Topic: Leveraging Artificial Intelligence to Improve Skin Color Diversity in Cancer Detection
Date of Presentation: Wednesday, September 28, 2022
Location: BSS 121 and Online
The paucity of dark skin images in dermatology textbooks and atlases hinders the accurate diagnosis of skin lesions in dark skin tones. For conditions such as skin cancer, in which early diagnosis makes a difference between life and death, people of color have a worse prognosis and lower survival rates than people with lighter skin tones as a result of delayed or incorrect diagnoses. Artificial intelligence applications have further disadvantaged people of color because those applications are mainly trained with light skin color images. In our work, we develop a deep learning framework to diversify the mostly light-skin image repositories by generating images for darker skin tones. Thus, facilitating the development of inclusive cancer early diagnosis systems that are trained and tested on diverse images that truly represent human skin tones.