Scribd Logo
Upload

Welcome to Scribd!

  • 27 views

    Brain Tumor

    Uploaded by

    hananth1513
    The document describes a proposed system for brain tumor detection using a convolutional neural network (CNN). The proposed system uses a 23-layer CNN architecture to classify tumor types directly from MRI images without needing feature extraction. This improves accuracy and makes identification easier and faster compared to existing systems. The CNN model is trained on preprocessed MRI images and can help with computer-aided diagnosis, research, and medical education.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd

    Brain Tumor

    Uploaded by

    hananth1513
    27 views19 pages
    The document describes a proposed system for brain tumor detection using a convolutional neural network (CNN). The proposed system uses a 23-layer CNN architecture to classify tumor types directly from MRI images without needing feature extraction. This improves accuracy and makes identification easier and faster compared to existing systems. The CNN model is trained on preprocessed MRI images and can help with computer-aided diagnosis, research, and medical education.

    Copyright

    © © All Rights Reserved

    Available Formats

    PPTX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The document describes a proposed system for brain tumor detection using a convolutional neural network (CNN). The proposed system uses a 23-layer CNN architecture to classify tumor types directly from MRI images without needing feature extraction. This improves accuracy and makes identification easier and faster compared to existing systems. The CNN model is trained on preprocessed MRI images and can help with computer-aided diagnosis, research, and medical education.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    27 views19 pages

    Brain Tumor

    Uploaded by

    hananth1513
    The document describes a proposed system for brain tumor detection using a convolutional neural network (CNN). The proposed system uses a 23-layer CNN architecture to classify tumor types directly from MRI images without needing feature extraction. This improves accuracy and makes identification easier and faster compared to existing systems. The CNN model is trained on preprocessed MRI images and can help with computer-aided diagnosis, research, and medical education.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    You are on page 1of 19

    BRAIN TUMOR DETECTION USING CNN

    S V COLLEGE OF ENGINEERING
    (AUTONOMOUS)
    Karkambadi Road, Tirupati
    DEPARTMENT OF ELECTRONICS AND COMMUNICATION ENGINEERING

    UNDER THE GUIDANCE OF PRESENTED BY


    Mr. P.SRINIVASULU,M.Tech P.Lohitha
    Assistant Professor, (20BF1A04H8)
    Department of ECE, B.Mallikarjun (20BF1A04C6)
    SVCE,Tirupathi. P.Bhagya Sree
    (20BF1A04G1)
    P.Mahesh (20BF1A04F9)
    S V COLLEGE OF P.RevanthTIRUPATHI
    ENGINEERING, Kumar (20BF1A04H3)
    CONTENTS

    •Abstract
    •Existing System and Drawbacks
    •Proposed System
    •Applications
    •Block Diagram
    •Image Processing
    •Image Enhancement
    •Binary Threshold
    •Tools to be Used
    •Architecture

    S V COLLEGE OF ENGINEERING, TIRUPATI


    ABSTRACT
    BRAIN TUMOR DETECTION USING CNN

     Brain tumor is considered as one of the aggressive diseases, among children and adults. Brain
    tumors grow very fast and if not treated well, the survival chances of the patient are very less.

     Early detection of brain tumors is very important. Proper treatment planning and accurate
    diagnosis is at the topmost priority to improve life expectancy of the patients.

     The best technique to detect brain tumor is Magnetic Resonance Imaging(MRI). The MRI
    images are examined by the radiologist.

     Manual examination can be error prone due to the level of complexities involved in brain
    tumors and their properties. Hence an automated brain tumor detection system is required to
    detect tumors at its early stage.

     This project uses deep learning based Depth wise Separable Convolution Neural Network to
    detect the tumor based on the MRI images.

    S V COLLEGE OF ENGINEERING,
    Existing System

    Brain tumor detection with segmentation based machine learning technique


    As a large volume of medical MRI imaging data is gathered through image
    acquisition, the researchers are now proposing different machine learning methods to
    identify brain tumors.
    Brain tumor detection through transfer learning
    • Transfer learning does well when the volume of data is limited since such a model
    is previously trained on a large dataset (e.g., the ImageNet database), containing
    millions of images..
    • Another benefit is that it does not require a massive amount of computational
    resources since only the model’s fully connected layers need to be trained.

