Retinal Layer Segmentation Research Articles

DIABETIC MACULAR EDEMA CLASSIFICATION WITH OCT IMAGES USING GENERATIVE ADVERSARIAL NETWORK AND ACTIVE CONTOUR MODEL

The major reason for blindness is diabetic macular edema (DME) and hence detection of DME at early stage using optical coherence tomography (OCT) is commonly employed for diagnosing retinal diseases. An accurate disease identification and classification poses a challenging task due to the difficulty in differentiating the abnormal and healthy regions. To overcome these issues and to accurately classify the DME, an effective DME classification approach named antlion spider monkey optimization-based generative adversarial network (ALSMO-based GAN) is proposed in this research for segmenting the retinal layers and to classify the DME more accurately. With the generator and the discriminator components of GAN, the DME is effectively classified so that the devised ALSMO algorithm can be used to train the process of GAN. The inspiration of the foraging and the hunting behavior enable the optimization to increase the rate of convergence and to achieve global optimal solution by reducing the local optima. With the segmented retinal layer, the classification process is progressed through the extraction of relevant features from the retinal layers. The performance of the developed method is verified using measures like accuracy, sensitivity, and specificity which attained values of 92.5%, 98%, and 92.3%, respectively.

Automated retinal layer segmentation in optical coherence tomography images with intraretinal fluid

We propose a novel retinal layer segmentation method to accurately segment 10 retinal layers in optical coherence tomography (OCT) images with intraretinal fluid. The method used a fan filter to enhance the linear information pertaining to retinal boundaries in an OCT image by reducing the effect of vessel shadows and fluid regions. A random forest classifier was employed to predict the location of the boundaries. Two novel methods of boundary redirection (SR) and similarity correction (SC) were combined to carry out boundary tracking and thereby accurately locate retinal layer boundaries. Experiments were performed on healthy controls and subjects with diabetic macular edema (DME). The proposed method required an average of 415[Formula: see text]s for healthy controls and of 482[Formula: see text]s for subjects with DME and achieved high accuracy for both groups of subjects. The proposed method requires a shorter running time than previous methods and also provides high accuracy. Thus, the proposed method may be a better choice for small training datasets.

MultiSDGAN: Translation of OCT Images to Superresolved Segmentation Labels Using Multi-Discriminators in Multi-Stages.

Optical coherence tomography (OCT) has been identified as a non-invasive and inexpensive imaging modality to discover potential biomarkers for Alzheimer's diagnosis and progress determination. Current hypotheses presume the thickness of the retinal layers, which are analyzable within OCT scans, as an effective biomarker for the presence of Alzheimer's. As a logical first step, this work concentrates on the accurate segmentation of retinal layers to isolate the layers for further analysis. This paper proposes a generative adversarial network (GAN) that concurrently learns to increase the image resolution for higher clarity and then segment the retinal layers. We propose a multi-stage and multi-discriminatory generative adversarial network (MultiSDGAN) specifically for superresolution and segmentation of OCT scans of the retinal layer. The resulting generator is adversarially trained against multiple discriminator networks at multiple stages. We aim to avoid early saturation of generator model training leading to poor segmentation accuracies and enhance the process of OCT domain translation by satisfying all the discriminators in multiple scales. We also investigated incorporating the Dice loss and Structured Similarity Index Measure (SSIM) as additional loss functions to specifically target and improve our proposed GAN architecture's segmentation and superresolution performance, respectively. The ablation study results conducted on our data set suggest that the proposed MultiSDGAN with ten-fold cross-validation (10-CV) provides a reduced equal error rate with 44.24% and 34.09% relative improvements, respectively (p-values of the improvement level tests .01). Furthermore, our experimental results also demonstrate that the addition of the new terms to the loss function improves the segmentation results significantly by relative improvements of 31.33% (p-value .01).

Evaluation of Retinal Layer Thicknesses in Patients with Keratoconus Using Retinal Layer Segmentation Analysis.

