Skin Lesion Classification

TLDR: The project consists in a binary classification over images of skin lesions in order to understand if the lesion is malign or benign. We used SAM to semantically segment the region of interest, and then experimented with a variety of deep learning models such as CNNs and ViTs. We also experimented with deep data augmentation by generating synthetic data via a StyleGAN.

Report • Presentation • GitHub Repo

Ops, details of the project are still work in progress!