Towards an Agent-based Model to Monitor Epidemics and Chronic

Diseases in DR Congo

Jean-Claude Tshilenge Mfumu

, Annabelle Mercier

, Christine Verdier

and Michel Occello

Grenoble Alps University, CNRS, LIG, F-38000 Grenoble, France

Grenoble Alps University, LCIS, 50 Rue Barthélémy de Laffemas, 26000 Valence, France

Keywords: Epidemic, District Health, Agent-based Model, Simulation.

Abstract: Many contagious diseases occurred around sub-Saharan countries in the last decade due to the inefficiency

of health structures to anticipate disease outbreaks. In a huge poorly-infrastructured country such as The

Democratic Republic of Congo (DRC) with insufficient health staff and laboratory facilities, to provide

quick response to an urgent case of epidemic is challenging especially facing the development of its rural

areas. As DRC’s Health System has three levels (peripheral, regional and national levels), from producing

health data at peripheral to national level that takes the decision, it can take time resulting in the spread of

disease. The lack of communication between health centers and laboratory facilities in the same health zone

does not contribute to regional riposte. This paper proposes to face this problem using an agent-centered

approach to study through simulation how to improve the process. An experiment is described by

agentifying two health zones on the same regional level to show how it can reduce the decision time.. It

consists of 2 peripheral coordination offices, 2 labs and 2 health zones the former with 12 health centers and

the latter with 20 health zones. The interaction between these agents will provide a first model to be

compared with the current system in other to reduce decision time.

1 INTRODUCTION

Access to health care is a major concern in

developing countries. The Democratic Republic of

Congo ranks among the poorest countries according

to its HDI

(UNDP, 2016). Despite its millions of

hectares of arable land, this vast country of Central

Africa is experiencing serious difficulties in

improving the living conditions of its population,

particularly in the field of basic health care. Life

expectancy at birth is 50 and 53 respectively for men

and women (WHO, 2014).

The country is currently divided into the city-

province of Kinshasa and 25 other provinces. The

provinces are subdivided into territories which are

divided into sectors. To facilitate the supervision of

health structures, the DRC health system is divided

into three levels (Ministère de la Santé Publique de

la République Démocratique du Congo, 2006):

central, intermediate and peripheral. The nearest

level to the population is the peripheral area

composed of 518 health zones (HZ) that coordinate

Human Development Index

the actions of the health facilities. A HZ is divided

into Health Area (HA). One or more Health Centers

(HC) can be found at an HA. A General Referral

Hospital (GRH) ensures the complementary

packages of health activities of the HZ. The Central

(national) level defines the policies, strategies and

resources of the sector. It enforces strategies and

policies at the peripheral level through the

intermediate level called the Provincial Health

Division (PHD), which coordinates primary health

care and technical support activities for health zones

in a province.

Each HZ has a Health Information Bureau (HIB)

which retrieves the aggregated data from all its

supervised HA to national level for decision

measures. The HIB organizes weekly meetings with

the Health Zone Executive Team (HZET) to decide

on suspicious cases to report to the hierarchy. HZET

manages health facilities (HF), that includes HC and

GRH, and Community-based organizations (CO).

Figure 1 represents three levels of DRC Health

System in with each HZET supervises many HC and

one GRH at peripheral level. PHD at intermediate

level provides technical assistance to HZET while

Mfumu, J-C., Mercier, A., Verdier, C. and Occello, M.

Towards an Agent-based Model to Monitor Epidemics and Chronic Diseases in DR Congo.

DOI: 10.5220/0006557900830093

In Proceedings of the 11th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2018) - Volume 5: HEALTHINF, pages 83-93

ISBN: 978-989-758-281-3

Direction of Disease Control (DDC) under National

Health Minister (NHM) gives national policies to 26

PHD at central level. Provincial Health Minister

plays the role of political authority.

Figure 1: Three levels of DRC health system.

As a WHO member country, DRC benefits from

the technical and financial support of the partners to

respond to epidemics under the conditions stipulated

in the International Health Regulations (IHR)

(WHO, 2005). All cases of these four diseases must

be automatically notified to WHO: smallpox,

poliomyelitis due to wild-type poliovirus, severe

acute respiratory syndrome (SARS) and cases of

human influenza caused by a new subtype. On the

ever-changing list of diseases provided by IHR, each

country is free to add other diseases with epidemic

potential or not, which constitute a public health

problem.

Access to basic care is difficult for a large part of

the population. People visit the health facilities in

case of extreme urgency. This is more evident in

rural areas where the diminishing resources of

farmers do not allow them to consult medical

services often. Often health care is provided during

free medical workers’ campaigns.

