Automated 3-D intraretinal layer segmentation of macular spectral-domain optical coherence tomography images

Optical coherence tomography shows neuroretinal thinning in myelopathy of adrenoleukodystrophy

Background

Progressive myelopathy is the main cause of disability in adrenoleukodystrophy (ALD). Development of therapies is hampered by a lack of quantitative outcome measures. In this study, we investigated whether myelopathy in ALD is associated with retinal neurodegeneration on optical coherence tomography (OCT), which could serve as a surrogate outcome measure.

Methods

Sixty-two patients (29 men and 33 women) and 70 age-matched and sex-matched controls (33 men and 37 women) were included in this cross-sectional study. We compared retinal nerve fiber layer (RNFL), ganglion cell layer (GCL) and peripapillary retinal nerve fiber layer (pRNFL) thickness between ALD patients and controls. In addition, we correlated these OCT measurements with clinical parameters of severity of myelopathy.

Results

Patients had significantly thinner RNFL (male group, p < 0.05) and pRNFL superior and temporal quadrant [both male (p < 0.005) and female (p < 0.05) groups] compared to controls. Comparing three groups (symptomatic patients, asymptomatic patients and controls), there were significant differences in RNFL thickness (total grid and peripheral ring) in the male group (p ≤ 0.002) and in pRNFL thickness (superior and temporal quadrant) in both male (p ≤ 0.02) and the female (p ≤ 0.02) groups. Neuroretinal layer thickness correlated moderately with severity of myelopathy in men (correlation coefficients between 0.29–0.55, p < 0.02), but not in women.

Conclusions

These results suggest that neurodegeneration of the spinal cord in ALD is reflected in the retina of patients with ALD. Therefore, OCT could be valuable as an outcome measure for the myelopathy of ALD. Additional longitudinal studies are ongoing.

Electronic supplementary material

The online version of this article (10.1007/s00415-019-09627-z) contains supplementary material, which is available to authorized users.

Deep learning based noise reduction method for automatic 3D segmentation of the anterior of lamina cribrosa in optical coherence tomography volumetric scans

A deep-learning (DL) based noise reduction algorithm, in combination with a vessel shadow compensation method and a three-dimensional (3D) segmentation technique, has been developed to achieve, to the authors best knowledge, the first automatic segmentation of the anterior surface of the lamina cribrosa (LC) in volumetric ophthalmic optical coherence tomography (OCT) scans. The present DL-based OCT noise reduction algorithm was trained without the need of noise-free ground truth images by utilizing the latest development in deep learning of de-noising from single noisy images, and was demonstrated to be able to cover more locations in the retina and disease cases of different types to achieve high robustness. Compared with the original single OCT images, a 6.6 dB improvement in peak signal-to-noise ratio and a 0.65 improvement in the structural similarity index were achieved. The vessel shadow compensation method analyzes the energy profile in each A-line and automatically compensates the pixel intensity of locations underneath the detected blood vessel. Combining the noise reduction algorithm and the shadow compensation and contrast enhancement technique, medical experts were able to identify the anterior surface of the LC in 98.3% of the OCT images. The 3D segmentation algorithm employs a two-round procedure based on gradients information and information from neighboring images. An accuracy of 90.6% was achieved in a validation study involving 180 individual B-scans from 36 subjects, compared to 64.4% in raw images. This imaging and analysis strategy enables the first automatic complete view of the anterior LC surface, to the authors best knowledge, which may have the potentials in new LC parameters development for glaucoma diagnosis and management.

Intravitreal Fluocinolone Acetonide May Decelerate Diabetic Retinal Neurodegeneration

Purpose

There is no prevention or treatment for diabetic retinal neurodegeneration (DRN), which is a complication of diabetes that can occur independently of diabetic retinopathy (DR). We hypothesized that an intravitreal fluocinolone acetonide (FAc) implant may affect the rate of DRN when used in patients with diabetic macular edema (DME).

