Multi-agent systems in epidemiology: a first step for computational biology in the study of vector-borne disease transmission

Leveraging artificial intelligence and software engineering methods in epidemiology for the co-creation of decision-support tools based on mechanistic models

Epidemiological modeling is a key lever for infectious disease control and prevention on farms. It makes it possible to understand the spread of pathogens, but also to compare intervention scenarios even in counterfactual situations. However, the actual capability of decision makers to use mechanistic models to support timely interventions is limited. This study demonstrates how artificial intelligence (AI) techniques can make mechanistic epidemiological models more accessible to farmers and veterinarians, and how to transform such models into user-friendly decision-support tools (DST). By leveraging knowledge representation methods, such as the textual formalization of model components through a domain-specific language (DSL), the co-design of mechanistic models and DST becomes more efficient and collaborative. This facilitates the integration of explicit expert knowledge and practical insights into the modeling process. Furthermore, the utilization of AI and software engineering enables the automation of web application generation based on existing mechanistic models. This automation simplifies the development of DST, as tool designers can focus on identifying users' needs and specifying expected features and meaningful presentations of outcomes, instead of wasting time in writing code to wrap models into web apps. To illustrate the practical application of this approach, we consider the example of Bovine Respiratory Disease (BRD), a tough challenge in fattening farms where young beef bulls often develop BRD shortly after being allocated into pens. BRD is a multi-factorial, multi-pathogen disease that is difficult to anticipate and control, often resulting in the massive use of antimicrobials to mitigate its impact on animal health, welfare, and economic losses. The DST developed from an existing mechanistic BRD model empowers users, including farmers and veterinarians, to customize scenarios based on their specific farm conditions. It enables them to anticipate the effects of various pathogens, compare the epidemiological and economic outcomes associated with different farming practices, and decide how to balance the reduction of disease impact and the reduction of antimicrobial usage (AMU). The generic method presented in this article illustrates the potential of artificial intelligence (AI) and software engineering methods to enhance the co-creation of DST based on mechanistic models in veterinary epidemiology. The corresponding pipeline is distributed as an open-source software. By leveraging these advancements, this research aims to bridge the gap between theoretical models and the practical usage of their outcomes on the field.

Epidemic prediction of dengue fever based on vector compartment model and Markov chain Monte Carlo method

Background

Dengue epidemics is affected by vector-human interactive dynamics. Infectious disease prevention and control emphasize the timing intervention at the right diffusion phase. In such a way, control measures can be cost-effective, and epidemic incidents can be controlled before devastated consequence occurs. However, timing relations between a measurable signal and the onset of the pandemic are complex to be discovered, and the typical lag period regression is difficult to capture in these complex relations. This study investigates the dynamic diffusion pattern of the disease in terms of a probability distribution. We estimate the parameters of an epidemic compartment model with the cross-infection of patients and mosquitoes in various infection cycles. We comprehensively study the incorporated meteorological and mosquito factors that may affect the epidemic of dengue fever to predict dengue fever epidemics.

Results

We develop a dual-parameter estimation algorithm for a composite model of the partial differential equations for vector-susceptible-infectious-recovered with exogeneity compartment model, Markov chain Montel Carlo method, and boundary element method to evaluate the epidemic periodicity under the effect of environmental factors of dengue fever, given the time series data of 2000–2016 from three cities with a population of 4.7 million. The established computer model of “energy accumulation-delayed diffusion-epidemics” is proven to be effective to predict the future trend of reported and unreported infected incidents. Our artificial intelligent algorithm can inform the authority to cease the larvae at the highest vector infection time. We find that the estimated dengue report rate is about 20%, which is close to the number of official announcements, and the percentage of infected vectors increases exponentially yearly. We suggest that the executive authorities should seriously consider the accumulated effect among infected populations. This established epidemic prediction model of dengue fever can be used to simulate and evaluate the best time to prevent and control dengue fever.

