ilanaliouchouche/bayesian-active-le...

This project provides an intuitive interface for medical professionals to integrate their domain expertise with Bayesian deep learning models, enabling active learning and model teaching for medical imaging tasks. Built with Marcelle and FastAPI, it allows doctors to label uncertain predictions, retrain models, and visualize performance metrics in real-time. The tool is designed to make advanced machine learning accessible to healthcare practitioners, improving model accuracy through iterative feedback and domain knowledge integration.