UNIT 5

INTERNET OF MEDICAL THINGS

Enhancing the Performance of Decision Tree Using NSUM Technique for Diabetes Patients

Diabetesisacommondiseaseamongchildrentoadultinthisera.Topreventthediseasesisveryimportantbecause it
saves the human lives. Data mining technique helps to solve the problem of predicting
diabetes.Ithasstepsofprocessestopredicttheillness.Featureselectionisanimportantphaseindataminingprocess.I
n feature selection when dimension of the data increases, the quantity of data required to deliver
adependable analysis raises exponentially. Numerous different feature selection and feature
extractiontechniques are present, and they are widely used filter-based feature selection method is
proposed whichtakes advantage of the wrapper, Embedded, hybrid methods by evaluating with a lower
cost and improves the performance of a classification algorithm like a decision tree, support vector
machine, logistic regression and soon. To predict whether the patient has diabetes or not, we introduce an
ovel filter method ranking technique called Novel Symmetrical Uncertainty Measure (NSUM). NSUM
technique experimentally shows that compared to the other algorithms in filter method, wrapper method,
embedded method and hybrid method it proves more efficient in terms of Performance, Accuracy, Less
computational complexity. The existing technique of symmetric uncertainty measure shows less
computational power and high performance, but it lacks in accuracy. The aim of the NSUM method is to
overcome the drawback ofthe filter method, i.e., less accuracy compared to other methods. NSUM
technique results show highperformance, improved accuracy, and less computational complexity. NSUM
method runs in 0.03 s with89.12%asaccuracy by usingWekatool.

SmartCityHealthcareCyberPhysicalSystem:Characteristics,Technologiesand Challenges

The recent pandemic has demanded a strong and smart healthcare system which can monitor the
patientsefficiently and handle the situation that arises from the outbreak of the disease. Smart healthcare
cyberphysical systems are the future systems as they integrate the physical and cyber world for
efficientfunctioning of medical processes and treatment through external monitoring and control of
patients,medicaldevicesandequipmentforcontinuouscommunicationandinformationexchangeofphysiologic
aldata.Technologieslike InternetofThings,MachinelearningandArtificialIntelligencehave givenbirthto
smart cyber physical systems like Smart Healthcare Systems, Smart Homes, Smart Vehicular Systemsand
Smart Grid. Such systems are interdisciplinary in nature with multitude of technologies contributingto its
effective working. This paper presents a case study on healthcare cyber physical systems presentingits
characteristics, role of various technologies in its growth and major challenges in
successfulimplementationofcyber physicalmedication systems.

Keywords:Artificialintelligence,Healthcarecyberphysicalsystems,Symbioticcyberphysicalsystems,Security

Introduction

In the current pandemic, a strong healthcare system is a backbone for any smart city. Effective
andefficientmonitoringofpatientstogetherwithregularsupplyofnecessarymedicinesandtreatmentwithaid of
medical devices is possible when a system can meet the demand and supply situation. Such ascenario is
possible with a predictive mechanism which forecasts the healthcare situation and workssmartlyfor
 handling themedicalemergency.

Smart Healthcare Cyber Physical Systems (SHCPS) are the future systems capable of supporting
themedical fraternity in handling the pandemic situation effectively. Such systems comprise of
physicalworldofpatients,medicaldevicesandequipment;
externallycontrolledandmonitoredmedicaltreatment,connected with cyber world through communication
networks for data transfer and information exchangeof physiological data which are analysed for
feedback and control signals. Such systems improve thequalityofmedicalcareby providingefficientand
smartservices[1].

Thedesignanddevelopmentofhealthcarecyberphysicalsystems havebroughtsomeopenissues fordiscussion

like autonomy level, security and reliability which are vital for healthcare cyber physicalsystems.

Autonomy Level

Cyber physical systems can be classified from low to high level of autonomy based on the
tasksperformedbythehumansintheloop.Technologieslikemachinelearningandartificialintelligenceplayamaj
orroleinshifting thecontrolfromhumantomachineandartificialdistributed networks [2].

Variousfactorsthatarecrucialindefiningtheautonomylevelinhealthcarecyberphysicalsystems(HCPS)arebro
adly categorizedintotwo domains.

 Equipment based Factors like the medical device type, interaction type and duration with the
patient
aredecidingfactorsforautonomylevel.Medicalrobotsforservingfoodandmedicinestocoronaviruspatie
ntscanworkautonomouslyforassistingtheMedicaresystems.TheteamofresearchersfromHongKongha
vecreatedahumanoidrobotcalled‘Grace’forinteractingwithisolatedcoronaviruspatientsandprovides
services like temperature recording and responsiveness through the thermal camera fitted in the
chest.Therobotwillalsosociallyinteractwiththecoronaviruspatientsinordertopreventthestressarisingfr
omisolation[3].
 PatientrelatedPatientrelatedfactorsincludethediseasetypeandpatientrisklevel.Communicable
diseaseslikecoronavirusnecessitatehighlevelofautonomousMedicarecyberphysicalsystemswhereas
high patient risk level require continuous medical expert support along with the autonomous
Medicaresystems.

SecurityMechanisms

A number of technologies play a vital role in the cyber physical system which can be classified
intodistinctiveareasrangingfromdatacollectionwithsensortechnologyandIOT;handlingbigdataanditsstorage
withCloudcomputing;dataanalytics anddecisionmakingusingartificialintelligence;
control andcoordination signals to smart machines and actuators thus leading to the commencement ofnew
era termed by Industry 4.0 [4] and specifically Healthcare 4 for medical applications. Acomprehensive
security mechanism for authenticated, confidential and secure communication
betweencyberandphysicalworldforms an integralpartofframeworkduring thedesignanddevelopmentof
cyberphysical systems. The criticality of healthcare applications with high levels of autonomy demands
apreventive approach for defence against cyber-attacks [5] and thus the Healthcare Cyber physical
systemsmustincorporatethefollowingin their securitymechanism:

 Identificationofcyber-attacksonsensors.
  Lightweightcryptographyfortransmittingphysiologicaldata.
 EncryptionmechanismsandBlockchaintechnologyforsecurestorageoncloud-basedsystems.
 Encryptedfeedbacksignalstoactuators.

The data transmitted from patients to medical expert is vulnerable to attacks and hence
requiresencryption techniques at the client end. Many cryptographic techniques exist in literature,
the broadclassification is symmetric cryptography and asymmetric cryptography. Biological
cryptographicapproachbasedonaminoacidcodes [6]is usedfortransmittingthedatafrompatients
tohealthcareexperts in telemonitoring healthcaresystem.

Reliability

Reliability is an important metric for cyber physical systems and of utmost significance to
healthcareindustry.Figure1showsthereliabilityofhealthcarecyberphysicalsystems(HCPS)whichisdependen
ton hardware units like health sensors and actuators, software reliability on software systems
computingpatient’s health status, and network reliability determined by communication networks for
transfer ofpatientdata.

Fig.1

Thecommencementofcyberphysicalsystemsrequiresoptimizationofresourcesandself-
adaptivebehaviourforefficient,reliableandimprovedservices[7].Theautonomoussystemsmustbeabletoidentif
ythe failure of different components in the closed loop system and take corrective measures in terms
ofhandlingthetasks.Self-
adaptivecomponentsofcyberphysicalsystemslearnfromthepastdataandbehaveinthecurrentscenario.Smartma
chineslikehealthcarerobotscanself-organizetodynamicenvironmenttomeetthechallengeof servicequality.

Thispaperhighlightsthefollowing:

 (i)Characteristicsofhealthcarecyberphysicalsystems
 (ii) Amalgamation of technologies that have contributed to the growth and implementation
ofhealthcarecyber physicalsystems.
 (iii)Realtimechallengeswhichprompttheresearchers,healthcareandmanufacturingsectortoretrosp
ect and consider during design and effective execution of healthcare cyber physicalsystems.

CyberPhysicalSystems
The need for continuous communication, control and collaboration for efficient and effective systems
forquality of service has brought a new era called Industry 4.0 that introduces cyber physical systems.
ThetermcyberphysicalsystemwasfirstcoinedbyHelenGillatNSFinUSintheyear2006.Acyberphysicalsystemi
ntegratescyberandphysicalworldwithsensors,whichactlikedatacollectorstogatherinformationliketemperatur
e,pressureorspeed/
activitytimefortransmissiontocyberworldforstorageinservers.Thisdataisfurtherprocessedandanalysedtoacta
sstimuliforcontrolandcoordinationsignalstoactuatorsthusformingaclosed loop systemasshownin Fig.2.

Fig.2

The systematic design and modelling of cyber physical systems play a key role in defining the
architectureofthesystemwithvarioussoftwareandhardwarecomponents,thefunctionalrolesofeachcomponent;
andthe communication and control mechanism between cyber and physical space. Thereafter, the
simulationofthedesignwithwelltestedandvalidationstrategiesareappliedforfinaladoptionofthemodelinthereal
time setup [8]. Unlike the conventional approach to design and deployment, the authors in [9] suggest
ashift from design to runtime and implementation phase for critical decisions, which depend on real
timeinputsandenvironment.Table
Table11presentsthevariousCPSsimulationsoftwareandtheircharacteristics, which can be used to test
various concepts of modularity, scalability and complexity ofcyberphysicalsystemsin theapplication
domain.
Somesystemshaveasymbioticrelationshipbetweenphysicalworldandvirtualworldgivingrisetoanewterm
“symbiotic cyber physical systems” [14]. The term symbiotic relationship has arisen from biologywhere
two organisms benefit from each other and such a relationship is symbiotic in nature. Smart Grid isa
perfect example of symbiotic cyber physical system where technology drives to generate energy
smartlyand the energy is the source of power for the technology elements in the cyber physical system.
The smarthealthcare cyber physical systems are symbiotic to various autonomous systems as shown in
Fig. 3. Smartgrid, smart home, smart vehicular systems, smart hospitals and smart manufacturing units
provide servicesfor a smart city healthcare system. In turn, the smart city healthcare system provides
healthcare to itsinhabitants. These inhabitants or humans are working in various organizations and
manufacturing
units,thusprovidingtheirserviceorroleinvariousautonomousunits.Todefinethepreciseroleofhumansintheauto
nomousCPSsystemsisabigandchallengingtask.Aneffectiveapproachinthisdirectionistoidentifythe control
strategies and interactions of humans in the closed loop CPS at early stages of
softwaredevelopmentandthereafter, validatethrough fastprototyping techniques[15].
Thecyberphysicalsystemsdifferintermsoftheircharacteristicslikeautonomylevelwhichdeterminethetaskscon
trolledandoperatedbyhumans,scalewhichdependsonthenumberofdevicesconnected andriskmanagement
which depends on the critical environment of the system. Table Table22presents the variousapplications
areas, characteristic features, technologies and services provided by the CPS in differentdomains in faceof
thevarioussecurity risksand challenges.

