International Journal of Trend in Scientific Research and Development (IJTSRD)

Special Issue: International Conference on Advances in Engineering, Science and Technology – 2021
Organized by: Uttaranchal Institute of Technology, Uttaranchal University, Dehradun
Available Online: www.ijtsrd.com e-ISSN: 2456 – 6470

Brain Tumor Detection using MRI Images

Deepa Dangwal1, Aditya Nautiyal1, Dakshita Adhikari1, Kapil Joshi2
1B.Tech
Student, 2Assistant Professor,
1,2UIT, Uttaranchal University, Dehradun, Uttarakhand, India

ABSTRACT How to cite this paper: Deepa Dangwal |

Brain tumor segmentation is a very important task in medical image Aditya Nautiyal | Dakshita Adhikari | Kapil
processing. Early diagnosis of brain tumors plays a crucial role in improving Joshi "Brain Tumor Detection using MRI
treatment possibilities and increases the survival rate of the patients. For the Images" Published in International
study of tumor detection and segmentation, MRI Images are very useful in Journal of Trend in Scientific Research
recent years. One of the foremost crucial tasks in any brain tumor detection and Development
system is that the detachment of abnormal tissues from normal brain tissues. (ijtsrd), ISSN: 2456-
Because of MRI Images, we will detect the brain tumor. Detection of unusual 6470, Special Issue |
growth of tissues and blocks of blood within the system is seen in an MRI International
Imaging. Brain tumor detection using MRI images may be a challenging task Conference on
due to the brain's complex structure. Advances in
Engineering, Science IJTSRD42456
In this paper, we propose an image segmentation method to detect tumors
and Technology –
from MRI images using an interface of GUI in MATLAB. The method of
2021, May 2021, pp.1-4, URL:
distinguishing brain tumors through MRI images is often sorted into four
www.ijtsrd.com/papers/ijtsrd42456.pdf
sections of image processing as pre-processing, feature extraction, image
segmentation, and image classification. During this paper, we've used various
Copyright © 2021 by author(s) and
algorithms for the partial fulfillment of the necessities to hit the simplest
International Journal of Trend in Scientific
results that may help us to detect brain tumors within the early stage.
Research and Development Journal. This
KEYWORDS: image processing; deep learning; brain tumor segmentation; MRI is an Open Access article distributed
under the terms of
the Creative
Commons Attribution
License (CC BY 4.0)
(http://creativecommons.org/licenses/by/4.0)

1. INTRODUCTION
Because brain tumors are the most common type of cancer Now that we have a scanned image of the brain, we must
in humans, research into them is crucial. precisely identify the tumor, its size, and its position. The
It is the primary cause of the rise in child and adult death rat Neurosurgeon will need all of this information to complete
es. Headaches, nausea, vomiting, personality and behavioral his diagnosis. This is where CIP (Computerized Image
changes, memory loss, sensory disruption, weakness, and Processing) can assist. We can accurately detect the tumor
numbness is among the signs of brain tumors. using various segmentation approaches and feature
extraction methods.
The term "tumor" refers to abnormal tissue growth. Brain
tumors are divided into two categories: benign and Tumors can damage all healthy brain cells directly, and they
malignant tumors. In the case of a brain tumor, MRI imaging can also harm healthy cells indirectly by crowding other
is crucial for analysis, diagnosis, and therapy planning. When sections of the brain. The proposed approaches are applied
compared to other commonly used X-ray scanners, the to an MRI scan of the human brain, which serves as the
images produced by an MRI scan are better at exhibiting system's input images or original images. At the same time
small details and so have a greater diagnostic quality. as histogram equalization methods are being processed, the
RGB input photos will be converted to grayscale in the pre-
In this paper, we will primarily focus on improving existing
processing stage.
image processing algorithms or designing a better strategy
for tumor diagnosis with a reliable graphical user interface.

