MENISCUS MICROPYON: An Ophthalmoscopic Sign Possibly Associated With Epiretinal Proliferation After Retinal Surgery With Gas Tamponade.

PURPOSE: To describe an ophthalmoscopic sign, termed a meniscus micropyon, and its possible association with proliferative vitreoretinopathy/epiretinal membrane (ERM) formation after retinal surgery with gas tamponade. METHODS: Patients with intravitreal gas were examined postoperatively by one of six vitreoretinal surgeons from four institutions. A micropyon was defined as a white-yellow, solid-appearing consolidation along the meniscus (i.e., the fluid-gas interface). RESULTS: A micropyon was visualized and photographed in 49 patients who received intravitreal gas. Preoperatively, retinal breaks were present in all 49 eyes and rhegmatogenous retinal detachment in 45 (92%). Postoperatively, 39 eyes (80%) developed epiretinal proliferation: 16 eyes (33%) developed recurrent rhegmatogenous retinal detachment from proliferative vitreoretinopathy, 6 eyes (12%) re-detached without frank proliferative vitreoretinopathy, 9 eyes (18%) developed postoperative ERM/worsening, and 8 eyes (16%) had postoperative ERM but no preoperative optical coherence tomography to determine if the postoperative ERM was new or worsening. The single-operation anatomical success in eyes with a micropyon was 51%, which was lower than that of a contemporaneous rhegmatogenous retinal detachment control group (91%) in which no micropyon was detected. In two patients, micropyons were biopsied during pars plana vitrectomy and examined histopathologically; they consist predominantly of white blood cells. CONCLUSION: The meniscus micropyon is an ophthalmoscopic sign that can occur after retinal surgery with gas tamponade. Features that distinguish a micropyon from postvitrectomy fibrin/fibrinoid syndrome include delayed appearance, hyperautofluorescence, absence of translucent strands or sheets in the anterior chamber or vitreous cavity, and the histopathologic identification of white blood cells. A clinically detectable micropyon may be a biomarker of proliferative vitreoretinopathy/ERM formation.

In Vivo Visualization of Intravascular Patrolling Immune Cells in the Primate Eye.

Purpose: Insight into the immune status of the living eye is essential as we seek to understand ocular disease and develop new treatments. The nonhuman primate (NHP) is the gold standard preclinical model for therapeutic development in ophthalmology, owing to the similar visual system and immune landscape in the NHP relative to the human. Here, we demonstrate the utility of phase-contrast adaptive optics scanning light ophthalmoscope (AOSLO) to visualize immune cell dynamics on the cellular scale, label-free in the NHP. Methods: Phase-contrast AOSLO was used to image preselected areas of retinal vasculature in five NHP eyes. Images were registered to correct for eye motion, temporally averaged, and analyzed for immune cell activity. Cell counts, dimensions, velocities, and frequency per vessel were determined manually and compared between retinal arterioles and venules. Based on cell appearance and circularity index, cells were divided into three morphologies: ovoid, semicircular, and flattened. Results: Immune cells were observed migrating along vascular endothelium with and against blood flow. Cell velocity did not significantly differ between morphology or vessel type and was independent of blow flood. Venules had a significantly higher cell frequency than arterioles. A higher proportion of cells resembled "flattened" morphology in arterioles. Based on cell speeds, morphologies, and behaviors, we identified these cells as nonclassical patrolling monocytes (NCPMs). Conclusions: Phase-contrast AOSLO has the potential to reveal the once hidden behaviors of single immune cells in retinal circulation and can do so without the requirement of added contrast agents that may disrupt immune cell behavior.

Are the Hypo-Reflective Clumps Associated With Age-Related Macular Degeneration in Adaptive Optics Ophthalmoscopy Autofluorescent?

