Melanoma is diagnosed in approximately 124,000 people and is responsible for about 10,000 deaths every year, in the USA. In Europe and Australia, about 97,000 new cases are detected every year. Dermatologists rely on visual and dermatoscopic examination to discriminate benign melanocytic lesions from malignant. With this low-resolution approach, benign-to-malignant biopsy ratios are high and variable, ranging from 4:1 to 600:1, resulting in millions of unnecessary biopsies of benign lesions. Reflectance confocal microscopy (RCM) imaging is a high-resolution optical sectioning approach that has been proven, in large trials, for noninvasively diagnosing melanoma, with sensitivity of 92-88% and specificity 71-84%. The specificity is 2 times superior to that of dermatoscopy. During 2013-2014, RCM imaging advanced to being implemented in clinics, to actively guide noninvasive diagnosis (i. e., rule out malignancy and biopsy), with promising initial impact: thus far, ~500 patients have been saved from biopsies. However, this success is currently confined to a small cohort of early adopter clinicians in a few leading academic centers. The current size and cost of the Vivascope (commercial version of the PI's original bench-top point-scanning microscope) and complexity of operation remain barriers against widespread adoption in diverse settings. Not surprisingly, the oft-repeated feedback from clinicians has been to produce a smaller, simpler, lower cost microscope that will be quick and easy to use with seamless integration into daily patient flow and routine clinical practice. Our academic-industrial partnership project (Memorial Sloan-Kettering Cancer Center and Caliber Imaging and Diagnostics (formerly, Lucid Inc.)) is intended to address these barriers. Line-scanning, with divided pupil configurations, miniaturized FPGA and GPU electronics, and video-mosaicing (innovations), is an approach that may compete with current point- scanning microscopy, and may offer a smaller, simpler and lower-cost alternative for noninvasive diagnosis. Preliminary results with a bench-top line-scanning confocal microscope (cost $15,000) show excellent imaging of nuclear and cellular detail in human skin in vivo. Instead of the standard full pupil, a divided pupil configuration provides strongly enhanced optical sectioning in scattering tissue. The optical sectioning is 1-2 m and lateral resolution is 0.5-1.0 m, which is competitive with that of existing point-scannin microscopes.
Our specific aims are (1) to optimize pupil configurations for line-scanning RCM; (2) to build a small and low cost microscope for imaging cellular morphology in melanoctyic skin lesions in vivo. The size will be 2x3x8 inches, weight about one pound, and cost about $5,000 (in commercial quantities); (3) to test performance for detecting melanocytic skin lesions in vivo, in a pilot study on 120 patients. The sensitivity and specificity of line-scanning RCM will be validated against that of current point-scanning RCM (Vivascope) and pathology. The validation will determine the feasibility of line-scanning for future commercialization and clinical trials. I successful, this approach may be useful also for detection of non-melanoma skin cancers (wider impact).
Reflectance confocal microscopy (RCM) is an optical imaging technology that can guide noninvasive diagnosis of melanoma and non-melanoma skin cancers, and reduce the need for biopsy. However, the current microscopes are large, expensive and complex to use. We propose to develop smaller, simpler and lower-cost microscopes for wider adoption of this technology by dermatologists and wider acceptance by patients.
| Abeytunge, Sanjee; Larson, Bjorg; Peterson, Gary et al. (2017) Evaluation of breast tissue with confocal strip-mosaicking microscopy: a test approach emulating pathology-like examination. J Biomed Opt 22:34002 |
