Brain Tumour Detection Using ML Techniques
Final Year Project Proposal
Session 2016-2020
A project submitted in partial fulfilment of the
COMSATS University Degree
of
BS in Computer Science / Software Engineering (CUI)
Department of Computer Science
COMSATS University Islamabad, Lahore Campus
18 March 2019
Project Registration
Project ID (for office use)
Type (Nature of project) [ ] Development [ ] Research [ * ] R&D
Area of specialization
Project Group Members
Sr.# Reg. # Student Name CGPA Email ID Phone # Signature
SP16-BCS- Sheikh Mohsin
(i) 176
3.46
[email protected] 0324-4999482
Baber
SP16-BCS-
(ii) Syed Asad
092
SP16-BCS- sp16-bcs-
(iii) Abdul Majid 2.95 03424417874
012 [email protected]
Name & Signature of Batch Advisor
(If students are eligible for FYP)
Plagiarism Free Certificate
This is to certify that, I am ________________________ S/D/o _______________________, group leader of FYP
under registration no CIIT/_________________/LHR at Computer Science Department, COMSATS Institute of
Information Technology, Lahore. I declare that my FYP proposal is checked by my supervisor and the similarity
index is ________% that is less than 20%, an acceptable limit by HEC. Report is attached herewith as Appendix A.
Date: ____________ Name of Group Leader: ________________________ Signature: _____________
Name of Supervisor: _____________________ Co-Supervisor (if any):____________________
Designation: _____________________ Designation: _____________________
Signature: _____________________ Signature: _____________________
Approval of FYP Management Committee
Committee Member 1: Name: _____________________
[ ] Accept [ ] *Defer [ ] *Reject Signature: _______________
*Remarks: _____________________________________________________________________
Committee Member 2: Name: _____________________
[ ] Accept [ ] *Defer [ ] *Reject Signature: _______________
*Remarks: _____________________________________________________________________
Convener: Name: _____________________
[ ] Accept [ ] *Defer [ ] *Reject Signature: _______________
*Remarks: _____________________________________________________________________
Project Abstract
Brain tumor is an unusual growth caused by cells that are reproducing themselves in an
unrestrained way. Curing cancer has been a major goal of medical researchers for decades. The
early detection of cancer can be helpful in curing the disease completely. A Machine Learning
based approach to identify brain tumor from Magnetic Resonance Imaging (MRI) images can
help in quicker more efficient detection. The main objective of the project is to help identify and
classify brain tumors by training a ML on MRI scans of various tumor free brains as well as
brains with tumors to allow the system to learn how to classify unseen MRI. This method
describes the strategy to detect and extraction of brain tumor from patient’s MRI scan images of
the brain. This method uses some noise removal functions, feature extraction and segmentation
operations from the basic concepts of image processing. The process of tumor detection and
elicitation of the tumor from the image is done using Python.
Introduction
Brain tumor is an abnormal mass of tissue in which cells grow and multiply uncontrollably
seemingly unchecked by the mechanisms that control normal cells. Many problems in computer
vision were saturating on their results before a decade. However, with the help of machine
learning (ML) techniques, the results of these problems drastically improved. Brain cancer
remains one of the most incurable forms of cancer. A primary brain is a tumor that starts in the
brain. It’s easy to diagnose brain cancer at initial stages. One of the major problem was that of
image classification, which is defined as predicting the class of the image. A well know
application of object detection is that it can be used in medical applications for tracking objects.
The major challenge in this problem is that of the variable dimension of the output which is
caused due to the variable number of objects that can be present in any given MR image.
Another important obstacle for widespread adoption of object detection systems is the
requirement of real-time while being accurate in detection.
Motivation and Scope
This year, an estimated 23,880 adults (13,720 men and 10,160 women) in the United States will
be diagnosed with primary cancerous tumors of the brain. Brain tumors account for 85% to 90%
of all primary CNS tumors. Also, about 3,560 children will be diagnosed with a brain tumor this
year. But in Pakistan diagnosed ratio is very low. A neurologist with high skills can identify
tumor stage from MR images but still there is chance of mistake to identify what type of tumor
exists, is it benign or malignant. ML techniques are available to identify more accurately. So, it is
good initiative to use some advance ML techniques to improve brain tumor diagnostic rate in
Pakistan.
Related Work
As traditional way of detecting brain tumor is used in this project, following are some of the
work done in this area:
Akwinder Kaur, Komal Sharma, Shruti Gujral, “Brain Tumor Detection based on
Machine Learning Algorithms”, International Journal of Computer Applications, 2014
proposed a traditional method for detecting brain tumor by first preprocessing MRI
images then, segmentation of image is done by using watershed segmentation technique.
Prachi Gadpayleand, P.S. Mahajani, “Detection and Classification of Brain Tumor in
MRI Images”, Internal Journal of Emerging Trends in Electrical and Electronics, July-
2013 proposed brain tumor detection and classification system. The tumor is extracted
using segmentation and then texture features are extracted using various classifiers to
classify image into normal or abnormal brain.
Mehdi Jafari, Reza Shafaghi, “A Hybrid Approach for Automatic Tumor Detection of
Brain MRI using Support Vector Machine and Genetic Algorithm”, Global Journal of
Science, Engineering and Technology, 2012 proposed a traditional method for detecting
brain tumor by first preprocessing the images then the images are segmented.
System Architecture
Goals and Objectives
To detect the size and location of brain tumours from the Magnetic Resonance Images.
To improve the performance of the segmentation of brain tumor images.
To reduce noise from brain tumor images using some efficient algorithms and machine
learning techniques.
The problem of false segmentation and segmentation with low accuracy that prevails in the
brain image segmentation can be overcome by using this approach.
Individual Tasks
Please mention your individual tasks in group project justifying one year effort.
Gantt chart
2019
Steps Task Name Performed Feb Mar Apr May Jun Jul Au Sep Oct Nov Dec
By g
1 Preparations All
(proposal, research) members
2 Project Plan (Scope All
phases, Articles members
Reading)
3 Pre-processing All
members
4 Extraction & All
Segmentation of members
Brain MR Image
5 Feature Extraction All
members
6 ML (Training - All
Testing) members
7 Extensive Work All
(Optional)
Future Work
More improvement can be brought into the system by training system with datasets that
are more complex.
Image processing algorithms can be used in future to increase the enhancement of images.
A mobile application can be developed to ease the use of system for the users.
The system can suggest the surgeons whether to operate on the tumour or prescribe
medicines to shrink or eliminate tumours.
Tools and Technologies
We will be using Python as a tool for implementing interface and backend. Different Machine
Learning and Image Processing algorithms will be used to implement system.
Appendix A
Include here the 1st page of Turnitin Report
Every supervisor has his/her own Turnitin account. If not then supervisors are requested to get the account from Library as
soon as possible.
