 for a two-week forecast horizon. The results demonstrate its applicability as a decision-support tool for healthcare managers, contributing to a more efficient allocation of hospital resources.

References

Ali, Y. A., Awwad, E. M., Al-Razgan, M., and Maarouf, A. (2023). Hyperparameter search for machine learning algorithms for optimizing the computational complexity. Processes, 11(2).

Almeida, J. F. d. F., Conceição, S. V., Pinto, L. R., Horta, C. J. G., Magalhães, V. S., and Campos, F. C. C. d. (2021). Estimating brazilian states’ demands for intensive care unit and clinical hospital beds during the covid-19 pandemic: Development of a predictive model. Sao Paulo Medical Journal, 139(2):178–185.

Bekker, R., uit het Broek, M., and Koole, G. (2023). Modeling covid-19 hospital admissions and occupancy in the netherlands. European Journal of Operational Research, 304(1):207–218. The role of Operational Research in future epidemics/pandemics.

Braga, M. d. B., Fernandes, R. d. S., Souza, Jr, G. N. d., Rocha, J. E. C. d., Dolácio, C. J. F., Tavares, Jr, I. d. S., Pinheiro, R. R., Noronha, F. N., Rodrigues, L. L. S., Ramos, R. T. J., Carneiro, A. R., Brito, S. R. d., Diniz, H. A. C., Botelho, M. d. N., and Vallinoto, A. C. R. (2021). Artificial neural networks for short-term forecasting of cases, deaths, and hospital beds occupancy in the covid-19 pandemic at the brazilian amazon. PLOS ONE, 16(3):1–27.

Car, Z., Baressi Šegota, S., Andelić, N., Lorencin, I., and Mrzljak, V. (2020). Modeling the spread of covid-19 infection using a multilayer perceptron. Computational and Mathematical Methods in Medicine, 2020(1):5714714.

Consoli, B., Viera, R., Bordini, R. H., and Manssour, I. H. (2024). Predicting inpatient admissions in brazilian hospitals. In Anais do 24º Simpósio Brasileiro de Computação Aplicada à Saúde (SBCAS), pages 284–295. Sociedade Brasileira de Computação.

Gomes, C. E. V., Clímaco, G., de Carvalho, J. A. F. N., and Quintanilha, D. B. P. (2024). Modelagem matemática para a alocação de leitos de uti com diferentes tipos de leitos e taxa de ocupação. In Anais do 24º Simpósio Brasileiro de Computação Aplicada à Saúde (SBCAS), pages 226–237. Sociedade Brasileira de Computação.

Marcilio, I., Lazar Neto, F., Lazzeri Cortez, A., Miethke-Morais, A., Dutilh Novaes, H. M., Possolo de Sousa, H., de Carvalho, C. R. R., Shafferman Levin, A. S., Ferreira, J. C., Gouveia, N., and Group, H. C.-. S. (2022). Mortality over time among covid-19 patients hospitalized during the first surge of the pandemic: A large cohort study. PLOS ONE, 17(9):1–12.

Ritter, M., Ott, D. V. M., Paul, F., Haynes, J.-D., and Ritter, K. (2021). Covid-19: A simple statistical model for predicting intensive care unit load in exponential phases of the disease. Scientific Reports, 11:5018.

Runge, M., Richardson, R. A. K., Clay, P. A., Bell, A., Holden, T. M., Singam, M., Tsuboyama, N., Arevalo, P., Fornoff, J., Patrick, S., Ezike, N. O., and Gerardin, J. (2022). Modeling robust covid-19 intensive care unit occupancy thresholds for imposing mitigation to prevent exceeding capacities. PLOS Global Public Health, 2(5):1–17.

Wacksman, J. (2021). Digitalization of contact tracing: balancing data privacy with public health benefit. Ethics and Information Technology, 23:855–861.