Methods

In this retrospective analysis, optical coherence tomography with neuroretinal analysis was obtained at 3-month intervals from 130 patients in the USER study both before (mean duration 903 days, range 35-4005 days) and after administration of FAc (mean 408 days, range 7 to 756 days). The rate of DRN was defined as the change over time on inner neuroretinal thickness using logistic regression. A DRN rate was calculated independently for two areas: region 1 located within 1.5 mm of the fovea, and region 2 from 1.5 mm to 3.0 mm from the fovea.

Results

In regions of the macula more than 1.5 mm from the central fovea, there was a statistically significant decrease in the rate of DRN in the post-FAc period. The pre-FAc neuroretinal loss in this area occurred at 4.0 μm/y, compared with a post-FAc loss rate of 1.1 μm/y (P = 0.001).

Conclusions

This retrospective study suggests that FAc may decelerate the rate of inner retinal thinning in patients with persistent DME. Further prospective studies are necessary to determine the effects of FAc on the rate of DRN in patients with DME.

Shape Decomposition of Foveal Pit Morphology using Scan Geometry Corrected OCT

The fovea is an important structure that allows for the high acuity at the center of our visual system. While the fovea has been well studied, the role of the foveal pit in the human retina is still largely unknown. In this study we analyze the shape morphology of the foveal pit using a statistical shape model to find the principal shape variations in a cohort of 50 healthy subjects. Our analysis includes the use of scan geometry correction to reduce the error from inherent distortions in OCT images, and a method for aligning foveal pit surfaces to remove translational and rotational variability between the subjects. Our results show that foveal pit morphology can be represented using less than five principal modes of variation. And we find that the shape variations discovered through our analysis are closely related to the main metrics (depth and diameter) used to study the foveal pit in current literature. Lastly, we evaluated the relationship between the first principal mode of variation in the cohort and the axial length from each subject. Our findings showed a modest inverse relationship between axial length and foveal pit depth that can be confirmed independently by existing studies.

Fully Convolutional Boundary Regression for Retina OCT Segmentation

A major goal of analyzing retinal optical coherence tomography (OCT) images is retinal layer segmentation. Accurate automated algorithms for segmenting smooth continuous layer surfaces, with correct hierarchy (topology) are desired for monitoring disease progression. State-of-the-art methods use a trained classifier to label each pixel into background, layer, or surface pixels. The final step of extracting the desired smooth surfaces with correct topology are mostly performed by graph methods (e.g. shortest path, graph cut). However, manually building a graph with varying constraints by retinal region and pathology and solving the minimization with specialized algorithms will degrade the flexibility and time efficiency of the whole framework. In this paper, we directly model the distribution of surface positions using a deep network with a fully differentiable soft argmax to obtain smooth, continuous surfaces in a single feed forward operation. A special topology module is used in the deep network both in the training and testing stages to guarantee the surface topology. An extra deep network output branch is also used for predicting lesion and layers in a pixel-wise labeling scheme. The proposed method was evaluated on two publicly available data sets of healthy controls, subjects with multiple sclerosis, and diabetic macular edema; it achieves state-of-the art sub-pixel results.

Accurate tissue interface segmentation via adversarial pre-segmentation of anterior segment OCT images

