Confocal Scanning Laser Ophthalmoscopy Imaging Research Articles

Large animal model species in pluripotent stem cell therapy research and development for retinal diseases: a systematic review.

Retinal cell therapy modalities, in the category of advanced therapy medicinal products (ATMPs), are being developed to target several retinal diseases. Testing in large animal models (LAMs) is a crucial step in translating retinal ATMPs into clinical practice. However, challenges including budgetary and infrastructure constraints can hinder LAM research design and execution. Here, to facilitate the comparison of the various LAMs in pluripotent retinal cell therapy research, we aimed to systematically evaluate the species distribution, reported scientific utility, and methodology of a range of LAMs. A systematic search using the words retina, stem cell, transplantation, large animal, pig, rabbit, dog, and nonhuman primate was conducted in the PubMed, Embase, Science Direct and GoogleScholar databases in February 2023. We included 22 studies involving pluripotent stem cells (induced pluripotent stem cells or human embryonic stem cells) in LAMs, including non-human primates (NHP), pigs, dogs, and rabbits. Nearly half of the studies utilized wild-type animal models. In other studies, retinal degeneration features were simulated via laser, chemical, or genetic insult. Transplants were delivered subretinally, either as cell suspensions or pre-formed monolayers (with or without biodegradable scaffolding). The transplanted cells dose per eye varied widely (40,000 - 4,000,000 per dose). Cells were delivered via vitrectomy surgery in 15 studies and by an "ab externo" approach in one study. Structural outcomes were assessed using confocal scanning laser ophthalmoscopy imaging. Functional outcomes included multifocal electroretinogram and, in one case, a measure of visual acuity. Generally, cell suspension transplants exhibited low intraretinal incorporation, while monolayer transplants incorporated more efficiently. Immune responses posed challenges for allogeneic transplants, suggesting that autologous iPSC-derived transplants may be required to decrease the likelihood of rejection. The use of appropriate LAMs helps to advance the development of retinal ATMPs. The anatomical similarity of LAM and human eyes allows the implementation of clinically-relevant surgical techniques. While the FDA Modernization Act 2.0 has provided a framework to consider alternative methods including tissue-on-a-chip and human cell culture models for pharmacologic studies, LAM testing remains useful for cell and tissue replacement studies to inform the development of clinical trial protocols.

Imaging the Area of Internal Limiting Membrane Peeling after Macular Hole Surgery.

Background: The aim of this study was to compare en-face optical coherence tomography (OCT) imaging and confocal scanning laser ophthalmoscopy (cSLO) imaging at different wavelengths to identify the internal limiting membrane (ILM) peeling area after primary surgery with vitrectomy and ILM peeling for macular hole (MH). Methods: In total, 50 eyes of 50 consecutive patients who underwent primary surgery with vitrectomy and ILM peeling for MH were studied. The true ILM rhexis based on intraoperative color fundus photography was compared to the presumed ILM rhexis identified by a blinded examiner using en-face OCT imaging and cSLO images at various wavelengths. To calculate the fraction of overlap (FoO), the common intersecting area and the total of both areas were measured. Results: The FoO for the measured areas was 0.93 ± 0.03 for en-face OCT, 0.76 ± 0.06 for blue reflectance (BR; 488 nm), 0.71 ± 0.09 for green reflectance (GR; 514 nm), 0.56 ± 0.07 for infrared reflectance (IR; 815 nm) and 0.73 ± 0.06 for multispectral (MS). The FoO in the en-face OCT group was significantly higher than in all other groups, whereas the FoO in the IR group was significantly lower compared to all other groups. No significant differences were observed in FoO among the MS, BR, and GR groups. In en-face OCT, there was no significant change in the ILM peeled area measured intraoperatively and postoperatively (8.37 ± 3.01 vs. 8.24 ± 2.81 mm2; p = 0.8145). Nasal-inferior foveal displacement was observed in 38 eyes (76%). Conclusions: En-face OCT imaging demonstrates reliable postoperative visualization of the ILM peeled area. Although the size of the ILM peeling remains stable after one month, our findings indicate a notable inferior-nasal shift of the overall ILM peeling area towards the optic disc.