Purpose: Hypo-reflective clumps (HRCs) are structures associated with age-related macular degeneration (AMD) that were identified using flood-illumination adaptive optics ophthalmoscopy (FIAO) and hypothesized to be either macrophages that have accumulated melanin through the phagocytosis of retinal pigmented epithelial (RPE) cell organelles or transdifferentiated RPE cells. HRCs may be autofluorescent (AF) in the near infrared (NIR) but clinical NIR autofluorescence imaging lacks the resolution to answer this question definitively. Here, we used near infrared autofluorescence (NIRAF) imaging in fluorescence adaptive optics scanning laser ophthalmoscopy (AOSLO) to determine whether HRCs are AF. Methods: Patients with AMD and HRCs underwent imaging with FIAO, optical coherence tomography (OCT), and multi-modal AOSLO (confocal, NIRAF, and non-confocal multi-offset detection using a fiber bundle). HRCs were segmented on FIAO and images, co-registered across modalities, and HRC morphometry and AF were quantified. Results: Eight patients participated (mean age = 79 years, standard deviation [SD] = 5.7, range = 69-89 years, and 5 female patients). Most HRCs (86%, n = 153/178) were autofluorescent on AOSLO. HRC AF signal varied but most uniformly dark HRCs on FIAO showed corresponding AF on AOSLO, whereas heterogeneous HRCs showed a smaller AF area or no AF. Conclusions: These findings are consistent with the hypothesis that HRCs contain AF RPE organelles. A small proportion of HRCs were not AF; these may represent macrophages that have not yet accumulated enough organelles to become AF. HRCs may have clinical significance but further study is needed to understand the interplay among HRCs, RPE cells, and macrophages, and their relationship to geographic atrophy (GA) progression in AMD.

Multimodal High-Resolution Imaging in Retinitis Pigmentosa: A Comparison Between Optoretinography, Cone Density, and Visual Sensitivity.

Purpose: Retinitis pigmentosa (RP), the most common inherited retinal disease, is characterized by progressive photoreceptor degeneration. It remains unknown to what extent surviving photoreceptors transduce light and support vision in RP. To address this, we correlated structure and functional measures using adaptive optics scanning laser ophthalmoscopy (AOSLO), adaptive optics microperimetry, and adaptive optics optical coherence tomography (AO-OCT)-based optoretinograms (ORGs). Methods: Four patients with RP were imaged with AOSLO across the visual field covering the transition zone (TZ) of normal to diseased retina. Cone density was estimated in discrete regions spanning the TZ. Visual sensitivity was assessed by measuring increment thresholds for a 3-arcmin stimulus targeted via active eye tracking in AOSLO. ORGs were measured at the same locations using AO-OCT to assess the cones' functional response to a 528 ± 20-nm stimulus. Individual cone outer segment (COS) lengths were measured from AO-OCT in each subject. Results: Cone density was significantly reduced in patients with RP. Density reduction correlated with TZ location in 3 patients with RP, while a fourth had patches of reduced density throughout the retina. ORG amplitude was reduced in regions of normal and reduced cone density in all patients with RP. ORG response and COS length were positively correlated in controls but not in patients with RP. Despite deficits in cone density and ORG, visual sensitivity remained comparable to controls in three of four patients with RP. Conclusions: ORG-based measures of retinal dysfunction may precede deficits in cone structure and visual sensitivity. ORG is a sensitive measure of RP disease status and has significant potential to provide insight into disease progression and treatment efficacy.

ADAPTIVE OPTICS AND MULTIMODAL IMAGING FOR INFLAMMATORY VITREORETINAL INTERFACE ABNORMALITIES.

PURPOSE: To investigate changes to the vitreoretinal interface in uveitis with multimodal imaging including adaptive optics. METHODS: Four eyes (four patients) affected by fovea-attached (subtype 1A) or fovea-sparing epiretinal membranes (ERMs) on spectral-domain optical coherence tomography or visible internal limiting membrane (ILM) on infrared scanning laser ophthalmoscope (SLO) fundus imaging were recruited in this pilot study. The microstructure of the vitreoretinal interface was imaged using flood-illumination adaptive optics (FIAO), and the images were compared with the cross-sectional spectral-domain optical coherence tomography data. RESULTS: Adaptive optics images revealed multiple abnormalities of the vitreoretinal interface, such as deep linear striae in ERM, and hyperreflective microstructures at the location of ERMs and ILMs. The cone mosaic was imaged by FIAO and was found altered in the four eyes with ERMs or visible ILM. The same four eyes presented alteration of photopic 30 Hz flicker that was reduced in amplitude indicating cone inner retinal layer dysfunction. CONCLUSION: FIAO imaging can identify specific patterns associated with ERMs and ILMs. Correlating FIAO imaging of the vitreomacular interface with the structural alterations seen in FIAO at the level of the outer retinal structures can help understand the cause of significant macular dysfunction associated with ERM.

