OPTIMAL CONTROL APPLIED TO RIFT VALLEY FEVER

Abstract Rift Valley Fever (RVF) virus is a mosquito‐born pathogen that infects livestock but it also has the capability to infect humans through direct or indirect contact with blood or organs of infected animals and by bites from infected mosquitos. The economic and social cost of the disease to rural populations can lead to a cascade of negative effects on the sustainability of animal and human populations. Vaccines exist to protect against this disease. Through a compartment model depicting the interactions leading to the spread of RVF in Aedes and Culex mosquitos and a livestock population, an optimal control problem is developed to minimize the number of vaccinated livestock at the final time while minimizing the negative effects of the infected Aedes and Culex mosquitos and the cost of the vaccination process. The unique optimal vaccination strategy is produced for given high transmission parameters and numerical results portray that vaccination depends on the level of effectiveness of the protocol.     

Modelling the $Wolbachia$ Strains for Dengue Fever Virus Control in the Presence of Seasonal Fluctuation

Consider that infection with $Wolbachiacan$ limit a mosquito''s ability to transmit Dengue fever virus through its saliva, a mathematical model describing the transmission of Dengue fever between vector mosquitoes and human, incorporating $Wolbachia$-carrying mosquito population and seasonal fluctuation, is proposed. Firstly, the stability and bifurcation of this model are investigated exactly in the case where seasonality can be neglected. Further, the basic reproductive number $\mathcal{R}_0^s$ for this model with seasonal variation is obtained, that is, if $\mathcal{R}_0^s$ is less than unity the disease is extinct and $\mathcal{R}_0^s$ is greater than unity the disease is uniformly persistent. Finally, numerical simulations verify the theoretical results. Theoretical results suggest that, compared with the mosquito reduction strategies (such as the elimination of mosquito breeding sites, killing of adult mosquitoes by spraying), introducing $Wolbachia$ strains is as effectual to fight against the transmission of Dengue virus.     

Simple discrete-time malarial models

We investigate dynamics of mosquito population models under two assumptions, respectively, and then formulate simple discrete-time compartmental susceptible-exposed-infective-recovered models for the malaria transmission based on the mosquito population models. We show that the mosquito population models either have robust dynamics or exhibit period-doubling bifurcation depending on the model assumptions. We derive a formula for the reproductive number of infection for the malaria model, which determines the stability of the infection-free fixed point. We then determine the existence of endemic fixed points for the malaria models. Using numerical simulations, we demonstrate that the dynamical characteristics of the mosquito populations, such as the global stability of the endemic fixed point and the appearance of a period-doubling bifurcation, are reflected in the dynamics of the malaria transmission.     

Comparing the efficiency of Wolbachia driven Aedes mosquito suppression strategies

Wolbachia is an endosymbiotic bacterium which manipulates host reproduction by cytoplasmic incompatibility, and restrains the transmission of dengue virus in Aedes mosquitoes. A novel strategy for dengue control involves releasing Wolbachia infected males into nature to suppress wild Aedes mosquito population. We develop a model of delay differential equations, integrating larval density-dependent competition and diapausing eggs, to compare the efficiency of different suppression strategies. The global asymptotical stability of the complete suppression state identifies the releasing amount threshold ascertaining suppression. Based on the experimental data for Aedes albopictus population in Guangzhou, our simulations show that the mosquito density in the highest peak season can be reduced by more than $95\%$ when the number of released males is above the releasing threshold. The best time to initiate the suppression is in early March, lasting until the end of June, followed by the parallel releasing policy from July to November. However, the egg bank has neglectable effects on the control of dengue vector in Guangzhou.     

Wolbachia-based biocontrol for dengue reduction using dynamic optimization approach

Aedes aegypti females mosquitoes are the principal transmitters of dengue and other arboviral infections. In recent years, it was disclosed that, when deliberately infected with Wolbachia symbiont, this mosquito species loses its vectorial competence and becomes less capable of transmitting the virus to human hosts. Thanks to this important discovery, Wolbachia-based biocontrol is now accepted as an ecologically friendly and potentially cost-effective method for prevention and control of dengue and other arboviral infections. In this paper, we propose a dengue transmission model that accounts for the presence of wild Aedes aegypti females and those deliberately infected with wMelPop Wolbachia strain, which is regarded as the best blocker of dengue and other arboviral infections. However, wMelPop strain of Wolbachia considerably reduces the individual fitness of mosquitoes, what makes rather challenging to achieve the gradual extrusion of wild mosquitoes and ensure their posterior replacement by Wolbachia-carriers. Nonetheless, this obstacle have been overcome by employing the optimal control approach for design of specific intervention programs based on daily releases of Wolbachia-carrying mosquitoes. The resulting optimal release programs ensure the population replacement and eventual local extinction of wild mosquitoes in the finite time and also entail a significant reduction in the number of expected dengue infections among human hosts under the long-term settings.     

