=Paper=
{{Paper
|id=Vol-2544/paper8
|storemode=property
|title=National Integrated Network for Remote Monitoring Of Patients in Benin
|pdfUrl=https://ceur-ws.org/Vol-2544/paper8.pdf
|volume=Vol-2544
|authors=Daton Medenou,Mêtowanou H. Ahouandjinou,Leandro Pecchia,Thierry R. Jossou,Roland C. Houessouvo,Davide Piaggio
|dblpUrl=https://dblp.org/rec/conf/irehi/MedenouAPJHP18
}}
==National Integrated Network for Remote Monitoring Of Patients in Benin==
https://ceur-ws.org/Vol-2544/paper8.pdf
National Integrated Network For Remote
Monitoring Of Patients In Benin
Daton Medenou Mêtowanou H. Ahouandjinou Leandro Pecchia
Electrotechnical Laboratory of Electrotechnical Laboratory of School of Engineering
Telecommunications and Applied Telecommunications and Applied University of Warwick
Informatic (Polytechnic School of Informatic (Polytechnic School of Warwick, United Kingdom
Abomey-Calavi) Abomey-Calavi) L.Pecchia@warwick.ac.uk
Unviversity of Abomey-Calavi Unviversity of Abomey-Calavi
Cotonou, Benin Cotonou, Benin Davide Piaggio
daton.medenou@epac.uac.bj heribert.metowanou@gmail.com School of Engineering
University of Warwick
Thierry R. Jossou Roland C. Houessouvo Warwick, United Kingdom
Electrotechnical Laboratory of Electrotechnical Laboratory of D.Piaggio@warwick.ac.uk
Telecommunications and Applied Telecommunications and Applied
Informatic (Polytechnic School of Informatic (Polytechnic School of
Abomey-Calavi) Abomey-Calavi)
Unviversity of Abomey-Calavi Unviversity of Abomey-Calavi
Cotonou, Benin Cotonou, Benin
thierry.djossou@gmail.com rolandchouessouvo@gmail.com
Abstract—Background: The Benin health system has Telecommunications, Networks and Information Processing.
challenges including: (i) the need to provide quality health care Among these communicating objects, we are interested in
at low cost to a growing population, (ii) the reduction of patients' sensors. Indeed, in recent decades, thanks to the Advanced
hospitalization time, (iii) and the optimization presence time of Embedded Systems and Wireless Technologies (SETSF), the
the nursing staff. Such challenges can be solved by remote Sensors Wireless Networks (WSN) are frequently used in
monitoring of patients. Methodology: To achieve this, five steps medical applications. Hence the emergence of Medical
were followed. 1) The identification of the different Wireless Sensor Networks (MWSN) used in Wireless Body
characteristics of the WBAN systems and the physiological Area Network (WBAN) systems, to improve the quality of
parameters monitored on a patient. 2) The modeling of the
care and record medical monitoring of patients.
national RIMP architecture in a cloud of Technocenters. 3)
Cross analysis between characteristics and functional The MWSN are characterized by their sensor nodes
requirements identified. 4) The simulation of the functionality of mobility, easy deployment and self-organization. Therefore,
each Technocenter through: a) the choice of design approach the MWSN are very convenient for monitoring elderly, the
inspired by the life cycle of V systems; b) functional modeling disabled, people at risk and people with chronic diseases and
through SysML Language; c) the study of the choice of to monitor their living environment [2]. By [3] [4] [5] today,
communication technology and different architectures of sensor the MWSN are used to monitor vital parameters such as
networks. 5) An estimate of the material resources of the national temperature, blood pressure or heart rate. The MWSN in the
RIMP according to physiological parameters. Findings: The
WBANs improve patient quality of life, real-time patient
main result is that it has designed a National Integrated Network
for Patient Monitoring (RNIMP) remotely, ambulatory or not,
follow-up and emergency decision-making [6] [7].
for the Benin health system. Conclusion: The implementation of In the implementation of RCSFM, the approaches are
the RNIMP will contribute to improve the care of patients in different according to the literature. The authors in [8]
Benin. The proposed network is supported by a repository that present a people monitoring network architecture accessible
can be used for its implementation, monitoring and evaluation. It via Internet called INSIGHT. Access collected data can be
is a table of 36 characteristic elements each of which must satisfy local or remote. The parameters monitored can be
5 requirements relating to: medical application, design factors,
reconfigured remotely. The authors justify the use of a
safety, performance indicators and materiovigilance.
single-hop architecture to reduce energy consumption. IEEE
Keywords— architecture, requirements, hospital, patient, 802.15.4 physical layer for the network deployement. The bit
repository, RNIMP, simulation, SysML, system, technocenter. rate is 250 kbps and the radio range is 100 meters. TmoteSky
platforms are used in experiments. The B-MAC layer (MAC
I. INTRODUCTION Berkeley) according to [9] is used to manage access to the
The health system in Benin faces challenges including: medium. To conserve energy, the nodes send data to base
(i) the need to provide high-quality, low-cost health care, station and spend the rest of the time in sleep mode. For this,
rapid growth, (ii) the reduction hospitalization time for a data reporting technique is used to define the delivery
patients, (iii) and optimization of the nursing staff presence intervals. In addition, the HPL (Hardware Presentation
time [1]. For a good sanitary opening of the population Layer) power management module and « watchdog timer »
including the rural one, any health policy in Benin must timers are used. The authors in [10] present one of the first
consider the 5295 villages and city districts which are experimental deployments of WSNs for remote monitoring
organized in 546 boroughs, 77 communes, 34 health zones, on Great Duck Island. The authors propose a multilevel
and 12 departments. To face these challenges, we can use the architecture, each providing a data management service. Two
new communicating tools and objects through the types of topologies are used: multi-jump (mesh) and a jump.
