Classifying Dermatological Skin Lesions

Classifying Dermatological Skin Lesions is a project that utilizes Computer Vision, Deep Learning, and Machine Learning techniques to analyze different pigmented skin lesions in a dermatoscopic image and accurately classify the type of potential skin cancer with which the lesion is associated. I primarily worked on the classical Computer Vision approach of the project, where I implemented the gradient lookups of the edges in different skin lesions using Canny edge detector, and analyzed the output of the edge detector for the Machine Learning approach.

Role

Programming
Research

Tools

Jupyter Notebook
Python
HTML
Excel

Duration

12 weeks
Sep - Nov 2020

Team

Alexander Guo Programmer
Rahil Patel Programmer
Sultan Sayezada Programmer
Christopher Turko Programmer
Yunni Zhu Programmer

Supervisor

Prof. Devi Parikh