Eyesi direct ophthalmoscope simulator: an effective training tool for medical undergraduates.

INTRODUCTION: Non-ophthalmologists often lack sufficient operational training to use a direct ophthalmoscope proficiently, resulting in a global deficit of basic ophthalmological skills among general practitioners. This deficiency hampers the timely diagnosis, referral, and intervention of patients. Consequently, the optimization of teaching tools and methods to enhance teaching efficiency is imperative. This study explores the effectiveness of the Eyesi Direct Ophthalmoscope Simulator (Eyesi) as an innovative tool for fundus examination training. METHODS: Medical undergraduates were randomly assigned to Group A or B (n = 168). All participants completed a pre-training questionnaire. Group A received Eyesi training, while Group B underwent traditional direct ophthalmoscope (TDO) training. Subsequently, participants answered questionnaires relevant to their respective training methods. Both groups exchanged training tools and completed a summary questionnaire. RESULTS: After training, 54.17% of participants believed that images presented by the Eyesi were consistent with the real fundus. Group A scored significantly higher than Group B in fundus structure recognition and self-confidence in examination. The degree of mastery over fundus theory score increased from 6.10 ± 0.13 to 7.74 ± 0.16 (P < 0.001) in Group A, but Group B did not demonstrate a significant difference. We also compared undergraduates' tendencies for different learning purposes, 75.59% of participants preferred the Eyesi to TDO as a training tool, and 88.41% of participants were receptive to introducing the Eyesi in training. CONCLUSION: According to subjective participant feedback, Eyesi outperformed TDO in fundus observation, operational practice, and theoretical learning. It effectively equips undergraduates with fundus examination skills, potentially promoting the use of direct ophthalmoscopes in primary medical institutions.

Multispectral label-free in vivo cellular imaging of human retinal pigment epithelium using adaptive optics fluorescence lifetime ophthalmoscopy improves feasibility for low emission analysis and increases sensitivity for detecting changes with age and eccentricity.

Significance: Adaptive optics fluorescence lifetime ophthalmoscopy (AOFLIO) provides a label-free approach to observe functional and molecular changes at cellular scale in vivo. Adding multispectral capabilities improves interpretation of lifetime fluctuations due to individual fluorophores in the retinal pigment epithelium (RPE). Aim: To quantify the cellular-scale changes in autofluorescence with age and eccentricity due to variations in lipofuscin, melanin, and melanolipofuscin in RPE using multispectral AOFLIO. Approach: AOFLIO was performed on six subjects at seven eccentricities. Four imaging channels ( λ ex / λ em ) were used: 473/SSC, 473/LSC, 532/LSC, and 765/NIR. Cells were segmented and the timing signals of each pixel in a cell were combined into a single histogram, which were then used to compute the lifetime and phasor parameters. An ANOVA was performed to investigate eccentricity and spectral effects on each parameter. Results: A repeatability analysis revealed < 11.8 % change in lifetime parameters in repeat visits for 532/LSC. The 765/NIR and 532/LSC had eccentricity and age effects similar to previous reports. The 473/LSC had a change in eccentricity with mean lifetime and a phasor component. Both the 473/LSC and 473/SSC had changes in eccentricity in the short lifetime component and its relative contribution. The 473/SSC had no trend in eccentricity in phasor. The comparison across the four channels showed differences in lifetime and phasor parameters. Conclusions: Multispectral AOFLIO can provide a more comprehensive picture of changes with age and eccentricity. These results indicate that cell segmentation has the potential to allow investigations in low-photon scenarios such as in older or diseased subjects with the co-capture of an NIR channel (such as 765/NIR) with the desired spectral channel. This work represents the first multispectral, cellular-scale fluorescence lifetime comparison in vivo in the human RPE and may be a useful method for tracking diseases.