To conduct an evaluation of the effects of irregular astigmatism on the retinal nerve fiber layer (RNFL) and the retinal layers observed using spectral-domain optical coherence tomography (SD-OCT) in patients who had keratoconus (KC). A total of 255 eyes from 255 individuals, comprising 72 eyes of KC patients, 70 eyes of patients with astigmia, and 113 eyes of healthy controls were included in the analysis. RNFL scan maps (comprising global, temporal, superotemporal, inferotemporal, nasal, inferonasal, and superonasal maps) and macular thickness (MT) maps of a standard from the Early Treatment Diabetic Retinopathy Study (ETDRS) grid were assessed. The measurements were segmented automatically using Spectralis software, and included the RNFL, inner and outer plexiform layers (IPL, OPL), inner and outer nuclear layers (INL, ONL), ganglion cell layer, retinal pigment epithelium (RPE) in the central 6-mm ETDRS subfield. The RNFL thickness in the KC group was lower when compared with the other two groups; however, statistically significant differences were noted in the global, temporal, superotemporal, and inferotemporal sectors (p < 0.05 for all). All of the central MT parameters showed significant variation among the groups, while a statistically significant decrease was noted in the KC group, except in the inferior outer sector (p = 0.741). In the segmentation analysis, the KC group had the significantly lowest IPL, ONL, RPE, and outer retinal layer (ORL) thickness among the groups (p < 0.05 for each). The astigmatic group was similar to the control group with regard to these parameters (p > 0.05 for each). The eyes in the KC group appeared to have a thinner RNFL and MT when compared to those in the astigmatic and control groups. The ORLs, especially the ONL and RPE, were the most affected component of the macula in the KC group.

Functional Correlates of Outer Retina Remodeling in Intermediate Age-Related Macular Degeneration Using Microperimetry.

PurposeTo assess the effect of drusen morphometric changes and choroidal vascular modifications on retinal sensitivity (RS) evaluated through microperimetry in intermediate age-related macular degeneration (iAMD).MethodsA retrospective review of 18 iAMD patients (18 eyes) with a 12-month follow-up was performed. Eye-tracked spectral-domain optical coherence tomography was obtained, with automatic segmentation of the outer retinal layer (ORL) delineating the drusen area from the external limiting membrane to Bruch's membrane and outer nuclear layer (ONL) thickness maps adjusted manually, as needed. Advanced retinal pigment epithelium analysis was also performed with a ZEISS PLEX Elite 900. Microperimetry obtained under mesopic conditions was overlaid with the corresponding thickness maps with Fiji software. The choroidal vascularity index (CVI) was calculated in the subfoveal b-scan and volumetric in the central 1-mm subfield.ResultsA reduced central ONL thickness was strongly associated with RS decline at the same region (r = 0.69, P = 0.002) and globally (r = 0.80, P < 0.001) at baseline, but also at 1 year in the central subfield (central: r = 0.70, P = 0.001). One-year subfoveal CVI variation, differently from volumetric CVI, directly influenced the central (r = 0.64, P = 0.004) and global RS (r = 0.59, P = 0.009), indicating that a CVI reduction negatively affected RS. A greater volumetric CVI within central 1-mm was associated with ORL thickening at 1 year (r = 0.61, P = 0.008).ConclusionsProgressive degeneration of the ONL is related to irreversible photoreceptor dysfunction in iAMD. Likewise, choroidal vascular modifications are associated with a significant functional decline in the central region and diffusely.

Hydroxychloroquine treatment alters retinal layers and choroid without apparent toxicity in optical coherence tomography

