STTR Phase I: Dermatologist-level detection of suspicious pigmented skin lesions from high-resolution full-body images

The broader impact/commercial potential of this Small Business Technology Transfer Program (STTR) Phase I project is to improve outcomes in dermatology and enable improved clinical care in the absence of specialists. The proposed technology enables cost-effective screening for cancer, psoriasis, atopic dermatitis, and other inflammatory skin conditions. In the United States, psoriasis affects about 8 million people while about 31.6 million people in the United States have some form of eczema, including atopic dermatitis. The proposed system generates a highly magnified image of the skin for analysis by a clinician or an artificial-intelligence based automated system. This Small Business Technology Transfer Program (STTR) Phase I project integrates several subsystems: 1) a scanner and software to capture and reconstruct super high-resolution (dermatoscope level) model of the entire skin surface of patients in a matter of minutes; 2) a multispectral illumination system to provide more information than currently available from a typical white-light systems, potentially leading to superior sensitivity and specify in lesion detection and classification; and 3) a system to find and classify lesions automatically from high-resolution images. This system advances the automated classification of images beyond those filtered previously by a clinician. Moreover, the system's UV illuminations system potentially can be used in a photodynamic treatment regimen. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.