    •
    Existing System

    • Data augmentation is a technique in machine learning used to reduce overfitting when


    training a machine learning model, by training models on several slightly-modified
    copies of existing data.
    • The above methodology achieved an overall prediction accuracy of 87.4% before the
    data augmentation. Finally, by applying the data augmentation technique the accuracy
    has been increased to 90.7%.

    S V COLLEGE OF ENGINEERING, TIRUPATI


    Drawbacks
    • Since all these methods are based on binary classification (normal and abnormal), it is
    not sufficient for the radiologist to decide the patient’s treatment concerning tumor
    grading.

    • Those methods are based on different hand-crafted feature extraction techniques,


    which are time-consuming, complex, and in many cases not effective.

    • Pre-trained models fail to obtain satisfactory results when training on imbalance


    datasets. They are more biased towards classes with a larger number of samples .

    • Proper fine-tuning is required in pre-trained models. Otherwise, the model will fail to
    achieve satisfactory result.

    S V COLLEGE OF ENGINEERING, TIRUPATI


    Proposed System
    • Brain tumor detection using convolution neural networks (CNN),CNN presents a
    segmentation-free method that eliminates the need for hand-crafted feature extractor
    techniques.
    • The proposed ‘‘23-layers CNN” architecture used to classify different tumor types,
    including meningioma, glioma, and pituitary. In the proposed architecture, we take
    MRI slices as input, process the slices in different layers, and differentiate them from
    one another. In this study, a total of 23 layers are used to process the slice.
    • Below is the description of each layer: One of the predominant building blocks of the
    CNN model is the convolutional layer.
    • One matrix relates to the kernel, while the other presents the pixel intensity values of

    the original image


    Proposed System

    • Proposed 23-layers CNN architecture

    S V COLLEGE OF ENGINEERING,
    Advantages

     Improved Accuracy
     Easy Identification
     Faster training

    S V COLLEGE OF ENGINEERING, TIRUPATI


    Applications

    • Computer-aided diagnosis (CAD) systems


    • Telemedicine
    • Research and clinical trials
    • Training and educating medical professionals
    • Radiomics analysis
    • Automated tumor segmentation

    S V COLLEGE OF ENGINEERING,
    BLOCK DAIGRAM
    IMAGE PREPROCESSING:
    • The MRI dataset consists of around 1900 MRI images, including normal, benign,
    and malignant.
    • These MRI images are taken as input to the primary step. The pre-processing is an
    essential and initial step in improving the quality of the brain MRI Image.
    • In the initial phase, we convert the brain MRI image into its corresponding
    gray-scale image
    • The removal of unwanted noise is done using the adaptive bilateral filtering
    technique to remove the distorted noises that are present in the brain picture
    • This improves the diagnosis and also increase the classification accuracy rate.
    IMAGE ENHANCEMENT:
    • Image enhancement is a technique used to improve the image quality and
    perceptibility by using computer-aided software.
    • This technique includes both objective and subjective enhancements.
    • This technique includes points and local operations. The local operations
    depend on the district input pixel values. Image enhancement has two types:
    spatial and transform domain techniques.
    • The spatial techniques work directly on the pixel level, while the transform
    technique works on Fourier and later on the spatial technique.
    BINARY THRESHOLD:
    • Image segmentation is a technique of segregating the image into many parts.
    The basic aim of this segregation is to make the images easy to analyze and
    interpret with preserving the quality.
    • This technique is also used to trace the objects borders within the images. This
    technique labels the pixels according to their intensity and characteristics.
    • Segmentation is used in machine perception, malignant disease analysis, tissue
    volumes, anatomical and functional analyses, virtual reality visualization, and
    anomaly analysis, and object definition and detection.
    Tools to be Used

    • Python
    • Pre-trained Models
    • Medical Imaging Datasets
    • Data Preprocessing Tools
    • Deep Learning Frameworks
    • Visualization Libraries
    • Evaluation Metrics

    S V COLLEGE OF ENGINEERING, TIRUPATI


    ARCHITECTURE
    -THANK YOU

    ##BATCH-9

    S V COLLEGE OF ENGINEERING,

    You might also like