Applicatio
Characteristics Technologies Services Security/risks/challenges
narea

Telemedicine,
Wearabl The accuracy of sensors
Healthcare4.0 edevices Dataanalyticso Roboti andother monitoring
, npatient’sdata csurger devices is
[4] y, abigchallengeasitdetermine
IOT of sthemedicationlevel
medicaldevic Mobilehealth
es

Autonomou
smachines Delaysabovesomethreshol
Manufacturin Smart inproductio dvalue lead to disruption
gunit machinesliker Machinetomachin nprocess, ofmanufacturingcycle
obots, ecommunication
[4,16–18] Self- Malfunctioningofnodes
Smartproduct organizingcap
ability
ofsmart
machines,
Applicatio
Characteristics Technologies Services Security/risks/challenges
narea

Detection
andmanageme
nt
ofindustrialha
zards,

Node
failuredete
ctionthroug
hsimilarity
iepatternm
atching

Reduce
Cameras,te powerconsu
mperatures mption, Attackonprivacyofresidents
Smart ensors by gathering data like
Remotecontrollingo Detect userinitiatedanddeviceactio
Smart fhomedevices andclassify ns
home[19– televisions,smar safetyhazard
t locks,smart s, Local network attacks
lights,smart bybringing devices
21]
switchesandsma Remote close tovicinityof
rtmeters healthmanage home
mentand
emergencyser
vices

Sensingtechnologyfor
obstacledetection, Stable
Navigationtrajector vehiclemove
Smartvehicles yundercontrolof mentwithcon
Cameras, tinuousmonit Softwaredefects
software,Prediction
sensors, mechanism oring
Autonomous ofvehicleheal
Vehiclestates todeterminemotion Difficulty in integration
th, ofdiagnostictoolsinframew
[22,23] ofothervehicles,
Proactivemechanisms ork
forfaultdetectionand Lane
detectionand
management
emissioncont
rolmechanis
ms

Vulnerabletocyberattacks
Powerline, Reliabledeliv affecting public life due
Smartgrid Remote control
ery ofpower todependency on power
andmonitoringofele
Communicatio withenergy forrunning of appliances at
[24] ctrical
nline, storageatlarge homeandequipmentinprofes
componentsingrid,
scale, sional
 sector
Applicatio
Characteristics Technologies Services Security/risks/challenges
narea

Sensors,Act Distributed Energycons

energyresources act ervationwit
asmicrogenerationun h
uators,Smart its, reducedloss
andgreenho
meters Distributed useemission
renewableenergygen s
eration

Big data analytics

onpower generated
andusage,

Predictionand

recommendationusin
gmachine learning
techniques

CharacteristicsofSmartCityHealthcareCyberPhysicalSystem

HealthcareCyberPhysicalSystemscanbedividedintovariouslevels:

 UnitlevelHCPS
 IntegrationlevelHCPS
 SystemlevelHCPS
 AcceptancelevelHCPS
 EvolutionarylevelHCPS

Unit level HCPS are basic or the first level of healthcare cyber physical systems which
providemonitoring and control of patients in intensive care units or at the hospital level. At this level
itcontinuously monitors the physiological parameters like temperature, pressure, heart rate etc. of
thepatients and feed the data to the intelligent systems which analyse and control the health
actuatorsconnectedtothepatient.Also,thehealthstaffis
intheHCPSloopforsupportandinformationtohealthexperts for immediatehealthcareaidtothepatients.

Integration level HCPS are the second stage of HCPS where the hospitals integrate with smart homes
toprovide remote monitoring and remote healthcare service to the patients. In case of transfer of high-
riskpatients in ambulances to hospitals, the latter can integrate with smart ambulances to
continuouslymonitorthepatient’shealthstatusandmakenecessaryemergencyservicesinhospital like
availabilityofbed,ventilator supportetc.
SystemlevelHCPS arethethirdstageofHCPS wheredifferentautonomoussystems supporttheHCPS toform a
Smart City Healthcare Cyber Physical System. The smart grid, the powerhouse of energy andbackbone for
various cyber physical system together with smart home, smart ambulances, smart hospitalmanufacturing
units and smart hospitals form a healthcare ecosystem providing quality of healthcareservicetothepatients.

Acceptance level HCPS is the level of HCPS where the researchers, technologists, engineers,
healthexperts,academicianscoordinatetomakethehealthcaresystemeffectivewithpoliciesandstandardsorient
edtowards successfulimplementationof thehealthcareecosystem.

Evolutionary level HCPS is the ideal future HCPS systems which have properties of self-adaptability
andself-management. Self-adaptive components of cyber physical systems learn from the past data
andbehaveinthecurrentscenario.Hence,theroleofevolutionarybehaviouriscriticalandofsignificanceforthedy
namicenvironmentof cyber physicalsystems.

The Cyber Physical System has physical components and processes which can be represented by a
statediagram and different states: healthy, unhealthy, critical and non-working state as shown in Fig.
4. Theoverall health of the CPS depends on the working conditions of different components in the closed
loopsystem[25].

Fig.4

The nature of devices in CPS is heterogeneous which range from sensors, actuators to physical
machinescontrolled by external inputs like mobile devices at dew layer and systems at cyber level. The
input andoutputofdataofvariousdevicesdifferinthedataformats,whichmaybestructured/
unstructuredinnature,whichnecessitatethe
conversionofformatsthroughinterfaces[16].Thehealthcarecyberphysicalsystemsare application domain
where time is a critical factor in determining the performance and reliability of thesystem. A delay in
seconds or microseconds in medication to patients can reduce the effectiveness
anddependabilityofthesystem.

Cyber physical systems are monitored and controlled continuously for effective operation and hence
thisgives rise to a new term called health-monitoring system (HMS) for CPS [25]. HMS can perform
dualfunction:tocheckthestatusofeachphysicalcomponentinCPSandtoapplypre-emptiveapproachin
determining probabilistic health of CPS. Study of behavioural models of physical components is a
passivemonitoring approach where as a stimulus-based response is an active strategy to determine for
anomalydetection[25].

Basedontheabovecharacteristics,theunitandtheintegrationlevelofHCPSareanalysedintermsofcharacteristics
of CPSin TableTable33.

Table3
CharacteristicsofunitandintegratedlevelofHCPS

Characteristics

Level
ofhealth
Statetransition&sta Heterogeneousde Seamlessm
CPS
tediagram vices/dataformats Timecriticalapplicati onitoring
ons &control

Healthy working state

ofall sensors and
actuators insmart Regular
hospitals as Medical Timecriticalandrequir monitoring
patientsareinriskzone applicationinterface esefficientdecisionma andcontrolinS
Unitleve stodealwithheterog king marthospitals
l eneousdata
Therefore,
earlyreplacement
of
faultymedicaldevi
ces

Healthy state depicts

goodworkingcondition
,

differentautonomous heterogeneous
Smart
systems Audio, image and datasets
ambulancesperiodica
forhealthyworkingsta video(behavioural)anal
lly checked
te, ysis,
forworking condition
Integratio
of itsmedical devices
nlevel[26 andconnectivity, Requires Different
] continuousmonitorin autonomoussystems
gforfailure connectedrequire
Coordination
medical API’s todeal
andcommunicationfo
with
rintegration of
 Monitoring
andcontrol
Home ofelderlypatients
monit
oringi
snotti Remindersyst
ems
me
critica
l Wearablesens
where ors
asmon
itorin
gofpat
ients
duringtra
nsitfromh
ometohos
pitalinsm
artambula
nceistime
critical
 Characteristics

Level
ofhealth
Seamlessm
CPS
Statetransition&state Heterogeneous onitoring
Timecriticalapplicatio &control
diagram devices/dataformats
ns

detection of nodes
andlinks

Section 4, presents the role of various technologies in seamless monitoring the time critical domain
withheterogeneous devices in cyber physical systems where each state is defined by a set of variables.
Thetechnologies playacrucialrolein definingthehealthofthesystem.

TechnologiesinHealthcareCyberPhysicalSystems

The recent pandemic has put an unprecedented pressure on the healthcare system of any city in the
worldand long working hours of medical experts and health workers. Hence, the future healthcare
systemsdemand Smart City Healthcare Cyber Physical Systems where technology plays a significant role
in itssuccessful implementation. Cyber Physical Systems is an application domain where there is
integration
ofplethoraoftechnologiesforsmartandefficientworkingofinterconnecteddevices.Internetisthebackbonefor
communication in cyber physical systems and is primarily the most crucial technology and enabler
forother technologies like IOT, cloud computing and blockchain [4]. This section presents the role of
varioustechnologies like digital twin, IOT, big data, cloud computing, blockchain, artificial intelligence,
machinelearningandrobotics inthefieldof healthcarecyberphysicalsystems as shownin Fig.5.
 Fig.5

DigitalTwin

Digital twin as shown in Fig. Fig.66is a virtual twin for a physical object or a process. In smart
healthcaresystems,thedigitaltwincreatesvirtualassetsincyberspacesothatthedigitalinformationofresourcesca
nbeusedforplanning,controlandcoordination[27].DigitaltwinforSHCPSisbasicallyasimulationmodelfor
medical devices and equipment, and for behavioural analysis of patient treatment process to assist
thehealthcareexpertstostudy, analyseand predictthehealthstatusofpatients.

