IAETSD JOURNAL FOR ADVANCED RESEARCH IN APPLIED SCIENCES, VOLUME 4, ISSUE 1, JAN-JUNE /2017

ISSN (ONLINE): 2394-8442

A REAL-TIME PREDICTIVE MODELLING FOR MITIGATION OF

CONTAGIOUS DISEASES- A MATHEMATICAL APPROACH
Samriddhi Vishwakarma #1, Shresth Agrawal #2
#
Bhavans RK Sarda Vidya Mandir
shresthagrawal.31@gmail.com

ABSTRACT.

This project was fabricated on the foundation of Computation Biology with the real time
predictive modeling for mitigating contagious diseases with a mathematical approach. The aim of this project
was to develop a control system that could compute and foresee an infectious disease in three stages: foremost,
quasi and hindmost prediction, accurately quantify the severity of an infectious disease. The idea of
accentuating this project emanated from the headlines of a daily newspaper quoting Zika prevailing all
around and India is on guard (the week). Surveys in local hospital proved that the world is going through a
serious threat of infectious diseases which had no effective real-time control system. On the contrary,
Mathematicians use the SIR model for dealing with this problem but which proved to have various cons. This
predictive system was built to perform on every aspect of a contagious disease which the existing models
couldnt. The processing of this system included a database of numerous diseases characterized on the basis of
mathematical factors which was obtained by the past two years report of INDIAN DISEASE SURVILLIANCE
PROGRAMME. This also included a dynamic application called DISEASOSAURUSthat can compute the
surveillance of an infectious diseases using the database and the real time data. It has unique features like
CYBER QUARANTINE and GOVERNMENT CORNER. This system was a requisite in Order to pull out
humanity from the clutches of infectious diseases.

Keywords: Emerging infectious diseases, event-based surveillance

I. INTRODUCTION

RonaldRossintheyear1857,formulatedthefirstmathematicalmodelling on malaria parasite.(www.ncbi.nlm.nih.gov/pmc/articles/PMC3162588/).

Mathematicians use the SIR model for forecasting epidemics, This was founded by A.G. Kendrick and W.O. Kermack in the year 1927 on the
research paper- A CONTRIBUTION TO THE MATHEMATICAL THEORY OF EPIDEMICS, in this paper a simple deterministic model
was formulated and was successful in predicting the behaviour of outbreaks very similar to that observed in many recorded epidemics. SIRis an
epidemiological model computes the theoretical number of people infected with a contagious illness in closed population overtime. Till date
myriad number of statistical model exists but has some loop holes:

It fails to explain the essential component of a society such as migration as it is a very uneven activity.
(www.sciencedirect.com/science/articles/pii/S1198743X14604080)

The transmission and recovery coefficient doesnt yield good accuracy on real time basis as it doesn't happen to take into consideration
the essential biological factors which change overtime and affect the prediction of a disease.

It fails to characterize a disease on the basis of severity: for e.g. : thousand cases of malaria in a tropical nation may not be an epidemic
but only a handful cases of malaria in Canada would be an epidemic, thats because thousand cases of malaria in endemic tropical area is
considered to be normal.

Mitigation Methods like vaccination, isolation and medication are not quantified on real time.

Moreover there exist many android application such as healthmaps.org, Geni-dp, Pro-med, etc. that act as an alert system for an
infectious disease but there exists no application to compute and predict infectious disease using mathematical modelling on real time basis.

To Cite This Article: Samriddhi Vishwakarma and Shresth Agrawal,. A REAL-TIME PREDICTIVE
MODELLING FOR MITIGATION OF CONTAGIOUS DISEASES- A MATHEMATICAL
APPROACH. Journal for Advanced Research in Applied Sciences ;Pages: 428-433
429. Samriddhi Vishwakarma and Shresth Agrawal,. A REAL-TIME PREDICTIVE MODELLING
FOR MITIGATION OF CONTAGIOUS DISEASES- A MATHEMATICAL APPROACH. Journal
for Advanced Research in Applied Sciences; Pages: 428-433

II. PROCEDURE
1. The data required for the processing of the predictive system will be aided from the INDIAN DISEASE SURVILLIENCE
PROGRAMMES (IDSP) WEBSITE.

