Scribd Logo
Upload

Welcome to Scribd!

  • 28 views

    Skin Cancer Detection

    Uploaded by

    Rishabh Sharma
    The document discusses skin cancer detection using deep learning models. It outlines the following phases of the project: 1) brainstorming technologies and concepts, settling on convolutional neural networks (CNN), 2) researching skin cancer detection and setting up dependencies, 3) obtaining a skin cancer dataset, 4) training models to compare accuracies. The goal is to classify images as benign or malignant to detect skin cancer early. So far, InceptionV3 achieved the highest accuracy of 85% while the custom CNN model achieved 83.48% accuracy. The next target is to improve the custom model's accuracy further.

    Copyright:

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

    Skin Cancer Detection

    Uploaded by

    Rishabh Sharma
    28 views7 pages
    The document discusses skin cancer detection using deep learning models. It outlines the following phases of the project: 1) brainstorming technologies and concepts, settling on convolutional neural networks (CNN), 2) researching skin cancer detection and setting up dependencies, 3) obtaining a skin cancer dataset, 4) training models to compare accuracies. The goal is to classify images as benign or malignant to detect skin cancer early. So far, InceptionV3 achieved the highest accuracy of 85% while the custom CNN model achieved 83.48% accuracy. The next target is to improve the custom model's accuracy further.

    Original Title

    Skin cancer detection

    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 discusses skin cancer detection using deep learning models. It outlines the following phases of the project: 1) brainstorming technologies and concepts, settling on convolutional neural networks (CNN), 2) researching skin cancer detection and setting up dependencies, 3) obtaining a skin cancer dataset, 4) training models to compare accuracies. The goal is to classify images as benign or malignant to detect skin cancer early. So far, InceptionV3 achieved the highest accuracy of 85% while the custom CNN model achieved 83.48% accuracy. The next target is to improve the custom model's accuracy further.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    28 views7 pages

    Skin Cancer Detection

    Uploaded by

    Rishabh Sharma
    The document discusses skin cancer detection using deep learning models. It outlines the following phases of the project: 1) brainstorming technologies and concepts, settling on convolutional neural networks (CNN), 2) researching skin cancer detection and setting up dependencies, 3) obtaining a skin cancer dataset, 4) training models to compare accuracies. The goal is to classify images as benign or malignant to detect skin cancer early. So far, InceptionV3 achieved the highest accuracy of 85% while the custom CNN model achieved 83.48% accuracy. The next target is to improve the custom model's accuracy further.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    of 7

    Skin cancer detection

    Rishabh Sharma I7
    41714803118
    TimeLine of Project

     Phase 1:
    1. Brainstorming the technologies and the concept using the project.  Phase 3.

    2. The technology was finalized to be convolutional neural networks 1. Found a suitable dataset for skin cancer detection.
    (CNN). 2. Found some suitable well known algorithms which would be
    3. The plan was to choose an idea that could be beneficial to the suitable for this project.
    society.
    4. Skin cancer detection was chosen as the topic.
     Phase 4. ( current)
     
    1. Trained the dataset on all the models to compare and get the
     Phase 2: best accuracy.
    1. Extensive research on the topic was done. 2. Tweaking the parameters to get better accuracy.
    2. Studied applications of CNN
    3. Set up the workspace and installed all the dependencies.
     
     
    Abstract

    Skin cancer is one of the most prevalently seen cancer type in human beings. Skin cancer
    occurs due to the uncontrollable growing of mutations taking place in DNAs owing to some
    reasons. Recognizing the cancer in early stages could increase the chance of a successful
    treatment. Now a days, computer aided diagnosis applications are used almost at every
    field. One of the mostly used areas is health sector. Biomedical datasets are created by
    saving the data of illness people in computers. Our goal is to obtain an effective way for
    early diagnosis of skin cancer by classifying our dataset images as benign or malignant.
    Our dataset consists of 2437 training images,660 test images and lastly 200 validation
    images.
    Problem Statement

    Skin cancer can lead to the death of an individual without proper treatment. The
    purpose of the project is to espy skin cancer in the initial stage and increase the
    possibility of a successful procedure and treatment. In the medical field, diagnosis with
    the help of computer applications is used and various datasets which are biomedical in
    nature are constructed from the data collected. Our aim is to acquire a capable course
    of action to diagnose skin cancer in its initial stages by labelling the images present in
    the dataset as malignant or benign. And to increase accuracy of our current model while
    comparing accuracies with other well known deep learning algorithms namely
    InceptionV3 and VGG16.
    Accuracy acquired till now

       

    CLASSIFICATION METHODS RESULT


     

    Deep CNN 83.48% accuracy


     

    VGG16 83.78% accuracy

    InceptionV3 85.00% accuracy


    Next Target

     Target is to improve the accuracy of our model further.


     Currently the accuracy is 83.48% .
    Thank You

    You might also like