Fig.6
The digital twin maintains a resource graph for various medical equipment which is stored as a three-
tuplevector represented by {resource id, allocation status, patient id}. The allocation status be busy or idle
andexpectedreallocationmaydependuponthepatienthealth.Thepatienttreatmentprocesscanberepresentedby a
graph where the nodes represent the health status at different intervals of time and the edges
representthetransitionbasedonchangeinphysiologicalparameters.Thesegraphscanbeusedformachinelearning
forcomputation andclassification of patient’s healthstatus.

InternetofThings(IOT)

TheInternetofThingsbroughtaneweraofmachine-to-
machinecommunication[4]whichcanbethroughwirelessnetworks,BluetoothandothertechnologieslikeNearFi
eldCommunication,radiocommunicationetc.TheIOTenablesinintegrationlevelsmarthealthcarecyberphysica
lsystemswherethesensornetworksgenerate high volumes of data which are then transmitted to remote
servers for analysis and controloperations. Since some devices in Medical IOT are resource constrained in
terms of processing power andmemory therefore lightweight authentication and lightweight cryptographic
schemes are essential forintegrity and confidentiality of data and information exchange. The Medical IOT
has helped in recentpandemic in remote monitoring of patients [28, 29] through smart wearables like
smart watch and
smartbandtocollectpatientsheartrateandbloodpressure;andsmartthermometerstomeasurebodytemperature.T
hesedevicesareconnectedtosmartphones,which ultimatelysend thedatatothecloudserversforhealthanalytics.

BigDataAnalytics

In healthcare cyber physical systems with IOT of Medical devices, machine to machine
communicationamong heterogeneous nodes and sensors capturing data continuously lead to large data sets
which may bestructured, unstructured or semi structured and hence require storage, processing and
analysis for medicaladvice. The big challenge is to deal with unstructured data, complexity and veracity
issues. The newcomputing paradigms with high processing power and big data technologies enable to
extract hiddenpatterns and relationships in large amount of data which are gathered from various sources
in healthcarecyberphysicalsystems.

Thepandemichasledtoregularcheck-upofphysiologicalparametersliketemperature,pressure,heartrateso that
the symptoms could be easily predicted. The different type of data generated related to coronavirusare the
human physiological parameters like temperature, pressure, heart rate; the hospital data
whichincludecurrentcoronapatientintake,patienthealthstatus,facilitiesavailablelikenumberofbeds,ventilator
s;thecitydatalikethenumberofresidentscurrentlycoronainfected,numberofpatientshealthy,number of corona
deaths, number of residents vaccinated. The researchers and analysts are interested insurvey of spread and
predictive analysis for future effect. High end computing devices and deep learningmodels enable to work
on large datasets and identify these hidden patterns. These hidden patterns can beunlockedusing
statistical,machinelearning and deeplearningtechniques.

The complexity and uncertainty of real world data generated in real time systems like healthcare CPS
canalso be dealt by methodology based on computational intelligence which includes fuzzy logic
approachbased on approximation techniques and fuzzy rules for decision making and inferences;
evolutionaryalgorithms like genetic programming and swarm intelligence for natural selection; and
artificial neuralnetworks which have many hidden layers with neurons, mimic human brain and trained on
large data setstosetthelearningparameters forpatternrecognition,predictionandclassificationofdatasets [30].
CloudComputing

Healthcare data silos are medical information of patients at discrete locations which may be redundant
ornon-
coherentbasedondatamanagementstrategies.Cloudcomputingisacomputingparadigmthatprovidesinfrastruct
ure,resourcesandservicestoendusersonpayperusebasiswherethecloudserversprovidedatastorageandcomputi
ngpowertousers.Inhealthcarecyberphysicalsystems,theElectronicHealthRecordscan be digitally stored in
encrypted format at the cloud server so that it can be shared and accessed by
thedifferententitieslikepatients,hospitalmanagement,
insurancecompaniesandbanks[31].Thechallengingissueistoensuresafetyofpatientrecordswhichdependsonth
esecurityframeworkadoptedforsafeguardingthekeygenerationcentrethatmaintainsprivatekeys of
allauthenticusers[31].

Besides the storage of encrypted data in cloud servers, industry 4.0 standards have defined new
servicemodels for cloud manufacturing like control as a service, machinery as a service and industry
automationas a service [4]. These new cloud-based models can be applied for hard real tasks with fog
computingresulting in service closed to end users with efficient real time communication in information
transfer andcontrolbetween thehealthcarelevelandcontrollers.

DewComputing

The cloud architecture provides services which can be enhanced by edge/fog computing paradigm,
adistributed service architecture which improves the efficiency of cyber physical systems by reducing
thetransmission delay of services provided by cloud model. Dew computing further reduces the delay
andprovides energy efficiency by introducing another level with smart interfaces or smart devices closer
innetworktoIOTdevicesascomparedtoedgedevices.Thesesmartsystemsprovideprocessingcapabilities,work
on data from physical components, control the actuators and have additional technology benefits
ofscalabilityandresilience[32].TheDewcomputinglayerisveryusefulforprovidingservicesinhealthcaredomai
n [33] where the patients can be monitored more effectively by providing processing and
analyticservicesclosetomonitoredarea.
AdewcomputingarchitecturewithIOTdevices;sensorsandactuatorsinfirst layer; smart devices like smart
phones and tablets in second layer called the dew computing layer;storage systems and network
equipment in edge device layer; cloudlets and servers at fourth layer callededge server layer which
represents the edge or fog computing distributed service and finally the fifth layerwith cloud servers
providing various infrastructure and software based services[32]; can be very
timeefficientforhealthcaredomains.ThelightweightapplicationsontabletsandsmartphonesinDewcomputingp
aradigmscansharepatientinformationandareinteroperablewhichhelpsincollaborationwithothersystemstobea
partof healthcarecyber physicalsystemof systems.

Blockchain

Blockchaintechnologyprovidesadecentralisedand
distributeddatabaseforsecureandauthenticaccesstoelectronichealthrecordsmaintainedbycloudservers[31].It
providesadecentralisedplatformformaintaining untampered records of events [4] for various medical
transactions that may be at device
levelorvaccines;andeventsatpatientlevel.Theblockchainisdefinedbyachainofblocks,each withanumberof
transactions or communications which are hashed and structured by a Merkle tree. Each block isidentified
by a hash value and contains the hash of previous block with the exception that the first block iscalled
genesis and has no parent hash value stored in it. Such a link with parent block gives an
immutablestructurewhichcannotbetampered.
Suchimmutablestructurescanstorethetransactionsof differentmedicaldevices andvaccines; andinformation
related to patients. Hence, this technology can have different blockchains for healthcare:Blockchain of
medical devices, Blockchain of coronavirus infected patients, Blockchain of vaccines in
ahospital.TheIBMblockchain[34]helpsintransparentdistributionofcoronavirusvaccine,bymaintainingthe
transactions safe and traceable. QuillTrace [35] is a blockchain based technology that helps in
trackingmedicinesinsupplychainandidentificationoffakemedicineswithhelpofQRcodeonmedicines.

ArtificialIntelligenceandMachineLearning

AI and machine learning can be applied at various aspects of healthcare which include medicines,
medicalequipment,patientanddisease.Theresearchers,academiciansandhealthcareexpertsareworkingtogethe
rtoextractusefulinformation,thehiddenpatternsfromlargedatabasesofpatientdata.Theselargedatabasesarealso
usedtotrainmachinelearningmodelswhichareusedtoclassifythepatientsandhelpinautomaticdisease detection
and thus support medical experts. The recent pandemic has seen the urgent need of
studyofdrugdiscoveryforcoronavirus,thehighdemandofventilatorswhichhavebecomethelifesavingdevicefor
thehigh-
riskcoronaviruspatients,thestudyofpatientswithhighrisklevelsandtheeffectofdiseaseonotherorgansof
thebody.

Basedontherisklevel,thecoronaviruspatientscanbedividedintolowrisklevelandhighrisklevel.Atthefirst stage
the patients can be isolated intheir homes and the various physiological parameters like theirtemperature,
pressure and heart rate at discrete time points can be sent to medical experts for predictingsymptoms
based on machine learning techniques and healthcare can be provided through telemonitoring.The next
stage demands hospitalization of patients with continuous medical support, care and monitoring.AI and
machine learning algorithms have been used to predict the mortality rate of high-risk patients [36]by
training the model with balanced dataset; and features are chosen based on wrapper and filter-
basedapproaches.Thesefeaturesincludesymptoms,pre-
healthstatusanddemographicfactorswhichsignificantlycontributeto thediseasestatusof thepatient.

Robots

Robots are autonomous machines which are programmed to perform a particular task with precision
andaccuracy. They are cognitive models based on artificial intelligence with capabilities to
continuouslycapture the environment data with sensors to work in complex environments and perform
pre-definedactions with high frequency. Robots have a vast role to play in cyber physical systems like
medical
robotstoassistinsurgeryandpatientcare,industrialrobotstoperformmanufacturingtasksandsurveillancerobotsf
orsecurity and safety.

The recent pandemic has seen rising number of coronavirus cases which resulted in patient overload
inhospitals.Thehealthcarerobotshavefoundakeyroleinpatientcareprovidingservicesfrompatienttestingtopati
entservicebydeliveringregularmedicines,foodetc.RobotslikeMoxiperformvariousserviceslikedelivering
PPE kits, covid 19 tests and provides pick/drop service to patients [37]. The robot Mitra assistshealth staff
by taking temperature readings of patients and helps patients in connecting with their
relativesthroughvideo conferencing [38].