Application of artificial intelligence in glaucoma

Glaucoma is the mainly cause of irreversible blindness characterized with structural changes of optic nerve head, retinal ganglion cell death and loss of visual acuity, with estimated 64.3 million patients aged from 40 to 80 worldwide. When patients seek medical advice because of visual acuity, glaucoma often has advanced to late stage. One major challenge is to identify large number of undiagnosed patients. The reason why screening programs are not employed regularly is that misdiagnoses can be made due to large amounts of false positives in the procedure.Artificial intelligence (AI) has brought new breakthroughs for automated screening for glaucoma，which can perform a task by classifying and identifying data without human intervention, including supervised and unsupervised machine learning two categories. Due to the limitation of manually designed features and disability in detection for high dimensionality of optic disc images, advances in techniques for evaluation of glaucoma came with the use of segmentation‐free approaches. We reported a deep learning system and assessed its generalizability in various data sets, with similar patterns in sensitivity (82.2%–96.1%) and specificity (70.4%–97.1%).Visual field data presents low dimensionality and high noise compared to optic disc images, whose hidden patterns can be detected by unsupervised learning better. Two most reported unsupervised algorithms were clustering and component analysis. Such technologies adopted in glaucoma screening have significant potential to increase accessibility in remote areas and decrease the cost for widely screening. Further studies similarly demonstrated that machine learning can perform comparable to or over human experts in mean deviation, pattern standard deviation and glaucoma hemifield test. A back‐propagation neural network was even reported to successfully detect visual field progression with Area Under Curve (AUC) of 0.92. Further advances were also made through algorithms based on CNN showing both higher sensitivity and specificity than traditional machine learning method.In terms of optical coherence tomography (OCT), early studies based on MLC using TD‐OCT showed an accuracy not worse than non‐artificial intelligence analysis. The latest OCT technologies, including SD‐OCT and SS‐OCT, combined with deep learning are reported more sensitive for early glaucoma detection, with a high AUC up to 0.937. Recently a study reported the ability of their model to detect RNFL thickness from OCT with the AUC of 0.944. Recent years, more parameters have been put in including OCT RNFL thickness, standard automated perimetry and confocal scanning laser ophthalmoscopy imaging, making an improvement in glaucoma detection.However, there remains several limitations and further studies is required. First, the exact mechanism how DL algorithms evaluate features and make predictions remains unknown, which is called ‘black boxes’. Another limitation is that it can be difficult for DL to classify glaucoma in eyes with less severe disease manifestations or multiple comorbid eye conditions, especially high myopia, which asks for a large images database. Furthermore, in order to develop a more dependent screening method, other clinical parameters including intraocular pressure and central corneal thickness should be integrated, and application and validation of such in a real‐world screening need additional research to support.

What artificial intelligence can bring us in the diagnosis and progression of glaucoma

Aims/Purpose: Our goal is to focus on the contribution of artificial intelligence in glaucoma from diagnosis to progression.Methods: By a literature review, we will review work using AI in the field of glaucoma, whether for screening, diagnosis or detection of progression.Results: For the diagnosis: Recently, two papers were published reporting results of DL algorithms to diagnose glaucoma from Visual Field data. Several authors have evaluated AI‐based fundus photograph analysis for its utility for detecting glaucoma such as the study of Liu et al which grouped 241 032 images of 68 013 patients with a sensitivity of 96.2% and specificity of 87.2% also for analysing OCT imaging data from peripapillary RNFL thickness maps and the macular ganglion cell complex for discriminating between glaucomatous and normal eyes with AROC values ranging from 0.69 to 0.99. Prediction of IOP trends from previous data and medications would be a useful and plausible use of AI. For the progression of glaucoma: Multiclass support vector machines (SVMs), have been used to simultaneously discriminate between normal, non progressing, and progressing eyes through the analysis of confocal scanning laser ophthalmoscopy (CSLO) images with a correct classification rate of 88% .Conclusions: Many AI strategies have shown promising results for glaucoma detection using fundus photography, optical coherence tomography, or automated perimetry. The combination of these imaging modalities increases the performance of AI algorithms, with results comparable to those of humans. Research in the coming years will need to address unavoidable questions regarding the clinical significance of such results and the explicability of the predictions.

Retinal Vascular Geometry in Hypertension: cSLO-Based Method.

