ABSTRACT: can affect the Aedes mosquitoes through multiple

ABSTRACT:

Dengue is the most important disastrous
disease by which many number of people are affected. The incidence of dengue is
associated particularly with the spread of the vector Aedes Aegypti and some of
the environmental variables such as trade, travel, demographic change,
inadequate domestic water supplies, warming temperature etc.., Therefore a
mobile app that gives the reliable real time information stating when and where
to expect the outbreak of dengue is essential which would use the GPS receiver
embedded in the mobile phone and this also gives the dengue risk index in and
around the exact location or at the remote location and the spread of the
disease in the near future can also be estimated.

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!


order now

KEYWORDS: Dengue; GPS ; Risk index

1.INTRODUCTION:

Dengue disease is caused by any one offour
closely related dengue viral serotypes of the genus Flavivirus, belonging to
the family Flaviviridae. The primary vector Aedes aegypti is closely associated
with humans and their dwellings.Water-holding containers in around homes are
used by the mosquitoes to complete their development while people provide the
blood meals for their egg development.Aedes aegypti preferentially rests in
dark, coolareas,such as closets,and generally bites indoors.Eggs are laid on
the side of water-holding containers and hatch into larvae after rain or
flooding.The larvae transform into pupae,and then adult mosquitoes,in little
over a week under favourable conditions.Females are predominantly infected with
dengue viruses after biting a viremichuman.Vertical transmission between
generations also occur to an extent.It takes between 5 and 33 days at 25 degree
Celsius with a mean of 15 days,for viruses to multiply,mature and migrate to
the salivary glands before the mosquito can transmit the virus to another
person. Variations in weather and climate can affect the Aedes mosquitoes
through  multiple mechanisms.Temperature
is an important determinant of biting rate,egg and immature mosquito
development,development time of virus in the mosquito.The ideal temperature
range for survival through all life phases is between 20? and 30?.

Dengue
has severe effects on human beings. It may even lead to death.Our project has three modules. Server-mobile
connectivity-> location access and processing -> database connectivity.
The user has to download the app and login with necessary details.  The database server is created for population
density data, dengue data, vector density data, surveillance data, GPS data
(from the GPS receiver embedded in the mobile phone) ,mobility data and the
Climate data .  It should be updated
periodically by the administrator. 
Dengue hotspots are indicated according to the severity levels in
particular areas using different contours based on user request location or
current location. The dengue risk index can be found out by a fuzzy logic based
decision engine that outputs a single value in a scale of 10.  The algorithm followed is Multi objective
decision making. Here the various inputs from the database server are
considered as fuzzy variables and suitable risk index is found. Automatic
notifications are enabled cautioning the user when the risk index is high. The
susceptibility to dengue for the user is indicated as high, low or intermediate
based on the dengue risk index. Forecasting and predicting the dengue risk
index in the near future (to a span of 6 weeks) using Probability distribution
model.  The mobile app is customized to
indicate the above details along with the risk maps and rate of spread. The
location access can be the current location or the user request location. The
user is given an additional feature to intimate the administrator regarding the
severity of the spread of dengue in his or her area by tagging his/her
location, which will be moderated by the administrator after thorough
inspection.

2. PREVIOUS WORKS:

Here is
an account of a few previous works on dengue surveillance in various developing
countries

A dengue
search index was constructed based on weekly dengue cases during 2011-2014 in
Guangdong, providing predictive models and climate factors. Several machine
learning algorithms, including Support Vector Regression(SVR) algorithm,
step-down linear regression model, gradient boosted regression tree algorithm
(GBM), negative binomial regression model(NBM), least absolute shrinkage and
selection operator(LASSO) linear regression model and generalised additive
model(GAM), were used as candidate models to predict dengue incidence.
Performance and goodness of fit of models were assessed using the
root-mean-square error (RMSE) and R-squared measures. The epidemics during the
peak of 2014 large outbreak were accurately forecasted by the SVR model selected
by a cross validation technique

Jakarta is one of the five provinces with the highest
Incidence Rate (IR) in Indonesia. Thus a Dengue Early Warning System (DEWS) was
developed to detect the potential outbreaks of dengue virus based on
statistical calculations and GIS. The four attributes identified for prediction
are house density, free larvae index, container potential nest larvae and
average rainfall in the last two months. The system achieved an accuracy of
97.05% in term of Geometric Mean. Further error analysis revealed that the
sensitivity, specificity, Positive Predicted Value and F1of the system were
94.52%,99.65%,98.57% and 96.50% respectively.

A real-time forecasting model for dengue hemorrhagic fever in
the 77 provinces of Thailand had been developed. A practical computational
infrastructure has been created which generated multistep predictions of dengue
incidence in Thai provinces every two weeks throughout 2014. These predictions
show mixed performance across provinces, out-performing seasonal baseline
models in over half of provinces at a 1.5 month horizon

 3.1. PROPOSED
WORK

The
surveillance data, GPS data, dengue  data,
vector density data are to be collected  and should be updated periodically. SURVEILLANCE
DATA includes the past data of dengue affected areas, population density,
vector density ,mobility of people .DENGUE DATA includes the data of dengue
affected cases in count segregated by the particular area. VECTOR DATA includes
the mosquito data which prone to dengue. GPSDATA  includes location data along with the latitude
and longitude values. On the back end of the
system, the database server is created containing the GPS data, Vector density
data, Population Density data,  Humidity
data ,Surveillance data (contains the data of the past affected victims), and
the Mobility data .All these data has to be periodically updated and maintained.

