This project is focused on the development of a DL model to address a dermatology classification problem. The dataset used in this project contains images, each corresponding to only one specific disease. The folder contains only the metadata of each image with some disease information such as malignant/benign, angle scores, and the URL of the images of skin lesions. The column "label" from the CSV file is your target column. The annotated images represent 114 skin conditions, with at least 53 images and a maximum of 653 images per skin condition.
The dataset of images was taken from the “FITZPATRICK17” repository. The Fitzpatrick scale is a numerical classification schema for human skin color. It was developed in 1975 by American dermatologist Thomas B. Fitzpatrick to estimate the response of different types of skin to ultraviolet light. The dataset is a combination of two dermatology datasets, “DermaAmin” and “ATLAS Dermatologico”. Your task is to build a model that performs well on unseen images. For that, you have to split the dataset into three splits: train, validation, and test.
Link to archive with images: https://drive.google.com/file/d/1C8m64W-Tv5R0VJHUtkUXOVt8wZi1coXc/view?usp=sharing