Melanoma, a deadly form of skin cancer, can be treated successfully if it is detected in its early stages. Since melanoma lesions are distinctive in their propensity to change, photographically assisted follow-up is an important tool in early diagnosis. Current photographic image based follow-up requires substantial photographic expertise, is labor intensive, costly, and time consuming, thereby limiting the process to highly specialized centers. Digital systems promise to reduce cost and increase efficiency of image handling. However, they do not capture the entire skin surface at adequate resolutions, and lengthy imaging sessions are required. The proposed research will produce a digital imaging system in which clinically acceptable resolution is achieved to allow analysis of change in individual moles in overview images while reducing duration, labor, and skill requirements of photographically assisted follow-up. The system achieves this by automating the capture process so that more images can be captured in less time, with greater resolution in terms of pixels per cm. The system includes a novel approach for automatically capturing the skin surface in the form of image tiles and for matching the borders of these tiles to create a complete composite overview of the skin.
