Skin Cancer Detection Mini Project With All the Details Attached!

To identify and detect Skin Cancer in patients

1. Presented by
ARIVUMATHI S(621320106007)
MADHUSHYA S(621320106056)
MUKILA K (621320106064)




9/16/2022 1
20EC506L – MINI PROJECT – I
FIRST REVIEW
SKIN CANCER DETECTION



DOMAIN OF THE PROJECT: Image Processing
ACADEMIC YEAR: 2022-2023 BATCH NO: 01 YEAR/SEMESTER: III/V
KONGUNADU COLLEGE OF ENGINEERING AND TECHNOLOGY
(AUTONOMOUS)
Tholurpatti (P.O), Thottiam –T.K, Trichy – 621 215.
(Approved by AICTE, New Delhi & Affiliated to Anna University, Chennai, Accredited by NBA (CSE, ECE & EEE), Accredited by
NACC with B++ Grade, Recognized by UGC with 2 (f) & 12(B) and ISO 9001:2015 certified Institution)
DEPARTMENT OF ELECTRONICS AND COMMUNICATION ENGINEERING


GUIDEDED BY
Mr. S. BASKAR, M.E., (Ph.D)
AP/ECE,
KNCET.

2. •
•
•
Deep learning algorithms are now being used for processing
medical imagery and pathological tools.
The features of the affected skin cells are extracted after the
segmentation of the dermoscopic images using feature
extraction technique.
A deep learning-based Convolutional Neural Network (CNN)
classifier is used for the stratification of the extracted features.
An accuracy of 89.5% and training accuracy of 93.7% have
been achieved using publicly available data set.

9/16/2022 2
INTRODUCTION

3. •
•
•
This is the method of detecting the skin
cancer cell using the image processing
which consumes less time than the
existing method
This Early detection of skin cancer cell
helps to increase the life time of the
human beings
9/16/2022
3
Contd.,

4. ABSTRACT
•
•
•
Skin cancer is an alarming disease among mankind. It is caused
by un-repaired DeoxyriboNucleicAcid(DNA) in skin cells, which
generates genetic defects and mutations on the skin.
The necessity of early diagnosis of skin cancer have been
increased because of the rapid growth rate of Melanoma skin
cancer, its high treatment costs and death rate.
These cancer cells are detected manually and it takes time to
cure in most of the cases. This project proposes a skin cancer
detection system using image processing machine learning.

9/16/2022
4

5. OBJECTIVES
•
•
We proposed a web based detection of skin cancer cell
detection using deep learning algorithm such as Convolutional
Neural Network(CNN) which includes image processing.

We give image of the skin cells from the test report.which will
be processed through image processing and the test result will
be displayed in the site.

9/16/2022
5

6. PROBLEM IDENTIFICATION
•
•
•
•
Risk of recurrence plays a major disadvantage
in manual detection Incase , if it spreads within
the detection then there will be no chance for
diagnosis.
Many melanomas can be cured with surgery, if
caught early. Melanoma that has metastasized
has a possible of curing by radio therapy but
after several stages it can’t be treated.
Time required for the detection takes too long
So, these risks of the existing system can be
over come by the proposed system

9/16/2022 6

7. 9/16/2022
7
LITERATURE REVIEW
TITLE
AUTHOR
&
YEAR
REMARKS

Melanoma
detection using
KNN

Vidhya and
Karki
&
2014
They downloaded 328 photos of benign
skin lesions and 672 images of
melanoma for their project from the
International Skin Imaging Collaboration
(ISIC). Using SVM classifiers, they got
classification results with 97.8%
accuracy and 0.94 area under the curve.
Furthermore, while employing KNN, the
sensitivity they achieved was 86.2
percent and the Specificity was 85
percent

8. 9/16/2022
8
Contd.,
TITLE AUTHOR &
YEAR
REMARKS
Skin cancer
detection

Sanketh &
2016
They concluded that in the future, the
researcher will need to use a larger
dataset to reduce the risk of overfitting.
Furthermore, in order to attain high
accuracy, CNN must learn to retrieve data
from people with dark skin. Age, gender,
and race must also be considered in order
to attain better results. However, boosting
the accuracy rate is still a work in progress.
Melanoma and
Squamous cell
carcinoma
detection
Roffman
&
2017
A new pixel-based fusion method is used
for their project. This stage addressed the
issues of uneven lesion shape, texture,
and size, as well as the presence of a
lesion on the border region. But the
execution time was too long in this project
for attaining moderate accuracy

9. EXISTING SYSTEM
9/16/2022 9


A skin biopsy is needed to diagnose skin cancer.
Your doctor removes a sample of skin tissue,
which is sent to a laboratory. In the laboratory, a
pathologist studies the sample under a
microscope.
The pathologist looks for abnormal cells that
indicate cancer. If it is cancer, the biopsy sample
provides important information about the cancer
stage

10. 




Processor : i5 intel core
RAM : 4GB
Hard disk : 4 GB
Keyboard : Standard keyboard
Monitor : 15.6 inch color monitor

9/16/2022
10
HARDWARE REQUIREMENTS

11. 





Operating system : Windows OS
Editor : Sublime Text Editor
Front End : HTML,CSS,JS
Back End : Flask
Programming Language : Python




9/16/2022
11
SOFTWARE REQUIREMENTS

12. REFERENCES





12
https://www.kaggle.com/datasets/kmader/skin-cancer-mnist-ham10000
https://ieeexplore.ieee.org/document/9397987
https://ieeexplore.ieee.org/document/9696874
https://skincancer.net/diagnosis
https://www.mayoclinic.org

[1] B. J. Janney, S. E. Roslin, and M. J. Shelcy, “A
Comparative Analysis of Skin Cancer Detection based on
SVM, ANN and Naive Bayes Classifier,” in 2018
International Conference on Recent Innovations in
Electrical, Electronics & Communication Engineering
(ICRIEECE), Bhubaneswar, India, Jul. 2018, pp. 1694–
1699. DOI: 10.1109/ICRIEECE44171.2018.9008943.

9/16/2022

13. QUESTIONS AND DISCUSSION
9/16/2022 13

14. THANK YOU
9/16/2022 14