Conclusions

Given our developed model, government epidemic prevention teams can apply this platform before they physically carry out the prevention work. The optimal suggestions from these models can be promptly accommodated when real-time data have been continuously corrected from clinics and related agents.

Cluster-Delay Mean Square Consensus of Stochastic Multi-Agent Systems with Impulse Time Windows

This paper investigates the cluster-delay mean square consensus problem of a class of first-order nonlinear stochastic multi-agent systems with impulse time windows. Specifically, on the one hand, we have applied a discrete control mechanism (i.e., impulsive control) into the system instead of a continuous one, which has the advantages of low control cost, high convergence speed; on the other hand, we considered the existence of impulse time windows when modeling the system, that is, a single impulse appears randomly within a time window rather than an ideal fixed position. In addition, this paper also considers the influence of stochastic disturbances caused by fluctuations in the external environment. Then, based on algebraic graph theory and Lyapunov stability theory, some sufficiency conditions that the system must meet to reach the consensus state are given. Finally, we designed a simulation example to verify the feasibility of the obtained results.

Modeling epidemic spread in transportation networks: A review

The emergence of novel infectious diseases has become a serious global problem. Convenient transportation networks lead to rapid mobilization in the context of globalization, which is an important factor underlying the rapid spread of infectious diseases. Transportation systems can cause the transmission of viruses during the epidemic period, but they also support the reopening of economies after the epidemic. Understanding the mechanism of the impact of mobility on the spread of infectious diseases is thus important, as is establishing the risk model of the spread of infectious diseases in transportation networks. In this study, the basic structure and application of various epidemic spread models are reviewed, including mathematical models, statistical models, network-based models, and simulation models. The advantages and limitations of model applications within transportation systems are analyzed, including dynamic characteristics of epidemic transmission and decision supports for management and control. Lastly, research trends and prospects are discussed. It is suggested that there is a need for more in-depth research to examine the mutual feedback mechanism of epidemics and individual behavior, as well as the proposal and evaluation of intervention measures. The findings in this study can help evaluate disease intervention strategies, provide decision supports for transport policy during the epidemic period, and ameliorate the deficiencies of the existing system.

•

Reviewed epidemic spread models and their applications in transportation networks.

•

Analyzed the advantages and limitations of epidemic spread model applications in transportation systems.

•

Summarized the emerging modeling requirements brought by the COVID-19 pandemic.

•

Proposed research trends and prospects for epidemic spread modeling in transportation networks.

Research perspectives on animal health in the era of artificial intelligence

Leveraging artificial intelligence (AI) approaches in animal health (AH) makes it possible to address highly complex issues such as those encountered in quantitative and predictive epidemiology, animal/human precision-based medicine, or to study host × pathogen interactions. AI may contribute (i) to diagnosis and disease case detection, (ii) to more reliable predictions and reduced errors, (iii) to representing more realistically complex biological systems and rendering computing codes more readable to non-computer scientists, (iv) to speeding-up decisions and improving accuracy in risk analyses, and (v) to better targeted interventions and anticipated negative effects. In turn, challenges in AH may stimulate AI research due to specificity of AH systems, data, constraints, and analytical objectives. Based on a literature review of scientific papers at the interface between AI and AH covering the period 2009-2019, and interviews with French researchers positioned at this interface, the present study explains the main AH areas where various AI approaches are currently mobilised, how it may contribute to renew AH research issues and remove methodological or conceptual barriers. After presenting the possible obstacles and levers, we propose several recommendations to better grasp the challenge represented by the AH/AI interface. With the development of several recent concepts promoting a global and multisectoral perspective in the field of health, AI should contribute to defract the different disciplines in AH towards more transversal and integrative research.