Industrialrobotshelpinmanufacturingproductslike covidtestingkitsandventilatorswithhighfrequencyand
precision; and short development life cycle. Industrial cyber physical working environments withhuman
robot collaboration aim for agile product development and demand for context awareness in robotswhich
identify human working zones and adjust their area and speed [39]. Unlike machines,
 humanbehaviourisflexibleandrequiresconstantremindersintheformofaudio-visualmessagesformaintaining
asafedistanceduringcollaborativeworkwithhigh-
speedautonomousmachines.Therefore,highprecisionalgorithms are required for depth estimation to send
control signals to robots and other autonomousmachines toavoidaccidentsduetoerroneouscomputations
andmovements.

Securityrobotscanhelpindetectingadherencetocovidprotocolslikewearingofmasks,socialdistancingetc.They
captureimages,identifyobjects,generatesreportsandcommunicateusingmultimodaltechnologies through
audio, videoandtextualdata.

Figure Figure77presents the various technologies at various domains and the benefits which helps
inautomation, resource control and intelligence-based decision making based on big data analytics.
Though,various technologies have contributed to the growth of modern cyber physical systems which
have led toan evolutionary process resulting in smart, self-aware and self-healing systems but still the
present
daycyberphysicalsystemsfacemanychallengeswhichhaveanadverseimpactonthehealthofcyberphysicalsyste
ms affecting its devices, their communication and collaboration resulting in human intervention
forreinstatingtheworking stateof system.

Fig.7

Section5,presentssomechallengesforasmartcityhealthcarecyberphysicalsystemlikeenergyflowinCPS,
integration of diverse devices/levels, the delay or latency which can affect the closed loop
systemandthecyber-attacks whichhavesurfacedforinformationtheftanddisruptionofdevices.

ChallengesinSmartCityHealthcareCyberPhysicalSystem
There are numerous challenges for a successful implementation of HCPS like energy
flow,integrationofheterogeneousdevicesandatvariouslevelsofHCPS,minimumacceptabledelayfortimecrit
icaloperationsandsecurityrisksinhealthcarecyberphysicalsystemsasshowninFig.8.

Fig.8

EnergyFlowinCyberPhysicalComponents

Theautonomyofcyberphysicalsystemsandtheirapplicationsincriticaldomainlikehealthcaredemandacontinuo
us flow of energy either from high-powered batteries or from direct power supply. During
thesystemdesignandmodellingofHCPSitbecomesimperativetoestimatethevariouscomponentsandtheirenerg
y requirement so that a continuous flow of energy is maintained for smooth functioning of thehealthcare
cyber physical system. Resource constraint devices with limited battery power is a challengingissue and
various energy saving mechanisms exist to manage energy efficiently like the smart sensors
[20]withinthesedevicesadapttotheneedofenvironmentandchangetheworkingmodeaccordinglybutgivesaripp
leeffectonenergyqualitytrade-
off.ThesuccessfulimplementationofCPSanditsintegrationtoformasmartcityhealthcareCPSdependsprimarily
ontheSmartGridwhichisabackboneofenergytothecyberphysicalsystems.Thesmarthome,smarthospitalsands
martmanufacturingunitscanaidtheSmartgridbyinstalling solar panels and contribute as energy production
units in order to meet the energy requirementswhichhaveincreased exponentiallyincyber physicalsystems.

IntegrationofHeterogeneousDevices/LevelsofHCPS

The smart city healthcare ecosystem requires the integration of cyber physical systems which depends
onthirdpartyserviceprovidersforcommunicationserviceandthird-
partycloudserviceprovidersforsafedatastorage which can be accessed by medical experts, insurance
companies, patients and researchers. At theinternal level, each cyber physical system has heterogeneous
nature of devices with different data formatswhich requires integration and encapsulation of the devices
for efficient and effective environment. Themajor challenge in a heterogeneous environment will be the
need of interfaces to communicating deviceswith different technology level in terms of hardware and
software resources. The integration and higherlevels of HCPS require data sharing and collaborative
functionality which demands system level planning,designand prototypetesting [26].
Delay/Latency

Lowlatencyisacriticalrequirementforrealtimeapplicationslikehealthcare.Thedelaysaboveathresholdvalue in
transmission of patient data are unacceptable and lead to disrupting the telemonitoring cycle ofcyber
physical systems affecting timely medication and care to the patients. Also, it is significant in realtime
applications like healthcare that the fault latency which is a measure of time delay between theoccurrence
of fault and its recognition must be a small value so that its timely management could
enhancethereliability ofthesystem[23].

Security

Security of healthcare cyber physical systems is a critical issue and needs to be addressed in view of
thedifferent types of attacks like denial of service, replay, false data injection and deception attacks.
Theauthors in [40] have proposed a tree-based attack model for cyber physical systems where the
branchescategorisetheattacksinvarioussubdomainslikefaultsignalinjectionandhardwaretamperingofsensors,
packetreplayattackandinformationtheftincommunicationchannelsofclosedloopwithsensors,controllers and
actuators; equipment failure and software malfunctioning in computing resources. Criticalinfrastructures
like smart healthcare systems require cyber security systems, which monitor continuouslyfor identifying
fault injections in system that lead to incorrect working of equipment and faulty readings[24].

Smarthealthcarecyberphysicalsystemsarefuturedigitalsystemswhichmustbeforensicreadytodealandcounter
with cyber security attacks [19, 41]. Artificial intelligence or machine intelligence have wide rolein
autonomous systems performing tasks like abnormal behaviour detection due to faults or cyber-
attacks[42]. Malware detection is an important task for smooth and efficient working of cyber physical
systems.Manymachinelearningapproachesbasedonsystemcalls,operationcodesandenergyconsumptionpatter
nsareused to identifythemalwares[43].

Along with challenges related to digital world, the healthcare systems are facing the major
physicalchallenge of smart waste management. The hospitals are generating the waste at an alarming rate
whichneeds effective management strategies includingits collection at generationsites,transportation
andhandling techniques for maintaining the health of city environment. The recent pandemic has seen the
riseofPPEkits and useofdisposablemasks which needssmartwastemanagementplan.

The Healthcare Cyber Physical systems are currently working at unit or integration level, with
continuousgrowth of technologies they will evolve into higher level CPS. Smart manufacturing systems
that producevaccines or medical equipment are advanced level CPS with subsystems having
characteristics of self-awareness and self-management, and can set to self-configuration mode to optimize
the various real timeproductionprocesses[20].

Conclusion

The pandemic has taught us that healthcare systems are the lungs of every society and a smart
healthcarecyber physical system provides an ecosystem which merges the two wheels cyber world and the
physicalworld connected by a closed loop and steered by various technologies like digital twin, IOT,
cloudcomputing, artificial intelligence, machine learning and big data analytics which play a major role in
itseffective functioning. The working and implementation in physical world controlled by cyber space
facemanychallengeslike heterogeneousnature ofphysical components,incompatibledata formatsexchanged
between components, resource constraint devices and vulnerability of devices to attacks. Apart from
thedigital challenges, the physical challenge of waste management still holds its critical place. Though,
thebirth of new technologies contributes to smart interconnected systems but the challenges of physical
andvirtualworldmustbeaddressed foritsgrowth, efficiencyand effectiveness.

AnIntelligentIoTBasedHealthcareSystemUsingFuzzyNeuralNetworks

Healthcare facilities in modern age are key challenge especially in developing countries where
remote areas face lack of high-quality hospitals and medical experts. As artificial intelligence
has revolutionized various fields of life, health has also benefitedfrom it. The existing
architecture of store-and-forward method of conventional telemedicine is facing some
problems, some ofwhich are the need for a local health center with dedicated sta ff, need for
medical equipment to prepare patient reports, timeconstraint of 24–48 hours in receiving
diagnosis and medication details from a medical expert in a main hospital, cost of
localhealthcenters,andneedforWi-
Ficonnection.Inthispaper,weintroduceanovelandintelligenthealthcaresystemthatisbasedonmodern
technologieslikeInternetofthings(IoT)andmachinelearning.Thissystemisintelligentenoughtosense
andprocessapatient’sdatathroughamedicaldecisionsupportsystem.Thissystemislow-
costsolutionforthepeopleofremoteareas;theycanuseittofindoutwhethertheyaresufferingfromaserio
ushealthissueandcureitaccordinglybycontactingnearhospitals.Theresultsoftheexperimentsalsosho
wthattheproposedsystemisefficientandintelligentenoughtoprovidehealthfacilities.Theresultsprese
ntedinthispaperaretheproofoftheconcept.
1. Introduction
Internet of things (IoT) is a network in which many devicesare connected, and these devices can
communicate by
computernetwork[1].Bythisworldwidenetwork,wecangetinformationthroughsensorswhichrelatetoit.Byusi
ngcomputernetwork,wecanaccessthisinformationanywhereinthisworld.Internetofthingscanconnectphysic
alobjectstoInternetandcanprovideopportunityofbuildingsystemswhicharebasedonvarioustechnologiessuc
hasnearfieldcommunication(NFC)andwirelesssensornetwork
(WSN).Inwirelesssensornetwork,sensorssensethe environmentandsendinformationtobasestation.
IoT has different methodologies such as smart dustbin, monitoring environment, IoT based irrigation
system, smarthealthcare system, and traffic control. In healthcare system, IoT brings gadget for
monitoring health [2]. Health data can beaccessed with the help of IoT by using sensors. Healthcareis a
system which is used to improve health and help in treatingdiseases[3].
 Health related issues/complications are increasing day by day, among which lung- and heart-related
issues are top-listed.Health can be monitored by wireless technology,which is a modern concept. In
wireless health monitoring systems, differenttechnologies are used, including wearablesensors, portable
remote health system, wireless commu- nications, and expertsystems. Life is precious; even a singlelife
is also valuable, but due to lack of health facilities, awareness about diseases, andproper access to
healthcare systems, people are dropping their lives. In all situations, Internet of things (IoT) helps in
theindicationofdiseasesandtreatmentofpatients[4].
In IoT healthcare system, there exit wireless systems in which di fferent applications and sensors are
attached to
patients,informationisobtained,andthisinformationisforwardedtoadoctororspecialistthroughanexpertsystem
[5].Medicaldevicesfor Internet of things (MD-IoT) are remotely accessed, where devices are connected to
the In- ternet and sensors, actuators, andother communication devices can monitor patient health [6].
Through these de- vices,thepatientinformation anddataaretransmitted bythe expert system via gateway
onto a secured cloud based platform where the information is stored and can beanalyzed.
In developing countries like Pakistan, telemedicine is used to handle health issues. Telemedicine
refers to the practice ofcaring for patients remotely when the provider and patient are not physically
present with each other. Telemedicine is simplydefined as “the remote delivery of healthcare services.”
Although telemedicine brings with it many benefits, it has somedownsides as well. Providers, payers,
and policymakers alike know that there are some gray areas that are di fficult to keep upwith. While the
field will grow exponentially over the next decade, it will bring with it both practical and
technologicalchallenges.