development technologies in the areas of In the one-hop architecture, a node called Sensor patch is
Copyright © 2019 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0)
IREHI 2018 : 2nd IEEE International Rural and Elderly Health Informatics Conference
used to send the data to a PDA (Personal Digital Assistant). encrypted key. It is based on symmetric cryptography. The
The latter relays the data to reach the base station. This choice of this biometry is based on heartbeat information
station makes data available on the Web. The called « interpulse interval (IPI) ». This solution achieves a
communications are bidirectional between the nodes. To high level of security with less calculation and memory. It is
reduce power consumption, the sensors are put into sleep an identification technique based on the physiological or
mode (off the radio and the processor (MCU)). A low power behavioral characteristics of the individual. This approach
MAC protocol « MAC Low power" » is developed, and makes possible to identify the sensor nodes and to secure the
hierarchical routing protocols are used. The authors in [11] distribution of the encrypted key.
proposed a monitoring system called WHMS. IEEE 802.15.4
standard is the Intra-WBAN communications support. They The authors in [16] have designed different types of
have developed several types of medical sensor nodes: sensor nodes for the WBAN (ECG, EEG, pulse, glucose).
accelerometers, ECG, pulse oximetry and reconfigurable The mechanical and thermal energy recovery means are used
breathing sensor. A PDA equipped with LINX transceiver is as supplements to solar energy (piezoelectric generators and
used to relay data to supervisor. thermal generators). The nodes of the WBAN are put in
specific locations of the body to better recover the energy
In [12] the authors present an energy efficient (from the temperature of the body). According to their
communication protocol for the WBAN. IEEE 802.15.4 experiments, an energy of 100 μW can be recovered by the
standard is the communications support. The platforms used batteries. In [17] the authors present the study and design of
are of Telos type. The authors propose protocol based on a an actimetric monitoring telemonitoring system. The
cyclic awakening of the nodes: « duty cycle ». The protocol architecture of the authors is a WBAN network. Works [3]
is based on a wake cycle called SFC (Super Frame Cycle). In present the detection of attacks in a WBAN remote medical
their experiments, the SFC period is set to 1 second. They surveillance system. According to [18] the evaluation of
evaluated the energy consumed by listening, transmission connected objects in health applications was presented. It
and sleep modes. The different consumptions measured are: shows the impact of connected objects on a sanitary system
1.53 mA in sleep mode, 17.4 mA in transmission mode and and their importance in the prevention of diseases. The work
19.7 mA in listening mode. According to [13] the authors of [19] show that the success of these health surveillance
proposed a sensor network, energy efficient, applied in the systems depend on data collecting and processing, to
military context. The surveillance system is based on inter- understand the environment of a subject, so that contextual
sensor cooperation and the organization of tasks in the care can be given to them. We note that the challenges for
network to detect and trace the positions and movements of any medical surveillance system lie in the proper design of
people and vehicles. The platforms used are of the type: the network architecture. This is the goal of this work. It
Mica 2. They used remote monitoring cameras controlled by aims at Modeling an Integrated Patient Monitoring Network
a laptop, to propose a solution that allows to reduce the delay (RIMP) in the Benin health system, through the use of
and improve the reliability of the data (minimization of the wireless medical sensor networks in WBAN systems. In the
number of alarms erroneous due to false readings). A remainder of this manuscript, we present the methodology
synchronization module of the clock of the nodes with the adopted for the work, the results obtained, the analysis of the
base station is also implanted. The selection of these nodes is results, the discussion and the envisaged perspectives.
done according to the quantity of their energy reserves. Then
the authors propose two models to control the cycles of sleep II. MATERIAL AND METHOD
and awakening of the nodes. A. Material
The authors in [14] present the necessary steps to build a In addition to resources from the literature, we used: MS
surveillance system in the habitat. They propose a model Visio for network architecture, SysML for modeling, a Dell
called « Frisbee ». This model is based on the creation of computer with 8 GB of RAM and 2 TB of disk, data on the
regions consisting of heterogeneous sensors that follow a health pyramid of Benin. In addition, we are based on the
given target. To save energy, nodes that are far from the model of the WBAN remote medical surveillance system,
target go into sleep mode. When an event is detected, soldier shown in Fig. 1, and the model of a WBAN comprehensive
nodes « sentries » support the mission to wake other sleeping medical surveillance system is divided into five subsystems.
nodes. Only the network area close to the event is in the [20] as shown in Fig. 2.
active state. Whenever the target moves, the "soldier" nodes
send wake-up signals to others (who must be in the listening
state). To recover solar energy, the nodes are equipped with
photovoltaic panels. They can be extinguished remotely via a
developed control software. Localization and
synchronization algorithms, as well as a mechanism that
allows the deletion of duplicate notifications are also
proposed. According to [15] a new approach is presented to
secure the exchanges between the sensor nodes of a WBAN.
