SMART HEALTH PREDICTION

INTRODUCTION
We are making an online web based application name as Smart Health
Prediction System. Here we propose a system that allows users to get instant
guidance on their health issues through an health care system online. The
system is fed with various symptoms and the diseases associated with that
system. The system is first taught with various symptoms and the disease
associated with each system. The Health Prediction system is an end user
support and online consultation project. The system contains data of various
symptoms and the disease/illness associated with those symptoms. It also has an
option for users of sharing their symptoms and issues. The system allows user to
share their symptoms and issues. It then processes users symptoms to check for
illness that could be associated with it. In that we use some data mining
techniques to guess the most accurate illness that could be associated with
patient’s symptoms. If the system is not able to provide suitable results, it
informs the user about the type of disease or disorder it feels user’s symptoms
are associated with. If users symptoms do not exactly match any disease in our
database, is shows the diseases user could probably have judging by his/her
symptoms. It also consists of doctor address, contacts along with Feedback and
administrator dashboard for system operations.
 OBJECTIVES

The aimed to build a fully functional system in order to achieve a efficiency in

faster health treatment and online consultations system. The overall mission of
system development is to make the primary treatment quickly and easily
complete online consultation system. It might have happened so many times
that you or someone yours need doctors help immediately, but they are not
available due to some reason. The Health Prediction system is an end user
support and online consultation project. Here we propose a system that allows
users to get instant guidance on their health issues through an intelligent health
care system online. The system is fed with various symptoms and the
disease/illness associated with those systems. The system allows user to share
their symptoms and issues. It then processes users symptoms to check for
various illness that could be associated with it. Here we use some intelligent
data mining techniques to guess the most accurate illness that could be
associated with patient’s symptoms. If the system is not able to provide suitable
results, it informs the user about the type of disease or disorder it feels user’s
symptoms are associated with. If users symptoms do not exactly match any
disease in our database, is shows the diseases user could probably have judging
by his/her symptoms. It also consists of doctor address, contacts along with
Feedback and administrator dashboard for system operations.
 EXISTING SYSTEM

The current system is a manual and file based one, we realize that system. We
are going to build must give the solutions for wastage of time and space which
affect the efficiency of the daily activities performed at the hospital. In previous
system there is no location tracker for patient and doctors. There is no any
feedback system in existing system for taking a feedback from patient. If the
patient requires an instant diagnosis on their disease then they have to go doctor
but it is not possible to everyone to identify disease at home instantly. Today’s
health prediction system is so much time consuming.

PROPOSED SYSTEM

The system is designed to use intelligent data mining techniques to guess the
most accurate illness based on patient’s symptoms. If user’s symptoms do not
exactly match any disease in the database, then it is shows the diseases user
could probably have based on his/her symptoms. It also consists of doctor
address, contacts along with Feedback and administrator dashboard for system
operations.
MODULES:

1. Patient management
Patient Registration: If Patient is a new user he will enter his personal
details and he will user Id and password through which he can login to the
system.
Patient Login: If Patient has already an account then he/she can log into the
system.
Diseases Prediction: Patient will specify the symptoms caused due to his
illness. System will ask certain question regarding his illness and system
predict. The disease based on the symptoms specifies by the patient and
system will also suggest doctor based on the disease.
Book appointment : Patients can book appointment.
Feedback: Patients give feedback about doctors.
2. Admin management
Admin Login : Admin can login to the system using his ID and Password.
Add Doctor: Admin can add new doctor details into the database.
Add Disease: Admin can add disease details along with symptoms and type.
View Doctor: Admin can view various Doctors along with their personal
details.
View Disease: Admin can view various diseases details stored in database.
Manage data : Admin can manage data of the database.
3. Doctor management

Doctor login : Doctor can login to the system using his ID and Password.
View appointment: Doctor can View appointment of the patients.
 ANALYSIS

Analysis is done using two different strategies:--

 Structure modelling and

 Object oriented modelling

Here structured approach is used. Analysis modelling uses a combination of text and
the diagrammatic forms to depict requirements for data, functions and behaviour in a
way that is relatively. Easy to understand and more important, straight forward to
review for correctness, completeness and consistency.

Objective of the analysis we have done:--

1. To describe what the organization requires.

2. To establish a basis for the creation of a software design.

3. To define a set of requirements that can be validated once the software is built.

Dynamic Web Application:-- Dynamic web application represent a new era in Node
js based communications and work tracking system development. They integrate a
wide range of web server and browser technology into powerful applications like
business that are easily deployed on internet and intranet. A dynamic web Node js
application can use a relational database of consolidated for reporting, time tracking
and status updating, as well as manages data and user access.

