Multiple Focus Reflectance Confocal Microscopy for In Vivo Imaging

Abstract

Light microscopy techniques provide a means to image interesting features on the very small scale. Advances in passive and active optical elements have driven progress in microscopy, including but not limited to super-resolution, large field of view, and three-dimensional microscopy. Medical diagnostics can benefit from this progress. Typically, for a suspected cancerous tissue, a biopsy is taken and analyzed by histology. Features of the tissue can be identified throughout the depth of the slice to inform diagnosis. Translating optical techniques to clinical applications presents a unique challenge. Typically, a microscopy sample is cut, sliced, and stained as part of the preparation needed for benchtop microscopes. In the clinic, the sample is inaccessible within a living, breathing human subject. Confocal microscopy is a well-known technique that can solve part of the problem, as it can image thin slices of tissue optically (i.e. no cutting necessary). Additionally, as a reflectance microscope, no exogenous contrast agents are needed. However, current techniques have not shown to be effective for imaging the entire tissue due to the cumbersome nature of in vivo imaging.

This dissertation reports solutions by designing a confocal microscope capable of imaging live bulk tissue at multiple depths by means of some non-mechanical tunable focus. First, we evaluate the technique, chromatic confocal microscopy, to simultaneously capture images at multiple depths. We present a chromatic confocal microscope with expanded range that can produce images from a highly scattering tissue sample. Design considerations and future work are discussed. Second, we discuss the use of a hand-held confocal microscope which employs a tunable lens for multi-depth imaging and present a second generation alignment-free design with improved image quality. Image quality and reliability are paramount to validating an imaging system for use as a clinical diagnostic tool. We show the capability of the system to detect features by means of optical biopsy and compare these images to the actual histological analysis.