3.2. METHOD OF FINDING A SOLUTION:

The
proposed model has three modules. Server-mobile connectivity  -> location access and processing -> database
connectivity. The user has to download the app and login with necessary
details.  The database server is created
for population density data, dengue data, vector density data, surveillance
data, GPS data (from the GPS receiver embedded in the mobile phone) ,mobility
data and the Climate data .  It should be
updated periodically by the administrator. 
Dengue hotspots are indicated according to the severity levels in particular
areas using different contours based on user request location or current location.
The dengue risk index can be found out by a fuzzy logic based decision engine
that outputs a single value in a scale of 10. 
The algorithm followed is Multi objective decision making. Here the
various inputs from the database server are considered as fuzzy variables and
suitable risk index is found. Automatic notifications are enabled cautioning
the user when the risk index is high. The susceptibility to dengue for the user
is indicated as high, low or intermediate based on the dengue risk index. Forecasting
and predicting the dengue risk index in the near future (to a span of 6 weeks)
using Probability distribution model. 
The mobile app is customized to indicate the above details along with
the risk maps and rate of spread. The location access can be the current
location or the user request location. The user is given an additional feature
to intimate the administrator regarding the severity of the spread of dengue in
his or her area by tagging his/her location, which will be moderated by the administrator
after thorough inspection.

3.3. SOFTWARE
USED:

The basic flowchart showing the implementation of
our project is  below

XAMPP stands for
Cross-Platform (X), Apache (A), MySQL (M), PHP (P) and Perl (P). It is a
simple, lightweight Apache distribution that makes it extremely easy for
developers to create a local web server for testing purposes. XAMPP is regularly updated
to the latest releases of Apache, MariaDB, PHP and Perl. It also comes with a number of other modules
including OpenSSL, phpMyAdmin, MediaWiki, Joomla, WordPress and
more. Self-contained, multiple instances of XAMPP can exist on a single
computer, and any given instance can be copied from one computer to
another. XAMPP is offered in both a full and a standard version.  Postman is a Chrome add-on and Mac
application which is used to fire requests to an API. It is very lightweight
and fast. It is basically a testing tool. It is possible to make different
kinds of HTTP requests – GET, POST, PUT, PATCH and DELETE. Sublime text is a
sophisticated text editor for code, mark up and prose. Sublime Text is built from custom components,
providing for unmatched responsiveness. From a powerful, custom cross-platform
UI toolkit, to an unmatched syntax highlighting engine, Sublime Text sets the
bar for performance.Ionic framework 
helps developers to build attractive mobile apps using HTML and Angular JS. Gis data- Google map
console javascript API: This helps us to design and add maps to our
applications.

 

3.4 ALGORITHM
FOR PREDICTION:

The generation time of dengue haemorrhagic fever (DHF) is
chosen as the discrete time interval for case reporting, thus the case reports
may more easily be used to model the reproductive rate of the disease . The
generation time for dengue is two weeks, hence we aggregated the line-list data
into biweekly intervals and interpolated the monthly counts into biweekly
counts. Interpolation was performed by fitting a monotonically increasing
smooth spline to the cumulative case counts in each province, and then taking
the differences between the estimated cumulative counts at each interval as the
number of incident cases in a given interval.

A key property of a surveillance system is the reporting
delay, defined for our purposes as the duration of time between symptom onset
and the case being available for analysis. To account for reporting delays,
our models specified a reporting lag l, in bi weeks. Data with
onset dates within last l bi weeks were considered to be not
fully reported and left out from the analysis. We present results from the
models with a lag of 6 bi weeks (about 3 months), as these produced stable
predictions across the entire region under observation.

We assumed the biweekly
province-level reported cases follow a Poisson distribution, where the previous
bi week’s reported cases serve as an offset term. Let the number of cases with
onset occurring within time interval t in province i be
represented as a random variable Yt, i,
then

where the lag-1 term yt ?
1, i is used as an offset in this model. We adopt the
convention of using lower-case yt, i to
indicate previously observed case counts that are treated as fixed inputs in
our model. We explicitly model the rate ? as

 _____ (1)

 

where C is the set
of j most-correlated provinces with province i and L
is the set of lag times used in the model; b(t) is the bi week
of time t; fi(b(t)) is assumed
to be a province-specific cyclical cubic spline with period of one year (i.e.
26 biweeks); and gi(t) is a province-specific
smooth spline to capture secular trends over time. Adding 1 to the numerator
and denominator of the correlated province covariates ensures that the
quantities are defined when no case counts are observed at a particular
province-biweek. This method of adjusting for zero counts has been interpreted
as an “immigration rate” added to each observation.