Mathematical Modeling Predicts That Strict Social Distancing Measures Would Be Needed to Shorten the Duration of Waves of COVID-19 Infections in Vietnam

Coronavirus disease 2019 (COVID-19) emerged in Wuhan, China in 2019, has spread throughout the world and has since then been declared a pandemic. As a result, COVID-19 has caused a major threat to global public health. In this paper, we use mathematical modeling to analyze the reported data of COVID-19 cases in Vietnam and study the impact of non-pharmaceutical interventions. To achieve this, two models are used to describe the transmission dynamics of COVID-19. The first model belongs to the susceptible-exposed-infectious-recovered (SEIR) type and is used to compute the basic reproduction number. The second model adopts a multi-scale approach which explicitly integrates the movement of each individual. Numerical simulations are conducted to quantify the effects of social distancing measures on the spread of COVID-19 in urban areas of Vietnam. Both models show that the adoption of relaxed social distancing measures reduces the number of infected cases but does not shorten the duration of the epidemic waves. Whereas, more strict measures would lead to the containment of each epidemic wave in one and a half months.

Risk prediction in surgery using case-based reasoning and agent-based modelization

Managing the risks arising from the actions and conditions of the various elements that make up an operating room is a major concern during a surgical procedure. One of the main challenges is to define alert thresholds in a non-deterministic context where unpredictable adverse events occur. In response to this problematic, this paper presents an architecture that couples a Multi-Agent System (MAS) with Case-Based Reasoning (CBR). The possibility of emulating a large number of situations thanks to MAS, combined with analytical data management thanks to CBR, is an original and efficient way of determining thresholds that are not defined a priori. We also compared different similarity calculation methods (Retrieve phase of CBR). The results presented in this article show that our model can manage alert thresholds in an environment that manages data as disparate as infectious agents, patient's vitals and human fatigue. In addition, they reveal that the thresholds proposed by the system are more efficient than the predefined ones. These results tend to prove that our simulator is an effective alert generator. Nevertheless, the context remains a simulation mode that we would like to enrich with real data from, for example, monitoring sensors (bracelet for human fatigue, monitoring, etc).

An agent-based model of tsetse fly response to seasonal climatic drivers: Assessing the impact on sleeping sickness transmission rates

BACKGROUND

This paper presents the development of an agent-based model (ABM) to incorporate climatic drivers which affect tsetse fly (G. m. morsitans) population dynamics, and ultimately disease transmission. The model was used to gain a greater understanding of how tsetse populations fluctuate seasonally, and investigate any response observed in Trypanosoma brucei rhodesiense human African trypanosomiasis (rHAT) disease transmission, with a view to gaining a greater understanding of disease dynamics. Such an understanding is essential for the development of appropriate, well-targeted mitigation strategies in the future.

METHODS

The ABM was developed to model rHAT incidence at a fine spatial scale along a 75 km transect in the Luangwa Valley, Zambia. The model incorporates climatic factors that affect pupal mortality, pupal development, birth rate, and death rate. In combination with fine scale demographic data such as ethnicity, age and gender for the human population in the region, as well as an animal census and a sample of daily routines, we create a detailed, plausible simulation model to explore tsetse population and disease transmission dynamics.

RESULTS

The seasonally-driven model suggests that the number of infections reported annually in the simulation is likely to be a reasonable representation of reality, taking into account the high levels of under-detection observed. Similar infection rates were observed in human (0.355 per 1000 person-years (SE = 0.013)), and cattle (0.281 per 1000 cattle-years (SE = 0.025)) populations, likely due to the sparsity of cattle close to the tsetse interface. The model suggests that immigrant tribes and school children are at greatest risk of infection, a result that derives from the bottom-up nature of the ABM and conditioning on multiple constraints. This result could not be inferred using alternative population-level modelling approaches.

CONCLUSIONS

In producing a model which models the tsetse population at a very fine resolution, we were able to analyse and evaluate specific elements of the output, such as pupal development and the progression of the teneral population, allowing the development of our understanding of the tsetse population as a whole. This is an important step in the production of a more accurate transmission model for rHAT which can, in turn, help us to gain a greater understanding of the transmission system as a whole.