Background To investigate whether single retinal layer, choroidal and peripapillary retinal fiber layer thicknesses were affected in patients under hydroxychloroquine treatment without retinal toxicityMethods Forty-nine patients under hydroxychloroquine treatment (HCQ group) and 39 healthy volunteers, a total of 88 individuals, were included in the study. Patients with HCQ treatment were further divided into two groups as less than 5 years of HCQ use (group 1, 29 patients) and 5 years or more of HCQ use (group 2, 20 patients). All participants underwent full ophthalmologic examination and optical coherence tomography (OCT) imaging as OCT is a recommended tool for screening toxicity by the American Association of Ophthalmology. The automatic segmentation of retinal layers included retinal nerve fiber layer (RNFL), ganglion cell layer (GCL), inner plexiform layer (IPL), inner nuclear layer (INL), outer plexiform layer (OPL), outer nuclear layer (ONL), retina pigment epithelium (RPE), inner retinal layers (IRL), outer retinal layers (ORL) and total retina thickness (TRT). Choroidal thickness (ChT), choroidal vascularity index and peripapillary RNFL were also measured.Results Both inferior and nasal parafoveal RNFL, temporal parafoveal GCL, and IPL layer thicknesses were decreased, while temporal parafoveal RPE layer thicknesses were increased in hydroxychloroquine users. Subfoveal ChTs were significantly reduced in both groups 1 and 2 compared to the control group(p = 0.042 and p = 0.009, respectively) and temporal ChT thicknesses were statistically decreased in group 2 compared to the control group(p = 0.018), but not for nasal ChT.Conclusions Hydroxychloroquine use may cause parafoveal retinal layer thickness alterations and also subfoveal and parafoveal ChT alterations even without evident retinal toxicity in OCT.

Abnormal Appearance of the Area Centralis in Labrador Retrievers With an ABCA4 Loss-of-function Mutation.

PurposeTo study retinal appearance and morphology in Labrador retrievers (LRs) heterozygous and homozygous for an ABCA4 loss-of-function mutation.MethodsOphthalmic examination, including ophthalmoscopy and simple testing of vision, was performed in five ABCA4wt/wt, four ABCA4wt/InsC, and six ABCA4InsC/InsC LRs. Retinas were also examined with confocal scanning laser ophthalmoscopy (cSLO) and optical coherence tomography (OCT). Infrared and fundus autofluorescence (FAF) images were studied, and outer nuclear layer (ONL) and neuroretinal thickness were measured in the central and peripheral area centralis.ResultsClinical signs in young ABCA4InsC/InsC LRs were subtle, whereas ophthalmoscopic findings and signs of visual impairment were obvious in old ABCA4InsC/InsC LRs. Retinal appearance and vision testing was unremarkable in heterozygous LRs regardless of age. The cSLO/OCT showed abnormal morphology including ONL thinning, abnormal outer retinal layer segmentation, and focal loss of retinal pigment epithelium in the fovea equivalent in juvenile ABCA4InsC/InsC LRs. The abnormal appearance extended into the area centralis and visual streak in middle-aged ABCA4InsC/InsC and then spread more peripherally. A mild phenotype was seen on cSLO/OCT and FAF in middle-aged to old ABCA4wt/InsC LRs.ConclusionsAbnormal appearance and morphology in the fovea equivalent are present in juvenile ABCA4InsC/InsC. In the older affected LRs, the visual streak and then the peripheral retina also develop an abnormal appearance. Vision deteriorates slowly, but some vision is retained throughout life. Older heterozygotes may show a mild retinal phenotype but no obvious visual impairment. The ABCA4InsC/InsC LR is a potential model for ABCA4-mediated retinopathies/juvenile-onset Stargardt disease in a species with human-sized eyes.Translational RelevanceThe ABCA4InsC mutation causes juvenile-onset abnormal appearance of the fovea equivalent in affected dogs that slowly spreads in the retina, while only a mild phenotype is seen in older carriers. This is the first non-primate, large-animal model for ABCA4-related/STGD1 retinopathies in a species with a fovea equivalent.

Photoreceptor Layer Thinning Is an Early Biomarker for Age-Related Macular Degeneration: Epidemiologic and Genetic Evidence from UK Biobank OCT Data