To collect data about suspicious cases to the

hierarchy for decision-making, the most widely used

ways are described as follows (Ministère de la santé,

2012):

-Health facilities;

-Information from community members

who are experienced in unusual symptoms

and signs;

-Pharmacies that report increasingly

purchases of the same drugs or detect

recurrent treatment;

-School reporting an unusual rate of pupil

absences due to bizarre signs and

symptoms;

- New suspect cases discovered during

medical consultations;

-Medical records providing information on

new diseases detected in the population;

-Media (radio, television or newspapers)

reporting unusual events;

-Civil registries indicating an increasing

rate of maternal deaths;

Despite great efforts to improve disease

surveillance and response, DRC faces big challenges

in identifying, diagnosing and reporting infectious

diseases properly due to the remoteness of

communities, the inadequate transport and

communication infrastructures and the lack of

qualified health staff and laboratory facilities to

ensure accurate diagnosis.

The challenge, in this paper, is to find new

technical solutions based on real population life and

situation to improve health services organization and

data sharing in order to detect infectious disease

very quickly, organize the response and prevent the

spread of the disease.

2 ISSUES

In this paper, we present a part of this challenge. We

propose a multi-agent system to simulate the

interactions between actors working together to

organize an optimal response to epidemic detection.

When a new case of infectious disease is suspected

in HC, actors will collaborate to report it to PHD

through HZET. The approach will be based on the

current DRC heath system processes to extract

relevant actors’ tasks. The identification of these

actors and their tasks will provide the opportunity to

simulate a new system that distributes the entire

competences of the old heath system to those actors

in order to improve their collaboration and

eventually shorten the making-decision time

response. Work-sharing protocols will be proposed

to simplify the complexity of the data sources.

A MultiAgent System (MAS) is a set of agents

situated in a common environment, which interact

and try to reach a set of goals. Through these

interactions, a global behavior, more intelligent than

the sum of the local intelligence of multiagent

system components, can emerge. By ‘agent’ we

mean a software entity evolving in an environment

that it can perceive and within which it reacts. It is

provided with autonomous behaviors and some

objectives. Autonomy is the main concept in the

agent issue: it is the ability of agents to control their

actions and their internal states. The agents’

autonomy implies no centralized control

(Wooldridge, 1999). One of the advantages of MAS

HEALTHINF 2018 - 11th International Conference on Health Informatics

is to model systems where a global description is not

possible at any given moment. Multiagent

conception is well suitable to model actors described

in Figure 1.

This paper focuses precisely on improving the

process of reporting health data from the peripheral

level to the hierarchy for rapid decision-making and

anticipate as much as possible the medical response..

Hierarchical dependency between levels forbids

periphery to directly transmit health data to National

level for quick decision. As information must pass

through intermediate level (PHD) combined with

often defective means of communications, it

drastically hampers the fight against a propagation

of a disease.

The next section shows the related work in

healthcare and multiagent domains. Section 3

describes the healthcare system and problematic in

DRC. The methodological approach and agent’s

models are explained in section 4. The model is

validated by a simulation presented in section 5.

Future research directions and conclusion are

developed in section 6.

3 RELATED WORK

Information and Communication Technology is a

powerful solution for health care in developing

countries (Greenberg, 2005). It made possible the

improvement of remote patient follow-up (Wouters

and all, 2009), controlling the progression of malaria

(Zurovac and all, 2012), improving the uptake of

information from health systems (Mutale and All,

2013).

Mobile phone coverage in Africa grew from 10

percent in 1999, 65 percent in 2008 to more than 70

percent in 2012(Aker and Mbiti, 2010). This

technology is used to cover numeric fraction. To

improve drug adherence and suppression of plasma

HIV-1 RNA load in Kenyan, mobile phone

communication between health-care workers and

patients starting antiretroviral therapy was set up

(Lester and all, 2010).Text-message reminders sent

to health workers' mobile phones improved and

maintained their adherence to treatment guidelines

for outpatient pediatric malaria (Zurovac and all,

2011).

Phone traces are powerful tools to estimate

population migration while investigating an

outbreak. These techniques were used to demonstrate

the feasibility of rapid estimates and to identify areas

at potentially increased risk of outbreaks in Haiti

(Bengtsson, 2011). They produced reports on SIM

card movements from a cholera outbreak area at its

immediate onset and within 12 hours of receiving

data. Results suggest that estimates of population

movements during disasters and outbreaks can be

delivered rapidly and with potentially high validity

in areas with high mobile phone.

A trial of mobile phone text messaging for

diabetes management in an eight-month period to

transmit data such as blood glucose levels and body

weight to a server that automatically answered with

a monthly calculated glycosylated hemoglobin

result. The trial results suggest that sms may provide

a simple, fast and efficient adjunct to the

management of diabetes ( Ferrer-Roca, 2004).

In developed countries SMS messages have been

widely used to remind patients of scheduled

appointments (Hasvold, 2011; Car and all, 2008).

Similarly, more complex mobile phone applications

have shown significant improvement in the follow-

up of malaria patients in Thailand (Meankaew,

2010). The same approaches have been tested in

Africa as part of the SMS reminder package to

improve patients’ adherence to antimalarial

treatment schedules in .six sub Saharan countries

(Zurovac, 2012).

Even if text messaging is the simplest and the

most widely used technology function for which all

of the reviewed studies have shown ease of use in

reporting periodic data from the health system

periphery to control managers, it however remains to

be proved for interventions targeting individual

patients, for whom a high facility workload or

illiteracy may present a barrier (Zurovac, 2012).

