Application of Physics-Informed Neural Networks (PINN) in the Kermack-McKendrick Approach to Epidemiological Models
DOI:
https://doi.org/10.5540/03.2026.012.01.0255Palavras-chave:
Epidemiology, Machine Learning, PINN, Forward Problems, Inverse ProblemsResumo
This work proposes the use of Physics-Informed Neural Networks (PINNs) to solve an inverse problem in the Kermack-McKendrick epidemiological model, fitting the parameters of the theoretical curve to the weekly mortality data from the bubonic plague epidemic in Bombay (1905-1906). The original model was revisited using PINNs, which integrate differential equations and machine learning to estimate parameters, minimizing the original mean squared error (MSE) on the data. Compared to the original fit (MSE ≈ 4273.42) and the Least Squares Method (MSE ≈ 3208.53), the PINN achieved an MSE of approximately 3204.00, reducing the error by about 25 % compared to the classical model. The implementation combined data normalization, hyperparameter optimization, and automatic differentiation, enabling greater numerical stability and accuracy in parameter estimation. The results highlight the potential of PINNs in epidemic modeling, with potential future applications to COVID-19 data in Brazil.
Downloads
Referências
N. Bacaër. “The model of Kermack and McKendrick for the plague epidemic in Bombay and the type reproduction number with seasonality”. In: Journal of Mathematical Biology 64 (2012), pp. 403–422. doi: 10.1007/s00285-011-0417-5.
A. G. Baydin, B. A. Pearlmutter, A. A. Radul, and J. M. Siskind. “Automatic differentiation in machine learning: a survey”. In: Journal of Machine Learning Research 18.153 (2018). Accessed on 2025-02-05, http://jmlr.org/papers/v18/16-107.html, pp. 1–43.
M. S. Cecconello, M. M. Diniz, and R. C. Bassanezi. “Analyzing the Sars-Cov-2 Pandemic Outbreak Using Fuzzy Sets and the SIR Model”. In: Applications of Fuzzy Techniques. Vol. 500. Springer International Publishing, 2023, pp. 40–51. doi: 10.1007/978- 3- 031- 16038-7_5.
D. J. D. Earn, S. W. Park, and B. M. Bolker. “Fitting Epidemic Models to Data: A Tutorial in Memory of Fred Brauer”. In: Bulletin of Mathematical Biology 86.109 (2024), pp. 1–32. doi: 10.1007/s11538-024-01326-9.
A. Farea, O. Yli-Harja, and F. Emmert-Streib. “Understanding Physics-Informed Neural Networks: Techniques, Applications, Trends, and Challenges”. In: AI 5.3 (2024), pp. 1534– 1557. doi: 10.3390/ai5030074.
S. Han, L. Stelz, H. Stoecker, L. Wang, and K. Zhou. “Approaching epidemiological dynamics of COVID-19 with physics-informed neural networks”. In: Journal of the Franklin Institute 361.6 (2024), pp. 1–29. doi: 10.1016/j.jfranklin.2024.106671.
W. O. Kermack and A. G. McKendrick. “A contribution to the mathematical theory of epidemics”. In: Proceedings of the Royal Society of London. Series A 115 (1927), pp. 700–721. doi: 10.1098/rspa.1927.0118.
D. P. Kingma and J. Ba. Adam: A Method for Stochastic Optimization. Accessed on 2025-02-05, https://arxiv.org/abs/1412.6980. 2017.
M. Raissi, P. Perdikaris, and G. E. Karniadakis. “Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations”. In: Journal of Computational Physics 378 (2019), pp. 686–707. doi: 10.1016/j.jcp.2018.10.045.
M. Raissi, N. Ramezani, and P. Seshaiyer. “On parameter estimation approaches for predicting disease transmission through optimization, deep learning and statistical inference methods”. In: Letters in Biomathematics 6.2 (2019), pp. 1–26. doi: 10.1080/23737867. 2019.1676172.
C. I. Siettos and L. Russo. “Mathematical modeling of infectious disease dynamics”. In: Virulence 4.4 (2013), pp. 295–306. doi: 10.4161/viru.24041.
The Advisory Committee Appointed by the Secretary of State for India, the Royal Society and the Lister Institute. “Reports on Plague Investigations in India”. In: The Journal of Hygiene 7.6 (1907). Accessed on 2025-01-16, http://www.jstor.org/stable/4619420, pp. 693–985.
S. Wang, Y. Teng, and P. Perdikaris. “Understanding and Mitigating Gradient Flow Pathologies in Physics-Informed Neural Networks”. In: SIAM Journal on Scientific Computing 43.5 (2021), pp. 1534–1557. doi: 10.1137/20M1318043.