Despite widespread use of OCT, an early-stage imaging biomarker for age-related macular degeneration (AMD) has not been identified. Pathophysiologically, the timing of drusen accumulation in relationship to photoreceptor degeneration in AMD remains unclear, as are the inherited genetic variants contributing to these processes. Herein, we jointly analyzed OCT, electronic health record data, and genomic data to characterize the time sequence of changes in retinal layer thicknesses in AMD, as well as epidemiologic and genetic associations between retinal layer thicknesses and AMD. Cohort study. Forty-four thousand eight hundred twenty-three individuals from the UK Biobank (enrollment age range, 40-70 years; 54% women; median follow-up, 10 years). The Topcon Advanced Boundary Segmentation algorithm was used for retinal layer segmentation. We associated 9 retinal layer thicknesses with prevalent AMD (present at enrollment) in a logistic regression model and with incident AMD (diagnosed after enrollment) in a Cox proportional hazards model. Next, we associated AMD-associated genetic alleles, individually and as a polygenic risk score (PRS), with retinal layer thicknesses. All analyses were adjusted for age, age-squared (age2), sex, smoking status, and principal components of ancestry. Prevalent and incident AMD. Photoreceptor segment (PS) thinning was observed throughout the lifespan of individuals analyzed, whereas retinal pigment epithelium (RPE) and Bruch's membrane (BM) complex thickening started after 57 years of age. Each standard deviation (SD) of PS thinning and RPE-BM complex thickening was associated with incident AMD (PS: hazard ratio [HR], 1.35; 95% confidence interval [CI], 1.23-1.47; P= 3.7× 10-11; RPE-BM complex: HR, 1.14; 95% CI, 1.06-1.22; P= 0.00024). The AMD PRS was associated with PS thinning (β, -0.21 SD per twofold genetically increased risk of AMD; 95% CI, -0.23 to -0.19; P= 2.8× 10-74), and its association with RPE-BM complex was U-shaped (thinning with AMD PRS less than the 92nd percentile and thickening with AMD PRS more than the 92nd percentile). The loci with strongest support for genetic correlation were AMD risk-raising variants Complement Factor H (CFH):rs570618-T, CFH:rs10922109-C, and Age-Related Maculopathy Susceptibility 2 (ARMS2)/High-Temperature Requirement Serine Protease 1 (HTRA1):rs3750846-C on PS thinning and SYN3/Tissue Inhibitor of Metalloprotease 3 (TIMP3):rs5754227-T on RPE-BM complex thickening. Epidemiologically, PS thinning precedes RPE-BM complex thickening by decades and is the retinal layer most strongly predictive of future AMD risk. Genetically, AMD risk variants are associated with decreased PS thickness. Overall, these findings support PS thinning as an early-stage biomarker for future AMD development.

Globally optimal OCT surface segmentation using a constrained IPM optimization.

Segmentation of multiple surfaces in optical coherence tomography (OCT) images is a challenging problem, further complicated by the frequent presence of weak boundaries, varying layer thicknesses, and mutual influence between adjacent surfaces. The traditional graph-based optimal surface segmentation method has proven its effectiveness with its ability to capture various surface priors in a uniform graph model. However, its efficacy heavily relies on handcrafted features that are used to define the surface cost for the "goodness" of a surface. Recently, deep learning (DL) is emerging as a powerful tool for medical image segmentation thanks to its superior feature learning capability. Unfortunately, due to the scarcity of training data in medical imaging, it is nontrivial for DL networks to implicitly learn the global structure of the target surfaces, including surface interactions. This study proposes to parameterize the surface cost functions in the graph model and leverage DL to learn those parameters. The multiple optimal surfaces are then simultaneously detected by minimizing the total surface cost while explicitly enforcing the mutual surface interaction constraints. The optimization problem is solved by the primal-dual interior-point method (IPM), which can be implemented by a layer of neural networks, enabling efficient end-to-end training of the whole network. Experiments on spectral-domain optical coherence tomography (SD-OCT) retinal layer segmentation demonstrated promising segmentation results with sub-pixel accuracy.

Multiscale Attention Gated Network (MAGNet) for Retinal Layer and Macular Cystoid Edema Segmentation

Multiscale Attention Gated Network (MAGNet) for Retinal Layer and Macular Cystoid Edema Segmentation

Artificial intelligence for assessment of Stargardt macular atrophy.