The Versatile Teaching Eye: an affordable, 3D-printed model eye for simulating ophthalmic examination.

Purpose: To describe the Versatile Teaching Eye (VT Eye), a 3D-printed model eye designed to provide an affordable examination simulator, and to report the results of a pilot program introducing the VT Eye and an ophthalmic training curriculum at a teaching hospital in Ghana. Methods: TinkerCAD was used to design the VT Eye, which was printed with ABS plastic. The design features an adapter that permits use of a smartphone as a digital fundus. We developed a set of digital flashcards allowing for an interactive review of a range of retinal pathologies. An analog fundus was developed for practicing traditional slit lamp and indirect examinations as well as retinal laser practice. The model was used for a period of 2 weeks by ophthalmic trainees at Komfo Anokye Teaching Hospital, Kumasi, Ghana, to practice indirect ophthalmoscopy, slit lamp biomicroscopy, smartphone funduscopy, and retinal image drawing. Results were assessed at by means of a pre-/post-training survey of 6 residents. Results: The VT Eye accommodates diverse fundus examination techniques. Its 3D-printed design ensures cost-effective, high-quality replication. When paired with a 20 D practice examination lens, the digital fundus provides a comprehensive, interactive training environment for <$30.00 (USD). This device allows for indirect examination practice without requiring an indirect headset, which may increase the amount of available practice for trainees early in their careers. In the Ghana pilot program, the model's use in indirect examination training sessions significantly boosted residents' confidence in various examination techniques. Comparing pre- and post-session ratings, average reported confidence levels rose by 30% for acquiring clear views of the posterior pole, 42% for visualizing the periphery, and 141% for capturing important pathology using personal smartphones combined with a 20 D lens (all P < 0.05). Conclusions: The VT Eye is readily reproducible and can be easily integrated into ophthalmic training curricula, even in regions with limited resources. It offers an effective and affordable training solution, underscoring its potential for global adoption and the benefits of incorporating innovative technologies in medical education.

Assessment of local sensitivity in incomplete retinal pigment epithelium and outer retinal atrophy (iRORA) lesions in intermediate age-related macular degeneration (iAMD).

Lesions of incomplete retinal pigment epithelium and outer retinal atrophy (iRORA) are associated with disease progression in age-related macular degeneration. However, the corresponding functional impact of these precursor lesions is unknown.We present a cross-sectional study of four patients employing clinical-grade MAIA (stimulus size: 0.43°, ~125 µm) and adaptive optics scanning light ophthalmoscope (AOSLO, stimulus size 0.07°, ~20 µm) based microperimetry (MP) to assess the specific impact of iRORA lesions on retinal sensitivity.AOSLO imaging showed overall reduced photoreceptor reflectivity and patches of hyporeflective regions at drusen with interspersed hyper-reflective foci in iRORA regions. MAIA-MP yielded an average retinal sensitivity loss of -7.3±3.1 dB at iRORA lesions compared with the in-eye control. With AOSLO-MP, the corresponding sensitivity loss was 20.1±4.8 dB.We demonstrated that iRORA lesions are associated with a severe impairment in retinal sensitivity. Larger cohort studies will be necessary to validate our findings.

SLO-Net: Enhancing Multiple Sclerosis Diagnosis Beyond Optical Coherence Tomography Using Infrared Reflectance Scanning Laser Ophthalmoscopy Images.