We aim to introduce a method using confocal scanning laser ophthalmoscopy (cSLO) images for measuring retinal vascular geometry, including vessel branch angle (BA), vessel diameter, vessel tortuosity, and fractal dimension (Df), and to elucidate the relationship between hypertension and these metrics. A total of 119 participants (119 eyes) were enrolled, among which 72 were normotensive and 47 were hypertensive. Infrared cSLO images were extracted from the circular scan around the optics disc using a commercial cSLO + optical coherence tomography instrument. Preprocessed cSLO images were further analyzed using the appropriate tool/macro/plugin of ImageJ. Intraclass correlation coefficients of selected methods used for conducting the cSLO-based geometric analyses were all higher than 0.80. Arterial/arteriolar BA, arteriolar vessel diameter, and total Df in normotensive subjects were 85.80 ± 7.79°, 116.80 ± 12.58μm, and 1.430 ± 0.037, respectively, significantly higher than those of hypertensive subjects (82.13 ± 10.83°, 108.2 ± 11.12μm, and 1.361 ± 0.044, all P < 0.05). The aforementioned metrics remained negatively correlated with hypertension even after adjusting for age alone or age and gender (P < 0.05). However, the difference between arteriolar tortuosity and all studied venous/venular geometric parameters in both subjects was insignificant (all P > 0.05). Proposed cSLO-based methods for assessing various vascular geometric parameters were highly repeatable and reproducible. Arterial/arteriolar BA, arteriolar vessel diameter, and total Df were retinal vascular parameters significantly correlated with hypertension in a negative manner.

Large Animal Models of Retinitis Pigmentosa in Therapy Development and Preclinical Testing.

Large animal models are valuable for developing and testing translational therapies for inherited retinal dystrophies such as retinitis pigmentosa (RP). Gene augmentation therapy has been developed utilizing such models. Adeno-associated viral (AAV) vectors have been frequently utilized and delivered by intravitreal or subretinal injection. In vivo longitudinal assessments of therapeutic outcomes are essential. These include regular ophthalmic examinations as well as detailed fundus assessments including confocal scanning laser ophthalmoscopy (cSLO) and high-resolution cross-sectional imaging of the retina by spectral domain-optical coherence tomography (SD-OCT). Retinal function assessment includes vision testing and electroretinography (ERG).

Prediction of visual impairment in retinitis pigmentosa using deep learning and multimodal fundus images

BackgroundThe efficiency of clinical trials for retinitis pigmentosa (RP) treatment is limited by the screening burden and lack of reliable surrogate markers for functional end points. Automated methods to determine...

Near-Infrared Confocal Reflectance Scanning Laser Ophthalmoscopy (SLO) and Short-Wavelength Autofluorescence Imaging in Cystic Diabetic Macular Edema.

Objective To characterize results of short-wavelength autofluorescent (SW-AF) and near-infrared confocal reflectance scanning laser ophthalmoscopy (NIR-cR SLO) imaging in cystic diabetic macular edema (DME). Design Cross-sectional study. Participants: 104 eyes from 52 naïve treatment patients with DME and mild to moderate nonproliferative diabetic retinopathy (NPDR) Methods: complete ocular examination, best-corrected visual acuity (BCVA), and imaging were performed. Results In NIR-cR SLO, small/medium and large-sized cysts presented with decreased and increased reflectance, respectively. In SW-AF, increased and decreased autofluorescence, corresponding to medium-/large- and small-sized cysts were noted. Mainly, the lower NIR reflectance was associated with petaloid edema pattern in SW-AF (P=0.011), BCVA (logMAR) (P=0.056), subretinal fluid (P=0.035), and the involved layers of retina by cysts (P < 0.001) in Pearson chi-square test. Fair agreement of 0.31 (P < 0.001) between NIR-cR SLO and late FA leakage was found by the weighted kappa test. In regression analysis, NIR-cR SLO abnormality is highly correlated with outer and inner nuclear layers location of the cystic changes. Conclusions The size of cysts and involved layers affect presenting features of SW-AF and NIR reflectance.

STRUCTURAL PATHOLOGY AFTER RETINAL DETACHMENT: Multicolor Confocal Scanning Laser Ophthalmoscopy Versus Color Fundus Photography.

