Skin cancer is the most common cancer in the United States. There are over 1 million cases per year of basal cell carcinoma, squamous cell carcinoma and malignant melanoma. The treatment of these cancers include topical chemotherapy, cryotherapy, radiation, and surgery. By far, surgery is the most common treatment modality. One of the limitations of the traditional excisional surgery is the lack of precise tumor margin control resulting in removal of excess normal skin. Polarized light imaging is a non-invasive skin imaging technique that uses polarized white light and a high-resolution digital camera to enhance visualization of the superficial skin where skin cancers arise. Using this technique, optical guidance of the surgical excision of skin cancers can be achieved by noninvasively visualizing skin cancer margins. The reflectance of obliquely illuminating polarized light from the skin is acquired by a CCD camera through a polarizing filter that either accepts or blocks the polarization of the illumination light, and two images are taken. The difference in the images yields the """"""""polarized image"""""""". Such imaging accepts photons backscattered from the subsurface and superficial layers of skin, but rejects surface glare and deeply penetrating photons that have been multiply scattered. Hence, the images use only about 10% of the photons reflected from the skin but these photons provide enhanced image contrast for visualizing the upper 300 um of the skin where skin pathology arises. Preliminary studies suggest that subtle disruption of the collagen matrix by either lateral or vertical movement of the cancer is visualized by the polarized images. This proposal will utilize this new technology in a pilot randomized clinical trial testing the efficacy of including the polarized light camera in visualizing the margins of skin cancer so as to guide the surgical excision. The hypothesis is that polarized light imaging can identify skin cancer margins not seen by normal vision and guide surgery. A library of images of normal and cancer will be accrued and an image analysis algorithm developed to assist the visualization of the cancer margin.
