VukIG / Cancer-Detection-AI

Custom CNN Model built using TensorFlow with HAM10000 dataset from Kaggle

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Cancer Detection App (CDA) Repository πŸš€

Overview

Welcome to the Cancer Detection App (CDA) repository! This app uses TensorFlow, Matplotlib, NumPy, and Pandas to detect cancer from the Human Against Machine 10000 dataset obtained from Kaggle.

Dataset

The HAM10000 dataset is organized into the HAM10000 folder, containing various skin lesion images.

Data Processing

To handle imbalanced data, we used TensorFlow's ImageDataGenerator to generate additional data. The dataset is split into training, validation, and test sets using the flow_from_directory function.

Model Creation

The model is created using TensorFlow's Sequential API, consisting of convolutional layers for image analysis and dense layers for classification.

Usage

A user-friendly interface is provided through a while True loop, allowing users to interactively choose images. The selected image is then loaded, plotted using Matplotlib, and labeled as benign or malignant.

How to Use

  1. Clone the repository:

    git@github.com:VukIG/Cancer-Detection-AI.git
  2. Install dependencies:

    pip install -r requirements.txt
  3. Run the app:

    python interface.py

Contributing πŸŽ‰

Feel free to contribute, report issues, or just explore the world of cancer detection with joy! πŸ˜„

Happy coding! πŸ‘©β€πŸ’»πŸ‘¨β€πŸ’»

About

Custom CNN Model built using TensorFlow with HAM10000 dataset from Kaggle


Languages

Language:Python 100.0%