To compare pathology detection using multicolor confocal scanning laser ophthalmoscopy with color fundus photography following macula-off rhegmatogenous retinal detachment. Postsurgery multicolor confocal scanning laser ophthalmoscopy and color fundus photography images from 30 rhegmatogenous retinal detachment patients were examined by 10 masked graders. Imaging was performed with the Heidelberg Spectralis HRA and the digital retinal camera Basler acA2500-14gc GigE. Swept-source optical coherence tomography was used as verification modality. Detection rates of ellipsoid zone disruption, foveal ellipsoid zone rosette, outer retinal folds, intraretinal cysts, subretinal fluid layer, subretinal fluid blebs, retinal striae, and retinal detachment line were compared. Intermodality and intergrader agreement were estimated. Overall pathology detection was significantly higher for multicolor confocal scanning laser ophthalmoscopy multicolor confocal scanning laser ophthalmoscopy (adjusted odds ratio = 7.39; 95% confidence interval, 1.64-33.30; P = 0.009). The intermodality and intergrader agreement on overall pathology detection were moderate. The intermodality agreement was 0.49 (95% confidence interval, 0.48-0.51; P < 0.0001) (Gwet's AC1). Intergrader agreement was 0.53 (95% confidence interval, 0.52-0.54; P < 0.0001) for multicolor confocal scanning laser ophthalmoscopy and 0.58 (95% confidence interval, 0.57-0.59; P < 0.0001) (Fleiss kappa) for color fundus photography. Multicolor confocal scanning laser ophthalmoscopy imaging is superior to color fundus photography in detecting and delineating structural retinal abnormalities following rhegmatogenous retinal detachment and can be a helpful tool in the visualization of retinal remodeling processes in patients recovering from rhegmatogenous retinal detachment surgery.

Characterization and allogeneic transplantation of a novel transgenic cone-rich donor mouse line

ObjectivesCone photoreceptor transplantation is a potential treatment for macular diseases. The optimal conditions for cone transplantation are poorly understood, partly because of the scarcity of cones in donor mice. To facilitate allogeneic cone photoreceptor transplantation studies in mice, we aimed to create and characterize a donor mouse model containing a cone-rich retina with a cone-specific enhanced green fluorescent protein (EGFP) reporter. MethodsWe generated OPN1LW-EGFP/NRL−/− mice by crossing NRL−/− and OPN1LW-EGFP mice. We characterized the anatomical phenotype of OPN1LW-EGFP/NRL−/− mice using multimodal confocal scanning laser ophthalmoscopy (cSLO) imaging, immunohistology, and transmission electron microscopy. We evaluated retinal function using electroretinography (ERG), including 465 and 525 nm chromatic stimuli. Retinal sheets and cell suspensions from OPN1LW-EGFP/NRL−/− mice were transplanted subretinally into immunodeficient Rd1 mice. ResultsOPN1LW-EGFP/NRL−/− retinas were enriched with OPN1LW-EGFP+ and S-opsin+ cone photoreceptors in a dorsal-ventral distribution gradient. Cone photoreceptors co-expressing OPNL1W-EGFP and S-opsin significantly increased in OPN1LW-EGFP/NRL−/− compared to OPN1LW-EGFP mice. Temporal dynamics of rosette formation in the OPN1LW-EGFP/NRL−/− were similar as the NRL−/− with peak formation at P15. Rosettes formed preferentially in the ventral retina. The outer retina in P35 OPN1LW-EGFP/NRL−/− was thinner than NRL−/− controls. The OPN1LW-EGFP/NRL−/− ERG response amplitudes to 465 nm stimulation were similar to, but to 535 nm stimulation were lower than, NRL−/− controls. Three months after transplantation, the suspension grafts showed greater macroscopic degradation than sheet grafts. Retinal sheet grafts from OPN1LW-EGFP/NRL−/− mice showed greater S-opsin + cone survival than suspension grafts from the same strain. ConclusionsOPN1LW-EGFP/NRL−/− retinae were enriched with S-opsin+ photoreceptors. Sustained expression of EGFP facilitated the longitudinal tracking of transplanted donor cells. Transplantation of cone-rich retinal grafts harvested prior to peak rosette formation survived and differentiated into cone photoreceptor subtypes. Photoreceptor sheet transplantation may promote greater macroscopic graft integrity and S-opsin+ cone survival than cell suspension transplantation, although the mechanism underlying this observation is unclear at present. This novel cone-rich reporter mouse strain may be useful to study the influence of graft structure on cone survival.

Quantifiable In Vivo Imaging Biomarkers of Retinal Regeneration by Photoreceptor Cell Transplantation.