Simulation has a broad application potential in

healthcare. The more general classification is

clinical, operational, managerial and educational

simulation (Barjis, 2011). Managerial and

operational of simulation are closely interrelated.

These two together are the core components for

healthcare process management. Some challenges

and trends of simulation models in healthcare in the

past two decades have been developed (Almagooshi,

2015). The design of a web-based clinical decision

support system that guides patients with Low Back

Pain in making suitable choices on self-referral has

been experienced in Netherlands (Nijeweme-

d’Hollosy and all, 2016).

MAS is used to describe an approach to the

analysis and development of telemedicine systems

(Mea, 2001), to manage communications in wireless

sensor networks (Jamont and Occello, 2007), the

epidemiological decision support system (Weber and

all, 2006), the care of seniors at home (Mercier and

all, 2013), decision-making for the monitoring and

Towards an Agent-based Model to Monitor Epidemics and Chronic Diseases in DR Congo

prevention of epidemics (Younsi, 2016), evaluation

of the disaster response system (Bae, 2017), medical

sensor modules in conjunction with

wireless communication technology supporting a

wide range of services including mobile

telemedicine, patient monitoring, emergency

management and information sharing between

patients and doctors or among the healthcare

workers (Byung-Mo and All, 2006) .

MAS can be considered a suitable technology for

the realization of applications for providing

healthcare and social services where the use of

loosely coupled and heterogeneous components, the

dynamic and distributed management of data and the

remote collaboration among users are often the most

requirements (Bergenti, Poggi, Tomaiuolo, 2013).

Cooperation in the Agent Technology can provide

better healthcare than the traditional medical system

(Jemal and All, 2015). Real programs built on the

multiagent paradigm are still evolving towards a

complete maturity, and the variety and complexity

of the e-health scenario make it one of most

interesting application fields, able of verifying the

advantages of their use of the conditioning their

evolution (Bergenti, Poggi, Tomaiuolo, 2016).

MAS was used to monitor a generic medical

contact center for chronic disease environment,

detect important cases, and inform the healthcare

and administrative personnel via alert messages,

recommendations, and reports, prompting them to

action (Koutkias, Chouvarda and Maglaveras,

2005). Developed MAS applications in healthcare

can provide a reasonable way to mitigate the cost

due to increased demand for services (Shakshuki and

all, 2015).

An Agent-Based Model (ABM) with Geospatial

and Medical Details was used to evaluate the

efficiency of disaster responders to rescue victims in

a mass casualty incident situation in South Korea

(Bae, 2017).

ABM can cooperate to share tasks between

sensors to observe a phenomena (Jamont, 2009), to

manage diabetes treatment between Caregivers and

Patients. The usability evaluation of a collaborative

information system for dementia assessment built

using a user-centred design approach was

experienced in Norway (Berglind and all, 2016). But

from several research papers we have reviewed we

didn’t find a paper addressing ABM in the sharing of

tasks of the actors involved in the processes of the

feedback of the multi-source health information and

the organization of the response to a disease with

high epidemic potential.

4 HEALTH DISTRICT IN DRC

4.1 Administrative Structures

The management of the patient and the reporting of

suspected cases are managed by the peripheral level

through health centers and the general reference

hospitals. The health data collected by the HF are

transmitted to the HZET for consolidation and

transmission of aggregated data from the HZ to the

PHD. This intermediate level structure convenes

weekly meetings to analyze data from each HZ,

decides on actions to be taken and produces

consolidated data from across the province.

The PHD must transmit the health data from its

province to the central level for a second analysis

and national consolidation. If suspected cases

reported by HZ require deeper investigation,

laboratory tests or kits, the PHD will seek technical

and financial support from the central level in the

event of its inability to provide the necessary means.

The Disease Control Direction (DCD) is the

central respondent. It also organizes weekly

meetings to analyze health data from all provinces. It

often provides advice and recommendations to PHD

for monitoring suspect cases in accordance with the

national policy of the sector. It can solicit

government authorities, special programs, partners

and even the international community. Figure 1

shows the data from the periphery to the national

level. Since decision-making is pushed back to the

central level, it can intervene belatedly at the risk of

witnessing an alarming spread of an epidemic with

high epidemic potential.

4.2 Structure Dependencies

The first difficulty in managing epidemics begins

with the processing of data from multiple sources at

the HZ level. National policy has expanded the list

of groups of individuals who can retrieve

information from suspect cases (Figure 2). This

information, which often comes in the form of phone

calls or narrative, is not exhaustive. Hence the

interest in diversifying the mode of communication

by adding text and voice sms, tweets and phone calls

on green lines.

A second difficulty in the accurate identification

of suspect cases is the insufficient number of

qualified health personnel (Ministère du Plan et de

Suivi de la Révolution de la République

Démocratique du Congo, 2014). In spite of the

training courses organized by the HZET for the

benefit of community relays and staff of health

HEALTHINF 2018 - 11th International Conference on Health Informatics

facilities, there are gaps in the implementation of the

information brought to their attention. For example,

the PHD conducts a thorough investigation by

qualified personnel as soon as the number of

suspected cases reaches the threshold for each

pathology. Lack of information on the list of the

nearest laboratories delays response time to confirm

cases and ensure accuracy of diagnosis.

