Speaker: Upender Kalwa, ECpE Graduate Student
Advisor: Santosh Pandey
Title: Democratizing Skin Cancer Diagnostics with an All-Inclusive Smartphone Application
Abstract: Among the different types of skin cancer, melanoma is considered the deadliest and is difficult to treat at advanced stages. As such, there is significant interest in developing portable, at-home melanoma diagnostic systems that can assess the risk of cancerous skin lesions. Here we present a smartphone application with capabilities of image capture, preprocessing, segmentation, extraction of ABCD features, and classification of the skin lesion malignancy based on a support vector machine (SVM) classifier. Adaptive algorithms are incorporated at each stage of image processing which are computationally light, smartphone friendly, and reliable in discriminating melanoma cases from the benign ones. The smartphone application is trained using a subpart of a public dataset and tested on the other subpart along with mole images collected, in real-time, from two individuals. The evaluated performance metrics and computation times are comparable or better than previous methods. This all-inclusive smartphone application is designed to be easy-to-download and easy-to-navigate for the end user, which is imperative for the eventual democratization of such medical diagnostic systems.