A Digital Diagnostic Aide for Skincare: The Role of Computer Vision and Machine Learning in Revealing Skin Texture Changes
Advances in Information and Communication: Proceedings of the 2020 Future of Information and Communication Conference (FICC), Volume 2
Kohei Arai, Supriya Kapoor, Rahul Bhatia
Skin disease is a serious disease and can also be a deadly disease. Prevalence of skin disease is high and is likely to increase as the population ages. Skin disease burdens Americans, their families and employers. According to American Academy of Dermatology (AAD), nearly 25% of the population ages 0–17 had diagnosed with skin disease in 2013 and the price tag for treatment was $75 billion. Worldwide, an estimated 1.9 billion people suffer from a skin condition at any given time, and shortage of dermatologists aggravating the issue. One of the chief early signs, per dermatologist, to a potential skin disease is change in the skin, ranging from discoloration to new growth. In this paper, we will discuss the application of machine learning algorithms and Computer Vision techniques to analyze skin texture changes that are invisible to the naked eye and provide actionable insights framework that would trigger preventive treatment procedures to address any impending skin disease. We will discuss several computer vision techniques and cognitive services to improve efficiencies of computer vision techniques. Our goal is to develop assistive Computer vision models that could potentially help dermatologists to take proactive healthcare actions to reduce occurrence skin diseases.
Azure & Amazon Cognitive Services, Cognitive Services, CV, HER, Local Binary Patter (LBP), Sanjeevani Electronic Health Records
Jaya Shankar Vuppalapati, Santosh Kedari, Anitha Ilapakurti, Chandrasekar Vuppalapati, Sharat Kedari, and Rajasekar Vuppalapati. "A Digital Diagnostic Aide for Skincare: The Role of Computer Vision and Machine Learning in Revealing Skin Texture Changes" Advances in Information and Communication: Proceedings of the 2020 Future of Information and Communication Conference (FICC), Volume 2 (2020): 898-915. https://doi.org/10.1007/978-3-030-39442-4_69