Confocal microscopy uses geometric principles to generate cross-sectional images of scattering biological tissue. Traditional reflectance confocal microscopes use serial pixel acquisition, thereby limiting acquisition speeds. Parallelization of pixel acquisition would dramatically increase imaging speeds and enable the study of new kinds of physiology. For example, massively parallelized reflectance confocal microscopy would be invaluable in the emerging field of cilia-driven mucus physiology in the respiratory system. Indeed, in terms of hypothesis-driven ciliary biology, there is a critical methodological gap in high-speed (0.1 to 1 kHz frame rates) imaging methods at the ~1 um scale. Traditional confocal microscopy uses physical pinholes in sample illumination and in photo detection to generate cross-sectional images by straightforward geometric principles. While the use of physical pinholes can be partially parallelized, it is impossible to have complete, scan-free parallelizatio using physical pinholes. However, parallel interferometric confocal microscopy allows an entire field-of-view to be imaged without scanning. In interferometric confocal microscopy, virtual interferometric pinholes are generated, and mutually incoherent spatial modes act as independent and parallelizable confocal imaging channels. However, the lack of sources with low spatial coherence (many independent spatial modes) and high brightness per spatial mode is a critical barrier to massively parallel interferometric confocal microscopy. Traditional lasers exhibit high spatial coherence, while low spatial coherence sources such as thermal sources and light-emitting diodes (LEDs) do not provide the necessary brightness. Recently, we have shown that degenerate Nd: YAG lasers can support as many as ~105 mutually incoherent lasing modes (independent imaging channels). Therefore, using a specifically-designed degenerate Nd: YAG laser, we will build a massively parallel interferometric confocal imaging system with a 100 Hz frame rate for imaging in the ~1 um resolution regime. We expect the principles of operation to allow future scaling into the >kHz frame rate regime. We will use frequency-doubled 532 nm light for interferometric reflectance confocal imaging in ~104 (100 x 100) parallel, independent imaging channels. ~104 imaging channels is chosen as a design specification because several commonly-used imaging fiber bundles used in endoscopy have ~104 imaging cores. As an initial demonstration, we will image ciliary physiology in Xenopus (frog) embryos. Like respiratory epithelial surfaces, the epidermis (skin) of Xenopus embryos is ciliated and generates directional fluid flow. We will image the motion of individual cilia as well quantify cilia-driven fluid flow using interferometry-enabled Doppler flow imaging. Our design-driven research will support the future development of novel diagnostics in respiratory ciliary physiology. Our development of a degenerate laser for confocal microscopy supports future research in degenerate lasers for other kinds of medical imaging (e.g. optical coherence tomography, HiLo structured illumination, holographic microscopy).
Lasers are an indispensable part of sophisticated microscopes. We will use a new kind of laser to enable much faster imaging at very small (~1 micrometer) size scales. Our work will advance the understanding, diagnosis, and treatment of certain respiratory diseases, which will benefit public health.