Purpose: Several machine learning studies have used optical coherence tomography (OCT) for multiple sclerosis (MS) classification with promising outcomes. Infrared reflectance scanning laser ophthalmoscopy (IR-SLO) captures high-resolution fundus images, commonly combined with OCT for fixed B-scan positions. However, no machine learning research has utilized IR-SLO images for automated MS diagnosis. Methods: This study utilized a dataset comprised of IR-SLO images and OCT data from Isfahan, Iran, encompassing 32 MS and 70 healthy individuals. A number of convolutional neural networks (CNNs)-namely, VGG-16, VGG-19, ResNet-50, ResNet-101, and a custom architecture-were trained with both IR-SLO images and OCT thickness maps as two separate input datasets. The highest performing models for each modality were then integrated to create a bimodal model that receives the combination of OCT thickness maps and IR-SLO images. Subject-wise data splitting was employed to prevent data leakage among training, validation, and testing sets. Results: Overall, images of the 102 patients from the internal dataset were divided into test, validation, and training subsets. Subsequently, we employed a bootstrapping approach on the training data through iterative sampling with replacement. The performance of the proposed bimodal model was evaluated on the internal test dataset, demonstrating an accuracy of 92.40% ± 4.1% (95% confidence interval [CI], 83.61-98.08), sensitivity of 95.43% ± 5.75% (95% CI, 83.71-100.0), specificity of 92.82% ± 3.72% (95% CI, 81.15-96.77), area under the receiver operating characteristic (AUROC) curve of 96.99% ± 2.99% (95% CI, 86.11-99.78), and area under the precision-recall curve (AUPRC) of 97.27% ± 2.94% (95% CI, 86.83-99.83). Furthermore, to assess the model generalization ability, we examined its performance on an external test dataset following the same bootstrap methodology, achieving promising results, with accuracy of 85.43% ± 0.08% (95% CI, 71.43-100.0), sensitivity of 97.33% ± 0.06% (95% CI, 83.33-100.0), specificity of 84.6% ± 0.10% (95% CI, 71.43-100.0), AUROC curve of 99.67% ± 0.02% (95% CI, 95.63-100.0), and AUPRC of 99.65% ± 0.02% (95% CI, 94.90-100.0). Conclusions: Incorporating both modalities improves the performance of automated diagnosis of MS, showcasing the potential of utilizing IR-SLO as a complementary tool alongside OCT. Translational Relevance: Should the results of our proposed bimodal model be validated in future work with larger and more diverse datasets, diagnosis of MS based on both OCT and IR-SLO can be reliably integrated into routine clinical practice.

Smartphone fundoscopy with 20 dioptres lens: our experience.

Objective: Assessment of the utility of smartphone fundoscopy in diagnosing posterior pole pathologies. Methods: An iPhone 12 and a 20D Volk lens were used for smartphone fundoscopy. Patients needing bedside consultation were examined with direct ophthalmoscopy and smartphone fundoscopy. Some patients were examined with this technique after slit lamp examination. Results: Over one year 23 bedside fundus examinations were performed and 2 papilledema were diagnosed. After initial slit lamp examination, photos of various pathologies were taken: age-related macular degeneration, branch retinal artery occlusion, arterial embolus, branch retinal vein occlusion, non-arteritic anterior ischemic optic neuropathy, myelinated retinal nerve fiber layer, choroidal naevus. Discussion: With the 20D lens, the image is overturned, magnified 3,13X, and the field of view is 46°. The utility was demonstrated in literature by teaching students this technique and using it in screening for retinopathy of prematurity. The weighted retinal irradiance was measured in two studies. It was 4,6 mW/cm2 in one and from 0,58 to 2,30 mW/cm2 in the other, within safe limits. Conclusions: Smartphone fundoscopy is a fast, accessible, and safe technique for fundus examinations. Other departments could use it for the diagnosis of papilledema.

Foveolar Drusen Decrease Fixation Stability in Pre-Symptomatic AMD.

