Cancer Detection Models (Breast & Skin Cancer)

    Developed classification models to predict breast cancer malignancy and built a CNN to classify skin lesions from medical images, deploying both using Flask.

    Muhammad Zeeshan

    Technologies Used

    Python
    TensorFlow
    Keras
    scikit-learn
    Flask
    OpenCV
    NumPy

    Key Features

    1

    Breast cancer classification using Wisconsin dataset

    2

    CNN-based skin lesion classification

    3

    Image preprocessing and augmentation pipeline

    4

    REST API deployment with Flask

    5

    Real-time prediction interface

    6

    Model performance metrics and visualization

    Implementation

    Developed two separate models: a traditional ML classifier for breast cancer diagnosis and a deep CNN for skin lesion classification. Implemented data augmentation, model training with cross-validation, and deployed as a Flask web service.

    Results & Impact

    Breast cancer model achieved 97% accuracy, skin lesion CNN achieved 92% accuracy. Successfully deployed as a web application for real-time medical image analysis.