PurposeShort-term improvements in retinal anatomy are known to occur in preclinical models of photoreceptor transplantation. However, correlative changes over the long term are poorly understood. We aimed to develop a quantifiable imaging biomarker grading scheme, using noninvasive multimodal confocal scanning laser ophthalmoscopy (cSLO) imaging, to enable serial evaluation of photoreceptor transplantation over the long term.MethodsPhotoreceptor cell suspensions or sheets from rhodopsin-green fluorescent protein mice were transplanted subretinally, into either NOD.CB17-Prkdcscid/J or C3H/HeJ-Pde6brd1 mice. Multimodal cSLO imaging was performed serially for up to three months after transplantation. Imaging biomarkers were scored, and a grade was defined for each eye by integrating the scores. Image grades were correlated with immunohistochemistry (IHC) data.ResultsMultimodal imaging enabled the extraction of quantitative imaging biomarkers including graft size, GFP intensity, graft length, on-target graft placement, intra-graft lamination, hemorrhage, retinal atrophy, and periretinal proliferation. Migration of transplanted material was observed. Changes in biomarker scores and grades were detected in 14/16 and 7/16 eyes, respectively. A high correlation was found between image grades and IHC parameters.ConclusionsSerial evaluation of multiple imaging biomarkers, when integrated into a per-eye grading scheme, enabled comprehensive tracking of longitudinal changes in photoreceptor cell grafts over time. The application of systematic multimodal in vivo imaging could be useful in increasing the efficiency of preclinical retinal cell transplantation studies in rodents and other animal models.Translational RelevanceBy allowing longitudinal evaluation of the same animal over time, and providing quantifiable biomarkers, non-invasive multimodal imaging improves the efficiency of retinal transplantation studies in animal models. Such assays will facilitate the development of cell therapy for retinal diseases.

Deep Learning Approaches Predict Glaucomatous Visual Field Damage from OCT Optic Nerve Head En Face Images and Retinal Nerve Fiber Layer Thickness Maps

To develop and evaluate a deep learning system for differentiating between eyes with and without glaucomatous visual field damage (GVFD) and predicting the severity of GFVD from spectral domain OCT (SDOCT) optic nerve head images. Evaluation of a diagnostic technology. A total of 9765 visual field (VF) SD OCT pairs collected from 1194 participants with and without GVFD (1909 eyes). Deep learning models were trained to use SD OCT retinal nerve fiber layer (RNFL) thickness maps, RNFL en face images, and confocal scanning laser ophthalmoscopy (CSLO) images to identify eyes with GVFD and predict quantitative VF mean deviation (MD), pattern standard deviation (PSD), and mean VF sectoral pattern deviation (PD) from SD OCT data. Deep learning models were compared with mean RNFL thickness for identifying GVFD using area under the curve (AUC), sensitivity, and specificity. For predicting MD, PSD, and mean sectoral PD, models were evaluated using R2 and mean absolute error (MAE). In the independent test dataset, the deep learning models based on RNFL en face images achieved an AUC of 0.88 for identifying eyes with GVFD and 0.82 for detecting mild GVFD significantly (P < 0.001) better than using mean RNFL thickness measurements (AUC= 0.82 and 0.73, respectively). Deep learning models outperformed standard RNFL thickness measurements in predicting all quantitative VF metrics. In predicting MD, deep learning models based on RNFL en face images achieved an R2 of 0.70 and MAE of 2.5 decibels (dB) compared with 0.45 and 3.7 dB for RNFL thickness measurements. In predicting mean VF sectoral PD, deep learning models achieved high accuracy in the inferior nasal (R2= 0.60) and superior nasal (R2= 0.67) sectors, moderate accuracy in inferior (R2= 0.26) and superior (R2= 0.35) sectors, and lower accuracy in the central (R2=0.15) and temporal (R2= 0.12) sectors. Deep learning models had high accuracy in identifying eyes with GFVD and predicting theseverity of functional loss from SD OCT images. Accurately predicting the severity of GFVD from SD OCT imaging can help clinicians more effectively individualize the frequency of VF testing to the individual patient.

Comparison of Blue and Green Confocal Scanning Laser Ophthalmoscope Imaging to Detect Retinal Nerve Fiber Layer Defects.

