Current active technology projects include the development of line-scanning confocal microscopes for imaging skin cancer in vivo and a confocal endoscope for imaging oral precancers in vivo. Another project that combines instrumentation with translational studies is the development of confocal mosaicing microscopy for rapid detection of basal cell carcinomas in excised tissue from Mohs surgery, with the goal of enabling real-time pathology at the bedside. Translational projects include (1) the preoperative mapping of lentigo maligna melanomas and amelanotic melanomas on patients in vivo to guide precise surgical excision, (2) mapping of residual basal cell carcinomas in shave biopsies, to guide biopsy and surgery and to serve as a test model for intraoperative imaging, and (3) in partnership with our head and neck surgeons, mapping of cancer margins in excised tissue to guide surgery.
A multicenter clinical study is in progress to determine the sensitivity and specificity of reflectance confocal microscopy for the in vivo diagnosis of melanoma. A collaboration with Northeastern University in Boston is focused on developing machine learning-based classification algorithms for “image understanding” analysis and detection of morphologic features in reflectance confocal images of skin cancers, to train and assist clinicians in reading and diagnosis. An exciting new idea is to combine imaging for diagnosis and image-guided laser surgery of non-melanoma skin cancers in vivo, to enable Mohs-like pathology, diagnosis and treatment directly on the patient in real-time.