1.1. Unclear Policies. Because technology is growing at sucha fast pace, it has been difficult for
policymakers to keep
upwiththeindustry.Thereisgreatuncertaintyregardingmatterslikereimbursementpolicies,privacyprotection,a
ndhealthcarelaws. Inaddition,telemedicinelaws varyfrom statetostate.
Therearecurrently29states
withtelemedicineparitylaws,whichrequireprivatepayerstoreimbursetelemedicineservicesin the same way
they would reimburse in-person visits. As additional states adopt parity laws, private payers may
institutemore guidelines and restrictions for tele- medicine services. Although it is a step in the right
direction,there is still uncertaintyregardingreimbursement rates, billingprocedures,andmore.

1.2.Fewer Face-to-Face Consultations. Several physicians and patients are finding it difficult to adapt
to telemedicine,especiallyolderadults.Physiciansareveryconcernedaboutpatientmismanagement.
Whileadvancesinmedicinehavemadeit more efficient to use technology, there are times when system
outages occur. There is also the potential for error astechnologycannotalways capture whatthehuman
touchcan.

1.3.Technology Is Expensive. Healthcare systems that adopt telemedicine solutions can attest that they require
a lot of time andmoney. Implementing a new system requires training, and sometimes staff members find
it difficult to welcome this change.Practice managers, nurses, physicians,and more have to learn how to
utilize the system so that practices can see the
benefits.Althoughtelemedicineisexpensiveinthebeginning,healthcaresystemsshouldseeapositivereturnoni
nvestmentovertimeduetomorepatientsandlessstaff.

The major components of healthcare systems are identification, location, sensing, and connectivity
as shown in Figure
1.Smarthealthcareisimplementedthroughawiderangeofsystems:emergencyservices,smartcomputing,sens
ors,labonchips,remote monitoring, wearabledevices,connectivitydevices,andbigdata.
The IoT based systems are equipped with body sensor networks within telemedicine systems. They
include
deviceswithspecialtypeofnodesthatsenseperiodicdifferenceofpatientdata;tochecktheventilationcondition
sforthepatientsinrooms,sensors are used to collect data for dif- ferent measures contributing to
ventilation process of a room. These sensors areprogrammed to assess data of di fferent ranges for
temperature, pressure, humidity, and othersignificantenvironmentalvariables.
Thesearrangementshelptomonitorthepatientcon-ditions remotely. The system can send periodic
reports to thehospital and maintain the patient history. The hospital sta ffcan view the data and prepare
the treatment plan for thepatientunder observation. The second type of devices usedin IoT healthcare
systems is based on wireless sensor net-works. Thesituation is more complex than the above sce-nario
in terms of remoteareapatient monitoring andmanagementtask. Insomesituations,IoTisthemostre-
liableandcheapestsolution,andtherelationshipbetweendifferent devices andinteractive communication
systemsalso needs to be investigated with more formal objectives.
Technologymakesiteasiertomonitorthepatient health by sending information to healthcare teams such as
adoctor, nurses, and specialists throughIoT (Internet
ofthings)andmobiletechnologies.Itwouldbehelpfulforprofessionals to save and gather patient datausing
store-and-forward method so that it is accessible at any time. Therole and services of IoT in modern
healthcare aredepictedinFigure2.
Internet of things (IoT) has different methodologies:smart healthcare, tra ffic control, smart dustbin,
and vehicle
parking.Thehealthofpatientismonitoredbyscreen,soitisdifficulttoexaminethepatientallthetime.Therefore,
here,patient’scurrentstatus,i.e.,pulserate,temperature,positionofbody, bloodglucose,and ECG,canbe
measuredinten-sivelybyusingsensors.
ThesensorsareattachedtoArduinoUNOsensorsthat,whenattachedtothebodyArduinoboard,getinformationa
ndtransmititto theserver.From thisserver,theinformation isforwarded to
thedoctorwhoadvisesformedicine.
Smarthealthcaresystemisactuallyatechnologyinwhichtreatmentsofpatientsarepossibleandcanimprovet
hestandardof life [7]. In the concept of smart health, the e-health concept is also included which has
commands on many technologieslike electronic record management, smart home services, and
intelligent and medical connected devices. Sensors, smartdevices,
andexpertsystemssupportthehealthpracticeforsmarthealthcaresystem.
Healthcare facilities in modern age are key challenge especially in developing countries where
remote areas face lack ofhigh-quality hospitals and medical experts. As ar- tificial intelligence has
revolutionized various fields of life, health has alsobenefitedfromit.Theexistingarchitectureof

Labonchip

Sensors Connectivity

Medical Bigdata
devices

Smart Remote
computing Security monitoring

FIgurE1:TypicalcomponentsofIoTbasedsmarthealthcare.

combination of sensors [8], imbedded electronics, and software components for decision support
systems [9]. The IoT alsosupportssmartsystemswiththehelpoflightweightnetworking
connectionsandsensorsdata[10].TheIoTcoversalmostalldomains such as home based smart systemslike
security, entertainment, and health; transport systems like smart parking,tra ffic, logistics emergency
services, andhighway management; community based systems like smartmetering,
businessintelligence,surveillance, environment,andretailsystems[10].
Healthcare is one of the primary concerns that need to beimproved by IoT and related technologies
[11]. IoT basedhealthcare consists of three stages of automation, namely, data collection by sensors,
analytics based on the collected
data,anddecisionmakingbasedonthecollecteddata[12].Healthcaresystemshavealotofpotentialtobeimpro
vedbyIoTbasedsystems.Therearemanydifferenttypesofhealthcareapplicationsproposedbytheresearchers
likee-health[13],communityhealth[14],homehealthmoni-
toring[15],telemedicine[16],andclinicalsupportfordoctors[17].Theprimarycontributionofthisresearchist
oprovidedevicesthathelpinmonitoring,management,andcommunicationbetweendi fferentstakeholdersin
thehealthcaredomain.
Table1showsthecomparisonofourworkwithrelated
work.ItshowsthatIoTbasedmodelsprovidemuchas-
sistancetopatients,butthetimeconstraintscanbereducedwiththehelpofCDSSintheabsenceofmedicalsta ff.
The methodology of IoTfor smart home, vehicle parking,
andtrafficcontrolisdifferentascomparedtothehealthstore-and-forwardmethod
ofconventionaltelemedicineisfacingsomeproblems:
 (i) Needforalocalhealthcenterwithdedicatedstaff.
(ii) Needformedicalequipmenttopreparepatientreports.
(iii) Timeconstraintof24–48hoursinreceivingdiag-
nosisandmedicationdetailsfromamedicalexpertinamainhospital.
(iv) Costoflocalhealthcenters.
(v) NeedforWi-Ficonnection.
Inthispaper,anovelandintelligenthealthcaresystemisproposed;itisbasedonmoderntechnologieslikeInter
netofthings(IoT) and machine learning. This system is intelligentenough to sense and process a patient’s
data through amedical decisionsupport system. This system is low-cost solution for the people of remote
areas; they can use it to findout whether they aresu fferingfromaserioushealth
issueandcureitaccordinglybycontactingnearhospitals.
 2. RelatedWork
The IoTterm was initially coined in 1999 and got attention ofcommunity asoneof the advanced
technologiesasit is
abenefitistomonitorthepatients24/7[26],whichisalmostimpossiblewithmanpower.ThesecondgoalachievedbyIo
Tbasedsolutionistomonitorprimarymeasuresneedstodeterminethe
patientconditions,andtreatmentplanmayincludepulserate,bodytemperature,respiratoryrate,bodyposition,b
loodpressure,ECG,andglucoselevel.ThesesensornetworksareconnectedthroughArduinoboardtocollectthe
informationthroughattachedsensors.Thecol-lectedinformationcanbetransmittedto
theserverandfurtherrefined fordecisionmaking ordecisionsupport systems.
Investigational experiments are made with the assistanceof sensors, and the patient’s health is traced
with Internet. Whatremains is the keen observation of pulse rate, eco of heart, pressure level,
temperature, etc. If there is any dis- turbance orchange in pulse rate or temperature, the system alerts
the person taking care of the patient. Through the Internet, the systemshowsthepulserate and
temperatureofthepatient.
The IoT with mobile technology provides smart andeasier ways to look after the patients under
observation, theirbodymovements, and health conditions and provides in- telligent mechanisms to
handle and share the relevant in- formation
withrelevantstakeholders.Thestudy[11]designedasystemwhichcollectspatientdataandsendsittocloud
forfurtherutilizationby people investigating health domain. The multipurpose application may also
provide thefamilies of patients with regularupdatesregardingpatient
health. Ghosh et al. [26] demonstrated a system to auto- matically gather data from patients and store the
gathered data intocloud for permanent use to help health profes- sionals. The system also helps the
guardians of the patients to know the healthinformation.
The study [27] proposes a system to track the patient records with the measures of pulse rate, ECO,
blood pres- sure, andbody temperature and maintain the patient his- tory. If the system detects any
abnormal behavior in the measures observed, itimmediately alerts the emergencyteam to handle the
situation. The article [28] provides a survey on the smart healthcare. Itdiscusses in detail the importance,
application, requirements, and classification of healthcare along with the challenges,vulnerability, and
se- curity attacks. Healthcare system plays a vital role in in- creasing application by using
connectivitytechnologies. The body sensor asamedicaldeviceisused toimplementsmart healthcare as
shown in Figure 3. Smarttelemedicinesystems
[15] are designed to monitor and manage the patient recordsby using sensors and microcontrollers. The
system observesthebodyconditionsandtransmitsthe datatocloudservers.The
patientconditionisobservedandstoredonserversforfurtheruseanddecisionmaking.
The study [29] investigates the challenges and conse- quences of remote health systems. The system
comprises wirelesstransmission system which collects ECG, bodytemperature, and pulse rate of the
patients in remote lo- cations for
severeproblemslikecardiogenicshock.Thepatientismonitored,anddataissenttothedoctortoanalyzethemandp
reparethetreatmentplan.Thisdataalsohelpsthesupportingstafftotakethenecessaryactions[30].
The study [31] states that health monitoring system is essential for a good health because health
problems
areincreasingdaybydaylikecardiacfailure,lungsfailure,andheartrelateddiseases.Nowadays,IoTbecameaplat
formformanyservicesandapplications in which sensor nodes are used. The monitoring of patients that is
continuously done by doctors is the base ofrevealedconsequencesofgenerichealthmonitoringsystem.