How does within‐host dynamics affect population‐level dynamics? Insights from an immuno‐epidemiological model of malaria

Malaria is one of the most common mosquito‐borne diseases widespread in the tropical and subtropical regions. Few models coupling the within‐host malaria dynamics with the between‐host mosquito‐human dynamics have been developed. In this paper, by adopting the nested approach, a malaria transmission model with immune response of the host is formulated. Applying age‐structured partial differential equations for the between‐host dynamics, we describe the asymptomatic and symptomatic infectious host population for malaria transmission. The basic reproduction numbers for the within‐host model and for the coupled system are derived, respectively. The existence and stability of the equilibria of the coupled model are analyzed. We show numerically that the within‐host model can exhibit complex dynamical behavior, possibly even chaos. In contrast, equilibria in the immuno‐epidemiological model are globally stable and their stabilities are determined by the reproduction number. Increasing the activation rate of the within‐host immune response “dampens” the sensitivity of the population level reproduction number and prevalence to the increase of the within‐host reproduction of the pathogen. From public health perspective this means that treatment in a population with higher immunity has less impact on the population‐level reproduction number and prevalence than in a population with less immunity.     

An impulsive model for Wolbachia infection control of mosquito-borne diseases with general birth and death rate functions

Mosquito-borne diseases are global health problems, which mainly affect low-income populations in tropics and subtropics. In order to prevent the transmission of mosquito-borne diseases, the intracellular symbiotic bacteria named as Wolbachia is becoming a promising candidate to interrupt the virus transmission. In this paper, an impulsive mosquito population model with general birth and death rate functions is established to study the cytoplasmic incompatibility (CI) effect caused by mating of Wolbachia-infected males and uninfected females. The dynamics of the spread of Wolbachia in mosquito population are studied, and the strategies of mosquito extinction or replacing Wolbachia-uninfected mosquitoes with Wolbachia-infected mosquitoes are analyzed. Moreover, the results are applied to models with specific birth and death rate functions. It is shown that strategies may be different due to different birth and death rate functions, the type of Wolbachia strains and the initial number of Wolbachia-infected mosquitoes. Furthermore, numerical simulations are conducted to illustrate our conclusions.     

Wolbachia infection dynamics by reaction-diffusion equations

Dengue fever is caused by the dengue virus and transmitted by Aedes mosquitoes.A promising avenue for eradicating the disease is to infect the wild aedes population with the bacterium Wolbachia driven by cytoplasmic incompatibility(CI).When releasing Wolbachia infected mosquitoes for population replacement,it is essential to not ignore the spatial inhomogeneity of wild mosquito distribution.In this paper,we develop a model of reaction-diffusion system to investigate the infection dynamics in natural areas,under the assumptions supported by recent experiments such as perfect maternal transmission and complete CI.We prove non-existence of inhomogeneous steady-states when one of the diffusion coefficients is sufficiently large,and classify local stability for constant steady states.It is seen that diffusion does not change the criteria for the local stabilities.Our major concern is to determine the minimum infection frequency above which Wolbachia can spread into the whole population of mosquitoes.We find that diffusion drives the minimum frequency slightly higher in general.However,the minimum remains zero when Wolbachia infection brings overwhelming fitness benefit.In the special case when the infection does not alter the longevity of mosquitoes but reduces the birth rate by half,diffusion has no impact on the minimum frequency.     

Regional control for a spatially structured malaria model

A two‐component reaction‐diffusion system to describe the spread of malaria is considered. The model describes the dynamics of the infected mosquitoes and of the infected humans. The spread of the disease is controlled by three actions (controls) implemented in a subdomain of the habitat: killing mosquitoes, treating the infected humans, and reducing the contact rate mosquitoes‐humans. To start with, the problem of the eradicability of the disease is considered, while the cost of the controls is ignored. We prove that it is possible to decrease exponentially both the human and the vector infective population everywhere in the relevant habitat by acting only in a suitable subdomain. Later, the regional control problem of reducing the total cost of the damages produced by the disease, of the controls, and of the intervention in a certain subdomain is treated for the finite time horizon case. An iterative algorithm to decrease the total cost is proposed; apart from the three controls considered above, the logistic structure of the habitat is taken into account. The level set method is used as a key ingredient for describing the subregion of intervention. Some numerical simulations are given to illustrate the applicability of the theoretical results.     