Purpose: This study aims at linking subtle changes of fixational eye movements (FEM) in controls and in patients with foveal drusen using adaptive optics retinal imaging in order to find anatomo-functional markers for pre-symptomatic age-related macular degeneration (AMD). Methods: We recruited 7 young controls, 4 older controls, and 16 patients with presymptomatic AMD with foveal drusen from the Silversight Cohort. A high-speed research-grade adaptive optics flood illumination ophthalmoscope (AO-FIO) was used for monocular retinal tracking of fixational eye movements. The system allows for sub-arcminute resolution, and high-speed and distortion-free imaging of the foveal area. Foveal drusen position and size were documented using gaze-dependent imaging on a clinical-grade AO-FIO. Results: FEM were measured with high precision (RMS-S2S = 0.0015 degrees on human eyes) and small foveal drusen (median diameter = 60 µm) were detected with high contrast imaging. Microsaccade amplitude, drift diffusion coefficient, and ISOline area (ISOA) were significantly larger for patients with foveal drusen compared with controls. Among the drusen participants, microsaccade amplitude was correlated to drusen eccentricity from the center of the fovea. Conclusions: A novel high-speed high-precision retinal tracking technique allowed for the characterization of FEM at the microscopic level. Foveal drusen altered fixation stability, resulting in compensatory FEM changes. Particularly, drusen at the foveolar level seemed to have a stronger impact on microsaccade amplitudes and ISOA. The unexpected anatomo-functional link between small foveal drusen and fixation stability opens up a new perspective of detecting oculomotor signatures of eye diseases at the presymptomatic stage.

Modeling Human Macular Cone Photoreceptor Spatial Distribution.

Purpose: The purpose of this study was to investigate the spatial distribution of human cone photoreceptors and examine cone density differences between the retinal meridians and quadrants. Method: Using adaptive optics scanning laser ophthalmoscopy, the maculae were imaged in 17 eyes of 11 subjects with normal chorioretinal health aged 54 to 72 years. We measured cone density at 325 points within the central 10 degrees radius of the retina. Cone density spatial distributions along the primary retinal meridians and in four macular quadrants (superior-nasal, superior-temporal, inferior-temporal, and inferior-nasal) were analytically modeled using the polynomial function to assess the meridional and quadrantal difference. Results: The mean and 95% confidence interval for the prediction of cone density along the primary retinal meridians was modeled with a 7-degree one-variable polynomial (R2 = 0.9761, root mean squared error [RMSE] = 0.0585). In the 4 retinal quadrants, cone density distribution was described by a 2-variable polynomial with X degree 3 and Y degree 4 (R² = 0.9834, RMSE = 0.0377). The models suggest no statistically significant difference between medians and between quadrants. However, cone density difference at corresponding spatial locations in different areas can be up to 25.6%. The superior-nasal region has more areas with high cone density, followed by quadrants of inferior-nasal, inferior-temporal, and superior-temporal. Conclusions: Analytical modeling provides comprehensive knowledge of cone distribution across the entire macula. Although modeling analysis suggests no statistically significant difference between medians and between quadrants, the remarkable cone density discrepancies in certain regions should be accounted for in applications requiring sensitive detection of cone variation.

Retinal peri-arteriolar versus peri-venular amyloidosis, hippocampal atrophy, and cognitive impairment: exploratory trial.

The relationship between amyloidosis and vasculature in cognitive impairment and Alzheimer's disease (AD) pathogenesis is increasingly acknowledged. We conducted a quantitative and topographic assessment of retinal perivascular amyloid plaque (AP) distribution in individuals with both normal and impaired cognition. Using a retrospective dataset of scanning laser ophthalmoscopy fluorescence images from twenty-eight subjects with varying cognitive states, we developed a novel image processing method to examine retinal peri-arteriolar and peri-venular curcumin-positive AP burden. We further correlated retinal perivascular amyloidosis with neuroimaging measures and neurocognitive scores. Our study unveiled that peri-arteriolar AP counts surpassed peri-venular counts throughout the entire cohort (P < 0.0001), irrespective of the primary, secondary, or tertiary vascular branch location, with a notable increase among cognitively impaired individuals. Moreover, secondary branch peri-venular AP count was elevated in the cognitively impaired (P < 0.01). Significantly, peri-venular AP count, particularly in secondary and tertiary venules, exhibited a strong correlation with clinical dementia rating, Montreal cognitive assessment score, hippocampal volume, and white matter hyperintensity count. In conclusion, our exploratory analysis detected greater peri-arteriolar versus peri-venular amyloidosis and a marked elevation of amyloid deposition in secondary branch peri-venular regions among cognitively impaired subjects. These findings underscore the potential feasibility of retinal perivascular amyloid imaging in predicting cognitive decline and AD progression. Larger longitudinal studies encompassing diverse populations and AD-biomarker confirmation are warranted to delineate the temporal-spatial dynamics of retinal perivascular amyloid deposition in cognitive impairment and the AD continuum.

