A Threshold-Weighted Mathematical Fusion Model for Epidemic Outbreak Prediction Using SEIR Residual Dynamics and Cloud-Based Machine Learning

Year : 2026 | Volume : 15 | Issue : 02 | Page : 12 19
By

Narendra Janardhanrao Padole,

Nitin S. Shrirao,

  1. Assistant Professor, Department of Computer, HVPM, Maharashtra, India
  2. Director, Siddhant Institute of Computer Application (SICA),Sudumbare, Pune, Maharashtra, India

Abstract

Accurate prediction of epidemic outbreaks is critical for effective public health management, resource planning, early warning generation, and timely intervention by municipal authorities. Traditional compartmental models such as Susceptible–Exposed–Infectious–Recovered (SEIR) offer valuable epidemiological insights and mathematical interpretability; however, they may not adequately capture the complex nonlinear relationships present in real-world urban health systems. Conversely, data-driven machine learning techniques can identify hidden patterns in large datasets but often lack epidemiological structure and interpretability. To address these limitations, this study proposes a threshold-weighted hybrid epidemic prediction framework for Amravati Municipal Corporation. The proposed approach integrates a mechanistic SEIR model with advanced machine learning algorithms, including Extreme Gradient Boosting (XGBoost), Long Short-Term Memory (LSTM) networks, and Prophet time-series forecasting. Epidemic incidence is decomposed into a deterministic epidemiological component and a residual component representing unexplained variations. A constrained non-negative fusion mechanism is introduced to combine the outputs of residual predictors while preserving model stability and interpretability. Furthermore, an effective reproduction tendency metric is employed as a threshold activation variable, enabling adaptive correction during periods of accelerated disease transmission. To support decision-making, a Composite Epidemic Risk Index is developed by integrating predicted incidence, transmission tendency, projected growth patterns, and healthcare burden indicators. The framework is designed for cloud-based deployment, allowing integration of clinical, demographic, environmental, and public surveillance data for ward-level epidemic forecasting and risk assessment. The proposed methodology establishes a mathematically rigorous and computationally scalable foundation for doctoral research in epidemic prediction, machine learning, and cloud-enabled public health analytics.

Keywords: Cloud computing, epidemic prediction, epidemic risk index, long short-term memory, mathematical modeling

[This article belongs to Research & Reviews : Journal of Statistics ]

How to cite this article: Narendra Janardhanrao Padole, Nitin S. Shrirao. A Threshold-Weighted Mathematical Fusion Model for Epidemic Outbreak Prediction Using SEIR Residual Dynamics and Cloud-Based Machine Learning. Research & Reviews : Journal of Statistics. 2026; 15(02):12-19.
How to cite this URL: Narendra Janardhanrao Padole, Nitin S. Shrirao. A Threshold-Weighted Mathematical Fusion Model for Epidemic Outbreak Prediction Using SEIR Residual Dynamics and Cloud-Based Machine Learning. Research & Reviews : Journal of Statistics. 2026; 15(02):12-19. Available from: https://journals.stmjournals.com/rrjost/article=2026/view=252326

References

  1. Kermack WO, McKendrick AG. A contribution to the mathematical theory of epidemics. Proc R Soc Lond A. 1927;115(772):700-721. doi:10.1098/rspa.1927.0118.
  2. Hethcote HW. The mathematics of infectious diseases. SIAM Rev. 2000;42(4):599-653. doi:10.1137/S0036144500371907.
  3. Brauer F. The Kermack-McKendrick epidemic model revisited. Math Biosci. 2005;198(2):119-131. doi:10.1016/j.mbs.2005.07.006.
  4. Girardi P, Gaetan C. An SEIR model with time-varying coefficients for analyzing the SARS-CoV-2 epidemic. Risk Anal. 2023;43(1):144-155. doi:10.1111/risa.13858.
  5. Mwalili S, Kimathi M, Ojiambo V, Gathungu D, Mbogo R. SEIR model for COVID-19 dynamics incorporating the environment and social distancing. BMC Res Notes. 2020;13(1):352. doi:10.1186/s13104-020-05192-1.
  6. Chen T, Guestrin C. XGBoost: A scalable tree boosting system. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM; 2016. p. 785-794. doi:10.1145/2939672.2939785.
  7. Hochreiter S, Schmidhuber J. Long short-term memory. Neural Comput. 1997;9(8):1735-1780. doi:10.1162/neco.1997.9.8.1735.
  8. Taylor SJ, Letham B. Forecasting at scale. Am Stat. 2018;72(1):37-45. doi:10.1080/00031305.2017.1380080.
  9. Kong L, Guo Y, Lee CW. Enhancing COVID-19 prevalence forecasting: A hybrid approach integrating epidemic differential equations and recurrent neural networks. AppliedMath. 2024;4(2):427-441. doi:10.3390/appliedmath4020022.
  10. Baccega D, et al. Enhancing COVID-19 forecasting precision through the integration of compartmental models, machine learning and variants. Sci Rep. 2024;14:19220. doi:10.1038/s41598-024-69660-5.
  11. Kumar N, Susan S. Epidemic modeling using hybrid of time-varying SIRD, particle swarm optimization, and deep learning. arXiv [Preprint]. 2024; arXiv:2401.18047.
  12. Bertozzi AL, Franco E, Mohler G, Short MB, Sledge D. The challenges of modeling and forecasting the spread of COVID-19. Proc Natl Acad Sci U S A. 2020;117(29):16732-16738. doi:10.1073/pnas.2006520117.
  13. Roda WC, Varughese MB, Han D, Li MY. Why is it difficult to accurately predict the COVID-19 epidemic? Infect Dis Model. 2020;5:271-281. doi:10.1016/j.idm.2020.03.001.
Regular Issue Subscription Original Research
Volume 15
Issue 02
Received 30/05/2026
Accepted 04/06/2026
Published 15/06/2026
Publication Time 16 Days

Login

My IP

PlumX Metrics