We note that the model can be
expressed as

=E (

)

?

____( 2)

 

which shows that ?t, i can
be interpreted as the expected reproductive rate at time t in
location i, or Rt, i.

These models were fit using the
Generalized Additive Model (GAM) framework (i.e. as generalized linear models
with smooth splines estimated by penalized maximum likelihood) . Each
province’s time-series was subset to remove any cases from the previous l biweeks.
The remaining data were smoothed before fitting the model and making
predictions.

Seasonal patterns were modeled
using a penalized cubic regression spline, constrained to have a cycle of one
year with continuous second derivatives at the endpoints. Secular trends were
modeled using penalized cubic splines with 5 equally spaced knots over 47 years
(roughly one knot per decade).

We approximated the predictive
distribution for all provinces using sequential stochastic simulations of the
joint distribution of the case counts for each province. We created M independently
evolving sequential chains of predictions by drawing, at each prediction time
point, from the province-specific Poisson distribution with means given
by eq (1). For
example, if data through time t* was used to fit the models for all
locations, then a single simulated prediction consisted of a simulated Markov
chain of dependent observations for timepoints t* + 1, t*
+ 2, …, t* + H, across all provinces, where H was the
largest horizon considered. To make a prediction for province i at
time t* + h in the mth chain,
we draw

Due to the assumed interrelations between the provinces, it is required
to simulate counts for all provinces at a single time point before moving on to
the next time point.

 

4.
WORKING:

It is a distributed
GIS system based on the client/server model, where the client is the element
that makes a request and the server the element that carries the request
through. There are different stages of the client/server model according to
where  the three main elements of an
application take place. These elements are the presentation, which is the user
interface, the logic, which is referring to the processing and the data, which
is the database or DBMS, and they can take place on either the client or the
server side

 

The below fig shows the
prediction of dengue in the various location and the other shows the first page
of the mobile application which shows the various features of  the application .

The above fig. shows the
use case diagram of the mobile application.

MOBILE APPLICATION
PHASE

Use
case name: Identify
Hotspots  Actor: End user

   
            Followed by the user’s login into the
app, GPS locates the user’s current location enabling the user to identify the
vector hotspots around . By identifying the nearest hotspot the user gets to
know about the risk index and susceptibility to dengue in that area.

Use
case Name: Forecast  Actor : End User

   
                 Forecasting and
predicting of dengue risk index in the near future (to a span of 6 weeks) using
Probability distribution model.

Use
case Name: Awareness
Notification  Actor: End User

   
               If the dengue risk index
is high in the user’s location, then the user gets a notification making the
user aware of the situation around.

SERVER PHASE

Use
case Name: Database
Creation  Actor: Administrator

   
              On the back end of the system,
the database server is created containing the GPS data, Vector density data,
Population Density data,  Humidity data
,Surveillance data (contains the data of the past affected victims), and the
Mobility data .All these data has to be periodically updated and maintained.

DENGUE RISK INDEX 

The dengue risk index can
be found out by a fuzzy logic based decision engine that outputs a single value
in a scale of 10.  The algorithm followed
is Multi objective decision making. Multi-objective
decision making occurs when there are several objectives to be realized. There
are following two issues in this type of decision making ?

·       
To acquire proper
information related to the satisfaction of the objectives by various
alternatives.

·       
To weigh the
relative importance of each objective.

 

 Here the various inputs from the database
server are considered as fuzzy variables and suitable risk index is found.

A rule set has been worked out that takes
into account all the above parameters in different combinations.

1 ) If the distance between our location
and the nearest hotspot is less than 1 km then the risk is very high. Between
1-2 km medium risk and if > 2km then the risk is less.

2) 
If located in an area (where lot of cases have been reported refer to
chloropeth map), then risk is high.

3) 
If 2 is true AND there is a school/hospital close by then the risk is
very high.

 4)  If
it is rainy season , summer , … and humidity

 As the user logins the app, GPS locates
his/her current location and he /she could identify the vector hotspots around
him. By identifying the nearest hotspot he gets to know about the risk index
and susceptibility to dengue

 Forecasting and predicting the dengue risk
index in the near future (to a span of 6 weeks) using Probability distribution
model.

 If the dengue risk index is high for the user
location, then the user gets the notification (stating the surveillance data
and the mobility data) as a message making him aware of the situation around.

5. RESULTS:

It has a very valuable application of predicting the
dengue risk index.Easy to use and gives real time information.Able to predict
the incidence of dengue with accuracy.Indicates the various hotspots of dengue
infected areas. The mapping system described above can be employed as a
low-cost management tool for the control of dengue in many developing countries

6. CONCLUSION:

Therefore,
by finding the dengue risk at a particular location or at a remote location the
spread of the disease in future can be determined and steps to avoid becoming
prone to the ailment can be efficiently avoided and the control measures can
also be taken. In future, the mobile application  can also be modified to find the occurring of
some other endemic diseases also and thus providing the a better preventive
caution to everyone.