Dynamic, spatial models of parasite transmission in wildlife: Their structure, applications and remaining challenges

Individual differences in contact rate can arise from host, group and landscape heterogeneity and can result in different patterns of spatial spread for diseases in wildlife populations with concomitant implications for disease control in wildlife of conservation concern, livestock and humans. While dynamic disease models can provide a better understanding of the drivers of spatial spread, the effects of landscape heterogeneity have only been modelled in a few well-studied wildlife systems such as rabies and bovine tuberculosis. Such spatial models tend to be either purely theoretical with intrinsic limiting assumptions or individual-based models that are often highly species- and system-specific, limiting the breadth of their utility. Our goal was to review studies that have utilized dynamic, spatial models to answer questions about pathogen transmission in wildlife and identify key gaps in the literature. We begin by providing an overview of the main types of dynamic, spatial models (e.g., metapopulation, network, lattice, cellular automata, individual-based and continuous-space) and their relation to each other. We investigate different types of ecological questions that these models have been used to explore: pathogen invasion dynamics and range expansion, spatial heterogeneity and pathogen persistence, the implications of management and intervention strategies and the role of evolution in host-pathogen dynamics. We reviewed 168 studies that consider pathogen transmission in free-ranging wildlife and classify them by the model type employed, the focal host-pathogen system, and their overall research themes and motivation. We observed a significant focus on mammalian hosts, a few well-studied or purely theoretical pathogen systems, and a lack of studies occurring at the wildlife-public health or wildlife-livestock interfaces. Finally, we discuss challenges and future directions in the context of unprecedented human-mediated environmental change. Spatial models may provide new insights into understanding, for example, how global warming and habitat disturbance contribute to disease maintenance and emergence. Moving forward, better integration of dynamic, spatial disease models with approaches from movement ecology, landscape genetics/genomics and ecoimmunology may provide new avenues for investigation and aid in the control of zoonotic and emerging infectious diseases.

Introduction to MAchine Learning & Knowledge Extraction (MAKE)

The grand goal of Machine Learning is to develop software which can learn from previous experience—similar to how we humans do. Ultimately, to reach a level of usable intelligence, we need (1) to learn from prior data, (2) to extract knowledge, (3) to generalize—i.e., guessing where probability function mass/density concentrates, (4) to fight the curse of dimensionality, and (5) to disentangle underlying explanatory factors of the data—i.e., to make sense of the data in the context of an application domain. To address these challenges and to ensure successful machine learning applications in various domains an integrated machine learning approach is important. This requires a concerted international effort without boundaries, supporting collaborative, cross-domain, interdisciplinary and transdisciplinary work of experts from seven sections, ranging from data pre-processing to data visualization, i.e., to map results found in arbitrarily high dimensional spaces into the lower dimensions to make it accessible, usable and useful to the end user. An integrated machine learning approach needs also to consider issues of privacy, data protection, safety, security, user acceptance and social implications. This paper is the inaugural introduction to the new journal of MAchine Learning & Knowledge Extraction (MAKE). The goal is to provide an incomplete, personally biased, but consistent introduction into the concepts of MAKE and a brief overview of some selected topics to stimulate future research in the international research community.

Individual-Based Models for Public Health

Today, infectious diseases represent a threatening concern for human health. Understanding their transmission, and possibly forecasting the dynamics of these pathogens, represents both a scientific and sanitary emergency. To this goal, mathematical modeling has been a widely used tool. Nevertheless, they have important limitations to explicitly model the mechanisms involved in the infectious processes at the individual level. Thanks to the increase of computing capacity, computational models such as individual-based models (IBMs) are very relevant for understanding the complexity of mechanisms at the individual level that can be involved in disease outbreaks. Their computational formalism allows a large flexibility, while they rely on the same philosophy than current models in mathematical epidemiology that have proved their relevance. In this chapter, we review the main qualities of IBMs, what kind of new knowledge they can bring and they have already produced in epidemiological modeling. Then, we highlight their caveats and what could be developed during the future years to make IBMs a more reliable and useful approach.

