$PG-NODE^{TB}$: Physics-Guided Neural Ordinary Differential Equations for Tuberculosis Transmission Dynamics

TL;DR

This paper introduces PG-NODE, a physics-guided neural ODE framework for TB modeling that adapts to time-varying dynamics and improves predictive accuracy over classical models.

Contribution

The paper develops a novel PG-NODE approach for TB transmission, integrating neural networks with compartmental models to learn unknown rate functions from data.

Findings

Lower 27% RMSE compared to classical SLIR model.

Effective in tracking time-varying transmission rates.

Potential for improved long-term intervention forecasting.

Abstract

Tuberculosis (TB) remains a leading global infectious disease, causing approximately 1.3 million deaths and 10.6 million new infections annually. Classical compartmental ODE models are the standard epidemiological tool for TB, yet their fixed-parameter structure cannot adapt to time-varying dynamics, unmodeled effects, or heterogeneous real-world data. This paper presents a methodological framework and proof-of-concept for applying Physics-Guided Neural Ordinary Differential Equations (PG-NODE) to TB transmission modeling within a SLIR (Susceptible, Latent, Infectious, Recovered) compartmental framework. We perform a rigorous mathematical analysis of the SLIR model, including derivation of the basic reproduction number $\mathcal{R}_0$, equilibrium analysis, and normalized sensitivity indices. We then reformulate the SLIR system as a PG-NODE, preserving compartmental conservation laws and biological constraints while enabling neural network components to learn unknown or time-varying rate functions from data. Three simulation scenarios illustrate the framework's intended capabilities: (i) adaptive tracking of time-varying transmission rates, (ii) correcting for unmodeled treatment and relapse dynamics with 27\% lower RMSE than the classical SLIR, and (iii) comparative forecasting of competing intervention policies over a 20-year horizon. Simulation results indicate that PG-NODE has strong potential for improving predictive accuracy while maintaining epidemiological interpretability; full adjoint-based training on real WHO surveillance data is identified as the key next step for empirical validation.