@ IJTSRD | Unique Paper ID – IJTSRD42456 | ICAEST-21 | May 2021 Page 1

 Special Issue: International Conference on Advances in Engineering, Science and Technology – 2021 (ICAEST-21)
Available online @ www.ijtsrd.com eISSN: 2456-6470
2. Methodology:

Data Preproces Feature

Collection sing extraction

Kaggle Dataset Input Image

Grayscale, Anisotropic
diffusion, Image
enhancement

Benign
Yes Image
Classification Tumor Segmentation
found?
Malignant

No

SVM, ANN, SOM, Threshold segmentation

Fuzzy, PNN, MRF, Watershed segmentation
RF Clustering K-mean
Clustering

2.1. Preprocessing:
The pre-processing stage improves an image by removing converting an RGB image to a grayscale image, only 8 bits are
undesired distortions and showing observable areas of the required to save a single pixel of the image. As a result,
image. The input image is converted to grayscale, which storing a grayscale image will take up 33% less memory than
improves image quality. It's a catch-all term for operations storing an RGB image.
on images at the most basic level of abstraction. To increase
Grayscale images are far more convenient to work within a
the detection of the problematic region from an MRI,
variety of situations, like as It is easier to deal with a single-
preprocessing techniques are applied.
layered image (Grayscale image) than a three-layered image
Before we begin processing our image, we must ensure that in many morphological processes and image segmentation
it is free of extraneous data and in the correct format for problems (RGB color image ) because it is easier to detect
processing in order for the results to be accurate. Pre- features in a grayscale image rather than MRI images
processing is the term used to describe this procedure.
2. Anisotropic diffusion
Conversion to grayscale, noise reduction and noise removal, Anisotropic diffusion reduces visual noise without deleting
picture reconstruction, image augmentation, and, in the case substantial portions of the image information, such as edges,
of medical pictures, steps such as skull removal from an MRI lines, or other crucial elements for visual interpretation.
are all examples of pre-processing.
3. Image enhancement
Converting a picture to grayscale is one of the most used pre- Before processing, image enhancement is the process of
processing techniques. A grayscale image is frequently increasing the quality and information content of raw data.
mistaken for a black and white image, but this is not the case. Contrast enhancement, density slicing, spatial filtering, are
Because a black and white image has only two hues, black all common techniques for image enhancement
and white, the intensity can be either 1 or 0. A grayscale
image, on the other hand, is made up of shades of grey with
no apparent color.
This means that without revealing any color, each pixel
indicates the intensity value at that pixel. A grayscale image,
unlike a black and white image, has a variety of colors, with
white being the lightest and black being the deepest. The
intensity values of a pixel in a grayscale image are also not
absolute and can be infractions.
So basically we are dividing preprocessing stage into 3 main 2.2. Feature extraction:
steps: The most important part of image processing is feature
1. Converting MRI images to grayscale image: extraction. It is done for reducing the original dataset by
We are converting MRI images into grayscale for the measuring and extracting certain features. Feature
following reasons - extraction starts from an initial set of data and builds
An RGB color image requires 8*3 = 24 bits to store a single features that are informative and non-redundant.
color pixel (8 bits for each color component); however, when