A Multi-Host Agent-Based Model for a Zoonotic, Vector-Borne Disease. A Case Study on Trypanosomiasis in Eastern Province, Zambia

BACKGROUND

This paper presents a new agent-based model (ABM) for investigating T. b. rhodesiense human African trypanosomiasis (rHAT) disease dynamics, produced to aid a greater understanding of disease transmission, and essential for development of appropriate mitigation strategies.

METHODS

The ABM was developed to model rHAT incidence at a fine spatial scale along a 75 km transect in the Luangwa Valley, Zambia. The method offers a complementary approach to traditional compartmentalised modelling techniques, permitting incorporation of fine scale demographic data such as ethnicity, age and gender into the simulation.

RESULTS

Through identification of possible spatial, demographic and behavioural characteristics which may have differing implications for rHAT risk in the region, the ABM produced output that could not be readily generated by other techniques. On average there were 1.99 (S.E. 0.245) human infections and 1.83 (S.E. 0.183) cattle infections per 6 month period. The model output identified that the approximate incidence rate (per 1000 person-years) was lower amongst cattle owning households (0.079, S.E. 0.017), than those without cattle (0.134, S.E. 0.017). Immigrant tribes (e.g. Bemba I.R. = 0.353, S.E.0.155) and school-age children (e.g. 5-10 year old I.R. = 0.239, S.E. 0.041) were the most at-risk for acquiring infection. These findings have the potential to aid the targeting of future mitigation strategies.

CONCLUSION

ABMs provide an alternative way of thinking about HAT and NTDs more generally, offering a solution to the investigation of local-scale questions, and which generate results that can be easily disseminated to those affected. The ABM can be used as a tool for scenario testing at an appropriate spatial scale to allow the design of logistically feasible mitigation strategies suggested by model output. This is of particular importance where resources are limited and management strategies are often pushed to the local scale.

Interactive machine learning for health informatics: when do we need the human-in-the-loop?

Machine learning (ML) is the fastest growing field in computer science, and health informatics is among the greatest challenges. The goal of ML is to develop algorithms which can learn and improve over time and can be used for predictions. Most ML researchers concentrate on automatic machine learning (aML), where great advances have been made, for example, in speech recognition, recommender systems, or autonomous vehicles. Automatic approaches greatly benefit from big data with many training sets. However, in the health domain, sometimes we are confronted with a small number of data sets or rare events, where aML-approaches suffer of insufficient training samples. Here interactive machine learning (iML) may be of help, having its roots in reinforcement learning, preference learning, and active learning. The term iML is not yet well used, so we define it as “algorithms that can interact with agents and can optimize their learning behavior through these interactions, where the agents can also be human.” This “human-in-the-loop” can be beneficial in solving computationally hard problems, e.g., subspace clustering, protein folding, or k-anonymization of health data, where human expertise can help to reduce an exponential search space through heuristic selection of samples. Therefore, what would otherwise be an NP-hard problem, reduces greatly in complexity through the input and the assistance of a human agent involved in the learning phase.

Predictive modeling of West Nile virus transmission risk in the Mediterranean Basin: how far from landing?

The impact on human and horse health of West Nile fever (WNF) recently and dramatically increased in Europe and neighboring countries. Involving several mosquito and wild bird species, WNF epidemiology is complex. Despite the implementation of surveillance systems in several countries of concern, and due to a lack of knowledge, outbreak occurrence remains unpredictable. Statistical models may help identifying transmission risk factors. When spatialized, they provide tools to identify areas that are suitable for West Nile virus transmission. Mathematical models may be used to improve our understanding of epidemiological process involved, to evaluate the impact of environmental changes or test the efficiency of control measures. We propose a systematic literature review of publications aiming at modeling the processes involved in WNF transmission in the Mediterranean Basin. The relevance of the corresponding models as predictive tools for risk mapping, early warning and for the design of surveillance systems in a changing environment is analyzed.

