Computational aberration correction for SD-OCT imaging of fish retina at 100 kHz A-scan rate

초록

While high axial resolution of Optical Coherence Tomography (OCT) provides cellular-level visualization of the retina layers, high-NA imaging is needed to visualize cellular level structures in en face plane. This, however, is often restricted by severe ocular aberrations present in imaging eye, including eyes of experimental animals, that preclude the attainment of diffraction-limited resolution. In this study, we propose a robust volumetric aberration correction framework integrated with a custom-built 100 kHz SD-OCT system for in vivo small animal retinal imaging. Instead of employing complex hardware-based adaptive optics, the proposed method utilizes a two-step computational optimization strategy: estimating a global wavefront map via Shannon entropy minimization on a representative structural layer and applying a depth-variant defocus model to the entire volume. The performance of this method is validated through a comparative intensity analysis and pixel intensity histograms, which demonstrate an enhancement in the signal intensity of individual photoreceptor cells. Furthermore, the enhancement of internal retinal structures is quantitatively confirmed through A-scan line profiling, showing marked improvements in signal contrast throughout the axial direction. This computational approach offers an efficient solution for high-speed volumetric imaging, facilitating high-resolution morphological studies of the small animal retina without additional hardware complexity.

키워드