The problem addressed is related to the confidentiality and
integrity of the data. The question is: how do the nodes of a
WBAN know that they belong to the same patient? To
answer this question, the authors proposed a solution based
on a « biometrics » approach. It is an identification technique
based on the physiological or behavioral characteristics of Fig. 1: WBAN monitoring system
the individual. This approach makes it possible to identify
the sensor nodes and to secure the distribution of the
These characteristics constitute a repository for the
design of a functional WBAN network of a technocenter,
noted fc (WBAN) . Thus the design function of a WBAN
network is a function: of the requirement function of the
medical application of the WBAN noted fEXappM ; of the
design factor function, noted ffacco (WBAN) ; of the
communication technology function, noted fcom and sensor
architecture function, noted farch . The mathematical model
of designing a functional WBAN network can be written as
following equation. (1) :
𝑓𝐸𝑋𝑎𝑝𝑝𝑀 (𝑊𝐵𝐴𝑁)
𝑓 (𝑊𝐵𝐴𝑁)
𝑓𝑐 (𝑊𝐵𝐴𝑁) = 𝑓𝑎𝑐𝑐𝑜 (1)
𝑓𝑐𝑜𝑚
Fig. 2: Architecture of a medical surveillance system
{ 𝑓𝑎𝑟𝑐ℎ
Several medical sensors are deployed on the patient's
body to measure several physiological parameters. These with
nodes are sensors capable of harvesting and transmitting
environmental data in an autonomous manner. The position 𝑛
of these nodes is not necessarily predetermined. 𝑓𝐸𝑋𝑎𝑝𝑝𝑀 (𝑊𝐵𝐴𝑁) = ∑ 𝐸𝑋𝑎𝑝𝑝𝑀(𝑖)
Method 𝑖=1
A five-step methodology was followed. 1) The 𝑛′
identification of the different characteristics of the WBAN 𝑓𝑓𝑎𝑐𝑐𝑜 (𝑊𝐵𝐴𝑁) = ∑ 𝑓𝑎𝑐𝑐𝑜(𝑗)
systems and the physiological parameters that can be
𝑗=1
monitored on a patient. 2) Modeling the national architecture
of the RIMP, in the form of a cloud of Technocentres at 6 𝑚
levels (National, Departmental, Health Zone, Communal,
𝑓𝑐𝑜𝑚 = ∑ 𝑝𝑟𝑜𝑡𝑜𝑐𝑜𝑙(𝑘)
Borough, Village and City District). 3) Cross analysis
between characteristics and functional requirements 𝑘=1
identified. 4) The simulation of the functionality of each 𝑚′
Technocentre through: a) the choice of design approach
𝑓𝑎𝑟𝑐ℎ = ∑ 𝑡𝑜𝑝𝑜𝑙𝑜𝑔𝑦(𝑙)
inspired by the life cycle of V systems; b) functional
𝑙=1
modeling through Language SysML; c) the comparative
study of the choice of communication technology and
different architectures of sensor networks. 5) An estimate of The 𝑬𝑿𝒂𝒑𝒑𝑴(𝒊) are the member elements of the
the material resources of the national RIMP according to WBAN medical application requirements.
physiological parameters. The 𝒇𝒂𝒄𝒄𝒐(𝒋) are the elements of the WBAN design
factors.
III. RESULTS
The identification of the different characteristics of The design of a functional WBAN network aims to
WBAN systems. We have listed in Table I, a total of 36 optimize care in the health systems and thus to have a smart
characteristics of WBAN systems. hospital (technocenter). We can therefore deduce, the
existence of a patient monitoring function noted 𝒇𝒔𝒖𝒗𝒑𝒂𝒕 and
Modeling Requierements a smart hospital function, noted 𝒇𝒉𝒐𝒔𝒊𝒏𝒕𝒆𝒍 . Thus the patient
TABLE I. CHARACTERISTICS IDENTIFIED FOR WBAN SYSTEMS
36 Characteristics identified for WBAN Systems
N° Designations N° Designations N° Designations
1 National Architecture 13 Robustness 25 Reliability
2 Local architecture 14 Usability 26 The passage ladder (scaling)
3 Dimension 15 Ergonomics 27 The flow
4 Environment / Obstacle 16 Energetic efficiency 28 The Deadline
5 Building material 17 interoperability 29 The Gigue/Jip
6 Size to watch 18 Precision 30 Loss rate
7 Mobility Management 19 Miniaturization 31 Life time
8 Respect for private life 20 Reduced detection time 32 The availability
9 Securing data 21 High security 33 Confidentiality
10 Low cost of deployment 22 Tolerances to breakdowns 34 Integrity
11 Easy installation 23 Sensitivity to Data Loss 35 Access control
12 Flexibility 24 High sensitivity 36 Authentication
monitoring function, noted 𝒇𝒔𝒖𝒗𝒑𝒂𝒕 is the equation (2) I1 I2 I3 I4 I5
formed by the performance indicators function, noted
Referential characteristics of a
𝒇𝒊𝒏𝒑𝒆𝒓 (𝑾𝑩𝑨𝑵) and the design function, noted 𝒇𝒄 (𝑾𝑩𝑨𝑵)
WBAN Performance Assessment
Key Design Factors for WBANs
added to the security function, noted 𝑓𝑠𝑒𝑐 , which is
WBAN security requirement
paramount in patient monitoring. So equation (2) :
Requirement of medical
N°
application of WBAN
𝑓 (𝑊𝐵𝐴𝑁)
WBAN network
Materiovigilance
𝑓𝑠𝑢𝑣𝑝𝑎𝑡 = { 𝑐 + 𝑓𝑠𝑒𝑐 (2)
𝑓𝑖𝑛𝑝𝑒𝑟 (𝑊𝐵𝐴𝑁)
Comments
Indicators
One of the major constraints of WBAN network
operation is energy. We then establish that the smart hospital
function, noted 𝒇𝒉𝒐𝒔𝒊𝒏𝒕𝒆𝒍 is expressed by the system of Low cost of
10 1 0 1 0 1 0 1 0 1 0
deployment
equation (3).