Hierarchical dependence does not favor

communication between structures of the same level.

This is the case for the health areas of a HZ,

contiguous health areas but belonging to different

PHD. This lack of dialogue can lead to the non-

detection of an epidemic for the simple reason that

the number of cases is not reached in a HZ.

However, by combining this number with that of the

contiguous health areas, we could detect the

pathology at the intersection of the provinces, which

constitutes a business lock.

Figure 2: Many structures are designated to produce

Health data at each Health Zone.

4.3 Collection and Response for

Epidemic Surveillance

The designated structures report information relating

to suspect cases to the FOSA or HZET. Apart from

the pathologies of the International Sanitary

Regulations (smallpox, poliomyelitis due to wild

polio virus, human influenza and Severe Acute

Respiratory Syndrome (SARS)), the authorities have

added to the list of diseases with epidemic

Eradication measures or Elimination and other

chronic diseases. Reporting of cases is immediate,

weekly, monthly or quarterly according to the

contagiousness of the pathologies.

4.4 Data Collection and Epidemics

Response

As soon as they appear, suspect-cases information

must be transmitted to HZET by all data providers

indicated on 3.2. . When the number of suspect cases

in HZ equals to the threshold according to the

pathology observed, a rapid riposte team (RRT) has

to investigate some HC and the population of the

concerned HZ to make sure the allegation was

correct. The investigation of RRT team could result

to laboratory tests of some samples. In case of

riposte many hierarchical structures such as PHD

and national level would intervene to provide

technical and financial support.

The process used to organize riposte (Ministère

de la Santé Publique de la RDC, 2012) is shortly

described in Figure 3. HZET analyze the report of

surveillance to determine if the number of suspect

case has reached the threshold to order an

investigation. RRT will research new cases at HA

according to the clinical definition of case. It will

find out new determinants of the outbreak to report

to PHD in other to realize the response.

Figure 3: Cycle of outbreak response.

A final evaluation of outbreak response

presented as a report of the process can be shared

with other HZ and Health Facilities.

This type of system is well suited to MAS using

an AEIO representation. The real system is analyzed

with four elements the Agent, the Environment, the

Organizations and the Interactions between the

agents. This model will be detailed in the next part.

Towards an Agent-based Model to Monitor Epidemics and Chronic Diseases in DR Congo

5 INDIVIDU-CENTERED

MODELS

“The process of designing a model of a real system

and conducting experiments with this model for the

purpose either of understanding the behavior of the

system or of evaluating various strategies (within the

limits imposed by a criterion or a set of criteria) for

the operation of the system” is a definition given by

(Shannon, 1977) for simulation activity.

At this stage of this research, the simulation's

objective is to understand the DRC healthcare

system for Epidemic diseases. Multiagent-based-

simulation (MABS) allows explicitly modeling the

behavior of each individual and viewing the

emergent system from the interactions between the

individuals.

In further work, on the one hand, we will

determine some metrics to analyze process

simulations and on the other hand, we plan to

develop modules in embedded systems (like phones

or tablets) to assist the end-user in the data

collection, coupled with the multiagent system.

(Morvan, 2013) proposes a survey on MABS and

presents several multiagents platforms. In these

existing platforms, we have not found solutions

which can act as both a simulation system and a tool

to end-users on embedded systems.

However, the DIAMOND method and the

MASH simulator developed in LCIS Laboratory will

be used to model and simulate the system (Jamont

and Occello, 2007, 2009).

The hierarchical organization for collecting data

in DRC (healthcare system) administration is a good

candidate for a multiagent model because there are

several kinds of agent with their own goals which

share the same global achievement. In the process

described previously (figure 3), the agents use some

knowledge and tasks to perform a main goal

together: collecting data in order to respond with

efficiency to epidemic. The process is modeled by

agents able to be simulated in the MASH simulation

platform. The advantage of doing this is to have an

individual-centered vision of the process. After that,

we will be able to contribute to the improvement of

the process with an exterior view provided by the

simulation and propose changes and ideas to

improve the response time, for example.

This section shows the steps to break down

multiagent system’s elements.

5.1 Agents’ Tasks and Knowledge:

The Internal Behavior

For starting the analyses, each individual agent’s

behavior is studied. It is a way of seeing things at a

micro level, the phenomenon (at macro level) does

not change and the process remains the same but the

observer’s level changes.

The goal is to be able to adjust the behavior of

each individual agent and possibly add skills to

certain nodes or node types.

The first step is to model into agent the elements

of the process. For the problem, it is the health

centers, the main hospital of the health areas,

province districts and national health entity related

to the administrative structures and HZET and RRT

for human team working group. The figure 4 shows

the internal behavior of the RRT.

For each agent, we have to list all their skills,

what information will be required to store and to

handle and how the agent acquires this information.

This information should be acquired directly by

perception (e.g. the user grasps something) or on

demand by asking other nodes (higher hierarchy or

same level nodes).

