Automated Classification of Second- and Third-Degree Burn Images Using Convolutional Neural Networks

This release contains the stable code and artifacts used in the paper "Automated Classification of Second- and Third-Degree Burn Images Using Convolutional Neural Networks". It includes the optimized green-channel CNN and all models and experiments necessary to reproduce the results. Highlights Optimized green-channel CNN for second- and third-degree burn classification. Local Gradio demo for inference with proposed and transfer-learning models. Complete experiment results and notebooks to reproduce figures and metrics from the paper. Included in this release Trained models (models/): Keras checkpoints of proposed and transfer-learning models. Gradio demo (interface/): gradio_burn_classification.py for local inference. Notebooks (notebooks/): Experiments, evaluation, and figure generation. Experiment results (data/results/): CSV files with raw results. Figures (figures/): Plots and figures from the manuscript. Quick start Install dependencies and run the demo: pip install -r requirements.txt python interface/gradio_burn_classification.py