Easy
11 1 0 1 0 1 0 1 0 1 0
installation
max 𝑓𝑠𝑢𝑣𝑝𝑎𝑡 12 Flexibility 1 0 1 0 1 0 1 0 1 0
𝑓ℎ𝑜𝑠𝑖𝑛𝑡𝑒𝑙 = { (3)
min 𝑓( 𝑒𝑛𝑒𝑟𝑔𝑖𝑒)
13 Robustness 1 0 1 0 1 0 1 0 1 0
A. Cross analysis between characteristics and functional 14 Usability 1 0 1 0 1 0 1 0 1 0
requirements identified 15 Ergonomics 1 0 1 0 1 0 1 0 1 0
We establish then in Table II, the binary matrix of the Energetic
16 1 0 1 0 1 0 1 0 1 0
requirements ( 𝐼𝑖 ) and characteristics of the requirements efficiency
( 𝐼𝑖𝑗 ) . To do this, we have added to the previous 17 interoperability 1 0 1 0 1 0 1 0 1 0
requirements, that relating to the Materiovigilance to 18 Precision 1 0 1 0 1 0 1 0 1 0
guarantee the maintenance and minimize the potential risks Miniaturi-
of the network. Thus we release the different validation 19 1 0 1 0 1 0 1 0 1 0
zation
matrices of a well-designed WBAN network. Reduced
20 1 0 1 0 1 0 1 0 1 0
detection time
TABLE II. THE BINARY VALIDATION MATRIX OF A 21 High security 1 0 1 0 1 0 1 0 1 0
FUNCTIONAL WBAN NETWORK
Tolerances to
22 1 0 1 0 1 0 1 0 1 0
I1 I2 I3 I4 I5 breakdowns
Sensitivity to
23 1 0 1 0 1 0 1 0 1 0
Data Loss
Referential characteristics of a
WBAN Performance Assessment
Key Design Factors for WBANs
High
24 1 0 1 0 1 0 1 0 1 0
WBAN security requirement
sensitivity
25 Reliability 1 0 1 0 1 0 1 0 1 0
Requirement of medical
N°
application of WBAN
The passage
WBAN network
Materiovigilance
26 ladder 1 0 1 0 1 0 1 0 1 0
(scaling)
Comments
Indicators
27 The flow 1 0 1 0 1 0 1 0 1 0
28 The Deadline 1 0 1 0 1 0 1 0 1 0
QdS
National 29 The Gigue/Jip 1 0 1 0 1 0 1 0 1 0
1 1 0 1 0 1 0 1 0 1 0
Architecture
30 Loss rate 1 0 1 0 1 0 1 0 1 0
Local
2 1 0 1 0 1 0 1 0 1 0
architecture 31 Life time 1 0 1 0 1 0 1 0 1 0
3 Dimension 1 0 1 0 1 0 1 0 1 0 The
Environment / 32 1 0 1 0 1 0 1 0 1 0
4 1 0 1 0 1 0 1 0 1 0 availability
Obstacle
Building 33 Confidentiality 1 0 1 0 1 0 1 0 1 0
5 1 0 1 0 1 0 1 0 1 0
material 34 Integrity 1 0 1 0 1 0 1 0 1 0
6 Size to watch 1 0 1 0 1 0 1 0 1 0
Mobility 35 Access control 1 0 1 0 1 0 1 0 1 0
7 1 0 1 0 1 0 1 0 1 0
Management
36 Authentication 1 0 1 0 1 0 1 0 1 0
Respect for
8 1 0 1 0 1 0 1 0 1 0
private life
B. Assessment of physiological parameters monitorable by
9 Securing data 1 0 1 0 1 0 1 0 1 0 a network of sensors
Medical sensors are used to monitor 16 different that the data are decentralized by sanitary zone and then to
groups of parameters Table III relating to: physiological interconnect the sanitary zones to have the RIMP. As a
variables, physical activities and movements of a person, result, we see that the RIMP-B is a continuation of the RIMP
TABLE III. PHYSIOLOGICAL CHARACTERISTICS MONITORABLE WITH SENSOR NETWORKS
N° Physiological sources or characteristics Sensor type, Methods, Technologies
Combining bioelectrical (EEG) and biooptical (NIRS)
1 A (M3BA) & (NIRS) technology & Brain-Computer Interfaces (BCI) [21]
neurophysiological measurements
2 (Real life environnement) EEG : monitoring Ear EEG Dry-Contact Electrode [22]. BCI and NeuroFeedback (NF) [23]