@ IJTSRD | Unique Paper ID – IJTSRD42456 | ICAEST-21 | May 2021 Page 2

 Special Issue: International Conference on Advances in Engineering, Science and Technology – 2021 (ICAEST-21)
Available online @ www.ijtsrd.com eISSN: 2456-6470
It entails reducing the number of resources required to 2.4. Classification:
accurately describe a huge quantity of data. When The tumor is additionally classified in two: malignant and
performing an analysis of complex data the major problems benign. Benign tumors are non-cancerous cells that do not
stem from a number of variables involved. When input data infiltrate healthy tissue around them. The growth is very
to an algorithm is too large to be processed and data is slow and the periphery or edge of the benign tumor can be
redundant, feature extraction is used. clearly visualized. Benign brain tumors are sometimes life-
threatening when they press the sensitive region of the
While handling images, feature extraction is done, if the
brain. Malignant tumors are cancerous cells, which do invade
feature dimension is high, then feature selection or feature
surrounding healthy tissue and spread to another part of the
transformation is done where high-quality discriminant
brain or spine. This tumor grows very faster and more fatal
features are extracted.
than a benign tumor. In order to treat tumors in the brain, a
In this stage, the input image is transformed into squashed classification technique is used which can easily detect the
form, it extracts features. These features include contrast, type of tumor.
energy, entropy, smoothness, variance, standard deviation,
Based on the existing papers on brain tumor classification,
etc which are useful in the classification stage. The output
various techniques can be used for classification such as
from the feature extraction stage can be used as an input to
SVM, ANN, SOM, Fuzzy, PNN, MRF, RF, and CNN. In this
the classification stage.
paper, the SVM technique is used. The classification method
The existence of tumors can be detected using features. categorizes all pixels in a digital image. SVM is a supervised
These features are analyzed and detection of tumor region is learning method that is used for data analysis and
done. This is how the tumor region is extracted in the MRI classification. It operates based on a decision plane,
Image. When features for tumor and non-tumor MRI images separating things with distinct class properties.
are extracted, they are used to train the different algorithms.
This technique can be used for very large data as well due to
2.3. Image segmentation: its fast learning speed. The classification of brain tumors is
The segmentation is the most important stage for analyzing used to identify the tumor class present in the MRI image. It
images properly since it affects the accuracy of the has two basic steps of training and testing. After this phase,
subsequent steps. Pathologists use hematoxylin and eosin to accurate tumor type is found in the input image. Because of
stain body tissue during cancer diagnosis to distinguish its high generalization, the Support Vector Machine (SVM)
between tissue types. As a result, we apply image approach is considered a good candidate performance,
segmentation to identify the different tissue types in their especially when the feature space's dimension is very large.
images. Converting an image into a collection of pixels
represented by a mask or a labeled image is the process of
image segmentation. By dividing an image into segments, we
can identify the important segments of the image. However,
proper segmentation of images is difficult because of the
great varieties of lesion shapes, sizes, and colors along with
different skin types and textures. In addition to this, some
lesions have irregular boundaries and in some cases, there is
a smooth transition between the lesion and the skin so this Normal Image Benign tumor Malignant tumor
creates problems while doing image processing. To solve
these problems, several techniques have been developed to 3. Conclusion & Future Scope:
overcome this problem. In the medical field, manual identification of brain tumors by
doctors referring to the MRI images is a very time-
They can be broadly classified as thresholding, region-based, consuming task and can be inappropriate for a large amount
supervised, and unsupervised classification techniques:- of data. Instead of manual identification, image processing
Threshold segmentation can be used to identify the tumor from the images.
Therefore, this model helps in understanding the creation of
Watershed segmentation a system that will carry out image processing and identify
Gradient Vector Flow (GVF) the brain tumor MATLAB tools. And also show the location of
the tumor in the highlighted area of the image.
Clustering
In near future, a database can be created for different
K-mean Clustering patients having different types of brain tumors and locate
them. Tumor growth can be analyzed by plotting a graph
which can be obtained by studying sequential images of
tumor effect. A possible extension of the presented work
could use more features. Connecting the system to the
hospital's cloud storage of patient data would be helpful.
This application can be enhanced to provide mobile phone
accessibility and usability.
If this application is developed to analyze all types of MRI
scans of the same patient and the results of all scans are
integrated, it can suggest appropriate treatment and
medication as well.