With this step, we obtain for each agent, a vision

of the relevant knowledge to perform its individual

tasks. This is all the necessary information for the

agent who works in the system. The result is a set of

tasks that the agent can perform. These tasks

correspond to the skills of each node. Some skills

are made locally without need to contact other

agents. But to achieve a goal, an agent should have

partial information and needs to ask other agents to

complete their goal. However, we will have

cooperative behavior in opposition to completely

individual behavior entirely internal to the agent.

This kind of social behavior reflecting an interaction

among several agents: either to gain information or

to share tasks.

HEALTHINF 2018 - 11th International Conference on Health Informatics

Figure 4: Investigating an outbreak.

5.2 Agent’s Sharing Data and

Interactions

5.2.1 Social Behavior

In this step, we will have to create interactions

between nodes for example to back up information

(HCR to HZET) or to receive orders (HZET from

PHD). These interactions should be between

different partner groups such as health areas. In the

implementation, we define very simple interaction

protocols for data exchange such as receiving

information, answers/queries or order to perform a

task. For some tasks, such as health alert

surrounding areas, it is no longer just a request for

information but cooperative behavior; that is to say

behavior that brings into play several kinds of

agents. In need thereof soliciting other agents, the

agent is led to use interaction protocols. We will

therefore define how this behavior will be realized

by defining a more sophisticated interaction protocol

than query/response. Diverse protocols are available

for negotiating, giving orders, waiting for answers.

The interaction patterns that will govern this

cooperative behavior will be organized between the

agents.

5.2.2 Interaction Protocols

The protocol is a part of the agents’ knowledge.

Agents have a list of protocols that they should

initiate or that they are able to use to answer others.

For the moment, we use a simple protocol with two

states as represented in figure 6. For example, Agent

A1 launches an instance of protocol P1, it is in the

state S1. The agent A2 receives a solicitation of A1

with the performative “information” in the state S1.

A2 knows the protocol and searches the next

transition; it passes in state S2 and sends an

acknowledgment to agent A1. A1 treats the message

and the conversation finishes. The ACL FIPA

compliant Performatives are used:

• RDCMessage.ACL_QUERY_REF for

queries/answers,

• RDCMessage.ACL_REQUEST for an order

to perform a task and,

• RDCMessage.ACL_INFORM for

inform/acknowledgment.

5.2.3 Position in the Organization

The last step is to take into account the position in

the organization when an agent will initiate

interactions with each other. An organization should

be a hierarchy or a simple group. As an example, to

alert neighboring HZ, health center agents will know

the surrounding areas of health, which is a group,

that is to say an organization in the multiagent

system and make a decision based on its position in

this group. The agent’s position in the organization

is integrated in the decision making loop.

5.2.4 Agent’s Internal Decision Making

Loop

The previous steps give the agent’s skills, the

agent’s complex behavior (internal and social), the

knowledge of interaction protocols. We will now

build the agent decision loop. On the one hand

purely individual behavior runs only with an agent’s

context information and does not need other agents

to complete the agent goal. On the other hand, social

behavior involves relationships with other agents.

Using an individual-centered approach defines the

agent at micro level so the interactions with other

agents have to be merged with the internal behavior

in the agent’s decision loop. The individual and

cooperative behaviors are both integrated into the

decision making loop. In the individual behavior,

there is a set of tasks that are launched in the internal

Towards an Agent-based Model to Monitor Epidemics and Chronic Diseases in DR Congo

decision loop. In its decision loop, the agent should

have to respond to the message from others’ which

are part of interaction protocols initiated by other

agents or parts of own launch of the agent

interaction patterns. These tasks have to be

synchronized with the messages received from the

others agents.

As an external view, basically, huge decision

loops, which are decentralized in the several kinds of

agents, seem to be synchronized at the system level,

but in fact, each agent decides what state it passes

according to its knowledge and the state of its

interactions.

5.3 Collaborative Tasks of Agents

Agents would collaborate to achieve some

objectives. To investigate on HA, RRT must wait for

an order from HZET. The later receives health data

every week from HC and checks if the threshold of

the followed pathology has been reached. The same

collaboration is needed between RRT and HA, RRT

and Laboratory. The sequence diagram (Figure 5)

gives a snapshot of the kind of collaboration found

in agents concerned with an outbreak investigation.

Figure 5: Sequence diagrams of investigating an outbreak.

5.3.1 Message Format for Interaction

The messages exchanged between agents contains

sender and receiver agents, protocol information and

data to manage like [sender; receiver; conversation;

perform; protocol; inst_prot; state_prot; data]. The

data follow a format according to the performative.

To interact through a message sent by another

agent, a simple protocol is established. For instance

when RRT asks a laboratory to perform exams, he

has to first check its state to be convinced that it can

answer his request. A simple protocol with

acknowledgment is used.

Figure 6: Protocol for information with acknowledgment.

The agent changing state when he asks for

information and when he receives the answer to his

request is represented in Figure 6. In a future

simulation, a negotiation protocol with a call for

proposal to several laboratories will be tested.