3 Decoding of covert somatosensory attention (SAO) somatosensory attentional orientation [24]
4 Pulmonary function testing (PFT) : Depth (and) Microsoft Kinect V2 RGB-D sensors. [25]
Machine learning model to accurately predict the blood-analog viscosity during
5 HCT of VAD patients support of a pathological circulation with a rotary ventricular assist device (VAD).
[26]
Biomedical Big Data analytics & multi-omic data & –Omic information into
6 Identifying disease biomarkers (Precision Medicine)
electronic health records (HER) [27]
7 Glucose Monitoring in Individuals With Diabetes Percutaneous glucose sensors with sending information by wirelessly [28]
[Monitoring frail elderly patients with chronic disease(s) and
Interoperable End-to-End Remote Patient Monitoring Platform Based on IEEE
8 patients with diabetes.]: blood pressure, weight, blood glucose
11073 PHD and ZigBee Health Care Profile [5]
and SpO2,
9 Person’s physical activity (PA) monitoring Smartwatch ZGPAX S8 [29]
38 features extracted from HRV, SC, and EEG SIGNAL
A wearable physiological sensors system (Sensors-Type : IMU, EDA, SpO2,
10 (SKIN conductance (SC ) : 16 / heart rate variability (HRV):
ECG, EDA, Microphone, Accelerometer, Proximity, Respiration, EMG, EEG) [4]
16 /SKIN CONDUCTANCE (SC : 16) )
11 Photoplethysmographic (PPG) signals : SpO2 ESPRIT-MLT:[30]
Cardiorespiratory system : Obstructive sleep apnea (OSA)
Wearable sensor measurement signals( sensors :One-lead ECG, SpO2) with the
12 detection (PaCO2), (SaO2), (ABP), (HR), (Vt), SpO2 , virtual
mathematical models-Gaussian processes [7]
oxygen saturation state (VSO2 ))
Insole Based, Wrist Worn Wearable Sensors (SmartStep and Wrist Sensor) and
Activities of Daily Living (ADL) : energy balance, and
13 ADL Sensors : Bi axial accelerometers, magnetometer, pressure sensors, heart rate
quality of life (understanding)
sensor, visual sensors [6], Complex Network Analysis [31]
Hemoglobin (HbT), concentration and tissue oxygen
14 Wearable optical device [32]
saturation (StO2)
Detection of Nocturnal Scratching Movements in Patients with
15 Accelerometers and Recurrent Neural Networks [33]
Atopic Dermatitis
Inertial Sensors (Accelerometers, Gyroscopes), electromyography (EMG) sensors,
16 Detect the onset and duration of freezing of gait (FOG)
force resistive sensors, video-based gait analysis. [34]
social inclusion of the elderly or living with disabilities. by Health Zone (RIMP-ZS).
From the point of view location, as in Fig.1, the Let 𝑖𝑛 be the number of communes constituting a
sensors can be placed at 17 different locations on a patient's sanitary zone with 𝑖1 , 𝑖2 … . . 𝑖𝑛 the communes.
body. [6] [21].
Let 𝑗𝑛′ be the rounding number of each commune of
From the point of view monitoring physical activities, a health zone with 𝑗1 , 𝑗2 … . . 𝑗𝑛′ .
sensors can monitor 63 kinds of physical activity in a
person's body. Let 𝑘𝑛" be the number of villages in each district.
From the point of view social inclusion, the network Let 𝑇𝐶𝑖𝑛 the municipal technocentres representing the
of medical sensors can monitor elderly people and living CSCs of a health zone and 𝑇𝐴𝑖𝑛𝑗 technocenters of districts
𝑛′
people with one of the 6 disabilities, namely: Cognitive representing the CSA of the districts of each commune with
disability, Disability in general, [2]. 𝑖1…… 𝑖𝑛 ; 𝑗1 … . . 𝑗𝑛′ .
From the point of view technologies and applications For example:
or services, 22 technologies and 75 applications / services are
available according to the literature [2], for the deployment For 𝑇𝐶𝑖1 the technocenters of the first commune of a
of medical sensor networks. sanitary zone, we have 𝑇𝐴𝑖1𝑗1 … … . . 𝑇𝐴𝑖1 𝑗 ′ technocenters of
𝑛
C. The modeling of the RIMP national architecture in the the boroughs of this commune.
cloud Technocenters form Let 𝑇𝑉𝑖𝑛 𝑗𝑛′ 𝑘𝑛" be the village technocentres
The health system of Benin is organized thirty and representing the UVS of each district with 𝑖𝑛 going from de
four (34) health zones. Each health zone is subdivided into: 𝑖1 to 𝑖𝑛 ; 𝑗𝑛′ going from 𝑗1 to 𝑗𝑛′ and 𝑘𝑛" going from de 𝑘1
village health unit (UVS), district health center (CSA), to 𝑘𝑛" . For example: for 𝑖1 and 𝑗1 we have them 𝑇𝑉𝑖1𝑗1𝑘1
municipal health center (CSC) and zone hospital (HZ). Let's to 𝑇𝑉𝑖1 𝑗1𝑘𝑛" .
call a health data monitoring center by technocenter. Thus,
the modeling of the Benin Integrated Patient Monitoring Let 𝑇𝑍𝑙 be the technocentres representing the
Network (RIMP-B), is to model first each health zone, so monitoring centers of the health zones with 𝑙 ranging from 1
to 34. Because Benin's health system has 34 sanitary zones. Let 𝑇𝐷𝑚 the departmental technocenter regrouping
The Technocenters cloud of the Integrated Patient the technocenters of the zones (𝑇𝑍𝑙 ) , representing the
Monitoring Network of a health zone (RIMP-ZS) is shown in departmental health departments (DDS). We then have the
Fig.3. technocentres cloud of the Departmental Integral Patient
Monitoring Network (RIMP-DDS), shown in Fig. 4.
