ABSTRACT

Polycystic Ovary Syndrome or PCOS is an endocrine disorder that occurs in women of

reproductive age. The condition once detected cannot be cured but treatment can help relieve its
effects.

The exact cause of PCOS is still unknown but there are certain factors that portray the risk of
getting PCOS. The factors that result in this syndrome are obesity, insulin resistance, blood
pressure, depression, inflammation. The symptoms include: hirsutism, Oligo-ovulation, acne, heavy
bleeding, skin darkening. Using the causes and symptoms, a model is prepared in order to accept
them as features and outputs the presence or absence of this condition. The machine learning
models used for supervised classification are Logistic Regression. The reason behind building
multiple models is to find out the best one for the given dataset, in the known scope of knowledge

● web- flask server

● access to all public
● refer the nearest gynecologist
 TABLE OF CONTENTS

Chapter No. TITLE Page No.

ABSTRACT v
LIST OF FIGURES viii
LIST OF ABBREVIATIONS ix
1 INTRODUCTION 1
1.1. OVERVIEW 2
. MACHINE LEARNING 3
MACHINE LEARNING STRATEGIES 3
SUPERVISED LEARNING 3
UNSUPERVISED LEARNING 4
DEEP LEARNING

2 LITERATURE SURVEY 6
RELATED WORK 6
3 METHODOLOGY 8
EXISTING SYSTEM 8
PROPOSED SYSTEM 8
OBJECTIVE 8
SOFTWARE AND HARDWARE REQUIREMENTS 9
SOFTWARE REQUIREMENTS 9
HARDWARE REQUIREMENTS 9
LIBRARIES 9
PROGRAMMING LANGUAGES 10
PYTHON 10
HTML
CSS
3.5.4. DOMAIN 10
SYSTEM ARCHITECTURE 11
ALGORITHMS USED 11
 MULTI LAYER PERCEPTRON 11
LOGISTIC REGRESSION 12
MODULES 13
DATASET COLLECTION 13
TRAIN AND TEST THE MODELS 13
DEPLOY THE MODELS 13
4. RESULTS AND DISCUSSION 14
WORKING 14
5. CONCLUSION 15
CONCLUSION 15
REFERENCES 16
APPENDICES 17
A. SOURCE CODE 17
B. SCREENSHOTS 20
C. PLAGIARISM REPORT 24
D. JOURNAL PAPER 25
 LIST OF ABBREVIATIONS

ABBREVIATIONS EXPANSION

ML Machine Learning

DL Deep Learning

MLP Multi-Layer
Perceptron
LG Logistic
Regression
 CHAPTER 1
INTRODUCTION
PCOS (Polycystic ovary syndrome) is one of the most common hormonal disorders
which most women suffer once in a lifetime during their reproductive age. This paper
considers the issue when it affects more than 70% of the population of females, it is not
that known among the people. Many time’s females ignore the symptoms and only
recognize it when it gets very serious when one has to take drastic measures against it.
PCOS is a hormonal disorder where women develop some follicles in their ovaries
called cysts and these follicles cause ovaries to fail in releasing eggs. Women suffering
from PCOS can release excessive androgen (male hormone) and can suffer from
extended or infrequent menstrual periods .

This paper focuses on the female population to make them aware about this
disorder so that they can take the symptoms into consideration as not everyone goes to
screenings as this disorder can be diagnosed through it, considering the symptoms can
be seen as a cause to go for screenings. Up until 2019, all that has been done
regarding PCOS are always based on ultrasound images and image recognition
techniques to diagnose the disorder using particle swarm optimization, automated
detection using follicle recognition, segmentation using poly cystic ovary using
ultrasound images, data mining techniques or speed gradient method.

Out of all the methods one of the problems is they are considering the symptoms
but focusing on the ultrasound images or recognizing follicles through any other
screening method and then giving the result that determines whether a person is
suffering from PCOS or not. The researcher is predicting the percentage of chances that
PCOS can occur in a female only through the symptoms using back propagation
algorithm in neural network and genetic algorithm as they are known to provide a good
result. These two methods can also help us recognize the disorder in the early
reproductive age of a female (9-15 years) to warn them about their future.

Polycystic Ovary Syndrome is a disease that occurs in women during their child-
bearing years. The reproductive organs of women called ovaries that produce
progesterone and estrogen-hormones that regulate the menstrual cycle, are affected.
Ovaries also produce little amounts of androgen also called as male hormones. The
basic features of PCOS are:

● Cysts in ovaries.

● High levels of hormone: androgen.

● Irregular Periods

● Excessive body hair growth

Since the condition is a syndrome, it has a collection of symptoms that suggest its
presence. These symptoms play a vital role in detection of this condition. Along with
these symptoms, causes which may lead to the risks of the syndrome can also be
considered. It is necessary to detect PCOS as early as possible since it holds the risk of
infertility, diabetes, endometrial cancer and cardiovascular disease at a later stage of
the condition. Here, a couple of machine learning models are built in order to determine
the presence of PCOS. Since the dataset does classify if the condition is present or not,
supervised machine learning algorithms are used called: K-Nearest Neighbor (K-NN)
and Logistic Regression. The former is a distance based technique and the latter is
probability based, this is why these two techniques, that are poles apart, are used and
their accuracy is compared. Logistic Regression is more accurate with accuracy 92%
while K-NN’s is 90.74%. But we tried to get more accuracy than above two
techniques.so we tried to use another algorithm which is Multi layer Perceptron from
DL.whuch gave us the accuracy of 98%.
 OVERVIEW
This report discusses the result of the work done in development of "YUNI-PCOS
DETECTOR" using machine learning and python. It aims at the development of an
application that assists females to check whether they are prone to pcos or not,and can
get some food tips and gynecologist nearby .