An agent-based model for the transmission dynamics of Toxoplasma gondii

Toxoplasma gondii (T. gondii) is a unicellular protozoan that infects up to one-third of the world's human population. Numerous studies revealed that a latent infection of T. gondii can cause life-threatening encephalitis in immunocompromised people and also has significant effects on the behavior of healthy people and animals. However, the overall transmission of T. gondii has not been well understood although many factors affecting this process have been found out by different biologists separately. Here we synthesize what is currently known about the natural history of T. gondii by developing a prototype agent-based model to mimic the transmission process of T. gondii in a farm system. The present model takes into account the complete life cycle of T. gondii, which includes the transitions of the parasite from cats to environment through feces, from contaminated environment to mice through oocysts, from mice to cats through tissue cysts, from environment to cats through oocysts as well as the vertical transmission among mice. Although the current model does not explicitly include humans and other end-receivers, the effect of the transition to end-receivers is estimated by a developed infection risk index. The current model can also be extended to include human activities and thus be used to investigate the influences of human management on disease control. Simulation results reveal that most cats are infected through preying on infected mice while mice are infected through vertical transmission more often than through infection with oocysts, which clearly suggests the important role of mice during the transmission of T. gondii. Furthermore, our simulation results show that decreasing the number of mice on a farm can lead to the eradication of the disease and thus can lower the infection risk of other intermediate hosts on the farm. In addition, with the assumption that the relation between virulence and transmission satisfies a normal function, we show that intermediate virulent lineages (type II) can sustain the disease most efficiently, which can qualitatively agree with the fact that the evolution of the parasite favors intermediate virulence. The effects of other related factors on transmission, including the latent period and imprudent behavior of mice, and prevention strategies are also studied based on the present model.

An agent-based model to study the epidemiological and evolutionary dynamics of Influenza viruses

Background

Influenza A viruses exhibit complex epidemiological patterns in a number of mammalian and avian hosts. Understanding transmission of these viruses necessitates taking into account their evolution, which represents a challenge for developing mathematical models. This is because the phrasing of multi-strain systems in terms of traditional compartmental ODE models either requires simplifying assumptions to be made that overlook important evolutionary processes, or leads to complex dynamical systems that are too cumbersome to analyse.

Results

Here, we develop an Individual-Based Model (IBM) in order to address simultaneously the ecology, epidemiology and evolution of strain-polymorphic pathogens, using Influenza A viruses as an illustrative example.

Conclusions

We carry out careful validation of our IBM against comparable mathematical models to demonstrate the robustness of our algorithm and the sound basis for this novel framework. We discuss how this new approach can give critical insights in the study of influenza evolution.

Mosquito Biting and Movement Rates as an Emergent Community Property and The Implications for Malarial Interventions

Malaria, a mosquito-vectored disease, continues to be one of the most important scourges afflicting humankind. In this paper, we take a mosquito-centric approach by studying mosquito states (i.e., energy, neurological health, and toxin information state) to demonstrate how key parameters of malaria, biting and movement rates and mosquito survival, are all emergent properties of those states when considered in the context of the background community interactions. We do so as follows: First, we develop a dynamic state variable model of mosquito biting and movement decisions that maximize mosquito expected reproductive success (fitness), and then we embed those optimal policies in a Monte Carlo simulation wherein mosquitoes attempt to feed on human hosts at domiciles where insecticide-treated bednets (ITNs) and insecticidal residual wall sprays (IRSs) are used. We find that biting rates, at the domicile level, are not impacted by mosquito state but that emigration rates from domiciles are determined by an interaction between mosquito energy state, information state, and risk of predation. This means that malaria incidence, at the village level at least, may be best understood as a response of mosquitoes to their ecological community that includes nectar-bearing plants, predators, the spatial arrangement of homes, and the protection of humans in those homes.