PurposeWe detected retinal nerve fiber layer (RNFL) defects using a confocal scanning laser ophthalmoscopy (CSLO) with both blue and green laser sources and evaluated image quality based on laser wavelength.MethodsThis was a retrospective observational case study. Blue and green CSLO images of 181 eyes with suspected glaucoma were evaluated and compared. Three independent observers identified the presence of RNFL defects and determined which CSLO imaging source provided superior visibility of the defect. After assessing the defect imaging by laser source, demographics and image quality indices of optical coherence tomography between blue better and green better groups were analyzed.ResultsBoth blue and green CSLO showed high discriminating ability for RNFL defects. The discriminating ability of blue CSLO was significantly greater than that of green CSLO (p = 0.004). Among eyes with a detectable RNFL defect, 61.8% were better visualized with the blue laser compared to the green laser. Compared with the blue better group, the green better group was significantly older (p = 0.009), had a greater proportion of females (p = 0.005), had poorer best-corrected visual acuity (p = 0.001), more severe cataracts (p = 0.001), lower signal strength (p = 0.003), and poor image quality indices (p = 0.001).ConclusionsBoth blue and green CSLO imaging was useful for detecting RNFL defects, but blue CSLO was superior to green CSLO in quality of RNFL defect imaging in most patients with clear media.

Possible Undiagnosed Glaucoma Detected by MultiColor Imaging of Retinal Diseases

To evaluate for the presence of visible retinal nerve fiber layer (RNFL) defects consistent with undiagnosed glaucoma in MultiColor images (Heidelberg Engineering, Heidelberg, Germany) obtained at clinicians' discretion for the diagnosis and management of retinal diseases. Retrospective chart review of MultiColor images obtained as part of routine clinical care between December 25, 2013 and June 24, 2014. Images were reviewed by 2 fellowship-trained glaucoma specialists (S.H.V.T., S.G.A.) for the presence of RNFL defects in the superotemporal or inferotemporal quadrants. No attention was given to the optic nerve head when identifying candidate images. Retinal nerve fiber layer defects were not considered glaucomatous if they traced to regions of retinal atrophy. A total of 157 retina clinic patients aged 18 years or older who carried at least 1 retinal diagnosis; patients were subsequently excluded if medical records review revealed a history of glaucoma or glaucoma suspect diagnoses at the time of imaging. MultiColor confocal scanning laser ophthalmoscopy images were obtained using the Heidelberg Spectralis HRA+OCT (Heidelberg Engineering, Heidelberg, Germany). The primary end point was the presence of at least 1 RNFL defect tracking toward the optic nerve. Among 157 reviewed charts, 134 eyes of 80 patients met inclusion criteria, of whom 13 eyes of 10 patients had visible RNFL defects consistent with possible undiagnosed glaucoma. The average age of patients with presumed glaucomatous eyes was 67 years (range, 35-88 years; standard deviation, 15.7 years). The retinal diagnoses in the glaucomatous eyes included branch retinal vein occlusion (3 eyes and 1 fellow eye), neovascular age-related macular degeneration (2 eyes), non-neovascular age-related macular degeneration (both eyes of 1 patient), proliferative diabetic retinopathy (2 eyes), nonproliferative diabetic retinopathy (both eyes of 1 patient), and epiretinal membrane (1 eye). Retinal nerve fiber layer defects consistent with glaucoma may be visible on MultiColor images obtained for the diagnosis and management of retinal diseases and may represent an underused opportunity to prevent additional visual dysfunction in patients already undergoing retinal disease management.

A Data Mining Framework for Glaucoma Decision Support Based on Optic Nerve Image Analysis Using Machine Learning Methods.

Ocular imaging instruments, such as Confocal Scanning Laser Ophthalmoscopy (CSLO), captures high-quality images of the optic disc (also known as optic nerve head) that help clinicians to diagnose glaucoma. We present an integrated data analytics framework to aid clinicians in interpreting CSLO optic nerve images to diagnose and monitor the progression of glaucoma. To distinguish between healthy and glaucomatous optic discs, our framework derives shape information from CSLO images using image processing (Zernike moment method), selects salient features (hybrid feature selection), and then trains image classifiers (Multilayer Perceptron, Support Vector Machine, Bayesian Network). To monitor glaucoma progression over time, our framework uses a mathematical model of the optic disc to extract morphological features from CSLO images and applies clustering (Self-Organizing Maps) to visualize subtypes of glaucomatous optic disc damage. We contend that our data analytics framework offers an automated and objective analysis of optic nerve images that can potentially support both diagnosis and monitoring of glaucoma. We validated our framework with CSLO optic nerve images and our data analytics approach detected glaucomatous optic discs with a sensitivity of 0.86, a specificity of 0.80, an accuracy of 0.838, and an AUROC of 0.913 with a Bayesian network classifier using the optimal subset of Zernike features (six moments). Furthermore, our framework identified, using morphological features, five clusters of CSLO images, where each cluster stands for a subtype of optic nerve damage (two healthy subtypes and three glaucoma subtypes). The characteristics of each cluster-the subtype of the image-were determined by experts who examined the morphology of the images within each cluster and provided subtype characteristics to each cluster.