There are also some remote villages which lack medical facilities. The dataset was
processed in DL models Multi Layer Perceptron. While processing the data, symptoms
are given as input and the disease was received as an output.This project helps to get the
idea about the disease of an individual based on the symptoms he/she have, and get the
treatment easily by contacting the concerned doctor.

1.2.MACHINE LEARNING
Machine learning could be a subfield of computer science (AI). The goal of machine
learning typically is to know the structure of information and match that data into models
which will be understood and used by folks. Although machine learning could be a field
inside technology, it differs from ancient process approaches.
In ancient computing, algorithms are sets of expressly programmed directions
employed by computers to calculate or downside solve. Machine learning algorithms
instead give computers to coach on knowledge inputs and use applied math analysis so
as to output values that fall inside a particular vary. thanks to this, machine learning
facilitates computers in building models from sample knowledge so as to modify decision-
making processes supported knowledge inputs.

1.3 Machine Learning Strategies

In machine learning, tasks square measure typically classified into broad classes.
These classes square measure supported however learning is received or however
feedback on the educational is given to the system developed. Two of the foremost wide
adopted machine learning strategies square measure supervised learning that trains
algorithms supported example input and output information that's tagged by humans,
and unattended learning that provides the algorithmic program with no tagged
information so as to permit it to search out structure at intervals its computer file.

Supervised Learning

In supervised learning, the pc is given example inputs that square measure

labeled with their desired outputs.The aim of this technique is for the algorithmic
program to be ready to “learn” by comparison its actual output with the “taught” outputs
to search out errors, and modify the model consequently. Supervised learning thus uses
patterns to predict label values on extra unlabelled information. For example, with
supervised learning, an algorithm may be fed data with images of sharks labeled as fish
and images of oceans labelled as water. By being trained on this data, the supervised
learning algorithm should be able to later identify unlabelled shark images as fish and
unlabelled ocean images as water.

A common use case of supervised learning is to use historical information to

predict statistically probable future events. It's going to use historical stock exchange
info to anticipate approaching fluctuations, or be used to filter spam emails. In
supervised learning, labeled photos of dogs are often used as input files to classify
unlabeled photos of dogs.

Unattended Learning

In unattended learning, information is unlabeled, that the learning rule is left to

seek out commonalities among its input file. The goal of unattended learning is also as
easy as discovering hidden patterns at intervals a dataset, however it should even have
a goal of feature learning, that permits the procedure machine to mechanically discover
the representations that square measure required to classify data.

Unsupervised learning is usually used for transactional information. You will have
an oversized dataset of consumers and their purchases, however as a person's you'll
probably not be able to add up of what similar attributes will be drawn from client profiles
and their styles of purchases.
 With this information fed into Associate in Nursing unattended learning rule, it
should be determined that ladies of a definite age vary UN agency obtain unscented
soaps square measure probably to be pregnant, and so a promoting campaign
associated with physiological condition and baby will be merchandised.

MACHINE LEARNING CLASSIFICATION

MACHINE LEARNING TASK

 DEEP LEARNING

Deep learning is a machine learning technique that teaches computers to do

what comes naturally to humans: learn by example. Deep learning is a key
technology behind driverless cars, enabling them to recognize a stop sign, or to
distinguish a pedestrian from a lamppost. It is the key to voice control in consumer

devices like phones, tablets, TVs, and hands-free speakers. Deep learning is getting
lots of attention lately and for good reason. It’s achieving results that were not
possible before.
In deep learning, a computer model learns to perform classification tasks directly
from images, text, or sound. Deep learning models can achieve state-of-the-art
accuracy, sometimes exceeding human-level performance. Models are trained by
using a large set of labeled data and neural network architectures that contain many
layers.

DEEP LEARNING
 CHAPTER 2

LITERATURE SURVEY

2.1 RELATED WORK

Zhang,2018 ,Tried to predict and identify the risk of pcos through

Hyperprolactinemia, Cushing’s syndrome and non-classic congenital adrenal
hyperplasia algorithm.

Norman,2007,paper determines the data analysis on the Risk of Obesity and

Overweight in a Women study on the disorder and its three diagnostic criteria in
depth giving us insights on not just pcos.
P. Mehrotra, 2011, used machine learning algorithms like Bayes and Logistic
Regression .Performance of both the models was compared and the best
classifier is used to predict the above diseases.

B.Cahyono Adiwijaya, M.s. Automated Detection of PCOS using follicle

recognition(fr) has tried to recognize PCOS based on the FR through pre-
processing to increase the quality of image.

SharvariS.Deshpande ,Asmita’s implementations of Convolutional Neural

Network on Pcos classifies patients into Pcos and non-Pcos class by using
convolutional neural network algorithms.

Honglei Jing,Xiang Xu, Jinzhu Wang, provided a prospect that Pcos can be
prevented at an early age .

Xiangyang Liu, Juan Bao, Yan, used various types of machine learning
techniques. In that Rf algorithm is found superior.

Akinrotimi,Akinyemi Omololu, paper determined that patient suffering from obese

are more likely to get hypertension and more.so by controlling obesity in patients
makes them more responsive towards clomiphene.