Fig. 3: Technocenters cloud of the Integrated Network for Patient Monitoring of a Health Zone (RIMP-ZS)
Fig. 4. Technocenters Cloud of the Integrated Patient Monitoring Network of a Department (RIMP-DDS)
A series of these clouds gives the national network D. The simulation of the functionality of each
shown in Fig. 5. Technocentre: software architecture
The software architecture of the smart hospital shown
in Fig. 6, shows the various management software modules
from the patient embalmed that will allow better monitoring.
This architecture also shows the exchanges between the function of the different elements involved. Let's designate
different servers. The data server Fig. 6 is responsible for by 𝑓𝑚𝑎𝑡 the material resources function. This function 𝑓𝑚𝑎𝑡
collecting the data (physiological and actimetric parameters) is size dependent 𝑇 data to monitor which itself depends on
and storing them in a technocenter database via the the size 𝑁 population and number of sensors 𝑁𝑐 placed on
acquisition module and / or the network. This same module the patient. This hardware function also depends on the
sends this data to the display module in order to follow the number of simultaneous data access (𝑁𝑝 + 𝑁𝑐𝑚 + 𝑁𝑎𝑑𝑚) ,
patients in real time and to display the alerts in case of with 𝑁𝑝 the number of patients, 𝑁𝑐𝑚 the number of the
detection of critical cases. The omics data are sent to the medical profession and 𝑁𝑎𝑑𝑚 the number of administrative
calculation server via the send / receive module and stored in technocenters. Function 𝑓𝑚𝑎𝑡 would be equal to equation
a second database (zone, departmental, national). The (4).
delayed calculation module retrieves these data in order to
generate the thresholds of the behavioral deviation, nocturnal 𝑓𝑚𝑎𝑡 = 𝑓(𝑇, 𝑁𝑝, 𝑁𝑐𝑚, 𝑁𝑎𝑑𝑚) (4)
agitation, prolonged immobility, residence time in the
bathroom, difference between physiological parameters and IV. ANALYSIS AND DISCUSSION
others. These thresholds of the different physiological In the face of the challenges of the Benin health
parameters are therefore sent directly to the database of the system, our solution aims to make it efficient from the
local technocentre. This is to allow the diagnostics module to villages to the cities. The solution aims a powerful health
compare them with the current data and generate alerts (on system allowing to anticipate in view of several data that it
the real-time application and phones) in case of overruns. will provide. The implementation of this solution will go
through several stages (from the analysis of ICT potential in
E. Estimation of the material resources of the national the 5295 villages and city districts to the technological
RIMP according to physiological parameters. choice).
An analysis of the different parameters that can be
monitored with the population size of each village (or city Several design factors for WBAN networks
district), shows that the size of the RIMP resources would be (scalability, quality of service (QoS), power consumption,
unique for each health zone. Moreover, the size of the RIMP wireless technology) should be considered [22]. Many works
would also depend on the different services offered by each in the literature deal with the application of WBAN networks
branch of the sanitary system. (UVS, CSA, CSC, HZ). An for health [16] [11] [8] [18].
estimate of the RIMP material resources would then be a
Fig. 5. Technocenters cloud of the Benin Integrated Patient Monitoring Network
Fig. 6. Software architecture of the smart hospital
This work presents on the one hand the characteristics Compared to several works in literatures where
and the requirements of the medical application of WBAN technological choices are proposed [24] [20] [17] [25], our
networks, and on the other hand the characteristics and work presents a basic model for setting up a patient
design factors of these networks. monitoring network, especially in the case of the Benin
health system.
The design of WBAN networks also involves security
requirements. (WBAN and traditional networks have the V. CONCLUSION
same) security requirements [19].
Wireless Medical Sensor Networks (MWSN)/WSN are a
These works are different from ours since we propose revolution in wireless computer networks. Choosing a
a repository of 36 elements according to five requirements technology will depend strongly on the solutions offered and
that the design must follow for the patient monitoring the vision of the proposer. Features such as power, data flow
network. In addition, each requirement is a matrix block that and parameters related to scope, cost, security and number
serves as a compass for the design and / or evaluation of a of nodes should be considered. In the case of Benin, the
patient monitoring system. (Several technologies have been need to have a health system that responds to the many
used in) WBAN networks for patient monitoring. security
challenges and considers the population at the base is no
threats or attacks can occur such as: modifying and listening
longer to demonstrate.
to medical data, activity detection and location, counterfeit
security system is needed on different block [19]. This justifies the guidelines of this work which proposed a
reference system for the implementation of a patient
Our repository takes this into account in terms of monitoring system, which modeled a network for the Benin
security requirements. Network data flows and capacity are health system.
among the parameters that impact network performance. This work also presented a point of the sensors and the
high-speed wireless technology choice provides benefits to different physiological parameters that can be monitored
meet network scalability and increased numbers of people according to the services offered. The implementation of
being monitored. On the other hand, with some technologies this proposed RIMP-B will go through several stages.
we have low energy consumption but significant delays
(generation) and / or low transfer rates.
Future work will consist of a field survey across the country
The chosen technology will have flow and energy to:
consumption compromission. Several technologies are used 1) Validate the data of the sanitary cartography;
in patient monitoring architectures to provide multiple 2) Identify ICT potentials and different constraints of
services [23] [17]. each localized health mapping;
That is why we have started to identify all the 3) Propose the different technologies to be used in
technologies used with the different services. From there we each health locality for the proper functioning of
got a roadmap for any surveillance system with the different technocenters;
possible positions where the sensors can be put on a patient 4) Propose an algorithm for calculation the material
body. Here is expressed the strength of this work. resource applicable to each level.
