Deep Learning Project - Image Classification

Deep Learning and Artificial Intelligence

Classification of Pigmented Skin Lesions

In this project, I classified images of common pigmented skin lesions in either benign or cancerous. I built different deep neural network architectures both with and without transfer learned weights.

For this purpose, I used the “HAM10000” dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions which can serve as a training set for academic machine learning purposes.

The full project can be found here.