Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 8

Brain Tumor MRI Images Classification Using CNN

1 Introduction
1.1 Project Introduction

A brain is the most sensitive organ of our body, which controls the core functions and
characteristics in the human body and according to the National Brain Tumor Society, in the
United States, about 700,000 people live with a brain tumor, and the figure will rise to 787,000
by the end of 2021. Compared with other cancers such as breast cancer or lung cancer, a brain
tumor is not more common but, still, a brain tumor is the number 10th leading cause of deaths
worldwide. An estimated 18,021 adults will die this year from brain cancer. Brain tumor has a
lasting and psychological impact on patient life. The Brain tumor is caused by tissue abnormality
that develops within the brain or in the central spine, interrupting proper brain function. A Brain
tumor is marked as Benign and Malignant. Benign brain tumors do not contain cancer cells and
grow gradually. They do not spread and commonly stay in one region of the brain, whereas
malignant brain tumors contain cancer cells and grow quickly and spread through to other brain
and spine regions as well. A Malignant tumor is life-threatening and harmful. World Health
Organization (WHO) has graded brain tumors according to brain health behavior, into grade 1
and 2 tumors that are low-grade tumors also known as benign tumors, or grade 3 and 4 tumors
which are high-grade tumors also known as malignant tumors. The Brain tumor is diagnosed
using several techniques such as CT scan, EEG, but Magnetic Resource Image (MRI) is the most
effective and widely used method. MRI uses powerful and effective magnetic fields and radio
waves to generate internal images of the organs within the body. MRI provides more detailed
information on the internal organs and is, therefore, more effective than CT or EEG scanning.
In the past few years because of AI and Deep learning, significant advancement has been
made in the medical science like Medical Image processing technique which helps doctors to the
diagnose disease early and easily, before that, it was tedious and time-consuming. So, to resolve
such kind of limitations computer-aided technology is much needed because Medical Field needs
efficient and reliable techniques to diagnose life-threatening diseases like cancer, which is the
leading cause of mortality globally for patients. So, in our study with the help of Brain MRI
Images, we provide a method for classification of brain tumors into cancerous and non-cancerous
using data augmentation technique and convolutional neural network model.

1.2 Background of the Project

Artificial intelligence and deep learning are primarily used in image processing
techniques to segment, identify, and classify MRI Images and are also used to classify
and detect brain tumors. So many works have already been done on the classification and
segmentation of brain MRI images. Some of the international journals we reviewed on
the detection and classification of brain tumor using deep learning. Carlo, Ricciardi
presented an approach for classifying pituitary adenomas tumor MRIs by using
multinomial logistic regression and k-nearest neighbor algorithms. The approach
achieved an accuracy of 83% on multinomial logistic regression and 92% on a k-nearest
neighbor with an AUC curve of 98.4%. We will have presented some possible
approaches to develop a system for automatic classification of brain tumor images
classification. Globally, there is a drastic increase in the rate of brain tumor cases because
of several factors. So early detection plays a crucial role in detection and treatment. In
this system, a Convolutional Neural Networks based approach have been proposed for
melanoma classification. A system is developed that can help patients and doctors to be
able to detect or identify tumor classes whether it is benign or malignant. By taking some
random images any doctor can identify the accurate results but in traditional approach too
much time are taken to detect the cases correctly. We will have presented some possible
approaches to develop a system for automatic classification of skin cancer. We applied
the data augmentation technique along with the image processing technique on those
MRI images and then trained a CNN model from scratch on that augmented preprocessed
image data to determine whether the MRI image contains a tumor or not.

1.3 Objectives of the Project

 We are going to do the literature studies for this project, and it is based on the
research that will be helpful for further studies.

ii
 We will integrate the previous research projects in this study.
 The final objective is to assist the doctors in the detection of brain tumor at an
early stage by providing improved and reliable results

2 Proposed Solution

2.1 Project Functionality

 Online Dataset
 Preprocessing
 Feature extraction
 Image classification
 Evaluation

2.2 Users and Characteristics

This system will help in the early diagnosis and treatment of tumor just by looking at the image
using software rather than paying heavy exposure on x rays and other examination diagnosis.
The purpose of this research is to find improved and more effective ways to detect brain tumor
using digital image processing and machine learning techniques.
 This system will help the dermatologist in early identification of Tumor.
 The dermatologist can identify by just looking at the patient only with naked eye.
 Previous researches showed that this process id time saving.
 Only expert can use it

2.3 Operating Environment


Software: Python 3.10.0
System Type: 64-bit operating system, x64-based processor
OS Build: 19043.1348

iii
Processor: Intel(R) Core(TM) i5-5300U CPU @ 2.30GHz 2.30 GHz
RAM: 8.00 GB
Edition: Windows 11 Pro
Core i5
Tools and Technology

Tools Version Rational

Vs Code 1.71 Application Tool

Spyder 5.1 Ide

Python 3.7 Programming


Language

MS Word 2018 Documentation

MS Power Point 2018 Presentation

Label Img 1.8 Annotate

3 Specific Requirements

3.1 Functional Requirements

Skin cancer detection using SVM is basically defined as the process of detecting the presence of
cancerous cells in image. Skin cancer detection is implemented by using GLCM and Support
Vector Machine (SVM). Gray Level Co-occurrence Matrix (GLCM) is used to extract features
from an image that can be used for classification. SVM is machine learning technique, mainly
used for classification and regression analysis.