Reducing mosquito-borne disease outbreak size: The relative importance of contact and transmissibility in a network model

The complex biological and environmental factors involved in the transmission of mosquito-borne diseases in humans have made their control elusive in many instances. Conceptual models contribute to gain insight and help to reduce the risk of taking poor managerial decisions. The focus of this paper is to compare, using a contact network model, the impact that perturbation of the number infectious contacts and of transmissibility have on the size of an outbreak. We illustrate the analysis on a contact network parametrized with data that associates humans and the mosquito Culex quinquefasciatus, a vector for lymphatic filariasis. The model suggests that, if the values corresponding to transmissibility and number of infectious contacts is relatively large, variations in the size of an outbreak are significantly in favor of control measures to reduce infectious contacts.     

Effect of partial immunity on transmission dynamics of dengue disease with optimal control

Dengue fever is one of the most dangerous vector‐borne diseases in the world in terms of death and economic cost. Hence, the modeling of dengue fever is of great significance to understand the dynamics of dengue. In this paper, we extend dengue disease transmission models by including transmit vaccinated class, in which a portion of recovered individual loses immunity and moves to the susceptibles with limited immunity and hence a less transmission probability. We obtain the threshold dynamics governed by the basic reproduction number R₀; it is shown that the disease‐free equilibrium is locally asymptotically stable if R₀ ≤ 1, and the system is uniformly persistence if R₀ > 1. We do sensitivity analysis in order to identify the key factors that greatly affect the dengue infection, and the partial rank correlation coefficient (PRCC) values for R₀ shows that the bitting rate is the most effective in lowering dengue new infections, and moreover, control of mosquito size plays an essential role in reducing equilibrium level of dengue infection. Hence, the public are highly suggested to control population size of mosquitoes and to use mosquito nets. By formulating the control objective, associated with the low infection and costs, we propose an optimal control question. By the application of optimal control theory, we analyze the existence of optimal control and obtain necessary conditions for optimal controls. Numerical simulations are carried out to show the effectiveness of control strategies; these simulations recommended that control measures such as protection from mosquito bites and mosquito eradication strategies effectively control and eradicate the dengue infections during the whole epidemic.     

Vaccination strategies based on feedback control techniques for a general SEIR-epidemic model

This paper presents several simple linear vaccination-based control strategies for a SEIR (susceptible plus infected plus infectious plus removed populations) propagation disease model. The model takes into account the total population amounts as a refrain for the illness transmission since its increase makes more difficult contacts among susceptible and infected. The vaccination control objective is the asymptotically tracking of the removed-by-immunity population to the total population while achieving simultaneously that the remaining populations (i.e. susceptible plus infected plus infectious) tend asymptotically to zero.     

The threshold infection level for Wolbachia invasion in random environments

Dengue fever and Zika are mosquito-borne diseases threatening human health. A novel strategy for mosquito-borne disease control uses the bacterium Wolbachia to block virus transmission. It requires releasing Wolbachia infected mosquitoes to exceed a threshold level. Since an accurate forecast for temperature and rainfall, the major environmental conditions regulating the mosquito dynamics, is often not available over a long time period, it is important to explore how the threshold releasing level changes in random environments. In this work, we estimate the threshold level in a stochastic system of differential equations where the reproduction rates of mosquitoes change randomly. We prove that the threshold level is, surprisingly, defined by a deterministic curve that does not fluctuate with environmental conditions. The major difficulty in the proof is to construct various auxiliary curves to limit the dynamic behaviors of the whole family of innumerable solutions satisfying a given initial condition.     

On vaccination controls for the SEIR epidemic model

This paper presents a simple continuous-time linear vaccination-based control strategy for a SEIR (susceptible plus infected plus infectious plus removed populations) disease propagation model. The model takes into account the total population amounts as a refrain for the illness transmission since its increase makes more difficult contacts among susceptible and infected. The control objective is the asymptotically tracking of the removed-by-immunity population to the total population while achieving simultaneously the remaining population (i.e. susceptible plus infected plus infectious) to asymptotically tend to zero.     

Birth-pulse models of Wolbachia-induced cytoplasmic incompatibility in mosquitoes for dengue virus control

Dengue fever, which affects more than 50 million people a year, is the most important arboviral disease in tropical countries. Mosquitoes are the principal vectors of the dengue virus but some endosymbiotic Wolbachia bacteria can stop the mosquitoes from reproducing and so interrupt virus transmission. A birth-pulse model of the spread of Wolbachia through a population of mosquitoes, incorporating the effects of cytoplasmic incompatibility (CI) and different density dependent death rate functions, is proposed. Strategies for either eradicating mosquitoes or using population replacement by substituting uninfected mosquitoes with infected ones for dengue virus prevention were modeled. A model with a strong density dependent death function shows that population replacement can be realized if the initial ratio of number of infected to the total number of mosquitoes exceeds a critical value, especially when transmission from mother to offspring is perfect. However, with a weak density dependent death function, population eradication becomes difficult as the system’s solutions are sensitive to initial values. Using numerical methods, it was shown that population eradication may be achieved regardless of the infection ratio only when parameters lie in particular regions and the initial density of uninfected is low enough.     