Intervisit Reproducibility of Foveal Cone Density Metrics.

Purpose: To assess longitudinal reproducibility of metrics of foveal density (peak cone density [PCD], cone density centroid [CDC], and 80th percentile centroid area) in participants with normal vision. Methods: Participants (n = 19; five male and 14 female) were imaged at two time points (average interval of 3.2 years) using an adaptive optics scanning light ophthalmoscope (AOSLO). Foveally centered regions of interest (ROIs) were extracted from AOSLO montages. Cone coordinate matrices were semiautomatically derived for each ROI, and cone mosaic metrics were calculated. Results: On average, there were no significant changes in cone mosaic metrics between visits. The average ± SD PCD was 187,000 ± 20,000 cones/mm2 and 189,000 ± 21,700 cones/mm2 for visits 1 and 2, respectively (P = 0.52). The average ± SD density at the CDC was 183,000 ± 19,000 cones/mm2 and 184,000 ± 20,800 cones/mm2 for visits 1 and 2, respectively (P = 0.78). The average ± SD 80th percentile isodensity contour area was 15,400 ± 1800 µm2 and 15,600 ± 1910 µm2 for visits 1 and 2, respectively (P = 0.57). Conclusions: Foveal cone mosaic density metrics were highly reproducible in the cohort examined here, although further study is required in more diverse populations. Translational Relevance: Determination of the normative longitudinal changes in foveal cone topography is key for evaluating longitudinal measures of foveal cone topography in patients with progressive retinal dystrophies.

Discrimination of multiple sclerosis using scanning laser ophthalmoscopy images with autoencoder-based feature extraction.

OBJECTIVE: Optical coherence tomography (OCT) investigations have revealed that the thickness of inner retinal layers becomes decreased in multiple sclerosis (MS) patients, compared to healthy control (HC) individuals. To date, a number of studies have applied machine learning to OCT thickness measurements, aiming to enable accurate and automated diagnosis of the disease. However, there have much less emphasis on other less common retinal imaging modalities, like infrared scanning laser ophthalmoscopy (IR-SLO), for classifying MS. IR-SLO uses laser light to capture high-resolution fundus images, often performed in conjunction with OCT to lock B-scans at a fixed position. METHODS: We incorporated two independent datasets of IR-SLO images from the Isfahan and Johns Hopkins centers, consisting of 164 MS and 150 HC images. A subject-wise data splitting approach was employed to ensure that there was no leakage between training and test datasets. Several state-of-the-art convolutional neural networks (CNNs), including VGG-16, VGG-19, ResNet-50, and InceptionV3, and a CNN with a custom architecture were employed. In the next step, we designed a convolutional autoencoder (CAE) to extract semantic features subsequently given as inputs to four conventional ML classifiers, including support vector machine (SVM), k-nearest neighbor (K-NN), random forest (RF), and multi-layer perceptron (MLP). RESULTS: The custom CNN (85 % accuracy, 85 % sensitivity, 87 % specificity, 93 % area under the receiver operating characteristics [AUROC], and 94 % area under the precision-recall curve [AUPRC]) outperformed state-of-the-art models (84 % accuracy, 83 % sensitivity, 87 % specificity, 92 % AUROC, and 94 % AUPRC); however, utilizing a combination of the CAE and MLP yields even superior results (88 % accuracy, 86 % sensitivity, 91 % specificity, 94 % AUROC, and 95 % AUPRC). CONCLUSIONS: We utilized IR-SLO images to differentiate between MS and HC eyes, with promising results achieved using a combination of CAE and MLP. Future multi-center studies involving more heterogenous data are necessary to assess the feasibility of integrating IR-SLO images into routine clinical practice.

