One Liner:

Developing a machine learning model to classify pigmented skin lesions using the HAM10000 dataset, aiming to expedite skin cancer diagnosis.

Abstract:

Our Skin Cancer Classification project aims to develop a machine-learning model with the HAM1000 dataset. This data set contains over 10,000 images of different pigmented skin lesions. We aim to use this model to diagnose pigmented skin lesions automatically. With various deep learning techniques, we strive to classify conditions like melanoma, basal cell carcinoma, and other pigmented lesions, which will lessen the time for diagnosis.

Description:

Our Skin Cancer Classification project aims to develop a machine-learning model using the HAM10000 dataset to accurately distinguish melanoma from other skin lesions and classify various subtypes, including basal cell carcinoma and benign lesions. Utilizing deep learning, specifically convolutional neural networks (CNNs) in TensorFlow, the model aims to provide quick diagnostic support to physicians. An interactive user interface will allow healthcare providers to upload images and receive classification results. This tool is designed to aid in early detection, ultimately enhancing patient care by reducing diagnosis times and supporting clinical decisions.

Video: https://1513041.mediaspace.kaltura.com/media/Skin%20Cancer%20Classification%20-%20Presentation/1_5oa79yal

Print Poster: View Poster

Digital: View Poster

Team Members

Advisors

Stakeholders