Stargardt disease (also known as juvenile macular degeneration or Stargardt macular degeneration) is an inherited disorder of the retina, which can occur in the eyes of children and young adults. It is the most prevalent form of juvenile-onset macular dystrophy, causing progressive (and often severe) vision loss. Images with Stargardt disease are characterized by the appearance of flecks in early and intermediate stages, and the appearance of atrophy, due to cells wasting away and dying, in the advanced stage. The primary measure of late-stage Stargardt disease is the appearance of atrophy. Fundus autofluorescence is a widely available two-dimensional imaging technique, which can aid in the diagnosis of the disease. Spectral-domain optical coherence tomography, in contrast, provides three-dimensional visualization of the retinal microstructure, thereby allowing the status of the individual retinal layers. Stargardt disease may cause various levels of disruption to the photoreceptor segments as well as other outer retinal layers. In recent years, there has been an exponential growth in the number of applications utilizing artificial intelligence for help with processing such diseases, heavily fueled by the amazing successes in image recognition using deep learning. This review regarding artificial intelligence deep learning approaches for the Stargardt atrophy screening and segmentation on fundus autofluorescence images is first provided, followed by a review of the automated retinal layer segmentation with atrophic-appearing lesions and fleck features using artificial intelligence deep learning construct. The paper concludes with a perspective about using artificial intelligence to potentially find early risk factors or biomarkers that can aid in the prediction of Stargardt disease progression.

DW-Net: Dynamic Multi-Hierarchical Weighting Segmentation Network for Joint Segmentation of Retina Layers With Choroid Neovascularization.

Choroid neovascularization (CNV) is one of the blinding factors. The early detection and quantitative measurement of CNV are crucial for the establishment of subsequent treatment. Recently, many deep learning-based methods have been proposed for CNV segmentation. However, CNV is difficult to be segmented due to the complex structure of the surrounding retina. In this paper, we propose a novel dynamic multi-hierarchical weighting segmentation network (DW-Net) for the simultaneous segmentation of retinal layers and CNV. Specifically, the proposed network is composed of a residual aggregation encoder path for the selection of informative feature, a multi-hierarchical weighting connection for the fusion of detailed information and abstract information, and a dynamic decoder path. Comprehensive experimental results show that our proposed DW-Net achieves better performance than other state-of-the-art methods.

LF-UNet - A novel anatomical-aware dual-branch cascaded deep neural network for segmentation of retinal layers and fluid from optical coherence tomography images.

Computer-assistant diagnosis of retinal disease relies heavily on the accurate detection of retinal boundaries and other pathological features such as fluid accumulation. Optical coherence tomography (OCT) is a non-invasive ophthalmological imaging technique that has become a standard modality in the field due to its ability to detect cross-sectional retinal pathologies at the micrometer level. In this work, we presented a novel framework to achieve simultaneous retinal layers and fluid segmentation. A dual-branch deep neural network, termed LF-UNet, was proposed which combines the expansion path of the U-Net and original fully convolutional network, with a dilated network. In addition, we introduced a cascaded network framework to include the anatomical awareness embedded in the volumetric image. Cross validation experiments showed that the proposed LF-UNet has superior performance compared to the state-of-the-art methods, and that incorporating the relative positional map structural prior information could further improve the performance regardless of the network. The generalizability of the proposed network was demonstrated on an independent dataset acquired from the same types of device with different field of view, or images acquired from different device.

Robust region encoding and layer attribute protection for the segmentation of retina with multifarious abnormalities.

To robustly segment retinal layers that are affected by complex variety of retinal diseases for optical coherence tomography angiography (OCTA) en face projection generation. In this paper, we propose a robust retinal layer segmentation model to reduce the impact of multifarious abnormalities on model performance. OCTA vascular distribution that is regarded as the supplements of spectral domain optical coherence tomography (SD-OCT) structural information is introduced to improve the robustness of layer region encoding. To further reduce the sensitivity of region encoding to retinal abnormalities, we propose a multitask layer-wise refinement (MLR) module that can refine the initial layer region segmentation results layer-by-layer. Finally, we design a region-to-surface transformation (RtST) module without additional training parameters to convert the encoding layer regions to their corresponding layer surfaces. This transformation from layer regions to layer surfaces can remove the inaccurate segmentation regions, and the layer surfaces are easier to be used to protect the retinal layer natures than layer regions. Experimental data includes 273 eyes, where 95 eyes are normal and 178 eyes contain complex retinal diseases, including age-related macular degeneration (AMD), diabetic retinopathy (DR), central serous chorioretinopathy (CSC), choroidal neovascularization (CNV), and so forth. The dice similarity coefficient (DSC: %) of superficial, deep and outer retina achieves 98.92, 97.48, and 98.87 on normal eyes and 98.35, 95.33, and 98.17 on abnormal eyes. Compared with other commonly used layer segmentation models, our model achieves the state-of-the-art layer segmentation performance. The final results prove that our proposed model obtains outstanding performance and has enough ability to resist retinal abnormalities. Besides, OCTA modality is helpful for retinal layer segmentation.

