In central serous chorioretinopathy (CSC), the macula is detached because of fluid leakage at the level of the retinal pigment epithelium. The fluid appears to originate from choroidal vascular hyperpermeability, but the etiology for the fluid is controversial. The choroidal vascular findings as elucidated by recent optical coherence tomography (OCT) and wide-field indocyanine green (ICG) angiographic evaluation show eyes with CSC have many of the same venous patterns that are found in eyes following occlusion of the vortex veins or carotid cavernous sinus fistulas (CCSF). The eyes show delayed choroidal filling, dilated veins, intervortex venous anastomoses, and choroidal vascular hyperpermeability. While patients with occlusion of the vortex veins or CCSF have extraocular abnormalities accounting for the venous outflow problems, eyes with CSC appear to have venous outflow abnormalities as an intrinsic phenomenon. Control of venous outflow from the eye involves a Starling resistor effect, which appears to be abnormal in CSC. Similar choroidal vascular abnormalities have been found in peripapillary pachychoroid syndrome. However, peripapillary pachychoroid syndrome has intervortex venous anastomoses located in the peripapillary region while in CSC these are seen to be located in the macular region. Spaceflight associated neuro-ocular syndrome appears to share many of the pathophysiologic problems of abnormal venous outflow from the choroid along with a host of associated abnormalities. These diseases vary according to their underlying etiologies but are linked by the venous decompensation in the choroid that leads to significant vision loss. Choroidal venous overload provides a unifying concept and theory for an improved understanding of the pathophysiology and classification of a group of diseases to a greater extent than previous proposals.

To benchmark the human and machine performance of spectral-domain (SD) and swept-source (SS) optical coherence tomography (OCT) image segmentation, i.e., pixel-wise classification, for the compartments vitreous, retina, choroid, sclera. A convolutional neural network (CNN) was trained on OCT B-scan images annotated by a senior ground truth expert retina specialist to segment the posterior eye compartments. Independent benchmark data sets (30 SDOCT and 30 SSOCT) were manually segmented by three classes of graders with varying levels of ophthalmic proficiencies. Nine graders contributed to benchmark an additional 60 images in three consecutive runs. Inter-human and intra-human class agreement was measured and compared to the CNN results. The CNN training data consisted of a total of 6210 manually segmented images derived from 2070 B-scans (1046 SDOCT and 1024 SSOCT; 630 C-Scans). The CNN segmentation revealed a high agreement with all grader groups. For all compartments and groups, the mean Intersection over Union (IOU) score of CNN compartmentalization versus group graders’ compartmentalization was higher than the mean score for intra-grader group comparison. The proposed deep learning segmentation algorithm (CNN) for automated eye compartment segmentation in OCT B-scans (SDOCT and SSOCT) is on par with manual segmentations by human graders.

Purpose: To study a novel and fast optical coherence tomography (OCT) device for home-based monitoring in age-related macular degeneration (AMD) in a small sample yielding sparse OCT (spOCT) data and to compare the device to a commercially available reference device. Methods: In this prospective study, both eyes of 31 participants with AMD were included. The subjects underwent scanning with an OCT prototype and a spectral-domain OCT to compare the accuracy of the central retinal thickness (CRT) measurements. Results: Sixty-two eyes in 31 participants (21 females and 10 males) were included. The mean age was 79.6 years (age range, 69–92 years). The mean difference in the CRT measurements between the devices was 4.52 μm (SD ± 20.0 μm; range, −65.6 to 41.5 μm). The inter- and intrarater reliability coefficients of the OCT prototype were both >0.95. The laser power delivered was <0.54 mW for spOCT and <1.4 mW for SDOCT. No adverse events were reported, and the visual acuity before and after the measurements was stable. Conclusion: This study demonstrated the safety and feasibility of this home-based OCT monitoring under real-life conditions, and it provided evidence for the potential clinical benefit of the device. Translational Relevance: The newly developed spOCT is a valid and readily available retina scanner. It could be applied as a portable self-measuring OCT system. Its use may facilitate the sustainable monitoring of chronic retinal diseases by providing easily accessible and continuous retinal monitoring.

Feasibility testing of a novel volume renders technology to display optical coherence tomography data (OCT) in a virtual reality (VR) environment. A VR program was written in C++/OpenGL to import and display volumetric OCT data in real time with 180 frames per second using a high-end computer and a tethered head-mounted display. Following exposure, participants completed a Simulator Sickness Questionnaire (SSQ) to assess for nausea, disorientation, and oculomotor disturbances. A user evaluation study of this software was conducted to explore the potential utility of this application. Fifty-seven subjects completed the user testing (34 males and 23 females). Mean age was 48.5 years (range, 21–77 years). Mean acquired work experience of the 35 ophthalmologists (61.40%) included in the group was 15.46 years (range, 1–37 years). Twenty-nine participants were VR-naïve. The SSQ showed a mean total score of 5.8 (SD = 9.44) indicating that the system was well tolerated and produced minimal side effects. No difference was reported between VR-naïve participants and experienced users. Overall, immersed subjects reported an enjoyable VR-OCT presence effect. A usable and satisfying VR imaging technique was developed to display and interact with original OCT data. An advanced high-end VR image display method was successfully developed to provide new views and interactions in an ultra high-speed projected digital scenery using point-cloud OCT data. This represents the next generation of OCT image display technology and a new tool for patient engagement, medical education, professional training, and telecommunications.

Machine learning has greatly facilitated the analysis of medical data, while the internal operations usually remain intransparent. To better comprehend these opaque procedures, a convolutional neural network for optical coherence tomography image segmentation was enhanced with a Traceable Relevance Explainability (T-REX) technique. The proposed application was based on three components: ground truth generation by multiple graders, calculation of Hamming distances among graders and the machine learning algorithm, as well as a smart data visualization (‘neural recording’). An overall average variability of 1.75% between the human graders and the algorithm was found, slightly minor to 2.02% among human graders. The ambiguity in ground truth had noteworthy impact on machine learning results, which could be visualized. The convolutional neural network balanced between graders and allowed for modifiable predictions dependent on the compartment. Using the proposed T-REX setup, machine learning processes could be rendered more transparent and understandable, possibly leading to optimized applications.

Click on the article title to read more.