2. The data collected is converted into accessible format

3. Then by formulating a code through MATLAB, the data acquired is processed.

4. It will be observed that certain common factors affect the spread of an infectious disease. Further, these factors are then enlisted and
then will be stored in a database.

5. A mathematical model is built and implemented on real time basis by including some essential components like- migration and
mitigation solutions

6. An app is also built for the visualization, alerting, functioning and forecasting of this model.

III. MATERIAL REQUIRED

1. Mat lab- It is a high performance language for technical computing this software is used because the testings required huge
mathematical calculations which couldnt be performed manually

2. Android Studio- It is an open source android development platform which is used in our project for the development of the android
application

3. PHP(personal home page) and MySQL (My Structured Query Language) was used for database management and server site
scripting

IV. DATA LISTS

The initial components that will be required for the working of the real time predictive system: (all these components will be collected using
android app on real time basis)

1. Total population of a given area

2. Infected population at a given time

3. Recovered population at a given time

4. Rate of medication and vaccination available

5. Total number of isolations possible

6. Population Migrated

7. Population Cyber Quarantined

V. LIMELIGHT

The prediction of this system will be characterized into three levels Foremost-prediction: It is often observed that there is a lot of
trepidation as soon as any outbreak of new infectious disease takes place so this system will help to combat the situation to fight an unknown
disease by the predicting its behaviour. Quasi-prediction: As soon as a new disease is identified there may exist mitigating method to it.In this
situation quasi prediction will be taken into account. Hindmost-prediction: in a given area if there is an outbreak of a disease we can use quasi-
prediction for its control but if the same disease has a re-outbreak in any other given area we will use the existing data that we obtained earlier
and this stage will be known as hind most prediction.

The transmission and recovery coefficient can vary from place to place and climate to climate in sort it may vary according to the
biological factors and will chance with respect with time. Therefore it is essential to include the biological factors to maintain the accuracy
which can be done by using real-time.

In the course of an infectious disease there can be following possible solutions they are medication, vaccination, quarantine and
isolation. This system will quantify the Mitigation solutions using mathematical modelling.

It is not possible to completely quarantine the susceptible population a method of alerting a susceptible about an approaching infection
using android app will be emanated and will hence help people to quarantine themselves. This method is termed as CYBER QUARANTINE.
430. Samriddhi Vishwakarma and Shresth Agrawal,. A REAL-TIME PREDICTIVE MODELLING
FOR MITIGATION OF CONTAGIOUS DISEASES- A MATHEMATICAL APPROACH. Journal
for Advanced Research in Applied Sciences; Pages: 428-433

Migration: This essential feature is included in this system by the help of GPS. This feature will be enabled by tracking the moment of
a people in and out of a place on an android app.

Government Corner: Hoverer, the government has the resources and capability to provide mitigating solutions but where it lacks is to
where, when and how much to provide these solution. Therefore this system will help solve this problem.

VI. WORKING

We will try to modify the SIR model according to real-time application in the following way:

A. The following will be the final equations obtained-

= 1 +

= 1 2 3 +


= 2 + 3


=


= 3

431. Samriddhi Vishwakarma and Shresth Agrawal,. A REAL-TIME PREDICTIVE MODELLING
FOR MITIGATION OF CONTAGIOUS DISEASES- A MATHEMATICAL APPROACH. Journal
for Advanced Research in Applied Sciences; Pages: 428-433

B. Factors

Alpha-
No of people
infected per day
till max condition

Mathematical
Factors to
determine
severity
Tp/Th ratio-
Beta-
The ratio
No of people
Between the
recovered per
total population
day after max
and threshold
condition
value

The definition of an epidemic is indecisive and contradicting as it differs from place to place. For example- 10000 cases of malaria in Africa
will not be termed as an epidemic, but few cases of malaria in Canada will be termed as an epidemic. Therefore, it is more apt to quantify an
infectious disease on the basis of its severity. Moreover, the severity of an infectious disease can be determined by the following factors that this
system introduces.

