Dissertation Date: April 8, 2025
Ophthalmoscopy, or the in vivo imaging of the eye, provides a unique “window to the brain,” aiding in the diagnosis of both retinal and systemic conditions such as diabetes, head trauma, and dementia. Despite the ubiquity of ophthalmic devices, oftentimes the visualization of smaller features is limited, even with the advent of adaptive optics (AO) to correct distortions caused by the eye’s innate aberrations and movements. To remedy this, AO was integrated into an AO scanning light ophthalmoscope (AOSLO) and allowed single-cell characteristics of the living retina to be resolved – a microscope for the living eye. Despite these advancements, AOSLOs still struggle to capture both single-cell and tortuous structures, like vasculature, and current AOSLO modifications are prohibitively extensive. Altogether, these challenges hinder both researchers and clinicians, underscoring the need for improved AOSLO devices. This dissertation explores novel illumination and detection strategies in AOSLO to enhance its’ resolution and expand patient accessibility.
Current AOSLOs fail to resolve features that lie at irregular orientations, such as vasculature, limiting their resolution to a fraction of the total retinal morphology. Here, this limitation was overcome by implementing a Dove prism into an AOSLO to rotate refracted light and resolve structures at irregular orientations in the eye. Through AOSLO integration and using the Dove’s intrinsic rotation properties, features from various structures were captured at multiple angles, increasing final visualization of features previously hidden using conventional methods. Simultaneously, this small-footprint design (4”) enables extant systems to be altered with minimal complexity. Results, both in animal and human in vivo retinae, confirmed the Dove prism to be an easily adapted method for resolving structures that lie at any angle in the eye as compared to current methods.
Another major limitation of standard en face imaging devices is diffraction, the process of light bending around an obstacle or diverging after passing through a small obstacle. This restricts AOSLO resolution, particularly in resolving individual cells like rod photoreceptors. To overcome this, a two-pronged approach was performed to concentrate imaging illumination and increase image resolution: (1) Using Fresnel zone plates (FZPs), and (2) Through light axicons. Through ophthalmoscope-specific generalization of established formulae for optical element construction and placement, the phase and amplitude of the light being delivered to the eye was manipulated, transforming it into concentrated (Bessel) beams that overcome diffraction. The manipulated beams improved resolution and depth of focus (DOF) and were assessed in a phantom retina and human in vivo retinae. This translatable model of optical elements allowed for increased visualization of the eye’s microstructures, as well as insights into how varying light profiles influence final image characteristics. Moreover, Bessel beam illumination allowed for simultaneous resolution of the vascular and photoreceptor layers through increased DOF, a hitherto largely unexplored method of study. Ultimately, integration of these elements into an AOSLO forms a basis on which en face ophthalmic imaging can expand beyond current patient populations while simultaneously providing images with increased feature resolution and enhanced focus.
Cumulatively, this work provides three key advancements in retinal imaging: (1) Enhanced resolution of retinal structures at various orientations; (2) Improved AOSLO focus for multi-layer retinal imaging; and (3) A flexible method for achieving sub-cellular resolution in existing AOSLO systems. These findings expand both the accessibility and quality of ophthalmic imaging to broader studies and populations. Moreover, this work demonstrated the first use of Dove prisms, axicons, and FZPs in an AOSLO, laying the groundwork for future novel experiments and studies.