However, the agent launching the conversation

should negotiate among laboratories which one is

available, near or powerful.

6 SIMULATION RESULTS

In order to test and evaluate our approach, we

adapted MASH simulator developed for a wireless

sensor multiagent system (Jamont and Occello,

2009). We focus our simulation to the former

Equateur Province that was split in 2015 into the

new, smaller Equateur province, as well as

the Tshuapa, Mongala, Nord-Ubangi and Sud-

Ubangi provinces.

Located in the north of the country, the province

bordered the Republic of the Congo to the west,

the Central African Republic to the north, to the east

the Orientale province, and to the south the Kasai-

Oriental, Kasai-Occidental.

The former Equateur registered 18% of cholera

outbreak suspect cases on 2016 with 31% of death

(Ministère de la Santé Publique de la RDC, 2017).

The crossing of many rivers facilitates the spread of

a disease such as cholera if health staffs don’t alert

the population as soon as a suspect case is noticed.

The simulation concerns precisely Tshuapa

Province that doesn’t register cholera outbreak

suspect case on 2016 while its neighbors Equateur

and Mongala provinces have respectively 2751 and

1781 suspect cases with deaths. The main idea is to

see how the future system would react if each actor

of health system could perform his own task with

autonomy. These experiences could result to many

scenarios and the best of them will be proposed to

DRC’s Health System to reduce time decision as

HEALTHINF 2018 - 11th International Conference on Health Informatics

each actor can execute his talks according to the

knowledge of environment and outbreak he will be

provided with.

We chose two HZ of Tshuapa PHD for the

simulation. Befale and Boende are HZ that didn’t

bordered with two provinces crossed by rivers that

encountered many deaths due to Cholera at the same

period.

To respond to an outbreak noticed at a HC,

health staff of the concerning HC must refer to

HZET. In their turn HZET must refer to PHD and

PHD to central level. This chain of hierarchical

contacts can enlarge time decision.

In our simulation we worked with this

hypothesis: each actor must contact immediately the

nearest one able to answer to his request or to use his

information to make decision. We considered only

three HA of each HZ and twenty HC in blue with

two HZET (in yellow for each HS) and two medical

test laboratories (Labo in green square).The

suspicious case detected in HC #6 is in red. Figure

illustrates those actors working as Agents in two HZ

named Befale and Boende. The map represents the

former Equateur Province.

Figure 7: Collecting data system simulation in Befale and

Boende.

The simulation trace file (Figure 8) shows the

communication between the agents. As they are

autonomous they perform their own tasks like

“conduct an investigation” and manage messages

like “receive sample to analyze” or “send suspicious

case detected”. The HZET Agents (#1 and #2) are

waiting for a suspect case message from any HC.

Whenever HZET Agent receives such a message he

builds a RRT Agent able to investigate in HAs

around suspect case. RRT Agent could send samples

to LaboAgent while searching for new cases in HAs.

The organization of outbreak riposte depends on the

results from LaboAgent and investigation report

from RRT Agent. .

The agents operate independently: RRTAgent

completes a full investigation, LaboAgent conducts

quality medical testing and PHD manages all

informations from HZET under it supervision.

Message synchronization between kinds of agent is

done in the agent's decision loop. A protocol with

two states is used and implements KQML-like

performatives. The four numbers in the message are

“1” for inform (give information), “2” for query

information (ask for an information) and “3” for

request (ask for a task to be done). The agents

communicate and achieve their goal by reacting to

messages from others or executing their inner task as

response to a query.

Figure 8: Simulation trace.

7 CONCLUSIONS

In this paper, we presented how ABM can be used to

monitor an epidemic outbreak in two Health zones

(Befale and Boende). The process of collecting

health data in health Zones and the response given

by provincial and national levels were analyzed. A

multiagent model for healthcare system in DRC has

been built to represent the use case. This first

experience of our research and the results described

in the section 5 convince us that it is possible to

make a social simulation of the real system. It’s

feasible to simulate the hierarchical organization of

the administrative structures, to collect and manage

health data from health system actors.

Towards an Agent-based Model to Monitor Epidemics and Chronic Diseases in DR Congo

Our self-centric analysis and simulation give a

promising result we would like to implement to a

real system by making comparison with the manual

process used to respond to an epidemic disease in

DRC. With certainty, we have some perspectives to

use the simulations. From the study of the current

process and the analyses of its simulation, the

process will be improved by extracting best practices

for a subset of tasks and actors. This paper

represents a first step and a positive sign that shows

that the multiagent solution represents a good

approach to help in decreasing the delay for

riposting.

The multiagent model must be improved. At

first, we are going to integrate analyses criteria on

the agents like the response time or the agent’s load

that is the number of requests resulting from other

agents to analyze the process and propose

improvements. We should simplify or improve

communication and information management.

Applying enriched agents’ behaviors will give us the

opportunity to simulate an operational system with a

new distribution of powers between agents so as to

improve collaboration and shorten the response time

between the actors.

A second axis of improvement is situated at the

level of the interaction protocols. For the moment,

the protocols described in section 4 have two states.