The significance of the above mentioned factors will be as follows:

1. Alpha and beta: by knowing the rate at which an infectious disease is growing and decaying or say spreading we can plan the
mitigation solutions prior to that with its correspondence which will result to the decay of the infectious disease a lot sooner.

2. Tp/th ratio: this is the total population by threshold ratio (threshold is a point obtained on the susceptible curve when the alpha has
reached its highest limit) this would hence help in comparing various diseases on the basis of its severity.

The above mentioned terms are the mathematical factors that may highly affect the spread of an infectious disease.
Email address is compulsory for the corresponding author.

The processing of this system includes a database which has numerous infectious diseases characterised on the basis of mathematical factors
obtained by the past 4 years report of Indian disease surveillance program.
432. Samriddhi Vishwakarma and Shresth Agrawal,. A REAL-TIME PREDICTIVE MODELLING
FOR MITIGATION OF CONTAGIOUS DISEASES- A MATHEMATICAL APPROACH. Journal
for Advanced Research in Applied Sciences; Pages: 428-433

Moreover this is the architect for hindmost prediction- The last 2 years data of various infectious diseases is taken from Indian disease
surveillance program and a database is formed. If an unknown infectious disease is registered, then its factors are found and are compared with
the factors of the existing diseases and the disease with which it matches the most is selected. The precautionary measures of the existing
infectious disease are updated for combating the unknown disease and this is the foremost prediction. When the unknown disease is identified
then, mitigating solutions are quantified. This is known as quasi prediction. Now if there is an outbreak of the same disease in some other area
then the previous data of the same disease is used and then the predictions are made accordingly. This is known as hind most prediction.

The algorithm will be performed by the following steps:

The algorithm which is the centrepiece of this system works as follows: Lets assume that there is an outbreak of an infectious disease
433. Samriddhi Vishwakarma and Shresth Agrawal,. A REAL-TIME PREDICTIVE MODELLING
FOR MITIGATION OF CONTAGIOUS DISEASES- A MATHEMATICAL APPROACH. Journal
for Advanced Research in Applied Sciences; Pages: 428-433

The initial parameters primarily required by the system are: Total population, initial infected population, infected population at nth day,
removed population at nth day.

Processing these initial parameters this system will find the severity of the infectious disease .It will then determine the severity using the
tp/th ratio and if the severity is low then the system will stop else, the system will proceed to its next step.

It will then find the other factors and by using the existing database it will try to compare it with other available infectious diseases. Further, the
combat system will start functioning. Then the cyber quarantine and warning system will be launched, mitigating solutions will be quantified
and the government corner will start its prediction and quantification. This loop will keep running until the severity is reduced.

Android app (DISEAOSAURUS):

When it comes to the practical implementations of this model, this ANDROID APP DISEASEOSAURUS is hard to beat. It boasts a
massive database of numerous infectious diseases, putting combat and control system of the infectious diseases in your hand. If you are infected
by certain illness you can simply fill a form that requires the date, zip code and the disease. After the submission of your form youll be allotted
with a serial number. Once you are registered as an infected person your data would be manipulated to the database. After which the system will
start-the cyber quarantine will be launched, migrations would be tracked, mitigating solutions will be quantified, government corner will start
processing data. With this an infectious disease will have no other option than surrendering itself.

VII. CONCLUSION
This system will be the requisite in todays world in Order to take humanity out of the hand of infectious diseases. It will make the best use of
technology available and will be reliable and will practically implement and hence this system will capable of irradiating the ultimate curse /
Infectious Diseases from the human society

REFERENCES

[1] www.ncbi.nlm.nih.gov/pmc/articles/PMC3162588/
[2] www.sciencedirect.com/science/articles/pii/S1198743X14604080
[3] https://www.youtube.com/watch?v=w5obThfgEyw
[4] https://en.wikipedia.org/wiki/Numerical_integration
[5] http://mathworld.wolfram.com/First-OrderOrdinaryDifferentialEquation.html
[6] http://mathworld.wolfram.com/SIRModel.html