The data analytics with big data enhanced the capacitiesof healthcare management system. The IoT
healthcare is based onsensors, data collection devices, cloud services or connectivity provider devices,
and mobile applications. Themain concernof the physician is to separate the informationof one patient
from all other massive information of patients in the
healthcaresystem.Fromsuchhugeinformation,thephysicianmakescriticaldecisionaboutthepatienthealthan
dsuggeststhetreatment.He etal.concluded thatalteringpatientinformationinrealtimeisveryimportant[32].
In order to build a smart system or application, thephysical objects are connected by using IoT
(Internet of things). Thestudy used IoT for smart resource managementsystem (SRMS) and intelligent
chair system (ICS). An Arduino board isattached to the sensors, user ID is con- nected through RFID
reader with the chairs, and chair al- lotment is managed andmonitoredbythissystem.
The study [33] investigates the influence of medicalsystem with remote patients. The patient
monitoring is the mainpurpose of the system with a prototype application. The main service provided
by this prototyping system is themonitoringof vital signs of patient health in ICU. Thesystem is more
effective for patients undergoing surgical procedures or
othertreatmentsthatneedintensivecareandmonitoring.Themajorbenefitsclaimedbytheauthorsofthestudyar
elowpowerprofilesensor, wireless communication,and gateway for cheap communication. The system is
also available on web domain for thepatientcaretakersto ob-serveandbeinformedaboutthepatientstatus.
A smart health monitoring system [34] is designed and implemented for ambulance coupled with
communicationchannels. The IoT was used with this smart healthcare system with the capability of low
power sensors. The humanbodysensorsareconsidered anefficientmodeofcommu-
nicationfornearfieldbodysensornetworkapplication.
The study [35] presents a generic model for IoT based healthcare system. The model identifies key
components with anend-to-endIoThealthcaresystem.Theauthorsclaimthattherewasnoend-to-
endIoTbasedremotepatientmonitoringsystem.Thesystemconsistsoffivesensors,threeofwhichwereformoni
toringpatientconditionslikepulserate,respiratoryrate,andbody temperature. The other two sensors were
used to monitor blood pressure and bloodoxygen. The paper also
identifiestechnologicalchallengesandpotentialopportunitiesforremotehealthcaremoni-
toringandmanagementsystemwithIoT.
Thestudy[36]presentstheapplicationsandpotentialusageofIoTinhealthcaresystems.Themajortaskper-
formedby
 thisexperimentalstudywastomonitorthepatientconditionsandmakeitpossibletousemoreoptimizedand
accuratemedicalequipment.Thebasicarchitectureof
Smarthealthcare

Applications Enduser Managemen

connectivity

Smartdevice Hospitals BodysensorsStationarydevices

DB
WiFiNFC
Earlywarning Homebased
systems patientsGovernment Network Mobliephon
bodiesClinics Security VitrosensorsVivosensors
Monitoringsystem

Forecasti
ngsyst
Classificationofsmarthealthcaresystem.

the proposed system was based on sensor data and analyzesthe patient data to make it possible to take
basic decision
forthepurpose.Theproposedsystemwascollectingbodytemperature,respirationrate,andheartbeatsandobserv
ingthepatientbodymovements.
Wearablebodydevices[37]wereusedtodesignsystemsformonitoringandmanagement ofpatientshealth.
ThestudyusedbodywearablesensorswithIoTforsmartpatientmonitoringandmanagement.Thefocusofthisstu
dywastoobservethepatienthealth during surgical procedures, considering more useful and reliable data
collection during such complex situations
wherehumanobservationalskillsarenotenough.Thesecondmajorbenefitwasthereductionofequipmentsizefo
rpatientmonitoringand wireless en- vironment for patient care. Devices like Fitbit health monitor, Pebble
smart watch, and Google glass are con-sidered as modern devices for health monitoring and body care
solutions. The important wearable devices are mea- suring theblood pressure which is used to assess the
stress on a human mind. IoT imparts a valuable role to electronicsand electricaldevicestomonitorand
managehealthcareforhumankind.

3. ArchitectureofSmartHealthcareSystem
The proposed smart healthcare system has the capability of decision making as per the observed
conditions of the pa- tientbased on body temperature, pulse rate, and heartbeats. This architecture is also
energy efficient solution because it does notturnonallthe sensorsallthe time.The
algorithmusedinthesystemwillhandle the usage ofthesensorsandcontroltheircostand lifetime. The
proposed system ad- dresses the issue of remote monitoring of patients and provides them with
necessarytreatmentthrough expertsinthehospital.
Thesmarthealthcaremonitoringandpatientmanage-mentsystemproposedinthisstudyconsistsofcommuni-
cationchannels,embeddedinternaland
externalsensors,IoTserver,andcloudstorageandissupportedbyagateway.
These activities are performed at different levels of refine- ment named application layer, management
layer, network layer,anddevicelayer. ThearchitectureoftheproposedsystemispresentedinFigure4.
The architecture shown in Figure 4 is revised to show more details. The use of sensors and decision
support systemintelemedicineisanovelideathatimprovesworking
performanceoftelemedicineinruralareas.

3.1.Data Collection through Sensors for Smart Healthcare System. With the help of IoT (Internet of
things), the proposedsystem will be designed to implement a device in remote clinic. The device will
take data of patient’s heart- beats,bodytemperature,andbloodpressureasinputand
willsendittothedoctorconcernedinthehospital.Withthehelpofthedata,thedoctorwillanalyzetheconditionoft
hepatientandwillinformtheremoteareacliniccrewaboutthenecessarystepsforpatient’sbesttreatment.
The architecture presented in Figure 5 shows physical view with necessary components of the
proposed system. Thesystem consists of three sensors: body temperaturesensor, pulse rate sensor, and
heartbeat sensors. These threesensors
areconnectedthroughArduinoboardtocollectandclassifythepatientdata.Thedatatransmissionismanaged
bycommunicationandnetworkingdevices.Thedataana-lyticsprovidesthedecision-
makingfacilities,andthefuzzylogicsystemisusedinthisarrangement to provide decisionmaking. The doctor
view provides the facility to hospital staff to monitor and communicate withthepatient atremoteplace.
The next subsection explains the fuzzy logic system implemented in this smart patient monitoring
and man-
agementsystemfordecisionmaking.Thefuzzysystemisplacedattheserveranditwillorderthedecisionsregardi
ngpatientconditionsand treatment and alert the doctor about the situation of the patient. The system is
fully automated. The last subsection givesthetechnicaldetailsanddescriptionoftheproposedsystem.
3.2. Fuzzy Logic-Based Smart Healthcare Monitoring and Management. There are the following
problems: a single model isnot enough, so two or more models are combined to solve that problem [38].
When different models are combined,
theyprovideaneffectivesolutiontotheproblem,referredtoasahybridsystem.Ahybridsystemisusedtoobtainind
oorairquality usingthefuzzylogicsystemand neuralnetworksrepresentedasthefuzzyneural network(FNN).
Neural networks focus on perceiving patterns, not on thelogic of how the decision is made [39]. The
fuzzy logic systemsare good at explaining how the decision is made, butthe inference rules are di fficult
task as prior knowledge is required
[40].Theselimitationsleadtothefuzzyneuralnetwork.Rulesoffuzzysystemsareacquiredfromtheneuralnetworkspa
tterns[41].
Thisprocessbeginswitha “fuzzyneuron,”
andtheprocessofthefuzzyneuronisdividedintotwostepsasfollows[42]:
(i) Evolutionofafuzzyneuronmodel.
(ii) Developmentofthemodelanditsalgorithmthatconsolidatefuzzinessintotheneuralsystem.
Figure 6 indicates that neural inputs are provided for neural network that provides neural outputs.
Neural outputsare theinference rules for the fuzzy interface that are storedin the system as a database
and used for decision making
andprovidelearningalgorithmsfortheneuralnetworkaspriorknowledge.Dataofneuralnetworksisgatheredby
propa-gationalgorithm,sotheprocedureisslow.Includingspecificdataintothe
neuralnetworktoclarifylearningtechniquesisadifficulttask.Fuzzyrules are explained, and they provide
better performance, so fuzzy systems are used in restricted systems and knowledgeacquisitionis a
difficulttask.Tosolve these problems insolutiondesign,the fuzzyrulesaredesignedfromnumericaldata.
The neural network model named Approximate Rea- soning Intelligent Control (ARIC) (see Figure
7) uses fuzzyneuronsystem. This fuzzy neuron system is trained by physical system forecast. It applies
a fine-tuning refreshing data method tocontroltheinformationbase.
Database