The social behavior of the agents should be

complicated by implementing for example a call for

proposal in the appropriate situations (medical

analyses in agent laboratory).

At the model level, additional work will be done

on the organization: the information will not only be

managed in a hierarchical way but with groups

dynamically built with explicit criteria such as

geographical location.

The last perspective is to couple the simulation

system with embedded devices to collect data. A

part of the agent perception should be the result of

the automation of a portion of the process embedded

in a phone or a tablet.

REFERENCES

Aker, J. C., and Mbiti, I. M., 2010. Mobile phones and

economic development in Africa. The Journal of

Economic Perspectives, 24(3), 207-232.

Almagooshi, S., 2015. Simulation modelling in healthcare:

Challenges and trends. Procedia Manufacturing,

3(supplement C):301–307. 6th International

Conference on Applied Human Factors and

Ergonomics and the Affiliated Conferences.

Bae, Jang Won, Shin, Kyohong, LEE, Hyun-Rok, 2017.

Evaluation of Disaster Response System using Agent-

Based Model with Geospatial and Medical Details.

IEEE Transactions on Systems, Man, and Cybernetics:

Systems.

Barjis, Joseph, 2011. Healthcare simulation and its

potential areas and future trends. SCS M&S Magazine,

vol. 2, no 5, p. 1-6.

Berglind Smaradottir, Santiago Martinez, Elisabeth Holen-

Rabbersvik, TorunnVatnøy and Rune Fensli, 2016.

Usability Evaluation of a Collaborative Health

Information System. Lessons from a User-centered

Design Process. Healthinf 2016.

Bengtsson, L., Lu, X., Thorson, A., Garfield, R., and von

Schreeb, J., 2011. Improved response to disasters and

outbreaks by tracking population movements with

mobile phone network data: A post-earthquake

geospatial study in Haiti. PLoS Med8, 8.

Bergenti, F., Poggi, A., & Tomaiuolo, M., 2016. Multi-

Agent Systems for E-Health and Telemedicine. In M.

Cruz-Cunha, I. Miranda, R. Martinho, & R. Rijo

(Eds.), Encyclopedia of E-Health and Telemedicine

(pp. 688-699). Hershey, PA: IGI Global.

Bergenti, F., Poggi, A., & Tomaiuolo, M. , 2013. Using

Multi-Agent Systems to Support e-Health Services. In

M. Cruz-Cunha, I. Miranda, & P. Gonçalves (Eds.),

Handbook of Research on ICTs for Human-Centered

Healthcare and Social Care Services (pp. 549-567).

Hershey, PA: IGI Global.

Byung-Mo Han, Seung-Jae Song, Kyu Min Lee, Kyung-

Soo Jang, Dong-Ryeol Shin, 2006. Multi Agent

System based Efficient Healthcare Service, 2006. In

the 8th International Conference on Advanced

Communication Technology, 2006. ICACT.

Car J, Gurol-Urganci I, De Jongh T, Vodopivecjamsek V,

Atun R., 2008. Mobile Phone Messaging Reminders

For Attendance At Scheduled Healthcare

Appointments. Cochrane Database of Systematic

Reviews.

Ferrer-Roca, O., Cardenas, A., Diaz-Cardama, A., and

Pulido, P., 2004. Mobile phone text messaging in the

management of diabetes. Journal of Telemedicine and

Telecare, 10(5):282–285.

Greenberg, A. , 2005. ICTs for poverty alleviation: basic

tool and enabling sector. Technical report, Sida 2005

ICT for Development Secretariat, Department for

Infrastructure and Economic Cooperation.

Hasvold, P. E. and Wootton, R., 2011. Use of telephone

and sms reminders to improve attendance at hospital

appointments: a systematic review. Journal of

Telemedicine and Telecare, 17(7):358-364.

Jamont, J.-P. and Occello, M. , 2007. Designing embedded

collective systems: The diamond multiagent method.

In International Conference on Tools with Artificial

Intelligence, pages 91–94, Patras Greece.

Jamont, J.-P. and Occello, M., 2009. A multiagent tool to

simulate hybrid real/virtual embedded agent societies.

In IEEE International Conference on Intelligent Agent

Technology.

HEALTHINF 2018 - 11th International Conference on Health Informatics

Jamont, J.-P., Occello, M., and Lagreze, A. , 2010. A

multiagent approach to manage communication in

wireless instrumentation systems. Measurement,

43(4):489–503.

Jemal, H., Kechaou, Z., Ayed, M. B., and Alimi, A. M. ,

2015. A multi agent system for hospital organization.

International Journal of Machine Learning and

Computing, 5(1):51–56.

Koutkias, V. G., Chouvarda, I., and Maglaveras, N., 2005.

A multiagent system enhancing home-care health

services for chronic disease management. IEEE

Transactions on information technology in

biomedicine, 9(4).

Lester, R. T., Ritvo, P., Mills, E. J., Kariri, A., Karanja, S.,

Chung, M. H., Jack, W., Habyarimana, J., Sadatsafavi,

M., Najafzadeh, M., Marra, C. A., Estambale, B.,

Ngugi, E., Ball, T. B., Thabane, L., Gelmon, L. J.,

Kimani, J., Ackers, M., and Plummer, F. A. , 2010.

