You are on page 1of 9

Skin Cancer Detection Application using Machine Learning

Submitted By
Hayat Ahmad Khan (14486)
Shah Zeb (14798)

Session 2018-2022

Supervised By
Rimsha Khan

A project report submitted to The IQRA National University, Peshawar in partial


fulfillment of the requirements for the degree of

BACHELOR OF SCIENCE (BS) IN COMPUTER SCIENCE

DEPARTMENT OF COMPUTER SCIENCE


IQRA NATIONAL UNIVERSITY, PESHAWAR-PAKISTAN
DECEMBER 2018

i
FINAL APPROVAL
This is certified that we have studied this report, title “Skin Cancer Detection” submitted by Hayat
Ahmad Khan and Shah Zeb under the university registration No.14486, 14798. We conclude that this
project report is of sufficient standard to warrant its acceptance by the IQRA National University,
Peshawar for the award of degree of Bachelor of Science in Software Engineering BS(SE).

Approved By:

__________
Supervisor Name Supervisor

__________
External Examiner Name External Examiner

__________
Internal Examiner Name Internal Examiner

__________
Dr. Atif Ishtiaq

Chairman Name Chairman

ii
AUTHOR’S DECLARATION
I hereby declare that I am the sole author of this thesis. This is a true copy of the thesis, including any
required final revisions, as accepted by my examinations, its further declared, that I have fulfilled all
the requirements in line with the Quality Assurance Guidelines of the Higher Education Commission.

iii
ACKNOWLEDGEMENT

All acclamation and appreciations are for Al mighty Allah, who bestowed mankind with
knowledge and wisdom. I bow before his compassionate endowment. I also pay my respect to last
Holy prophet Hazrat Mohammad (PBUH), who becomes to well-wisher of mankind.
I feel great pleasure and honor to express our deepest sense of gratitude and sincere feelings of
reverence and regards to my supervisor Mss. Rimsha Khan for his efficient guidance and kind
supervision throughout our project report. I am extremely thankful to all of my teachers, friends
and all those respondents who took part in my project.

Hayat Ahmad Khan


I.D. #14486

Shah Zeb
I.D #14798

iv
TABLE OF CONTENTS

Approval..........................................................................................................................................ii
Acknowledgement .........................................................................................................................iv
Table of Content ............................................................................................................................ v
Abstract...........................................................................................................................................ix

CHAPTER 01 INTRODUCTION ............................................................................................... 1


1.1 Overview ............................................................................................................................. 1
1.2 Introduction ......................................................................................................................... 1
1.3 Motivation ........................................................................................................................... 3
1.4 Problem Description ........................................................................................................... 3
1.5 Objective and Goals ............................................................................................................ 3
1.6 Significance of the Study .................................................................................................... 4

CHAPTER 02 LITERATURE REVIEW .................................................................................. 4


2.1 Overview ............................................................................................................................. 4
2.2 Literature review ................................................................................................................. 4

CHAPTER 03 METHODOLOGY .............................................................................................. 7


3.1 Overview ............................................................................................................................. 7
3.2 Proposed framework ........................................................................................................... 7
3.3 Preprocessing ...................................................................................................................... 7
3.4 Machine learning models architecture .............................................................................. 11
3.5 SDLC Model ..................................................................................................................... 20
3.6 Flow Chart ........................................................................................................................ 21
3.7 Use Case............................................................................................................................ 22

CHAPTER 04 RESULTS AND DISCUSSIONS ..................................................................... 34


4.1 Overview ........................................................................................................................... 34
4.2 System configuration ........................................................................................................ 34
4.3 Dataset description ............................................................................................................ 35

v
4.4 Results and model evaluation ........................................................................................... 35

CHAPTER 05 Conclusions and Future work .......................................................................... 39


5.1 Overview ........................................................................................................................... 39
5.2 Conclusions ....................................................................................................................... 39
5.3 Future Work ...................................................................................................................... 39
5.4 References……………...………………………………………………………………...41

vi
LIST OF FIGURES
Figure 3.1 Proposed framework.....................................................................................................07
Figure 3.2 Visual understanding of how a 36-feature vector form................................................10
Figure 3.3 KNN model visual understanding................................................................................12
Figure 3.4 Visual presentation of possible hyperplanes................................................................14
Figure 3.5 A hyperplane in 2D and 3D features space..................................................................14
Figure 3.6 Figure show support vectors.........................................................................................15
Figure 3.7 Diagram of Decision Trees...........................................................................................18
Figure 3.8 Diagram of Random Forest Classifier..........................................................................19
Figure 3.9 Diagrammatically understanding of Random Forest Classifier...................................19
Figure 3.10 Incremental Model......................................................................................................20
Figure 3.11 Flow Chart..................................................................................................................21
Figure 3.12 Use Case Diagram......................................................................................................22
Figure 3.13 Registration Page........................................................................................................23
Figure 3.14 Login Page.....................................................................................,,...........................24
Figure 3.15 Profile Page................................................................................................................26
Figure 3.16 Take Photo or Import Image Page..............................................................................27
Figure 3.17 Signs and Symptoms Page.........................................................................................28
Figure 3.18 History of Skins Page.................................................................................................29
Figure 3.19 History of Skins in Database......................................................................................30
Figure 3.20 Storage in Database....................................................................................................31
Figure 3.21 Users in Database.......................................................................................................32
Figure 3.22 Authentication of User in Database............................................................................33
Figure 4.1 Samples of Skin cancer................................................................................................33
Figure 4.2 Learning curves of KNN..............................................................................................34
Figure 4.3 Learning curves of SVM..............................................................................................34
Figure 4.4 Learning curves of Naïve Bayes...................................................................................34
Figure 4.5 Learning curves of Random Forest..............................................................................34
Figure 4.6 Naïve Bayes..................................................................................................................35

vii
Figure 4.7 KNN Confusion matrix................................................................................................35
Figure 4.8 SVM Confusion matrix................................................................................................36
Figure 4.9 Random Forest Confusion matrix................................................................................36

viii
ABSTRACT

Skin cancer is now one of the most dangerous and fatal types of cancer. Skin cancer is caused by
unrepaired deoxyribonucleic acid (DNA) in skin cells that create genetic defects or mutations in
the skin. Skin cancer tends to spread gradually to other parts of the body and is therefore more
curable in its early stages, which is why it is best detected in its early stages. The rising number of
skin cancer cases, the high mortality rate and the expensive medical treatment require early
diagnosis of the symptoms. The identification of these diseases is generally done through manual
inspection by medical experts. This can be a troublesome and time-consuming process. For this
reason, machine learning model were used to automatically classify this disease to avoid issues as
well as to save time of doctors. In this study we had used different supervised machine learning
models and done comparative analysis of these model on skin cancer screening dataset.
Additionally, we had done some visualization of these accuracy metrics in which include learning
curves of these models and confusion matrix that help us to understand performance of these
models.

ix

You might also like