Search

In recent years, with the frequent movement of livestock and related products on the Chinese mainland that have resulted in a rise in brucellosis cases in certain regions, the Chinese government has implemented "vaccination + culling" strategy to control the spread of brucellosis, but the epidemiological situation and the intensity of prevention and control measures vary from province to province, and much of the situation has not yet been reversed. Thus, it is crucial to understand and assess the combined effects of vaccination and culling on brucellosis transmission behind the data from multiple provinces and cities. In this paper, we combine an ordinary differential equation model and a deep neural network model to develop a hybrid model that not only characterizes the propagation mechanism of brucellosis, but also describes control strategies as time-varying parameters to track the dynamic evolution of brucellosis transmission under control measures with varying intensity. In addition, the real-time reproduction number is calculated and taken as an evaluation index. We then train and test the hybrid model using actual data from four provincial administrative regions in Northern China from 2010 to 2020. It is found that in Inner Mongolia, the sheep vaccination rate (θ (t)) and the culling rate of infected sheep (c (t)) have a weak effect on reducing the brucellosis transmission. In Gansu, the incomplete culling of infected sheep was the main reason for the increase in human brucellosis cases after the introduction of whole-flock vaccination in 2016. Culling and vaccination contributed to the decrease in human cases in both Xinjiang and Jilin. Prevention and control measures for human brucellosis are inadequate in Xinjiang. However, prevention and control measures for human brucellosis are inadequate, while indirect transmission of Brucella from the environment to humans and sheep poses a greater risk than direct contact with infected sheep in Jilin. The above analysis can provide theoretical foundation for decision-making on public health policies to control the pandemic. • This study conducted a hybrid model combining a mechanistic model with a data-driven model to quantify the impact of prevention and control measures on brucellosis. • The intensity of implementation of prevention and control measures were represented as time-varying parameters in the dynamical model to show the evolution of the impact of prevention and control measures on brucellosis transmission. • The real-time reproduction number was calculated and used as an outbreak evaluation index to measure the real-time transmissibility of an outbreak. • The different transmission characteristics of the brucellosis outbreak are due to the different intensity of the control measures implemented in the different regions. • Sufficient attention should be paid to controlling human brucellosis while continuing to strengthen the prevention and control of brucellosis in the sheep population. [ABSTRACT FROM AUTHOR]