RETINAL VASCULAR ABNORMALITIES IN A LARGE COHORT OF PATIENTS AFFECTED BY NEUROFIBROMATOSIS TYPE 1: A Study Using Optical Coherence Tomography Angiography.

To evaluate the prevalence, the vascular features, and the clinical diagnostic implication of retinal vascular abnormalities (RVAs) associated with neurofibromatosis Type 1 (NF1) in a large cohort of patients. Two hundred and ninety-four patients affected by NF1 were consecutively enrolled. The presence of RVAs was detected by means of infrared confocal scanning laser ophthalmoscopy images. Three hundred age- and race-matched healthy subjects were enrolled as a healthy control group. Fluorescein angiography, indocyanine green angiography, and optical coherence tomography angiography were also performed in patients with RVAs. Retinal vascular abnormalities were detected in 18 patients with NF1 (6.1%) and in none of the healthy subjects. Retinal vascular abnormalities appeared in all cases as well-defined, small, tortuous retinal vessels with a spiral aspect, originating from small tributaries of retinal veins. The presence of RVAs did not correlate with the presence of other specific ocular or systemic NF1 features (P > 0.05). On optical coherence tomography angiography, RVAs appeared as an isolated tortuous vessel of the superficial vascular plexus in all cases, associated with localized anomalous crowded and congested capillary network of the deep vascular plexus in 75% of cases. Retinal vascular abnormalities are present in a limited proportion of patients affected by NF1 and can be considered an additional distinctive sign of the disease.

Optical coherence tomography in diagnostics and treatment of central serous chorioretinopathy

To identify the signs of leakage points in central serous chorioretinopathy (CSC) patients using optical coherence tomography (OCT) allowing focal laser photocoagulation (FLP) to be performed without the need for fluorescein angiography (FA) in their treatment; to evaluate the effectiveness and safety of the approach. The retrospective part of the study included 54 patients (54 eyes) with distinctive signs of CSC further confirmed by the presence of a leakage point as could be seen with FA. The number of pigment epithelial detachments (PED), the thickness of the photoreceptor outer segments (PROS) layer, maximum horizontal length of the photoreceptor thinning area, and morphological changes in the outer retinal layers were evaluated with OCT. Using FA for each leakage point, its localization and coincidence with PED were evaluated in retro-mode confocal scanning laser ophthalmoscopy and OCT images, respectively. The prospective part of the study included 11 CSC patients (11 eyes) who underwent en face OCT-guided FLP of the leakage point with NAVILAS 532 system. The leakage points were found to coincide with PED in 50 eyes (92.6%). Mean distance from the superior border of the retinal detachment area to the leakage point was 28.8±14.1% of the retinal detachment height. In 90.3% of cases the PROS layer in the area above the PED was thinner (23.3±7.8 µm) compared to the area outside the leakage point (64.0±9.4 µm) (p<0.001). FA-free OCT-guided FLP required in average 3.3±0.5 laser burns per leakage point. Complete resolution of subretinal fluid occurred in all treated eyes within 7.5±1.0 weeks after the surgery, and no adverse events were observed. Coincidence of PED in the superior half of the area of retinal detachment with thinning of the PROS layer as seen on OCT can be considered a reliable sign of the leakage point. The FA-free FLP based on OCT appears to be an effective and safe procedure for the treatment of CSC.

The predictability of ocriplasmin treatment effects: is there consensus among retinal experts? Results from the EXPORT study.