REFERENCES [18] F.-A. Allaert et N.-J. Mazen, «Évaluation des objets connectés et des
applications de santé,» Elsevier Masson SAS., n° %1556, pp. 29-
[1] M. d. l. S. Bénin, «Plan national de dévéloppement sanitaire,» 32, 2016.
Ministère de la Santé Bénin, Cotonou,Bénin, 2009.
[19] H. Mshali , T. Lemlouma , M. Moloney et D. Magoni, «A survey on
[2] M. Manzoor et V. Vimarlund, «Digital technologies for social health monitoring systems for health smart homes,» International
inclusion of individuals with disabilities,» Health and Journal of Industrial Ergonomics, n° %166, pp. 26-56, 2018.
Technology, vol. 8, pp. 377-37790, 2018.
[20] H. Alemdar et C. Ersoy, «Wireless sensor networks for healthcare: A
[3] A. Makke, «Détection d’attaques dans un système WBAN de survey,» The International Journal of Computer and
surveillance médicale à distance,» Paris, 2014. Telecommunications Networking, vol. 54, n° %115, pp. 2688-
[4] S. Betti, R. M. Lova,, E. Rovini, G. Acerbi, L. Santarelli, M. Cabiati, 2770, October 2010.
S. Del Ry et F. Cavallo, «Evaluation of an integrated system of [21] N. Jalloul, F. Por´ee, G. Viardot, P. L’Hostis et G. Carrault, «Activity
wearable physiological sensors for stress monitoring in working Recognition using Complex Network Analysis,» IEEE Journal
environments by using biological,» IEEE Transactions on of Biomedical and Health Informatics, vol. vol.6, n° %1NO.1,
Biomedical Engineering, pp. 1-12, 2017. pp. 2168-2194, 2017.
[5] M. Clarke, J. de Folter, V. Verma et H. Gokalp, «Interoperable End- [22] I. Akyildiz, T. Melodia et K. Chowdhury, «A Survey on Wireless
to-End Remote Patient Monitoring Platform Based on IEEE Multimedia SensorNetworks,» Computer Networks Journal
11073 PHD and ZigBee Health Care Profile,» IEEE (Elsevier), March 2007.
TRANSACTIONS ON BIOMEDICAL ENGINEERING, vol.
Vol.65, n° %1NO.5, pp. 1014-1025, 2018. [23] M. Chen, S. Gonzalez, A. Vasilakos, H. Cao et V. C. Leung, «Body
Area Network:A Survery,» Mob.Netw.Appl.Journal, pp. 171-
[6] N. Hegde, M. Bries, T. Swibas, E. Melanson et E. Sazonov, 193, April 2011.
«Automatic Recognition of Activities of Daily Living utilizing
Insode Based and Wrist Worn Wearable Sensors,» EEE Journal [24] I. Akyildiz, T. Melodia et K. Chowdury, «Wireless Multimedia Sensor
of Biomedical and Health Informatics, pp. 2168-2194, 2017. Networks:Applications and Testbeds,» Proceedings of the IEEE
(invited paper), vol. 96, n° %110, pp. 1588-1605, October 2008.
[7] S. Gutta, Q. Cheng, H. D. Nguyen et B. A. Benjamin,
«Cardiorespiratory Model-based Data-driven Approach for Sleep [25] C. S. Bingham, K. Loizos, G. J. Yu et A. Gilbert, «Model-Based
Apnea Detection,» IEEE Journal of Biomedical and Health Analysis of Electrode Placement and Pulse Amplitude for
Informatics, pp. 1-10, 2017. Hippocampal Stimulation,» IEEE TRANSACTIONS ON
BIOMEDICAL ENGINEERING, vol. VOL.65, n° %1NO.10,
[8] M. Demirbas, K. Chow et C. Wan, «INSIGHT: Internet-sensor pp. 2278-2288, 2018.
integration for habitat monitoring,» chez International
Symposium on a World of Wireless, Mobile and Multimedia [26] A. v. L¨uhmann, H. Wabnitz, T. Sander et K.-R. Muller, «M3BA: A
Networks(WoWMoM'06), 2006. Mobile, Modular, Multimodal Biosignal Acquisition
Architecture for Miniaturized EEG-NIRS-Based Hybrid BCI and
[9] J. Polastre, J. Hill et D. Culler, «Versatile low power media access for Monitoring,» IEEE TRANSACTIONS ON BIOMEDICAL
wireless sensor networks,» chez In Proceedings of the 2nd ENGINEERING, vol. VOL.64, n° %1NO.6, pp. 1199-1210,
international conference on embedded networked Sensor 2017.
Systems (SenSys'04), New York, 2004.
[27] S. L. Kappel, M. L. Rank, H. O. Toft, M. Andersen et P. Kidmose,
[10] A. Mainwaring, D. Culler , J. Polastre et R. Szewc, «Wireless sensor «Dry-Contact Electrode Ear-EEG,» IEEE Transactions on
networks for habitat monitoring,» chez Proceedings of the 1st Biomedical Engineering, 2017.