 Dynamic Web application is share several common features:- A server based

application architectures in which most of the application logic and functionality
is stored on the server.
 The integration of open internet standards and protocols for communication.
 The use of browser as a universal client and browser technologies such as
HTML and Java. This creates rich and flexible user interfaces that can be
delivered across multiple plate forms.

SOFTWARE REQUIREMENT ANALYSIS

Requirement analysis provides the software designer with a representation of

information function and behaviour that can be translated to data, architectural
interface and component level designs.

Software requirements are studied wear five headings:

1. Problem Recognition: -- There is software which manages the data they are
working but the problem with existing software solution is that they don’t have
any facility to take opinion of the common reader. This is new feature in the
existing new website. This will work as a public notice bard where anyone can
write his/her opinion.
2. Evaluation and Synthesis: -- there are other document and opinion gathering
systems but they are not working as proper notice board. They have their own
limitations. The basic problem with existing website. So there is only option to
develop their own. This software may be used in much organization where there
is a use of common notice board. The appropriate market of the product is still
to be explored. This software is developed to keep it affordable for Indian
market and with least functionality which is required by Indian market.
3. Modelling: -- The model serves as a foundation for software design as the
basic for orating of specification for the software. The model used to develop
the software fall models the development requirement of the software suits this
model.

4. Specification: -- The please is explained in detail. Quality

management technique to translate the needs of customer into technical
requirements for the software. The detail DFD is given under separate
headings.

TOOLS AND PLATFORM

Technical requirements: -

Programming Productivity: -- The technology we adopt this website will have

potential and appropriate by integrated with other relevant technologies. Thus,

the ability to develop & then deploy application as effectively & effectively &

as quickly as possible is important. Reliability & availability:

Scalability: the architecture adopted to develop this project must have technical

flexibility to adopt the changes in the future. To scale effectively requires not

only the ability to handle a large increase in the number of clients but also

effective use of the system resources.

Integration: all components must be properly integrated with each other so that

performance of the system must be good.

Hardware & Software Requirement:--

HARDWARE
• Pentium4
• 4 GB Hard Disk
• 512 MB RAM
SOFTWARE:-
• Windows XP/7/Vista
• Database MySQL server
• Development Tools: JDK, Netbeans
• Development Language: JSP, HTML
ALGORITHM
Naive Bayes :

Naive Bayes is a simple, yet effective and commonly-used, machine

learning classifier. It is a probabilistic classifier that makes classifications
using the Maximum A Posteriori decision rule in a Bayesian setting. It can
also be represented using a very simple Bayesian network. Naive Bayes
classifiers have been especially popular for text classification, and are a
traditional solution for problems such as spam detection.

1.2 The Model

The goal of any probabilistic classifier is, with features x_0 through x_n
and classes c_0 through c_k, to determine the probability of the features
occurring in each class, and to return the most likely class. Therefore, for
each class, we want to be able to calculate P(c_i | x_0, …, x_n).

In the context of classification, you can replace A with a class, c_i, and B
with our set of features, x_0 through x_n. Since P(B) serves as
normalization, and we are usually unable to calculate P(x_0, …, x_n), we
can simply ignore that term, and instead just state that P(c_i | x_0, …,
x_n) ∝ P(x_0, …, x_n | c_i) * P(c_i), where ∝ means “is proportional to”.
P(c_i) is simple to calculate; it is just the proportion of the data-set that
falls in class i. P(x_0, …, x_n | c_i) is more difficult to compute. In order
to simplify its computation, we make the assumption that x_0 through x_n
are conditionally independent given c_i, which allows us to say that
P(x_0, …, x_n | c_i) = P(x_0 | c_i) * P(x_1 | c_i) * … * P(x_n | c_i). This
assumption is most likely not true — hence the name naive Bayes
classifier, but the classifier nonetheless performs well in most situations.
Therefore, our final representation of class probability is the following:

Calculating the individual P(x_j | c_i) terms will depend on what

distribution your features follow. In the context of text classification,
where features may be word counts, features may follow a multinomial
distribution. In other cases, where features are continuous, they may
follow a Gaussian distribution.

Note that there is very little explicit training in Naive Bayes compared to
other common classification methods. The only work that must be done
before prediction is finding the parameters for the features’ individual
probability distributions, which can typically be done quickly and
deterministically. This means that Naive Bayes classifiers can perform
well even with high-dimensional data points and/or a large number of data
points.