 Online Dataset
Approximately 23907 images are collected from Kaggle. Kaggle is an online community
platform for data scientists and machine learning enthusiasts. Kaggle allows users to

iv
collaborate with other users, find and publish datasets, use GPU integrated notebooks,
and compete with other data scientists to solve data science challenges.

Dataset:
https://www.kaggle.com/code/jaykumar1607/brain-tumor-mri-classification-tensorflow-cnn

Preprocessing
Goal of pre-processing is an improvement of image data that reduces unwanted
distortions and enhances some image features important for further image processing.
Image pre-processing involves three main things
1) Gray scale conversion
2) Noise removal
3) Image enhancement. Feeding the preprocessed data to convolutional neural network
(CNN).
Three types of layers are present in a convolutional Neural Network. That are given in
following part-
• Convolution layer
• Pooling layer
• Fully connected layer
 Feature extraction
Feature extraction plays an important role in extracting information present in given
image. Here we are using GLCM for texture image analysis. GLCM is used to capture
spatial dependency between image pixels. GLCM works on gray level image matrix to
capture most common feature such as contrast, mean, energy, homogeneity the purpose
of feature extraction (GLCM) is to suppress the original image data set by measuring
certain values or features that helps to classify different images from one another. The
system was trained using the SVM classifier. Main reason for selecting SVM is that is
empirically proved that SVM outperforms among different machine learning techniques
for skin cancer segmentation

v
 Image classification
Classifier is used to classify cancerous image from other brain diseases. For simplicity
Support Vector machine classifier is used here. SVM takes set of images and predicts for
each input image belongs to which of the two categories of cancerous and non-cancerous
classes. The purpose of SVM is create hyper plane that separates two classes with
maximum gap between them. In our proposed system output of GLCM is given as input
to SVM classifier which takes training data, testing data and grouping information which
classifies whether given input image is cancerous or non-cancerous All images are
without any distracting elements like jewelry, clothes and any background. For each
image the region of interest was manually selected.

4 Non-Functional Requirements

4.1 Performance Requirements

 Each image must load within 2 seconds.


 The process must finish within 3 hours so data is available by 8 a.m. local time after an overnight
update.
 Users shall be prompted to provide an electronic signature before loading a new page.

4.2 Safety and Security Requirements

The safety requirements are those requirements that are defined for the purpose of risk reduction. Like
any other requirements, they may at first be specified at a high level,
This non-functional requirement assures that all data inside the system or its part will be
protected against malware attacks or unauthorized access. But there’s a catch. The lion’s share of
security non-functional requirements can be translated into concrete functional counterparts. If
you want to protect the admin panel from unauthorized access, you would define the login flow
and different user roles as system behavior or user actions. So, the non-functional requirements
part will set up specific types of threats that functional requirements will address in more detail.
But this isn’t always the case. If your security relies on specific standards and encryption

vi
methods, these standards don’t directly describe the behavior of a system, but rather help
engineers with implementation guides.

4.3 Software Quality Attributes

Reliability:

 Reliability of a software system derives from

 Correctness
 Availability

The behavior over time for the fulfillment of a given specification depends on the reliability of
the software system.

Reliability of a software system is defined as the probability that this system fulfills a function
(determined by the specifications) for a specified number of input trials under specified input
conditions in a specified time interval (assuming that hardware and input are free of errors).

Maintainability:

 Maintainability = suitability for debugging (localization and correction of errors) and for
modification and extension of functionality.

The maintainability of a software system depends on its:

 Readability
 Extensibility
 Testability

The portability of a software system depends on:

 Degree of hardware independence


 Implementation language
 Extent of exploitation of specialized system functions
 Hardware properties
 Structured:

vii
System-dependent elements are collected in easily interchangeable program components.

Gant Chart (Timeline)

S.NO Task Dur

November December January February March April

1 Requirement 1 Month
Analysis

2 Design Half
month

3 Programming 2 Month

4 Testing 2 Month
Presentation

5 Implementation 10 Days

6 Documentation 20 days

Project close 2 days

viii

You might also like