Monkeypox Transmission Dynamics Using Fractional Disease Informed Neural Network: A Global and Continental Analysis

The global Monkeypox outbreak has become a significant concern, as highlighted by the World Health Organization (WHO). Understanding its transmission dynamics is vital for projecting progression and designing effective mitigation strategies. This study addresses two major gaps: the inability of classical models to capture epidemiological patterns and the lack of research on transmission dynamics across all continents. By integrating fractional calculus and Neural Networks (NN), the study develops a unified model to analyze Monkeypox transmission worldwide using comprehensive WHO data. The hybrid Fractional Disease-Informed Neural Network (DINN) model captures disease transmission's memory effect and nonlocality through the Caputo fractional derivative. Key findings reveal high progression rates of exposure to infectious humans (β) in Oceania (0.1558) and Asia (0.1497), with Africa having the lowest rate (0.0703). The clinically ill rate (γ) is highest in Asia (0.2694), followed by South America (0.2503), and lowest in Africa (0.1225). Asia also has the highest contact rate (η<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub>) between infected rodents and susceptible humans. Fractional-order parameter (q<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">6</sub>) values indicate low memory effects and nonlocality across all regions. These findings highlight the potential of fractional-order models in understanding and controlling Monkeypox globally. However, the computational expense of the hybrid DINN model presents a challenge, warranting further research to enhance efficiency.