We present a deep-learning based approach for automated qualitative assessment of lesion volumes using OCT images to enable real-time assessment of injury severity and longitudinal tracking of tissues response to photodamage. The network has been trained to quantify photodamage between the outer plexiform layer (OPL) and retinal pigmented epithelium (RPE) accurately without the need for extensive image pre- and post-processing. Manually annotated OCT cross-sections were used as ground-truths to train a U-Net convolutional neural network. The network was designed and implemented in PyTorch based on the multi-scale U-Net architecture.
Mice are commonly used to model human retinal disease because of similarities in ocular anatomy and function and availability of transgenic phenotypes. Recently, targeted delivery of photo-lesions has been demonstrated using fundus imaging and a combined optical coherence tomography (OCT) and scanning laser ophthalmoscopy (SLO) system; however, these systems precluded simultaneous multimodality imaging. Here, we present multimodality OCT+SLO imaging combined with a photocoagulation laser and custom software that allows for targeted delivery of focal retinal laser injury under real-time en face OCT guidance. Longitudinal imaging results show lesions with varying degrees of retinal damage during an injury titration study.
Rodent models are robust tools for understanding human retinal disease and function because of their similarities with human physiology and anatomy and availability of genetic mutants. Optical coherence tomography (OCT) has been well-established for ophthalmic imaging in rodents and enables depth-resolved visualization of structures and image-based surrogate biomarkers of disease. Similarly, fluorescence confocal scanning laser ophthalmoscopy (cSLO) has demonstrated utility for imaging endogenous and exogenous fluorescence and scattering contrast in the mouse retina. Complementary volumetric scattering and en face fluorescence contrast from OCT and cSLO, respectively, enables cellular-resolution longitudinal imaging of changes in ophthalmic structure and function. We present a non-contact multimodal OCT+cSLO small animal imaging system with extended working distance to the pupil, which enables imaging during and after intraocular injection. While injections are routinely performed in mice to develop novel models of ophthalmic diseases and screen novel therapeutics, the location and volume delivered is not precisely controlled and difficult to reproduce. Animals were imaged using a custom-built OCT engine and scan-head combined with a modified commercial cSLO scan-head. Post-injection imaging showed structural changes associated with retinal puncture, including the injection track, a retinal elevation, and detachment of the posterior hyaloid. When combined with imagesegmentation, we believe OCT can be used to precisely identify injection locations and quantify injection volumes. Fluorescence cSLO can provide complementary contrast for either fluorescently labeled compounds or transgenic cells for improved specificity. Our non-contact OCT+cSLO system is uniquely-suited for concurrent imaging with intraocular injections, which may be used for real-time image-guided injections.
Intraocular injections are routinely performed for delivery of anti-VEGF and anti-inflammatory therapies in humans. While these injections are also performed in mice to develop novel models of ophthalmic diseases and screen novel therapeutics, the injection location and volume are not well-controlled and reproducible. We overcome limitations of conventional injections methods by developing a multimodality, long working distance, non-contact optical coherence tomography (OCT) and fluorescence confocal scanning laser ophthalmoscopy (cSLO) system for retinal imaging before and after injections. Our OCT+cSLO system combines a custom-built spectraldomain OCT engine (875±85 nm) with 125 kHz line-rate with a modified commercial cSLO with a maximum frame-rate of 30 fps (512 x 512 pix.). The system was designed for an overlapping OCT+cSLO field-of-view of 1.1 mm with a 7.76 mm working distance to the pupil. cSLO excitation light sources and filters were optimized for simultaneous GFP and tdTomato imaging. Lateral resolution was 3.02 µm for OCT and 2.74 μm for cSLO. Intravitreal injections of 5%, 10%, and 20% intralipid with Alex Fluor 488 were manually injected intraocularly in C57BL/6 mice. Post-injection imaging showed structural changes associated with retinal puncture, including the injection track, a retinal elevation, and detachment of the posterior hyaloid. OCT enables quantitative analysis of injection location and volumes whereas complementary cSLO improves specificity for identifying fluorescently labeled injected compounds and transgenic cells. The long working distance of our non-contact OCT+cSLO system is uniquely-suited for concurrent imaging with intraocular injections and may be applied for imaging of ophthalmic surgical dynamics and real-time image-guided injections.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.