A Hybrid Model Composed of Two Convolutional Neural Networks (CNNs) for Automatic Retinal Layer Segmentation of OCT Images in Retinitis Pigmentosa (RP).

PurposeWe propose and evaluate a hybrid model composed of two convolutional neural networks (CNNs) with different architectures for automatic segmentation of retina layers in spectral domain optical coherence tomography (SD-OCT) B-scans of retinitis pigmentosa (RP).MethodsThe hybrid model consisted of a U-Net for initial semantic segmentation and a sliding-window (SW) CNN for refinement by correcting the segmentation errors of U-Net. The U-Net construction followed Ronneberger et al. (2015) with an input image size of 256 × 32. The SW model was similar to our previously reported approach. Training image patches were generated from 480 horizontal midline B-scans obtained from 220 patients with RP and 20 normal participants. Testing images were 160 midline B-scans from a separate group of 80 patients with RP. The Spectralis segmentation of B-scans was manually corrected for the boundaries of the inner limiting membrane, inner nuclear layer, ellipsoid zone (EZ), retinal pigment epithelium, and Bruch's membrane by one grader for the training set and two for the testing set. The trained U-Net and SW, as well as the hybrid model, were used to classify all pixels in the testing B-scans. Bland–Altman and correlation analyses were conducted to compare layer boundary lines, EZ width, and photoreceptor outer segment (OS) length and area determined by the models to those by human graders.ResultsThe mean times to classify a B-scan image were 0.3, 65.7, and 2.4 seconds for U-Net, SW, and the hybrid model, respectively. The mean ± SD accuracies to segment retinal layers were 90.8% ± 4.8% and 90.7% ± 4.0% for U-Net and SW, respectively. The hybrid model improved mean ± SD accuracy to 91.5% ± 4.8% (P < 0.039 vs. U-Net), resulting in an improvement in layer boundary segmentation as revealed by Bland–Altman analyses. EZ width, OS length, and OS area measured by the models were highly correlated with those measured by the human graders (r > 0.95 for EZ width; r > 0.83 for OS length; r > 0.97 for OS area; P < 0.05). The hybrid model further improved the performance of measuring retinal layer thickness by correcting misclassification of retinal layers from U-Net.ConclusionsWhile the performances of U-Net and the SW model were comparable in delineating various retinal layers, U-Net was much faster than the SW model to segment B-scan images. The hybrid model that combines the two improves automatic retinal layer segmentation from OCT images in RP.Translational RelevanceA hybrid deep machine learning model composed of CNNs with different architectures can be more effective than either model separately for automatic analysis of SD-OCT scan images, which is becoming increasingly necessary with current high-resolution, high-density volume scans.

Fully-automated atrophy segmentation in dry age-related macular degeneration in optical coherence tomography