Neuralputnetwork Fuzzyinference

Neural Decisions
Neuralin

Learningalgorithm

FIgurE6:Modeloffuzzyneuralnetwork

ancestor and acquires a unit u. The unit u communicates with the activity control. It is called
defuzzified mixture whichfinishes the process.The
informationlayerinthisframeworkisfuzzified;itismonotonicinnatureandhasthecapacitiestoutilize its
components in ARIC model. The fuzzy tag is used in rules to balance local standards. The ancestor’s
enrolmentis estimated by these standards and then duplicated and uses load joining with the
association ofinformation component.The qualities base of this system produces the final input. Each
unit which was obscured is exceptional monotonic workcommunicating with final standard. The
monotonicity of this function yieldsthe output. The process is effortlesslydetermined by the op- posite
capacity. This esteem is produced with the function ofheaviness and with the association
ofhiddenunit. Theyieldvalueisfinallydeterminedbyweightedaveragemethod.
Theactionoperatorsusedtoevaluatethenetwork,whichtriesto forecastthe model
activity.Theneuralnetowrkmethodused in this system is a typical feedforward neural network system.
This feedforward neural network system isbased onshrouded layer which collects the model states as
information. It uses the blunder flag r from the physical model as a
pieceofhelpingdata.Theprocessgetsvt, tofthe proposed system produced as a forecast for future. Thissystem
relies on loadoftimetandthemodelconstraints.The
tiseithertort+1.Theconditionsinthissystemare
[ ]
 portrayed by fortification of higher values for information collected for decision making. The change
in load is man- agedby support method that uses the output of the system states of the network and
action state evaluation method.
TheengineeringofARICwasconnectedtopostadjustment.Itisalsodemonstratedthatthemodelwiththecom
pre-hensionanditsassignments.
Thesignsandweightsarerealnumberswithinputneurons.
This is a perfect combination of fuzzy system with neuralnetworks which boost the advantages of
both decision-
makingmethods.Theframeworkcanlearn,andinformationutilizedintheframeworkhasthetypeofif-
thenfuzzysystem.Therulesaredefined in advance, and the system canstart without outside help, so it
adapts quicker than a standard neural system.
TheframeworknamedARICconsistsofAENactionpositionwhichisusedtoevaluatethenetworkconstructedthr
oughinformationbase.TheARICalsocontainsASNoperationusedfornetworkselection.Itisamultilayerneural
networktechniquewithfuzzycontrol system. The ASN component has two separate fuzzy in- terfaces in
the first layer of the proposed system. The neuralnetworkisplaced in thesecond layer.
Theneuralsystemfindsf[a,a+1]operation,apartofconfidenceacquired
throughfuzzy inference. Itshould gainp (a+1)using
theamountoftimedenotedbytandtheconditionofframeworkt+1.Amodificationmodulewhichisstochasticinn
ature
improvesthecontrolwithp(t)offluffypartandtheexpectedlikelihoodregradingdecisionsandproducesthefinalout
put.
u′ (t) r(u(t),d[t,t+1]. (1)
The unitcioffuzzyinference isorganizedtoassessthefuzzyguideline.Theunitforinformationajisastandard
Theinformationdoesnotchangethesesigns.Theyieldisverymuchequivalenttotheinformation.Thesignalaima
ycol-laboratewiththeloadtowitoconstructtheseitems.
d wiai, i 1,2. (2)
The inputdata diiscollected,byaddition, todelivertheinformation,
net d1+d2 w1a1+w2a2, (3)

to the neuron. The neuron utilizes its exchange work f, whichcould be a sigmoidal functionf(x)(1 +e−x)
−1
, to figureouttheoutput:
y f(net) fw1 a1+w2 a2. (4)
This basic neural net, which utilizes duplication, addi- tion, and sigmoidal f, will be called a(nordinary
neural net asshowninFigure8.
Inthisevent,weutilizedifferentactivitieslikeat-norm,orat-
conorm,tojointheapproachinginformationtoaneuron;weget what we call hybrid neural net. These
modifications lead to fuzzy neural engineering dependent onfuzzy arithmetic
tasks.Thisgivesusachancetoexpressthesourcesofinfoa1,a2andtheweightsw1,w2overtheunitintervals[0,1].The
a1
w1

w2

a2

y=f (w1a1+w2a2)
 TaBlE2:Directfuzzificationofneuralnetwork.

Fuzzyneuralnet Weights Inputs TargetType1 Crisp Crisp Fuzzy

Type2 Fuzzy Crisp Crisp

Type3 Fuzzy Fuzzy Crisp

Type4 Crisp Fuzzy Crisp

Type5 Crisp Fuzzy Fuzzy

Type6 Fuzzy Fuzzy

FuzzyFIgurE8:Neuralnet.
systemsisto broadenassociationweightsandadditionalinputstofuzzynumbersasshowninTable2.
A set of fuzzy rules were defined for the clinical decisionsupport system used for IoT based
telemedicine. These rulesarebasedon the factsand fuzzydatashown in
Table2.Followingareafewexamplesoffuzzyrulesdefined.

IF(Temperature High)AND(Pulse_Rate Low)AND(Blood_Pressure Very_High)

THENDecision High
IF(Temperature High)AND(Pulse_Rate Low)AND(Blood_Pressure High)
THENDecision High
IF Normal) AND (Pulse_R-ate High)AND(Blood_Pressure Mediu
m)
(Temperature
THENDecision Low

IF(Temperature Low)AND High)AND(Blood_Pressure Medium)

THENDecision (Pulse_Rate
Low
IF Normal) AND (Pulse_R-ate Low)

(Temperature Normal)AND(Blood_Pressure
THENDecision High

3.3. Implementation Details. A microcontroller board (Arduino) (see Figure 9), which has model
numberATmega328, has
4digitalpinsforinputandoutputsources.Thesixi/opinsarePWMoutput.Themicroprocessorhas16MHzwithap
owerjack,USB connection. The other components on this microcontroller chip are analog input and reset
button with ICSP header. Thepower is supplied bya USB interface, and Arduino is designed as open
electronicplatform. The basic settings on Arduino
boardareinput/output,set/resetbutton,sensorlights,andactivating motorwithoutputLED.
HC-05Bluetoothmodule:Toaddwirelessfunctionalityoftwoways(fullduplex)toyourproject,HC-
05isverycoolmodule.If communication is required between two microcontrollers, Bluetooth module is
used as Arduino and can communicate
withanydevicewiththefunctionalityofBluetoothlikealaptoporaphone.BluetoothSSP(serialportprotocol)mo
duleisdesignedfor wireless transport. HC-05 can be used in a master or slave configuration that will
begreatsolutionforwirelesscommunication.
Temperaturesensor(seeFigure10)isusedtodetectheatstroke,bodytemperature,andfever.Inwearablehealthc
are
system, body temperature is used as a diagnostic tool. For themeasurement of body temperature,
thermistor type sensors areused.Temperaturesensingaccuracyislimited.
The temperature sensor (see Figure 11) is integrated circuit which is used to measure the body
temperature in
centigrade.Thetemperatureisshownasvoltageoutput.ThemodelnumberofthissensorisLM35.Thismodelofb
odytemperaturesensorisconsideredbetterinperformancethanlineartemperaturesensor.Thereasonisthatuser
neednotconvertKelvinscaletocentigradescalebyusingthismodel.Thesensorunderthissetupisveryusefulforr
emotesensingandcalibratesCelsiusscale.Theemergencyconditionsaremeasuredthroughcardiacarrest,pulm
onaryembolism,vasovagalsyncope,andpulsesensor.
Thepulserateisprimarymeasureforcriticalmedicalconditionsandbodyfitnessconditions.Thepulseratesenso
risthemostused and researched sensor in patient care and managementdomain.It isused toassess
heartbeats and complex diseaseslike heart attack. The sensor works when theobject places finger on input
panel. The output is detected onoutput panel. The
powerrequiredforthissensoris5voltdirectcurrent.Theworkingprincipleofthismoduleisbasedonbloodflowra
tethroughfinger.The heartbeat sensor normal reading was 60–100 bpm. Figure 12 shows the used blood
pressure sensor to measure the bloodpressure
ofthepatientandrecorditinanExcelsheetforfurtherprocessing.
1. ExperimentalResults
1.1.Experiment Setup. The system is tested under the su- pervision of medical sta ff. Samples are
collected from dif- ferentareas of South Punjab using the proposed device. Thedata collected through
sensors was forwarded to the server. The resultsare presented at the Arduino application and webbrowser.
Table 3 shows the information about locations thatwe selected totesttheproposedmodel.Almosteightdif-
ferentlocationsareselectedfortesting.Thedistancefrom
BVHandtestingperiodofselectedlocationsaredifferent.

1.2.Dataset. Table 4 shows the report sample of the patient that is generated on the server after receiving
data collectedthrough sensors and forwarded through smart device. Thereport has three sections:
patient’s data, sensor data, andsymptomsofthepatient.
Table 5 shows the comparison of response time of the queries that are responded to by CDSS and by
physician. Almost270 queries were received on the server from se- lective areas. Most of the queries
were treated by CDSS. Table 5 clearlyshowsthataverageresponsetimeofthe
queriesthatarerespondedtobyCDSSisquiteshort ascomparedtotheresponsetimeofthe
queriesrespondedtobyphysician. The proposed systemislow-costand
efficientsolutionforthepeopleofremoteareas;they can use it to findout whether they are su ffering from a
serious health issue and cure it accordingly by contacting nearhospitals.
 Usingsensorsanddecisionsupportsystemintele-
medicineisanewidea,andTable5showshowitminimizestimeconstraintincomparison to
theclassicaltelemedicinemethod.

