Cancer Detection App (CDA) Repository π
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
Clone the repository:
git@github.com:VukIG/Cancer-Detection-AI.git
Install dependencies:
pip install -r requirements.txt
Run the app:
python interface.py
Contributing π
Feel free to contribute, report issues, or just explore the world of cancer detection with joy! π
Happy coding! π©βπ»π¨βπ»