Optical Coherence Tomography (OCT) is an imaging modality that has been widely adopted for visualizing corneal, retinal and limbal tissue structure with micron resolution. It can be used to diagnose pathological conditions of the eye, and for developing pre-operative surgical plans. In contrast to the posterior retina, imaging the anterior tissue structures, such as the limbus and cornea, results in B-scans that exhibit increased speckle noise patterns and imaging artifacts. These artifacts, such as shadowing and specularity, pose a challenge during the analysis of the acquired volumes as they substantially obfuscate the location of tissue interfaces. To deal with the artifacts and speckle noise patterns and accurately segment the shallowest tissue interface, we propose a cascaded neural network framework, which comprises of a conditional Generative Adversarial Network (cGAN) and a Tissue Interface Segmentation Network (TISN). The cGAN pre-segments OCT B-scans by removing undesired specular artifacts and speckle noise patterns just above the shallowest tissue interface, and the TISN combines the original OCT image with the pre-segmentation to segment the shallowest interface. We show the applicability of the cascaded framework to corneal datasets, demonstrate that it precisely segments the shallowest corneal interface, and also show its generalization capacity to limbal datasets. We also propose a hybrid framework, wherein the cGAN pre-segmentation is passed to a traditional image analysis-based segmentation algorithm, and describe the improved segmentation performance. To the best of our knowledge, this is the first approach to remove severe specular artifacts and speckle noise patterns (prior to the shallowest interface) that affects the interpretation of anterior segment OCT datasets, thereby resulting in the accurate segmentation of the shallowest tissue interface. To the best of our knowledge, this is the first work to show the potential of incorporating a cGAN into larger deep learning frameworks for improved corneal and limbal OCT image segmentation. Our cGAN design directly improves the visualization of corneal and limbal OCT images from OCT scanners, and improves the performance of current OCT segmentation algorithms.

Deep learning based topology guaranteed surface and MME segmentation of multiple sclerosis subjects from retinal OCT

Optical coherence tomography (OCT) is a noninvasive imaging modality that can be used to obtain depth images of the retina. Patients with multiple sclerosis (MS) have thinning retinal nerve fiber and ganglion cell layers, and approximately 5% of MS patients will develop microcystic macular edema (MME) within the retina. Segmentation of both the retinal layers and MME can provide important information to help monitor MS progression. Graph-based segmentation with machine learning preprocessing is the leading method for retinal layer segmentation, providing accurate surface delineations with the correct topological ordering. However, graph methods are time-consuming and they do not optimally incorporate joint MME segmentation. This paper presents a deep network that extracts continuous, smooth, and topology-guaranteed surfaces and MMEs. The network learns shape priors automatically during training rather than being hard-coded as in graph methods. In this new approach, retinal surfaces and MMEs are segmented together with two cascaded deep networks in a single feed forward propagation. The proposed framework obtains retinal surfaces (separating the layers) with sub-pixel surface accuracy comparable to the best existing graph methods and MMEs with better accuracy than the state-of-the-art method. The full segmentation operation takes only ten seconds for a 3D volume.

Topographic correlation and asymmetry analysis of ganglion cell layer thinning and the retinal nerve fiber layer with localized visual field defects

Purpose

To evaluate the accuracy of the measurement of the ganglion cell layer (GCL) of the posterior pole analysis (PPA) software of the Spectralis spectral-domain (SD) optical coherence tomography (OCT) device (Heidelberg Engineering, Inc., Heidelberg, Germany), the asymmetry of paired GCL sectors, the total retinal thickness asymmetry (RTA), and the peripapillary retinal nerve fiber layer (pRNFL) test to discriminate between healthy, early and advanced glaucoma eyes.

Methods

Three hundred eighteen eyes of 161 individuals with reliable visual fields (VF) were enrolled in this study. All participants were examined using the standard posterior pole and the pRNFL protocols of the Spectralis OCT device. VF impairment was graded in hemifields, and the GCL sectors were correlated with this damage. Thicknesses of each GCL, the GCL map deviation asymmetry and the pRNFL were compared between control and glaucomatous eyes. The area under the receiver operating characteristic curve (AUC) of these analyses was assessed.

Results

Fourteen of the 16 sectors of the GCL and pRNFL were significantly thinner in eyes with glaucoma than in control eyes (p<0.006). Similarly, the GCL map deviation showed a significant difference between these eyes and both the control eyes as well as the eyes with early glaucoma (p = 0.001 and p = 0.039, respectively). The highest values of AUC to diagnose both early and advanced glaucoma corresponded to the average pRNFL analysis and the GCL map deviation (AUC>0.823, p<0.040 and AUC>0.708, p<0.188, respectively).

Conclusions

Although 16 central sectors of the GCL observed with PPA showed good correlation with VF damage, the pRNFL and the GCL map deviation were more effective for discrimination of glaucomatous damage.