Unique yellow shifts for small and brief stimuli in the central retina.

The spectral locus of unique yellow was determined for flashes of different sizes (<11 arcmin) and durations (<500 ms) presented in and near the fovea. An adaptive optics scanning laser ophthalmoscope was used to minimize the effects of higher-order aberrations during simultaneous stimulus delivery and retinal imaging. In certain subjects, parafoveal cones were classified as L, M, or S, which permitted the comparison of unique yellow measurements with variations in local L/M ratios within and between observers. Unique yellow shifted to longer wavelengths as stimulus size or duration was reduced. This effect is most pronounced for changes in size and more apparent in the fovea than in the parafovea. The observed variations in unique yellow are not entirely predicted from variations in L/M ratio and therefore implicate neural processes beyond photoreception.

Detection of capillary abnormalities in early diabetic retinopathy using scanning laser ophthalmoscopy and optical coherence tomography combined with adaptive optics.

This study tested if a high-resolution, multi-modal, multi-scale retinal imaging instrument can provide novel information about structural abnormalities in vivo. The study examined 11 patients with very mild to moderate non-proliferative diabetic retinopathy (NPDR) and 10 healthy subjects using fundus photography, optical coherence tomography (OCT), OCT angiography (OCTA), adaptive optics scanning laser ophthalmoscopy (AO-SLO), adaptive optics OCT and OCTA (AO-OCT(A)). Of 21 eyes of 11 patients, 11 had very mild NPDR, 8 had mild NPDR, 2 had moderate NPDR, and 1 had no retinopathy. Using AO-SLO, capillary looping, inflections and dilations were detected in 8 patients with very mild or mild NPDR, and microaneurysms containing hyperreflective granular elements were visible in 9 patients with mild or moderate NPDR. Most of the abnormalities were seen to be perfused in the corresponding OCTA scans while a few capillary loops appeared to be occluded or perfused at a non-detectable flow rate, possibly because of hypoperfusion. In one patient with moderate NPDR, non-perfused capillaries, also called ghost vessels, were identified by alignment of corresponding en face AO-OCT and AO-OCTA images. The combination of multiple non-invasive imaging methods could identify prominent microscopic abnormalities in diabetic retinopathy earlier and more detailed than conventional fundus imaging devices.

Unveiling the clinical incapabilities: a benchmarking study of GPT-4V(ision) for ophthalmic multimodal image analysis.

PURPOSE: To evaluate the capabilities and incapabilities of a GPT-4V(ision)-based chatbot in interpreting ocular multimodal images. METHODS: We developed a digital ophthalmologist app using GPT-4V and evaluated its performance with a dataset (60 images, 60 ophthalmic conditions, 6 modalities) that included slit-lamp, scanning laser ophthalmoscopy, fundus photography of the posterior pole (FPP), optical coherence tomography, fundus fluorescein angiography and ocular ultrasound images. The chatbot was tested with ten open-ended questions per image, covering examination identification, lesion detection, diagnosis and decision support. The responses were manually assessed for accuracy, usability, safety and diagnosis repeatability. Auto-evaluation was performed using sentence similarity and GPT-4-based auto-evaluation. RESULTS: Out of 600 responses, 30.6% were accurate, 21.5% were highly usable and 55.6% were deemed as no harm. GPT-4V performed best with slit-lamp images, with 42.0%, 38.5% and 68.5% of the responses being accurate, highly usable and no harm, respectively. However, its performance was weaker in FPP images, with only 13.7%, 3.7% and 38.5% in the same categories. GPT-4V correctly identified 95.6% of the imaging modalities and showed varying accuracies in lesion identification (25.6%), diagnosis (16.1%) and decision support (24.0%). The overall repeatability of GPT-4V in diagnosing ocular images was 63.3% (38/60). The overall sentence similarity between responses generated by GPT-4V and human answers is 55.5%, with Spearman correlations of 0.569 for accuracy and 0.576 for usability. CONCLUSION: GPT-4V currently is not yet suitable for clinical decision-making in ophthalmology. Our study serves as a benchmark for enhancing ophthalmic multimodal models.