Brain Tumor Detection

Using
Deep learning Techniques

Under the Esteemed Guidance of: Presented by:

R Poojitha - 206W1A4356
B.Susmitha M.Tech M NandiniDevi - 206W1A4343
Assistant Professor N Kavya - 206W1A4345
CH Gayatri - 206W1A4314
 Agenda

 Introduction
 Abstract
 Existing System
 Disadvantages of Existing System
 Proposed System
 Advantages of Proposed System
 Architecture
 Software Requirement Specification
 Hardware Requirement Specification
 Conclusion
Abstract
Now a day’s tumor is second leading cause of cancer. Due to cancer large no of patients are in
danger. The medical field needs fast, automated, efficient and reliable technique to detect tumor
like brain tumor. Detection plays very important role in treatment. If proper detection of tumor is
possible then doctors keep a patient out of danger. Various image processing techniques are used in
this application. Using this application doctors provide proper treatment and save a number of
tumor patients.tumor is nothing but excess cells growing in an uncontrolled manner. Brain tumor
cells grow in a way that they eventually take up all the nutrients meant for the healthy cells and
tissues, which results in brain failure. Currently, doctors locate the position and the area of brain
tumor by looking at the MR Images of the brain of the patient manually. This results in inaccurate
detection of the tumor and is considered very time consuming. A tumor is a mass of tissue it grows
out of control. We can use a Deep Learning architectures CNN (Convolution Neural Network)
generally known as NN (Neural Network) for detect the brain tumor. The performance of model is
predict image tumor is present or not in image. If the tumor is present it return yes otherwise return
no.
Introduction
o The human body is made up of many organs and brain is the most critical and vital organ of
them all. One of the common reasons for dysfunction of brain is brain tumor.
o A tumor is nothing but excess cells growing in an uncontrolled manner. Brain tumor cells grow
in a way that they eventually take up all the nutrients meant for the healthy cells and tissues,
which results in brain failure.
o Currently, doctors locate the position and the area of brain tumor by looking at the MR Images
of the brain of the patient manually. This results in inaccurate detection of the tumor and is
considered very time consuming.
o A Brain Cancer is very critical disease which causes deaths of many individuals. The brain
tumor detection and classification system is available so that it can be diagnosed at early stages.
Cancer classification is the most challenging tasks in clinical diagnosis.
o This project deals with such a system, which uses computer, based procedures to detect tumor
blocks and classify the type of tumor using Convolution Neural Network Algorithm for MRI
images of different patients.
.
o Different types of image processing techniques
like image segmentation, image enhancement
and feature extraction are used for the brain
tumor detection in the MRI images of the
cancer-affected patients.
o Detecting Brain tumor using Image Processing
techniques its involves the four stages is Image
Pre-Processing, Image segmentation, Feature
Extraction, and Classification.
o Image processing and neural network
techniques are used for improve the
performance of detecting and classifying brain
tumor in MRI images
Existing System

o In the first stage, there is a computer based

procedures to detect tumor blocks and classify the
type of tumor using Artificial Neural Network
Algorithm for MRI images of different patients.

o The second stage involves the use of different

image processing techniques such as histogram
equalization, image segmentation, image
enhancement, morphological operations and
feature extraction are used for brain tumor
detection in the MRI images for the cancer-
affected patients.

o This work is introduced one automatic brain

tumor detection method to increase the accuracy
and decrease the diagnosis time.
7
Disadvantages of Existing System
o Manual Intervention in Feature Extraction: Existing system relies on manual
feature extraction, time-consuming and may not capture all relevant tumor
characteristics accurately.
o Limited Adaptability to Complex Tumor Patterns: SVM-based classification may
struggle to adapt to complex tumor patterns, impacting detection accuracy.
o Dependency on Predefined Features: SVMs require predefined features for
classification, may not fully represent the variability of brain tumors, leading to
suboptimal performance.
o Complexity of Workflow: Existing system involves multiple steps such as
preprocessing, skull stripping, segmentation, morphological operations, and feature
extraction, increasing the complexity and computational requirements.
o Limited Performance with Imbalanced Data: SVMs may struggle with imbalanced
datasets, leading to biased classification and reduced sensitivity to tumors, especially in
scenarios where tumor samples are sparse.

