Advisor: Dr. Leonid Shmuylovich
To diagnose and manage skin disease, clinicians primarily use visual inspection aided with dermatoscopes, which are cross-polarized magnifiers. However, current diagnosis relies heavily on subjective visual assessment, while the gold standard for margin determination—Mohs micrographic surgery—is iterative, time-consuming, and inherently reactive.
The proposed project consists of developing a handheld, cross-polarized digital system that displays the collagen network of skin. Unlike existing multispectral systems that are bulky and rely on motorized filter wheels, this device utilizes a dual-camera snapshot design. This enables simultaneous acquisition of spectrally distinct images without inter-frame delay, minimizing motion artifacts.
We developed a dual-wavelength, cross-polarized dermoscopy system that repurposes a standard handheld dermatoscope as the illumination front end. The system uses a dichroic beamsplitter and two monochrome cameras to acquire 450 nm (collagen-sensitive) and 650 nm (reference) images simultaneously.
The central innovation of this technology is not just the creation of a new imaging device, but a new paradigm for augmenting ubiquitous, existing clinical tools. Instead of replacing the dermatoscope, our system enhances it, adding physiologically-specific information to a tool clinicians already trust and use daily.
This low-cost, modular approach could democratize access to advanced optical diagnostics, moving it from specialized centers to routine clinical practice. By providing rapid, non-invasive, point-of-care margin assessment, this device could help guide excisions before the first cut, potentially reducing positive margin rates and improving patient outcomes.
The following slides from the thesis presentation outline the physical and optical architecture of the clinic-ready device. They highlight the transition from existing bulky systems to our handheld, low-cost solution.