[go: up one dir, main page]
Scribd
Upload
79 views17 pages

B9 - Throat Cancer Detection Using Image Processing

This document describes a proposed system for throat cancer detection using image processing. The system uses MRI images of the throat and segments the images to detect cancerous cells. It then uses neural networks to classify the cancer cells. The system aims to more accurately detect cancer at early stages compared to existing biopsy and watershed algorithm methods. Key steps include preprocessing images, extracting features using discrete wavelet transform, segmentation, and cancer detection classification. The system is designed to help doctors more efficiently diagnose and treat throat cancer.

Uploaded by

Ramesh Daravath
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
79 views17 pages

B9 - Throat Cancer Detection Using Image Processing

This document describes a proposed system for throat cancer detection using image processing. The system uses MRI images of the throat and segments the images to detect cancerous cells. It then uses neural networks to classify the cancer cells. The system aims to more accurately detect cancer at early stages compared to existing biopsy and watershed algorithm methods. Key steps include preprocessing images, extracting features using discrete wavelet transform, segmentation, and cancer detection classification. The system is designed to help doctors more efficiently diagnose and treat throat cancer.

Uploaded by

Ramesh Daravath
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

THROAT CANCER DETECTION

USING
IMAGE PROCESSING
PRESENTED BY
              Ajinkya        221710402002                  Deepika   221710402017
              Hari Gopal  221710402036                  Shashank 221710402054
GITAM SCHOOL OF TECHNOLOGY (DEPARTMENT OF ECE)
Under the guidance of Ms. Arunjyothi E 
CONTENTS

 Introduction
 Existing System
 Proposed System
 System Requirements
 Block Diagram
 Modules
 Output Screens
 Advantages
 Applications
 Conclusion
INTRODUCTION
 Cancer occurs in salivary glands, tonsils and also in neck, head, face and throat cavity.
 There are various diagnosis methods to find the throat cancer such as biopsy method and
screening methods.
 In biopsy method a small sample of tissues being removed from a part of body and tested
in microscope.
 By this we couldn’t classify how much cells are affected by cancer, so we are going to detect
and classify the affected cancerous cell in the throat region by digital Image processing techniques.
Screening Methods:
Ø Most of the cancer cells in Throat region are low
contrast w.r.t the other surrounding
cells. These abnormalities in throat image are identified by
scanning the throat regions using
Magnetic Resonance Imaging (MRI) technique. 
Ø In current medical scanning method, MRI
scanning procedure is superior to Computer
Tomography (CT) due to its high sensitivity and high
contrast with respect to various intensity tissues in
throat image.
Here we use,
Ø Segmentation to detect the type of cancer in the MRI image 
Ø Neural Networks to classify cancer cells by using MATLAB
Existing System

 The existing system of Oral cancer detection using tumour based marker using
watershed algorithm.
 The watershed algorithm is based on the concept of visualizing an image as a
topographic surface where high-intensity values denote peaks and hills while the
low intensity denotes valleys.
 Watershed algorithm is proposed to preserve the image details as well as
prominent ones to identify tumors in dental radiographs. 
 SVM classifier used to classify tumors and performance evaluated using receiver
operating characteristic curve.
  Proposed System

Ø A  new neuronal approach is


proposed for the classification of
Throat cancers from MR images. 
Ø Neural Networks are a set of
algorithms and have been modeled
loosely after the human brain.
Computer scientists have designed
them to recognize patterns. We
also call them Artificial Neural
Networks or ANNs.
Ø The three steps namely, the
preprocessing step, features
extraction step and segmentation.
 HARDWARE REQUIREMENTS:  
Ø System : Pentium Dual Core.
Ø Hard Disk : 120 GB.
Ø Monitor : 15’’ LED
Ø Input Devices : Keyboard, Mouse
System
Requirements  SOFTWARE REQUIREMENTS:
Ø Operating system : Windows 7/8/10
Ø Coding Language : MATLAB
Ø Tool : MATLAB R2013A
Block Diagram
Modules

Image Preprocessing

Feature Extraction

DWT Method

Image Segmentation

Cancer Detection
Output Screens

Step 1: Input the image


Applying wavelet Transform​
Loading the image into database
Cancer Detection
Segmentation
Advantages and Applications

ADVANTAGES:
 High accuracy
 High precision value
 Easy to find in the early stages

APPLICATIONS:
 Bio medical Imaging
 Medical Test of testability
 Laboratory test
Conclusion

Ø The result of this project is to detect the cancerous cells in the Throat cavity and to
classify the cancer affected position to give results such as in percentage for
easier approach to the doctors to start their treatment efficiently.

Ø  The project is applying various techniques in the field of computer  technology using
the image processing.
THANK YOU

You might also like