Age-related macular degeneration (AMD) is a progressive retinal disease, causing vision loss. A more detailed characterization of its atrophic form became possible thanks to the introduction of Optical Coherence Tomography (OCT). However, manual atrophy quantification in 3D retinal scans is a tedious task and prevents taking full advantage of the accurate retina depiction. In this study we developed a fully automated algorithm segmenting Retinal Pigment Epithelial and Outer Retinal Atrophy (RORA) in dry AMD on macular OCT. 62 SD-OCT scans from eyes with atrophic AMD (57 patients) were collected and split into train and test sets. The training set was used to develop a Convolutional Neural Network (CNN). The performance of the algorithm was established by cross validation and comparison to the test set with ground-truth annotated by two graders. Additionally, the effect of using retinal layer segmentation during training was investigated. The algorithm achieved mean Dice scores of 0.881 and 0.844, sensitivity of 0.850 and 0.915 and precision of 0.928 and 0.799 in comparison with Expert 1 and Expert 2, respectively. Using retinal layer segmentation improved the model performance. The proposed model identified RORA with performance matching human experts. It has a potential to rapidly identify atrophy with high consistency.

Assessing the validity of a cross-platform retinal image segmentation tool in normal and diseased retina

Comparing automated retinal layer segmentation using proprietary software (Heidelberg Spectralis HRA + OCT) and cross-platform Optical Coherence Tomography (OCT) segmentation software (Orion). Image segmentations of normal and diseased (iAMD, DME) eyes were performed using both softwares and then compared to the ‘gold standard’ of manual segmentation. A qualitative assessment and quantitative (layer volume) comparison of segmentations were performed. Segmented images from the two softwares were graded by two masked graders and in cases with difference, a senior retina specialist made a final independent decisive grading. Cross-platform software was significantly better than the proprietary software in the segmentation of NFL and INL layers in Normal eyes. It generated significantly better segmentation only for NFL in iAMD and for INL and OPL layers in DME eyes. In normal eyes, all retinal layer volumes calculated by the two softwares were moderate-strongly correlated except OUTLY. In iAMD eyes, GCIPL, INL, ONL, INLY, TRV layer volumes were moderate-strongly correlated between softwares. In eyes with DME, all layer volume values were moderate-strongly correlated between softwares. Cross-platform software can be used reliably in research settings to study the retinal layers as it compares well against manual segmentation and the commonly used proprietary software for both normal and diseased eyes.

Optic neuritis in Asian type opticospinal multiple sclerosis (OSMS-ON) in a non-Asian population: A functional-structural paradox

BackgroundBiomarkers have improved the classification of autoimmune inflammatory disorders, including optic neuritis (ON) as a frequent presentation of multiple sclerosis, neuromyelitis spectrum disorders, MOG antibody-related disease (MOGAD), and opticospinal multiple sclerosis (OSMS). The phenotype of OSMS in non-Asian populations is less well known. ObjectiveWe investigated the clinical features and prognosis of OSMS–ON in a Brazilian cohort. MethodsThis was a single-center cohort study of patients from Rio de Janeiro (Brazil) with OSMS. All individuals were MOG- and AQP4-seronegative, clinically diagnosed with ON, and had magnetic resonance imaging-confirmed transverse myelitis (TM). Subjects and healthy controls (HCs) were assessed for visual acuity (logMAR VA), automated perimetry mean deviation (MD), intraocular pressure, and spectral-domain optical coherence tomography (OCT), followed by automated retinal layer segmentation of the peripapillary retinal nerve fiber layer (pRNFL) and macular ganglion cell and inner plexiform layer (mGCIPL). Receiver operator characteristic curves were plotted and the area under the curve (AUC) was calculated for group comparisons of retinal asymmetry of the pRNFL and mGCIPL. ResultsThe 30 patients with OSMS were predominantly female and white. The mean age was 48 years (range 20–70 years). Unilateral ON was the index event in 83.3% of patients. Over the average 18-year follow-up period, there were 89 relapses of ON. In individuals with OSMS, the average VA was 0.07±0.14 in the right eye (RE) and 0.13±0.30 in the left eye (LE). The MD was −5.37±5.88 dB and −5.23±3.34 dB for the RE and LE, respectively. There was a significant cumulative loss of VA (p = 0.0003) and MD (p = 0.0001) with a higher number of recurrent episodes. Atrophy of the pRNFL thickness was significant in OSMS (RE, 78.62 ± 16.01 µm; LE, 79.86 ± 13.79 µm) relative to the HC group (RE, 98.87 ± 10.68 µm; LE, 97.87 ± 10.85 µm, p = 0.0001). Likewise, there was significant mGCIPL atrophy in patients with OSMS (RE, 74.96 ± 14.46 µm; LE, 73.88 ± 13.79 µm) relative to the HC group (RE, 90.50 ± 6.74 µm; LE, 90.41± 6.89 µm; p = 0.0001). Retinal asymmetry, inter-eye percentage, and absolute differences accurately separated patients with unilateral ON from HCs (AUC=0.89 and AUC=0.85, respectively). ConclusionA structural-functional paradox was found in OSMS with a high diagnostic value for a novel metric based on retinal asymmetry. The functional visual outcome are excellent despite significant structural damage to the inner retinal layers in patients with a high ON relapse rate and long-term bilateral sequential involvement.