Sample Footer Text 20XX

Proposed
8
System
o The proposed system has mainly five modules. Dataset, Pre-
processing, Split the data, Build CNN model train Deep
Neural network for epochs, and classification.

o In dataset we can take multiple MRI images and take one as

input image. In pre-processing image to encoded the label and
resize the image. In split the data we set the image as 80%
Training Data and 20% Testing Data.

o Then build CNN model train deep neural network for epochs.
Then classified the image as yes or no if tumor is positive then
it returns yes and the tumor is negative the it returns no.

Sample Footer Text 20XX

Advantages of Proposed System
9
o Automatic Feature Learning: CNN models automatically learn discriminative features from
raw data, reducing the need for manual feature extraction and potentially improving detection
accuracy.
o Adaptability to Complex Patterns: CNN models can learn hierarchical representations of
data, enabling better adaptation to complex tumor patterns and variations in medical images.
o Comprehensive Representation of Tumor Characteristics: CNNs learn features directly
from raw data, allowing for more comprehensive representation of tumor characteristics and
potentially enhancing classification performance.
o Simplified Workflow: Proposed system streamlines the workflow by directly training a CNN
model on preprocessed data, reducing the number of intermediate steps and simplifying
implementation.
o Enhanced Performance with Imbalanced Data: CNN models are inherently robust to class
imbalances, potentially leading to improved performance in detecting tumors, even in
scenarios with limited tumor samples.

Sample Footer Text 20XX

System Architecture
11
Modules of Proposed System
o Image Pre-Processing Pre Processing

o Image Segmentation
Segmentation
o Image Classification

Classification

Sample Footer Text 20XX

Image Preprocessing
Image preprocessing refers to the techniques used to enhance the quality of input
images before feeding them into the brain tumor detection system.
Preprocessing Techniques Used:

o Filtering: Gaussian and bilateral filters are applied to reduce noise and preserve
edges.

o Color Space Conversion: Conversion to grayscale simplifies processing and

reduces computational complexity.

o Clustering: KMeans clustering segments images into tumor and non-tumor

regions for better feature extraction.
13

Sample Footer Text 20XX

Image Segmentation
Image segmentation is the process of partitioning an image into multiple segments to simplify its
representation and make it more meaningful for analysis.

o K-Means Clustering:
KMeans clustering is applied to group pixels in the image into clusters based on their
similarity in intensity values.
o Feature Extraction:
Extracting features such as texture, intensity, and spatial information helps in differentiating
between tumor and non-tumor regions.
o Thresholding:
Thresholding techniques are used to convert the segmented image into a binary mask, where
pixels belonging to the tumor are represented as foreground (tumor) and the rest as
background.
15

Sample Footer Text 20XX

16

Sample Footer Text 20XX

17

Image Classification
o Convolutional Layers:
o Extract features from tumor images.
o Find patterns like edges and textures.
o Pooling Layers:
o Shrinks feature maps to focus on important details.
o Helps reduce computation and speed up processing.
o Dense Layers:
o Processes flattened features.
o Makes predictions based on learned patterns.

Sample Footer Text 20XX

18

Sample Footer Text 20XX

 Output Screen’s
19

Sample Footer Text 20XX

20

Sample Footer Text 20XX

21

Sample Footer Text 20XX

Software Requirements
22
1.Functional Requirement
1. Python enabled system
2. Support for all given libraries
2.Software Requirement
1. Python 3.7
2. Anaconda Navigator
3. Google Colob
4. Spyder
3.Libraries used:
1. Keras
2. Scikit-Learn
3. Numpy 1.16.4
4. Tensorflow 1.13.1
5. Keras.models.load_model
6. OpenCV 3.4.2

Sample Footer Text 20XX

23
Hardware Requirements

o Processor – 400 MHz or above

o RAM – 8 GB or above

o Storage : High Performance SSD

Sample Footer Text 20XX

Conclusion

o The Input Database has been taken out to the preprocessing and images has been
filtered and noise has been removed.

o Image Segmentation has to be carried out by making use of cluster technique

called K-means, then followed by Image Classification where the MRI scanned
images are classified using CNN algorithm to obtain more accuracy when
compared to other algorithms and finally it shows the kind of tumor after
performing Classification.

o We can enhance work in future, where we can get the accurate tumor percentage
and in detecting tumor to which the input data been affected
Thank you