ACM international workshop on Wireless sensor networks and
applications, Atlanta, Georgia, USA, 2002. [28] C. Jeunet, F. Lotte, J.-M. Batail, P. Philip et J.-A. Micoulaud-Franchi,
«Using recent BCI literature to deepen our understanding of
[11] E. Jovanov, A. Milenkovic, C. Otto et P. C. de Groen, «A wireless clinical neurofeedback: A short review,» Neuroscience, Elsevier
body area network of intelligent motion sensors for computer - International Brain Research Organization2018, n° %1378, pp.
assisted physical rehabilitation,» Journal of NeuroEngineering pp.225-233., 2018.
and Rehabilitation, vol. 2, n° %16, 2005.
[29] L. Yao, X. Sheng, N. Mrachacz-Kersting, X. Zhu, D. Farina et N.
[12] A. Mlenkovic, C. Otto et E. Jovanov, «Wireless sensor networks for Jiang, «Decoding Covert Somatosensory Attention By a BCI
personal health monitoring:Issues and an implementation,» system calibrated with tactile sensation,» IEEE Transactions on
Computer Communications, Special issue: Wireless Sensor Biomedical Engineering, 2017.
Networks:Performance, Reliability, Security, and Beyond, vol.
29, n° %113-14, pp. 2521-2533, 2006. [30] V. Soleimani, M. Mirmehdi, D. Damen, J. Dodd, S. Hannuna, C.
Sharp, M. Camplani et J. Viner, «Remote, Depth-Based Lung
[13] T. He, S. Krishnamurthy, J. Stankovic , T. Abdelzah, L. Luo, R. Function Assessment,» IEEE TRANSACTIONS ON
Stoleru, T. Yan , L. Gu, J. Hui et B. Krogh, «Energy-efficient BIOMEDICAL ENGINEERING, vol. VOL.64, n° %1NO.8, pp.
surveillance system using wireless sensor networks,» chez In 1943-1958, 2017.
2nd International Conference on Mobile Systems, Applications,
and Services (MobiSys04), Boston, 2004. [31] A. Petrou, M. Kanakis, S. Boës, P. Pergantis, M. Meboldt et M. S.
Daners, «Viscosity Prediction in a Physiologically Controlled
[14] A. Cerpa, J. Elson, D. Estrin, L. Girod , M. Hamilton et J. Zhao, Ventricular Assist Device,» IEE Transactions on Biomedical
«Habitat monitoring: Application driver for wireless Engineering, 2018.
communications technology,» chez In Proceedings of the 2001
ACM SIGCOMM Workshop on Data Communications, 2001. [32] P.-Y. Wu, C.-W. Cheng, C. D. Kaddi, J. Venugopalan, R. Hoffman et
M. D. Wang, «–Omic and Electronic Health Record Big Data
[15] C. Y. Poon, Y. T. Zhang et S. D. Bao, «A Novel Biometrics Method Analytics for Precision Medicine,» IEEE TRANSACTIONS ON
to Secure Wireless Body Area Sensor Networks for BIOMEDICAL ENGINEERING, vol. Vol 64, n° %1N°2, pp.
Telemedicine and M-Health,» IEEE Communication Magazine, 263-273, 2017.
vol. 44, pp. 73-81, 2006.
[33] J. Y. Lucisano, T. L. Routh, J. T. Lin et D. A. Gough, «Glucose
[16] B. Gyselinckx , C. V. Hoof , J. Ryckaert , R. F. Yazicioglu , P. Fiorini Monitoring in Individuals With Diabetes Using a Long-Term
et V. Leonov, «Human++:autonomous wireless sensors for body Implanted Sensor/Telemetry System and Model,» IEEE
area networks,» chez Custom Integrated Circuits Conference, TRANSACTIONS ON BIOMEDICAL ENGINEERING, vol.
2005,Proceedings of the IEEE 2005, San Jose, CA, USA, 2005. vol 64, n° %1N° 9, pp. 1982-1993, 2017.
[17] W. Bourennane, «etude et conception d'un système de télésurveillance [34] J. D. Amor et C. J, «Validation of a Commercial Android Smartwatch
et de detection de situations critiques par suivi actimetrique des as an Activity Monitoring Platform,» IEEE Journal of
personnes à risques en milieu indoor et outdoor,» 2013. Biomedical and Health Informatics, 2017.
[35] F. Fan, Y. Yan, K. Zhao, F. Long et H. Zhang, «Estimating SpO2 via
Time-efficient High Resolution Harmonics Analysis and
Maximum Likelihood Tracking,» JOURNAL OF
BIOMEDICAL AND HEALTH INFORMATICS, pp. 1-12,
2017.
[36] Y.-K. Huang, C.-C. Chang, P.-X. Lin et B.-S. Lin, «Quantitative
Evaluation of Rehabilitation Effect on Peripheral Circulation of
Diabetic Foot,» IEEE Journal of Biomedical and Health
Informatics, pp. 2168-2194, 2017.
[37] A. Moreau, P. Anderer, M. Ross, A. Cerny, T. H. Almazan et B.
Peterson, «Movements in Patients with Atopic Dermatitis Using
Accelerometers and Recurrent Neural Networks,» IEEE Journal
of Biomedical and Health Informatics, pp. 2168-2194, 2016.
[38] G. Prateek, I. Skog, M. E. McNeely, R. P. Duncan, G. M. Earhart, A.
Nehorai et L. Fellow, «Modeling, Detecting, and Tracking
Freezing of Gait in Parkinson Disease using Inertial Sensorsd to
Derive Respiratory Signals from ECG,» IEEE Transactions on
Biomedical Engineering, 2017.