Assignment Flow for Order-Constrained OCT Segmentation

At the present time optical coherence tomography (OCT) is among the most commonly used non-invasive imaging methods for the acquisition of large volumetric scans of human retinal tissues and vasculature. The substantial increase of accessible highly resolved 3D samples at the optic nerve head and the macula is directly linked to medical advancements in early detection of eye diseases. To resolve decisive information from extracted OCT volumes and to make it applicable for further diagnostic analysis, the exact measurement of retinal layer thicknesses serves as an essential task be done for each patient separately. However, manual examination of OCT scans is a demanding and time consuming task, which is typically made difficult by the presence of tissue-dependent speckle noise. Therefore, the elaboration of automated segmentation models has become an important task in the field of medical image processing. We propose a novel, purely data driven geometric approach to order-constrained 3D OCT retinal cell layer segmentation which takes as input data in any metric space and can be implemented using only simple, highly parallelizable operations. As opposed to many established retinal layer segmentation methods, we use only locally extracted features as input and do not employ any global shape prior. The physiological order of retinal cell layers and membranes is achieved through the introduction of a smoothed energy term. This is combined with additional regularization of local smoothness to yield highly accurate 3D segmentations. The approach thereby systematically avoid bias pertaining to global shape and is hence suited for the detection of anatomical changes of retinal tissue structure. To demonstrate its robustness, we compare two different choices of features on a data set of manually annotated 3D OCT volumes of healthy human retina. The quality of computed segmentations is compared to the state of the art in automatic retinal layer segmention as well as to manually annotated ground truth data in terms of mean absolute error and Dice similarity coefficient. Visualizations of segmented volumes are also provided.

Depth-Resolved Visualization of Perifoveal Retinal Vasculature in Preterm Infants Using Handheld Optical Coherence Tomography Angiography

PurposeTo establish methods to visualize depth-resolved perifoveal retinal vasculature in preterm infants using handheld optical coherence tomography angiography (OCT-A).MethodsIn this exploratory study, eyes of preterm infants were imaged using an investigational noncontact, handheld swept-source OCT-A device as part of the prospective BabySTEPS infant retinal imaging study. We selected high-quality OCT-A volumes at two developmental stages for analysis. Customized MATLAB scripts were used to segment retinal layers, test offset parameters, and generate depth-resolved OCT-A slabs. The superficial (SCP), intermediate (ICP), and deep (DCP) capillary plexuses were visualized and qualitatively assessed by three image graders.ResultsSix eyes from six preterm infants were included in this analysis. A three-layered perifoveal retinal vasculature was successfully visualized in all three eyes (three infants) in the 40 weeks postmenstrual age (PMA) group (one of three eyes with treated type 1 retinopathy of prematurity [ROP]). No obvious ICP or DCP was found in good-quality scans of the three eyes (three infants) in the 35 weeks PMA group (three of three eyes developed type 1 ROP).ConclusionsCustom segmentation parameters are useful to visualize perifoveal retinal vasculature in preterm infants. At term age, a three-layered capillary structure is visible in most eyes, while prior to detectable flow within the ICP and DCP, the perifoveal vasculature may be better visualized in two layers.Translational RelevanceDevelopment of segmentation parameters for depth-resolved OCT-A of perifoveal retinal vasculature in preterm infants facilitates the study of human retinal vascular development and vascular pathologies of ROP.