Effects of a mobile phone short message service on

antiretroviral treatment adherence in Kenya (weltel

kenya1): a randomised trial. The Lancet,

376(9755):1838 – 1845.

Mea, D., 2001. Agents acting and moving in healthcare

scenario - a paradigm for tele medical collaboration.

IEEE Transactions on information technology in

biomedecine, 5(1).

Meankaew, P., Kaewkungwal, J., Khamsiriwatchara, A.,

Khunthong, P., Singhasivanon, P., and Satimai, W. ,

2010. Application of mobile-technology for disease

and treatment monitoring of malaria in the “better

border healthcare programme”. Malaria Journal,

9(1):237.

Mercier, A., Raievsky, C., Occello, M., and Genthial, D.,

2013. Solutions multi-agents pour la prise en charge

domicile des seniors. Ingénierie des systèmes

d’information, 18(6):83–112.

Morvan, G., 2013. Multi-level agent-based modeling: a

literature survey. The Computing Research Repository,

abs/1205.0561.

Mutale, W., Chintu, N., Amoroso, C., Awoonor-Williams,

K., Phillips, J., Baynes, C., Michel, C., Taylor, A., and

Sherr, K., 2013. Improving health information systems

for decision making across five sub-saharan African

countries: Implementation strategies from the African

health initiative. BMC Health Services Research,

13(2):S9.

Ministère de la Santé Publique de la RDC, 2006.Recueil

des normes de la Zone de Santé. Kinshasa.

Ministère de la Santé Publique de la RDC, 2012. Guide

Technique pour la Surveillance intégrée de la Maladie

et riposte. Kinshasa.

Ministère de la Santé Publique de la RDC, 2014. Plan de

Développement de l'Informatique de la Santé –

PNDIS. Kinshasa.

Ministère de la Santé Publique de la RDC, 2017. Rapports

annuels de surveillance des maladies épidémiques à

potentiel épidémique en République Démocratique du

Congo 2014-2016. Kinshasa.

Ministère du Plan et Ministère de la Santé Publique, 2014.

Deuxième enquête démographique et de santé

(eds-rdc II 2013-2014). Kinshasa.

Mutale, Chintu, Amoroso, Awoonor-Williams, Phillips,

Baynes, Michel, Taylor, Sherr, 2013. Improving health

information systems for decision making across five

sub-Saharan African countries: implementation

strategies from the African Health Initiative. BMC

health services research.

Nijeweme-d’Hollosy, W. O., van Velsen, L., Soer, R., and

Hermens, H., 2016. Design of a web-based clinical

decision support system for guiding patients with low

back pain to the best next step in primary healthcare.

In HEALTHINF 2016.

Shakshuki, E. and Reid, M. , 2015. Multi-agent system

applications in healthcare: Current technology and

future roadmap. Procedia Computer Science,

52(Supplement C):252 – 261. The 6th International

Conference on Ambient Systems, Networks and

Technologies (ANT-2015), the 5th International

Conference on Sustainable Energy Information

Technology (SEIT-2015).

Shannon, R. E., 1977. Simulation modeling and

methodology SIGSIM Simul. Dig., 8(3):33–38.

UNDP, 2016. Human development report 2016. Technical

report.

Weber, A., Rahmani, A., and Dupont, D., 2006.

Multiagent system for flow management in complex

systems: Development of a decision support system in

epidemiology. In European Simulation

Multiconference (ESM).

WHO, 2005. International Health Regulations.

WHO, 2014.World Health Statistics.

Wooldridge, M., 1999. Intelligent agents. In Weiss, G.,

editor, Multiagent Systems: A Modern Approach to

Distributed Artificial Intelligence. MIT Press.

Wooldridge, M., Jennings, N. R., and Kinny, D., 1999. A

methodology for agent-oriented analysis and design.

In Proceedings of the Third Annual Conference on

Autonomous Agents, AGENTS ’99, pages 69–76, New

York, ACM.

Wouters, B.; Barjis, J.; Maponya, G.; Martiz, J. and

Mashiri, M., 2009. Supporting home based healthcare

in South Africa rural communities using USSD

technology. In AMCIS.

Younsi, F., 2016 Mise en place d’un Système

d’Information Décisionnel pour le Suivi et la

Prévention des Epidémies. PhD thesis, Université

d’Oran.

Zurovac, D., Sudoi, R. K., Akhwale, W. S., Ndiritu, M.,

Hamer, D. H., Rowe, A. K., and Snow, R. W., 2011.

The effect of mobile phone text-message reminders on

kenyan health workers’ adherence to malaria treatment

guidelines: a cluster randomised trial. The Lancet,

378(9793):795 – 803.

Zurovac, D., Talisuna, A. O., and Snow, R. W., 2012.

Mobile phone text messaging: Tool for malaria control

in Africa. PLOS Medicine, 9(2):1–6.

Towards an Agent-based Model to Monitor Epidemics and Chronic Diseases in DR Congo