1.3.UsedToolsandDataAnalysis.Theuseofanalyticspotentiallyimprovestheaccuracyandpermitsearlydiseas
edetection,personalization,andcostreductioninmedical

12

1.ConclusionsandFutureWork
The proposedmethodconsistsofsensorsforbody tem-perature, pulserate,and blood pressure to assessthe
con-dition
ofthepatientunderobservation.Fordeterminingthepossibleconditionsandcure,thesystemusedaknowledge
baseandfuzzylogicsystemforintelligentdecision makingforpatientcare,monitoring,and management.

10

94.70

AdaBoostwithFeatureSelectionUsingIoTtoBringthePathsforS
omaticMutationsEvaluationinCancer

Abstract

Nowadays, the research in bioinformatics helps in finding out numerous ways in storing,
managingorganic information, and developing and analyzing the computational tools for better
understanding. Sofar, much of the research has been carried out to overcome the difficulties in
experimental methods whilestoring vast amounts of the data in different sequencing projects. In this
process, many of thecomputational methods and clustering algorithms were brought to light in the past to
diminish blocksbetween newly sequenced gene and genotypes by applying identified jobs. The latest
specific applicationsinvented in bioinformatics are paving way for more advancement by adding
developments in machinelearning and data mining fields. Because of a large quantity of applications
acquired by various featureencoding methods, the existing classification results remained inadequate.
Hence, the present study isintended to create awareness among the readers on the various possibilities
available in finding somaticmutations by using machine learning algorithm, AdaBoost with feature
selection, a classification invarious feature selection techniques with their applications, and detailed
explanation on the distinct typesof advanced bioinformatics applications. This study presents the
statistical metric-based AdaBoost featureselection in detail and how it helps in decreasing the size of the
selected feature vector, and it explainshow the improvement can be attributed through some
measurements using performance metrics:correctness, understanding, specificity, paths of mutations, etc.
The present study suggests some IOTdevices for early detection of breastcancer.

A Fuzzy-Based Expert System to Diagnose Alzheimer’sDisease

Abstract

Soft computing techniques came into reality to deal effectively with the emerging problems related
tomany fields. A medical diagnosis is totally based on human abilities, uncertain factors,
ambiguoussymptoms, high accuracy, and bulk of medical records. Soft computing techniques are suitable
to obtainresults in an efficient way in medical diagnosis. Fuzzy logic (FL) is one of the popular soft
computingtechniques. FL is a mathematical approach for computing and inferencing which generalizes
crisp
logicandsetstheoryemployingtheconceptoffuzzyset.Fuzzylogichasbeensuccessfullyappliedinthefieldsof
pattern recognition, image processing, knowledge engineering, medical diagnosis, control theory,
etc.,Alzheimer’s disease (AD) is the most popular dementia in aged people. AD is an irreversible
andprogressive neurodegenerative disorder that slowly destroys memory, thinking skill, and degrades of
theability of performing daily tasks. Hippocampus is a key biomarker for AD to identify the disease at
anearly stage. To detect and diagnose Alzheimer’s disease at an early stage, fuzzy logic is playing a
vitalrole. In this study, computerized system for classification of AD was constructed using fuzzy
logicapproach, i.e., fuzzy inference system (FIS) to classify the subjects into AD, mild cognitive
impairment(MCI),andnormalcontrol(non-AD)on thebasisofvisualfeaturesfromhippocampusregion
Secured Architecture for Internet of Things-Enabled
PersonalizedHealthcareSystems
Abstract
The Internet of Things (IoT) is the emerging area. This technology is made to connect any object
aroundus to the Internet with the unique IP, and these connected objects can be communicated each
otherremotely as per the user’s convenience. It has applications in all most all the fields like
industries,factories, environment, agriculture, transport, education, healthcare, energy, and retail. IoT
leads to thenewtechnologieslikebigdataandcyber-
physicalsystems.Connectinganyobject,fromanywhereatanytime,isnotsimple.Ithasvariouschallengeslikedi
scovery,scalability,softwarecomplexity,interoperability,faulttolerance,security,andprivacy.Oneofthemajo
rchallengesissecurity.Duetotheweak links used to connect the things to the Internet leads to many
security issues in different levels
oftheIoT.ThispaperpresentsthevarioussecurityissuesandnovelsecurityarchitecturefortheIoT-
enabledpersonalizedhealthcaresystems.KeywordsIoT·InternetofThings·Security·Architecture·Healthcar
e
7.1Introduction
Sinceadecade,InternetofThingsisevolving.TheenablingtechnologiesofIoTarethecloudcomputing,wireless
sensor networks, communication protocols, big data analytics, embedded systems, etc.
InternetofThingsisatrulyubiquitouscomputing,i.e.,anywhere,anytimeforeveryonecomputingsaidbyWeiser
[1]. The phrase “Internet of Things” is first coined by the Kevin Ashton in 1999 at MIT. Nowadays,
themobilephonesanddataratebecomeverycheaper,thewirelesscommunicationdevicesbecomingsmallerand
cheaperandtheprocessingcapabilitiesismore.Sothesmartphonebecomesthemediatorforthings,Internet,andp
eople[2].Ithasvastapplicationdomainsliketransportationandlogistics,healthcare,smartenvironments,
Personal, Social and Futuristic applications of IOT [3]. The major relevant scenarios forthis
domainareshown in Fig. 7.1
By 2025, the market share of the IoT applications is projected as healthcare 41%, manufacturing
33%,electricity7%,urbaninfrastructure4%,security4%,agriculture4%,resourceextraction4%,vehicles2%,an
d retail 1% [4]. The projected analysis shows that major application of IoT healthcare leads a goodmarket.
Remote monitoring of the health plays a key role in the domain of the healthcare. If many IoTdevices are
connected to Internet, it may lead to many attacks. Particularly in the healthcare, if any data ismisused,
then the patient’s life will be expired. The major attack in healthcare IoT is impersonate
attack.Inthisattack,theintruderwillpretendashe/sheisapatientordoctorforcollectingthedata,etc.Thisleadstom
anysecurityissues.Itis necessarytodevelopsecurityarchitectureintheareaofhealthcareIoT.
This paper proposed the novel security architecture for the IoT-enabled personalized healthcare
systems.The next sections in this paper are organized as follows: In Sect. 7.2, related works are discussed.
TheproposedarchitecturesarediscussedintheSect.7.3.Finally,inSect.7.4,conclusionandfutureworkwerediscu
ssed. 7.2 Related Work In this research work we have [5] proposed an architecture for an IoT, itconsists
of the IoT nodes and the protocols where we can plant the sensors and the actuators and
thisarchitecturecanbefitfordifferenthealthcaresystemsandvariousotherapplicationsindifferentdisciplines.
The experiments also proved that this is effective architecture for the IoT. In [6], authorsproposed a
detection method for various attacks in the household appliance. The analysis was done in thetheoretic
fields by the simulation methods. In terms of the accuracy and localization, the proposed methodproved
as good. In paper [7], authors proposed a novel architecture for 5G smart dieses, in the layer-wiseand
comparison of the various diabetic versions and discussed about the personalized healthcare systems.In
the paper [8], proposed the various medical care services like the metabolic syndrome with the cloud-
based personalized healthcare systems. Here service broker is responsible for the dynamic creations in
theperspectiveoftheusers.Thisarchitecturecanbeusedforthehealthcaresystemswiththeadditionofsomesecurit
y features. In paper [9] proposed the security features for the hybrid cloud architecture for the IoTwith
some security features. Particularly the authors had concentrated on the issues like scalability
andinteroperability, and some research challenges have been discussed. 7.3 Proposed Architecture In
theproposed system, the communication will be from the sensors that equipped with the patient’s body to
thecloud, and from the cloud to the corresponding doctor. The authorized doctor will access the data
fromcloud and treat the patient based on the obtained values. This will be done in layer-wise. The layers
in IoTare user-side layer, edge-side layer, and cloud-side layer [10]. The security should be provided in
andbetween those layers. The data flowed from one layer to other layer (Fig. 7.2). Here the goal is to
remotemonitor the patient that living in their home from the hospital. To achieve this goal, the sensor
values,communications links, data in cloud, etc., in all layers should be secured. To provide security for
thoselayers in
be equipped in the patient’s body and patient’s home for continuous monitoring the patient and
patient’shealth condition. The continuous data obtained by that equipped sensors will be stored in the
componentcloud. From that cloud component, the data can be accessed by the doctor module. The
doctor’s moduleconsists of the doctors and there have a facility to collect the data of a patient from the
cloud. In betweenall the three modules, doctor, patient, and cloud, secured communication links should
be established, i.e.,the security mechanisms module. For storing the data in cloud, the secured cloud
storage component isused. In the security mechanisms module, with the communication technology, there
is necessity tointegrate any one or more security mechanisms like the intrusion detection systems (IDS),
authorizationmechanisms, cryptographic mechanisms, and secured routing protocols. Simulataneously in
the securedstorage module need to use the secured data storage techniques like authorization and
cryptographicalgorithms.
7.4Conclusion
The Internet of Things is the emerging technology, and it has various applications in all disciplines.
Thispaperisfocusedonthepersonalizedhealthcaresystems,andnovelsecurityarchitecturefortheIoT-
enabledpersonalized healthcare systems is proposed. This architecture leads to many research challenges
likedeveloping novel IDS, cryptographic mechanisms, and secured routing protocols. In the future work,
wedevelopthenovelintrusiondetectionsystemsfortheRPL-
basednetworksformitigatingtheroutingattacksliketheblack holeattack,wormholeattack, and rankattack.