Spatial analysis of thickness changes in ten retinal layers of Alzheimer's disease patients based on optical coherence tomography

The retina is an attractive source of biomarkers since it shares many features with the brain. Thickness differences in 10 retinal layers between 19 patients with mild Alzheimer’s disease (AD) and a control group of 24 volunteers were investigated. Retinal layers were automatically segmented and their thickness at each scanned point was measured, corrected for tilt and spatially normalized. When the mean thickness of entire layers was compared between patients and controls, only the outer segment layer of patients showed statistically significant thinning. However, when the layers were compared point-by point, patients showed statistically significant thinning in irregular regions of total retina and nerve fiber, ganglion cell, inner plexiform, inner nuclear and outer segment layers. Our method, based on random field theory, provides a precise delimitation of regions where total retina and each of its layers show a statistically significant thinning in AD patients. All layers, except inner nuclear and outer segments, showed thickened regions. New analytic methods have shown that thinned regions are interspersed with thickened ones in all layers, except inner nuclear and outer segments. Across different layers we found a statistically significant trend of the thinned regions to overlap and of the thickened ones to avoid overlapping.

Enhanced Grid-Based Visual Analysis of Retinal Layer Thickness with Optical Coherence Tomography

Optical coherence tomography enables high-resolution 3D imaging of retinal layers in the human eye. The thickness of the layers is commonly assessed to understand a variety of retinal and systemic disorders. Yet, the thickness data are complex and currently need to be considerably reduced prior to further processing and analysis. This leads to a loss of information on localized variations in thickness, which is important for early detection of certain retinal diseases. We propose an enhanced grid-based reduction and exploration of retinal thickness data. Alternative grids are computed, their representation quality is rated, and best fitting grids for given thickness data are suggested. Selected grids are then visualized, adapted, and compared at different levels of granularity. A visual analysis tool bundles all computational, visual, and interactive means in a flexible user interface. We demonstrate the utility of our tool in a complementary analysis procedure, which eases the evaluation of ophthalmic study data. Ophthalmologists successfully applied our solution to study localized variations in thickness of retinal layers in patients with diabetes mellitus.

RETOUCH: The Retinal OCT Fluid Detection and Segmentation Benchmark and Challenge

Retinal swelling due to the accumulation of fluid is associated with the most vision-threatening retinal diseases. Optical coherence tomography (OCT) is the current standard of care in assessing the presence and quantity of retinal fluid and image-guided treatment management. Deep learning methods have made their impact across medical imaging, and many retinal OCT analysis methods have been proposed. However, it is currently not clear how successful they are in interpreting the retinal fluid on OCT, which is due to the lack of standardized benchmarks. To address this, we organized a challenge RETOUCH in conjunction with MICCAI 2017, with eight teams participating. The challenge consisted of two tasks: fluid detection and fluid segmentation. It featured for the first time: all three retinal fluid types, with annotated images provided by two clinical centers, which were acquired with the three most common OCT device vendors from patients with two different retinal diseases. The analysis revealed that in the detection task, the performance on the automated fluid detection was within the inter-grader variability. However, in the segmentation task, fusing the automated methods produced segmentations that were superior to all individual methods, indicating the need for further improvements in the segmentation performance.

Review on Retrospective Procedures to Correct Retinal Motion Artefacts in OCT Imaging

Motion artefacts from involuntary changes in eye fixation remain a major imaging issue in optical coherence tomography (OCT). This paper reviews the state-of-the-art of retrospective procedures to correct retinal motion and axial eye motion artefacts in OCT imaging. Following an overview of motion induced artefacts and correction strategies, a chronological survey of retrospective approaches since the introduction of OCT until the current days is presented. Pre-processing, registration, and validation techniques are described. The review finishes by discussing the limitations of the current techniques and the challenges to be tackled in future developments.