To evaluate the agreement and predictability of ocriplasmin treatment effects among retinal experts (raters) by assessment of retinal imaging data of eyes treated for vitreomacular traction in nine different centers in Germany and Austria. Retrospective cohort study. Combined confocal near-infrared scanning laser ophthalmoscopy and spectral-domain optical coherence tomography images (Spectralis® device, Heidelberg Engineering GmbH, Germany) from 136 eyes of 135 subjects were reviewed by 14 raters using an internet-based grading database and a standardized questionnaire. In addition to the images taken within 2 days prior to treatment, age, gender, and lens status were disclosed to the raters. Treatment success was defined as a complete cleavage of the posterior vitreous cortex at day 28±5. Main outcome was the agreement and predictability among raters for assessment of treatment success. Raters generally accepted starting ocriplasmin treatment (chance for treatment success ≥ 1%) in 22.4 to 69.1% (median 53.2%) of eyes (moderate intra- and interrater agreements with kappa-values of 0.6 and 0.48). The likelihood for a high potential treatment success (equal or higher than 25%) was judged by the raters in 43.4% to 86.0% (median 62.6%) of eyes (moderate intra- and fair interrater agreements with kappa-values of 0.56 and 0.22). Allocating eyes for high potential treatment success overall increased the odds by 3.07, with odds ratios of single raters up to 4.06 to 6.16. These results underscore the importance of training health care providers in the evaluation of retinal imaging data and also to define characteristic morphological features better in the presence of vitreoretinal interface diseases. The better results of single raters in the predictability of treatment success by the allocation of eyes in the high-potential group indicates the high relevance of the meticulous analysis of retinal images.

An Automated Approach for Localizing Retinal Blood Vessels inConfocal Scanning Laser Ophthalmoscopy Fundus Images.

In this work, we present a rules-based method for localizing retinal blood vessels in confocal scanning laser ophthalmoscopy (cSLO) images and evaluate its feasibility. A total of 31 healthy participants (17 female; mean age: 64.0 ± 8.2 years) were studied using manual and automatic segmentation. High-resolution peripapillary scan acquisition cSLO images were acquired. The automated segmentation method consisted of image pre-processing for gray-level homogenization and blood vessel enhancement (morphological opening operation, Gaussian filter, morphological Top-Hat transformation), binary thresholding (entropy-based thresholding operation), and removal of falsely detected isolated vessel pixels. The proposed algorithm was first tested on the publically available dataset DRIVE, which contains color fundus photographs, and compared to performance results from the literature. Good results were obtained. Monochromatic cSLO images segmented using the proposed method were compared to those manually segmented by two independent observers. For the algorithm, a sensitivity of 0.7542, specificity of 0.8607, and accuracy of 0.8508 were obtained. For the two independent observers, a sensitivity of 0.6579, specificity of 0.9699, and accuracy of 0.9401 were obtained. The results demonstrate that it is possible to localize vessels in monochromatic cSLO images of the retina using a rules-based approach. The performance results are inferior to those obtained using fundus photography, which could be due to the nature of the technology.

Retinal pathology in idiopathic moyamoya angiopathy detected by optical coherence tomography.

To investigate whether patients with moyamoya angiopathy without obvious retinal pathologies such as retinal infarctions or the congenital morning glory anomaly may have subtle subclinical retinal changes. In this cross-sectional study, spectral domain optical coherence tomography was used to analyze the retinal morphology of 25 patients with idiopathic moyamoya angiopathy and 25 age- and sex-matched healthy controls. We analyzed the retinal vasculature with blue laser autofluorescence, lipofuscin deposits with MultiColor confocal scanning laser ophthalmoscopy, and the optic nerve head (ONH) volume with a custom postprocessing algorithm. In addition to the total retinal thickness, semiautomated segmentation was used for segmentation of retinal layers in macular cross scans, macular volume scans, and peripapillary ring scans. The main finding was a pronounced reduction of the ONH volume in moyamoya angiopathy compared with controls (0.76 ± 0.45 mm(3) and 1.47 ± 0.50 mm(3), respectively; p < 0.0001), which was associated with a less pronounced reduction of the retinal nerve fiber layer in macular volume scans (0.97 ± 0.11 mm(3) and 1.10 ± 0.10 mm(3), respectively; p < 0.001). Autofluorescence and MultiColor confocal scanning laser ophthalmoscopy images revealed no pathologies except for one branch retinal artery occlusion. Our results indicate that even patients with moyamoya angiopathy who do not have obvious retinal abnormalities have retinal abnormalities. These can be detected by spectral domain optical coherence tomography, and the association of ONH abnormalities with the vascular changes may suggest that idiopathic moyamoya angiography is a systemic disease involving abnormalities of the early mesodermal development.