Published Paper Details:

Anthrax, caused by Bacillus anthracis, is a highly contagious zoonotic infection with important connotations for animal as well as human health. Compartmental epidemiological models and fractional order differential equations are traditionally used models to analyse anthrax dynamics. These models tend to face nonlinearity complexities, so high computational methods are needed for exact solutions. This paper is a critical analysis of various neural network- based intelligent computing processes for the solution of anthrax disease models. Sophisticated techniques like Radial Basis Bayesian Regularization Deep Neural Networks (RB-BRDNN), Mayer Wavelet Neural Networks (MW-NN), and supervised neural networks (SNNs) are investigated in terms of their effectiveness for the solution of nonlinear differential equations. These models, when paired with optimization methods like Particle Swarm Optimization (PSO) and Bayesian regularization, enhance numerical stability and accuracy. The review points out the benefits of neural network- based methods in forecasting anthrax outbreaks and aiding disease control measures. Directions for future research are presented, focusing on hybrid AI models for real-time disease surveillance and intervention planning.