Deep Learning Based Sub-Retinal Fluid Segmentation in Central Serous Chorioretinopathy Optical Coherence Tomography Scans

Development of an automated sub-retinal fluid segmentation technique from optical coherence tomography (OCT) scans is faced with challenges such as noise and motion artifacts present in OCT images, variation in size, shape and location of fluid pockets within the retina. The ability of a fully convolutional neural network to automatically learn significant low level features to differentiate subtle spatial variations makes it suitable for retinal fluid segmentation task. Hence, a fully convolutional neural network has been proposed in this work for the automatic segmentation of sub-retinal fluid in OCT scans of central serous chorioretinopathy (CSC) pathology. The proposed method has been evaluated on a dataset of 15 OCT volumes and an average Dice rate, Precision and Recall of 0.91, 0.93 and 0.89 respectively has been achieved over the test set.

A Mechanical Approach for Smooth Surface Fitting to Delineate Vessel Walls in Optical Coherence Tomography Images

Automated analysis of vascular imaging techniques is limited by the inability to precisely determine arterial borders. Intravascular optical coherence tomography (OCT) offers unprecedented detail of artery wall structure and composition, but does not provide consistent visibility of the outer border of the vessel due to the limited penetration depth. Existing interpolation and surface fitting methods prove insufficient to accurately fill the gaps between the irregularly spaced and sometimes unreliably identified visible segments of the vessel outer border. This paper describes an intuitive, efficient, and flexible new method of 3D surface fitting and smoothing suitable for this task. An anisotropic linear-elastic mesh is fit to irregularly spaced and uncertain data points corresponding to visible segments of vessel borders, enabling the fully automated delineation of the entire inner and outer borders of diseased vessels in OCT images for the first time. In a clinical dataset, the proposed smooth surface fitting approach had great agreement when compared with human annotations: areas differed by just 11 ± 11% (0.93 ± 0.84 mm²), with a coefficient of determination of 0.89. Overlapping and non-overlapping area ratios were 0.91 and 0.18, respectively, with a sensitivity of 90.8 and specificity of 99.0. This spring mesh method of contour fitting significantly outperformed all alternative surface fitting and interpolation approaches tested. The application of this promising proposed method is expected to enhance clinical intervention and translational research using OCT.

Automated segmentation of macular edema in OCT using deep neural networks

Macular edema is an eye disease that can affect visual acuity. Typical disease symptoms include subretinal fluid (SRF) and pigment epithelium detachment (PED). Optical coherence tomography (OCT) has been widely used for diagnosing macular edema because of its non-invasive and high resolution properties. Segmentation for macular edema lesions from OCT images plays an important role in clinical diagnosis. Many computer-aided systems have been proposed for the segmentation. Most traditional segmentation methods used in these systems are based on low-level hand-crafted features, which require significant domain knowledge and are sensitive to the variations of lesions. To overcome these shortcomings, this paper proposes to use deep neural networks (DNNs) together with atrous spatial pyramid pooling (ASPP) to automatically segment the SRF and PED lesions. Lesions-related features are first extracted by DNNs, then processed by ASPP which is composed of multiple atrous convolutions with different fields of view to accommodate the various scales of the lesions. Based on ASPP, a novel module called stochastic ASPP (sASPP) is proposed to combat the co-adaptation of multiple atrous convolutions. A large OCT dataset provided by a competition platform called "AI Challenger" are used to train and evaluate the proposed model. Experimental results demonstrate that the DNNs together with ASPP achieve higher segmentation accuracy compared with the state-of-the-art method. The stochastic operation added in sASPP is empirically verified as an effective regularization method that can alleviate the overfitting problem and significantly reduce the validation error.

Diagnostic Capability of Three-Dimensional Macular Parameters for Glaucoma Using Optical Coherence Tomography Volume Scans

Purpose

To compare the diagnostic capability of three-dimensional (3D) macular parameters against traditional two-dimensional (2D) retinal nerve fiber layer (RNFL) thickness using spectral domain optical coherence tomography. To determine if manual correction and interpolation of B-scans improve the ability of 3D macular parameters to diagnose glaucoma.