Understanding epidemics from mathematical models: Details of the 2010 dengue epidemic in Bello (Antioquia,Colombia)

Dengue is the most threatening vector-borne viral disease in Colombia. At the moment, there is no treatment or vaccine available for its control or prevention; therefore, the main measure is to exert control over mosquito population. To reduce the economic impact of control measures, it is important to focus on specific characteristics related to local dengue epidemiology at the local level, and know the main factors involved in an epidemic. To this end, we used a mathematical model based on ordinary differential equations and experimental data regarding mosquito populations from Bello (Antioquia, Colombia) to simulate the epidemic occurred in 2010. The results showed that the parameters to which the incidence of dengue cases are most sensitive are the biting and mortality rates of adult mosquitoes as well as the virus transmission probabilities. Finally, we found that the Basic Reproductive Number (R₀) of this epidemic was between 1.5 and 2.7, with an infection force (Λ) of 0.061, meaning that R₀ values slightly above one are sufficient to result in a significant dengue outbreak in this region.     

Avian-human influenza epidemic model with diffusion

A diffusive epidemic model is investigated. This model describes the transmission of avian influenza among birds and humans. The behavior of positive solutions to a reaction-diffusion system with homogeneous Neumann boundary conditions are investigated. Sufficient conditions for the local and global asymptotical stability are given by spectral analysis and by using Lyapunov functional. Our result shows that the disease-free equilibrium is globally asymptotically stable, if the contact rate for the susceptible birds and the contact rate for the susceptible humans are small. It suggests that the best policy to prevent the occurrence of a pandemic is not only to exterminate the infected birds with avian influenza but also to reduce the contact rate for susceptible humans with the individuals infected with mutant avian influenza. Numerical simulations are presented to illustrate the main results.     

Controlling Aedes aegypti populations by limited Wolbachia‐based strategies in a seasonal environment

In this work, the linear feedback limited control strategy is proposed to indicate how the Wolbachia‐infected mosquitoes should be introduced in the seasonal environment to reduce the non‐Wolbachia mosquito population. The numerical simulations show that the proposed strategy reduces the population level of non‐Wolbachia mosquitos, avoiding mosquito spread and, consequently, reducing the number of cases of vector‐borne diseases.     

Optimal bed net use for a dengue disease model with mosquito seasonal pattern

Controversial results concerning the effectiveness of bed net in reducing dengue fever transmission make further research necessary in this direction. At this aim, we consider a mathematical model of dengue transmission where the use by individuals of insecticide‐treated bed nets is taken into account, combined or not with insecticide spraying. Furthermore, as climatic factors play a key role in mosquito‐borne diseases, we model the effect of seasonality through a periodic mosquito birth rate. We numerically investigate some specific scenarios according to different rainfall and mean temperature values. We set an optimal control problem to minimize the number of human infections and the cost of efforts placed into bed net adoption and maintenance and insecticide spraying. To assess the most appropriate strategy to eliminate dengue with minimum costs, we perform a comparative cost‐effectiveness analysis, which also shows how the cost‐benefit of intervention efforts is affected by changes in the amplitude of seasonal variation. One general result is that in any case the combination of bed net use and insecticide spraying produces the highest ratio of infections averted, whereas in terms of cost‐benefit only spraying campaigns should be implemented in control programs for regions with no large seasonality.     

Mathematical Analysis of SIR Epidemic Model with Piecewise Infection Rate and Control Strategies

The limitation of contact between susceptible and infected individuals plays an important role in decreasing the transmission of infectious diseases. Prevention and control strategies contribute to minimizing the transmission rate. In this paper, we propose SIR epidemic model with delayed control strategies, in which delay describes the response and effect time. We study the dynamic properties of the epidemic model from three aspects: steady states, stability and bifurcation. By eliminating the existence of limit cycles, we establish the global stability of the endemic equilibrium, when the delay is ignored. Further, we find that the delayed effect on the infection rate does not affect the stability of the disease-free equilibrium, but it can destabilize the endemic equilibrium and bring Hopf bifurcation. Theoretical results show that the prevention and control strategies can effectively reduce the final number of infected individuals in the population. Numerical results corroborate the theoretical ones.