Fig.1.2
1.3 Naive Bayes Classifier

The Naive Bayes classifier is a simple classifier which uses probability to

make predictions. It is mainly built on Bayes theorem. The presumptions
it makes are strongly naive but still it has been proven to perform quite
well in many real world applications. The classifier is also referred to as
Idiot Bayes, Naive Bayes or Simple Bayes. A more descriptive term for
the underlying probability model would be independent feature model. In
simple terms, a Naive Bayes classifier assumes that the presence or
absence of a particular feature of a class is unrelated to the presence or
absence of any other feature. For instance, an object may be considered to
be a ball if it is round, bouncy, about 4 inches to 8 inches in diameter.
Even if these features depend on each other or upon the presence of other
features, a Naive Bayes classifier considers all of these properties to
independently contribute to the probability that the object is a ball.
Depending on the precise nature of the probability model, Naive Bayes
classifiers can be trained very efficiently in a supervised learning setting.
In many practical applications, parameter estimation for Naive Bayes
models uses the method of maximum likelihood in other words, one can
work with the Naive Bayes model without believing in Bayesian
probability or using any Bayesian methods. In spite of their naive design
and apparently over-simplified assumptions, Naive Bayes classifiers have
worked brilliantly in solving many complex problems. It is one of the
most effective and efficient learning algorithms for data mining and
machine learning.
 Overview of the System

Smart Health Prediction

User Admin Doctor

Add doctor Add disease View disease/ doctor

Diseases Prediction Book Appointment Feedback

( using naïve bayes)

View appointment Login

 DFD of Smart Health Prediction – Zero Level

Smart
User management Health Admin management
Prediction

Doctor management
 Level – 1 ( DFD for Users ) :-

Feedback

View doctors

User(patients)
Login

Search Symptoms
 Level – 1 ( DFD for Admin ) :-

View Users

Add diseases
Route Map

Add doctors
Admin Login
Manage data
 Level – 1 ( DFD for Doctor) :-

View appointment

Doctor Login
View users
Level – 2 ( DFD for Admin ) :-

Admin

Add doctor Add disease view disease

Add/delete Add/delete Add/delete

Store ACK Fetch ACK

ACK

Manage data
 Level – 2 ( DFD for User ) :-

User ( Patients)

Book appointment Feedback view disease

book/ Add view

Cancel

Store ACK

Manage data
 Level – 2 ( DFD for Doctor ) :-

Doctor

View appointment view patients

View View

Store

Manage data
 ER Diagram

hospital name
ID
name ID

Email
Addres
s add_docto
Add user
r
Specialist
email
Diseas Medicatio
id e n1
Timeto
Passwor has
d Timefrom
has
id precautions
Add name
disease
Doctor Medication ID
Name 2

Appointme
Spe Admin
nt details manage

Time Purpose
Pass email

ID U_name

Comment
diseases

Diseases
data has Feedback
F_id
 TESTING

Testing

Testing is the filter to catch defects before they are “discovered” by the
customer or end user. Every time a customer runs the program, he/she
generates a “test-case”. We tried the test cases to find the defects first
since software development is a human activity and there may be
potential errors. So testing before the product is delivered helped us to
assure the quality and saved resources.

Testing Objective

Objective of testing is to find errors, demonstrate that software functions

that satisfy the specification reduce the number of errors detected by
customer, and have “confidence” in the software system. A good test
always tries to discover undetected errors and is successful when errors
are found since zero defects is not possible. There is always a condition
or usage that can lead to an incorrect behaviour.

Testing Steps

We started testing each individual new component and worked out as unit
test, integration test, high order test, customer acceptance testing and
different testing techniques are used at different times in the process.

Testing Techniques

Following testing techniques are used :

White Box Testing

In white box testing we exercises the internal logic of a program, examine

internal program structure and derived tests from an examination of the
program’s logic, tested internal paths and working of the software. To
perform this we used to develop test cases for unit and integration testing.

Black Box Testing

In black box testing we exercises the input and output requirements of a

program, tested by conducting specifications of requirement for which the
software should do. This test is derived from the I/O specification and
used in most functional tests.
 FUTURE SCOPE

The system should be efficient to predict the diseases and suggestion of

medications using machine learning techniques like Classification,
Association rule mining. It has very wide range of scope in case of
diseases. Further the system can be extended to N number diseases
existing with proper medications. It can made available with the doctors
through live chat so the end user can freely interact with the doctor in
case of emergency

Bibliography
 Stackoverflow.com :- Logic for function
 expressjs.org:- express framework code for function define
 w3school.com:- html and css
 guru99.com:- Pug document code and ejs document code