Scribd
Upload
52 views

Machine Learning Approach-Based Gamma Distribution For Brain Tumour Detection and Data Sample Imbalance Analysis

Paper presentation on Machine learning approach-based gamma distribution for brain tumour detection and data sample imbalance analysis .

Uploaded by

kstorm11
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
52 views

Machine Learning Approach-Based Gamma Distribution For Brain Tumour Detection and Data Sample Imbalance Analysis

Paper presentation on Machine learning approach-based gamma distribution for brain tumour detection and data sample imbalance analysis .

Uploaded by

kstorm11
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 17

MACHINE LEARNING APPROACH-

BASED GAMMA DISTRIBUTION FOR


BRAIN TUMOUR DETECTION AND
DATA SAMPLE IMBALANCE ANALYSIS
TABLE OF CONTENTS

• ABSTRACT
• INTRODUCTION
• IMPORTANT TERMS
• IDEOLOGIES
• METHODS TO DETECT TUMOR
• COMPARISION OF METHODS
• CONCLUSION
ABSTRACT
• Brain tumor extraction and its analysis are challenging tasks in medical image processing
because brain image is complicated, Hence Artificial intelligence applications in magnetic
resonance imaging(MRI) have been applied in this clinical study.
• The analysis of brain tumors without human intervention is considered a significant area of
research because the extracted brain images need to be optimized using a segmentation
algorithm which automatically detects the region of interest (ROI).
• Here we use orthogonal gamma distribution-based machine-learning approach to analyze the
under- segments and over-segments of brain tumor regions to automatically detect
abnormalities in the ROI.
• The system is evaluated based on experimental results that showed the method of orthogonal
gamma distribution with the machine learning approach attained an accuracy of 99.55% in
detecting brain tumors.
INTRODUCTION
• The brain is the most important part of the central nervous system. Brain tumor is an intracranial
solid neoplasm. The only optimal solution for this is use of ‘Image Segmentation’.
• Early recognition of a brain tumor is an imperative factor in tumor treatment.
• Magnetic resonance imaging (MRI) is one of the medical imaging procedures utilized by doctors
to improve the accuracy of diagnoses [4]. This tumor detection methodology utilizes unaffected
radiation. X-rays are also utilized to analyze images.
• Brain abnormalities, such as injuries, damage, tumor related causes, affects, and symptoms, have
been analyzed for tumor recognition by using image processing, data mining, and machine
learning techniques.
• Abnormalities are analysed using image cryptography, computed tomography (CT), MRI, and
electroencephalogram (EEG) related data. These techniques capture information data in an
effective manner because they analyse every lobe in the brain.
IMPORTANT
TERMS
• Magnetic resonance imaging - Magnetic resonance imaging (MRI) is a medical imaging
technique used in radiology to form pictures of the anatomy and the physiological processes of the body.
MRI scanners use strong magnetic fields, magnetic field gradients, and radio waves to generate images of the
organs in the body.

• Gamma distribution - Gamma Distribution is a Continuous Probability Distribution that is widely


used in different fields of science to model continuous variables that are always positive and have skewed
distributions.

• brain tumor- A brain tumor is a collection, or mass, of abnormal cells in your brain. Your skull, which
encloses your brain, is very rigid. Any growth inside such a restricted space can cause problems. Brain
tumors can be cancerous (malignant) or noncancerous (benign).

• Machine learning- Machine learning is an application of artificial intelligence (AI) that provides
systems the ability to automatically learn and improve from experience without being explicitly
programmed. Machine learning focuses on the development of computer programs that can access data
and use it learn for themselves.
IDEOLOGIES USED
• Here, MRI segmentation is performed based on different algorithms and threshold methods.
• The segmentation method is used to conduct a qualitative analysis of the brain area for the
separation of tumor cells having high levels of sensitivity.
• Image processing is used in the medical field to detect abnormal cells in the body. here, the
method of fluid attenuated inversion recovery (FLAIR) is applied with MRI for the automatic
detection and prediction of tumor cells and normal cells in the brain.
• Automatic ROI (Region Of Interest) detection is considered a significant area of research in
brain tumor detection and analysis
• The best pixel technique and the classification of each pixel method is used.
• Coordinate matching using orthogonal gamma distribution and edge enhancement with
identifications are computed using the machine learning approach.
METHODS TO DETECT
TUMOR
OTS’S, LI’S, AND CHECHADE’S METHOD

L’s method, Chehade’s method, and Ots’s method, have drawbacks.


In Otsu’s method, L’s method, and Chechad’s method, the optimal threshold limit is not
calculated. Therefore, this algorithm (Ots’s, Li’s, and Chechade’s methods) is not suitable for
segmenting images for the accurate diagnosis of brain tumor edge.
ORTHOGONAL GAMMA DISTRIBUTION WITH MACHINE LEARNING
APPROACH (OGDMLA)

• The edge-based image segmentation coordinate matching with automatic ROI detection
is implemented using an orthogonal gamma distribution model with a machine learning
approach. Because this algorithm self-identifies the region of interest (ROI).
• This approach uses edge analysis and machine learning approach to identify and train the
edge coordinates. Initially, the fractional derivatives are analysed for the x and y axis of
the gray scale image, as shown in Eqs. (4) and (5).
ALGORITHM

Input (i): Brian tumor image of (Rx C)


//∗ R is the Row and ∗ C is the Column//
Output (O): Enhanced brain tumor image with proper black pixel reduction;
Begin
For i = 0 to ((R-n)-1)
For j = 0 to (C-n)-1)
where n = 0 to 255;
Source get (MRI data set)
If (variance > min variance)
Min variance ≠ variance;
Else Min variance = variance;
Go to Threshold check (for the values 0 to 255)
Ex(img) = [] = {Tu} τ (img){Tu};
//∗ Extract (Ex) the over and under segmentation region from the input tumor (Tu) image (img)∗// Return
(threshold value);
End
COMPARISION OF THE METHODS
MSE AND PSNR
MSE: Mean Square Error rate

PSMR: Peak signal-to-noise ratio


ratio between the maximum possible
power of a signal and the power of
corrupting noise.
THRESHOLD LIMIT
To perform image division, thresholding is the most straightforward technique
used to diagnose several brain tumors.
We need to estimate the optimal threshold to remove the black pixels present in
the background images which would give us the clear tumor image.
ACCURACY
CONCLUSION

• Accurate brain tumor analysis plays a vital role in the health care sector
• . The proposed technique is completely automated in identifying tumor images based on
training the edge-based image segmentation coordinates using orthogonal gamma
distribution with the machine learning approach.
• The significant features of OGDMLA are the self-identification of ROI with an enhanced
imaging segmentation approach.
• The proposed OGDMLA has potential for the field of real-time medical image diagnosis
in the health care sector.
• Further research will be carried out to accelerate realtime medical applications and
computation using machine learning techniques in the medical internet of things (MIoT).
THANK YOU

You might also like