@ IJTSRD | Unique Paper ID – IJTSRD42456 | ICAEST-21 | May 2021 Page 3

 Special Issue: International Conference on Advances in Engineering, Science and Technology – 2021 (ICAEST-21)
Available online @ www.ijtsrd.com eISSN: 2456-6470
4. References Tumor Detection and Feature Extraction Using
[1] Ms. Priya Patil, Ms. Seema Pawar, Ms. Sunayna Patil, Biologically Inspired BWT and SVM”, International
Prof. Arjun Nichal, “ A Review Paper on Brain Tumor Journal of Biomedical Imaging Volume 2017
Segmentation and Detection ”, International Journal of
[11] M. Havaei, A. Davy, D. Warde-Farley, A. Biard, A.
Innovative Research in Electrical, Electronics,
Courville, Y. Bengio, C. Pal, P.-M. Jodoin, and H.
Instrumentation and Control Engineering, Vol. 5,
Larochelle, “Brain tumor segmentation with deep
Issue 1, January 2017.
neural networks,” Medical image analysis, vol. 35, pp.
[2] Luxit Kapoor, Sanjeev Thakur, Professor, 2017 his 18-31, 2017.
paper “A Survey on Brain Tumor Detection Using
[12] H. Dong, G. Yang, F. Liu, Y. Mo, and Y. Guo, “Automatic
Image Processing Techniques", International
Brain Tumor Detection and Segmentation Using U-Net
Conference on Cloud Computing, Data Science &
Based Fully Convolutional Networks,” arXiv preprint
Engineering – Confluence 2017
arXiv: 1705.03820, 2017.
[3] Prabhjot Kaur Chahal1 · Shreelekha Pandey1 · Shivani
[13] Sehgal A, Goel S, Mangipudi P, Mehra A, Tyagi D,”
Goel2, “A survey on brain tumor detection techniques
Automatic brain tumor segmentation and extraction
for MR images”, © Springer Science+Business Media,
in MR images”, Conference on Advances in Signal
LLC, part of Springer Nature 2020, 22 March 2019 /
Processing (CASP) June 2016.
Revised: 11 February 2020 / Accepted: 27 March
2020. [14] Xiao Z, Huang R, Ding Y, Lan T, Dong R, Qin Z,”A deep
learning-based segmentation method for brain tumor
[4] Ali IúÕna, Cem Direko lub, Melike ùahc, “Review of
in MR images”, IEEE 6th International Conference on
MRI-based brain tumor image segmentation using
Computational Advances in Bio and Medical Sciences
deep learning methods”, 12th International
(ICCABS) 2016.
Conference on Application of Fuzzy Systems and Soft
Computing, ICAFS 2016, 29-30 August 2016 [15] Mohsen H, El-Dahshan ESA, El-Horbaty ESM, Salem
ABM,” Classification using deep learning neural
[5] J. Amin, M. Sharif, M. Yasmin, S.L. Fernandes,” Big data
networks for brain tumors”, Future Compututing and
analysis for brain tumor detection: Deep
information journal 3(2018).
convolutional neural networks”, Future Generation
Computer Systems (2018) [16] Joshi, K., Gupta, H., Chaudhary, P., & Sharma, P. Survey
on different Enhanced k-means Clustering Algorithm.
[6] Mahmoud Khaled Abd-Ellaha, Ali Ismail Awadb,c,,
International Journal of Engineering Trends and
Ashraf A.M. Khalafd, Hesham F.A. Hamedd,” A review
Technology (IJETT)–Volume, 27.
on brain tumor diagnosis from MRI images: Practical
implications, key achievements, and lessons learned”, [17] Chauhan, N., Dhaundiyal, R., & Joshi, K. K-MEANS ON
Magnetic Resonance Imaging 61(2019) SEARCH ENGINE DATASET THROUGH WEKA.
International Journal of Research Fellow for
[7] Sethuram Rao.G, Vydeki.D, “Brain Tumor Detection
Engineering (IJRFE)–Volume, 4.
Approaches: A Review”, International Conference on
Smart Systems and Inventive Technology (ICSSIT [18] Joshi, K., Rawat, S., & Chaudhary, S. ANALYSIS OF
2018) DIFFERENT OPTICAL SWITCHING TECHNIQUES IN
NOC ROUTER ARCHITECTURE. International Journal
[8] Kanmani, M. and Dr. Pushparani, M.,” . “Brain tumor
of Research Fellow for Engineering (IJRFE)–Volume, 4.
detection and classification”, International Journal of
Current Research, 8, (05), 31634-31637.0 [19] Joshi, K., Chaudhary, S., & Chauhan, N. HYBRID
CLUSTERING ALGORITHM USING K-MEANS
[9] Swapnil R. Telrandhe, Amit Pimpalkar, Ankita
CLUSTERING ALGORITHM. International Journal of
Kendhe,” Detection of Brain Tumor from MRI images
Research Fellow for Engineering (IJRFE)–Volume, 4.
by using Segmentation &SVM”, 2016 World
Conference on Futuristic Trends in Research and [20] Longkumer, M., Joshi, K. A Comprehensive Study on
Innovation for Social Welfare (WCFTR’16) Recent Botnet. International Journal of Science and
Research (IJSR)- Volume, 7.
[10] Nilesh Bhaskarrao Bahadure, Arun Kumar Ray,and
Har Pal Thethi,” Image Analysis for MRI Based Brain

@ IJTSRD | Unique Paper ID – IJTSRD42456 | ICAEST-21 | May 2021 Page 4