Methods

A total of 101 open angle glaucoma patients (29 with early glaucoma) and 57 healthy subjects had peripapillary 2D RNFL thickness and 3D macular volume scans. Four parameters were calculated for six different-sized annuli: total macular thickness (M-thickness), total macular volume (M-volume), ganglion cell complex (GCC) thickness, and GCC volume of the innermost 3 macular layers (retinal nerve fiber layer + ganglion cell layer + inner plexiform layer). All macular parameters were calculated with and without correction and interpolation of frames with artifacts. The areas under the receiver operating characteristic curves (AUROC) were calculated for all the parameters.

Results

The 3D macular parameter with the best diagnostic performance was GCC-volume-34, with an inner diameter of 3 mm and an outer of 4 mm. The AUROC for RNFL thickness and GCC-volume-34 were statistically similar for all regions (global: RNFL thickness 0.956, GCC-volume-34 0.939, P value = 0.3827), except for the temporal GCC-volume-34, which was significantly better than temporal RNFL thickness (P value = 0.0067). Correction of artifacts did not significantly change the AUROC of macular parameters (P values between 0.8452 and 1.0000).

Conclusions

The diagnostic performance of best macular parameters (GCC-volume-34 and GCC-thickness-34) were similar to or better than 2D RNFL thickness. Manual correction of artifacts with data interpolation is unnecessary in the clinical setting.

Persistent and Progressive Outer Retina Thinning in Frontotemporal Degeneration

OBJECTIVE

While Alzheimer's disease is associated with inner retina thinning measured by spectral-domain optical coherence tomography (SD-OCT), our previous cross-sectional study suggested outer retina thinning in frontotemporal degeneration (FTD) patients compared to controls without neurodegenerative disease; we sought to evaluate longitudinal changes of this potential biomarker.

METHODS

SD-OCT retinal layer thicknesses were measured at baseline and after 1-2 years. Clinical criteria, genetic analysis, and a cerebrospinal fluid biomarker (total tau: β-amyloid) to exclude likely underlying Alzheimer's disease pathology were used to define a subgroup of predicted molecular pathology (i.e., tauopathy). Retinal layer thicknesses and rates of change in all FTD patients (n = 16 patients, 30 eyes) and the tauopathy subgroup (n = 9 patients,16 eyes) were compared to controls (n = 30 controls, 47 eyes) using a generalized linear model accounting for inter-eye correlation and adjusting for age, sex, and race. Correlations between retinal layer thicknesses and Mini-Mental State Examinations (MMSE) were assessed.

RESULTS

Compared to controls, returning FTD patients (143 vs. 130 μm, p = 0.005) and the tauopathy subgroup (143 vs. 128 μm, p = 0.03) had thinner outer retinas but similar inner layer thicknesses. Compared to controls, the outer retina thinning rate was not significant for all FTD patients (p = 0.34), but was significant for the tauopathy subgroup (-3.9 vs. 0.4 μm/year, p = 0.03). Outer retina thickness change correlated with MMSE change in FTD patients (Spearman rho = 0.60, p = 0.02) and the tauopathy subgroup (rho = 0.73, p = 0.04).

CONCLUSION

Our finding of FTD outer retina thinning persists and longitudinally correlates with disease progression. These findings were especially seen in probable tauopathy patients, which showed progressive outer retina thinning.

Unsupervised Identification of Disease Marker Candidates in Retinal OCT Imaging Data

The identification and quantification of markers in medical images is critical for diagnosis, prognosis, and disease management. Supervised machine learning enables the detection and exploitation of findings that are known a priori after annotation of training examples by experts. However, supervision does not scale well, due to the amount of necessary training examples, and the limitation of the marker vocabulary to known entities. In this proof-of-concept study, we propose unsupervised identification of anomalies as candidates for markers in retinal optical coherence tomography (OCT) imaging data without a constraint to a priori definitions. We identify and categorize marker candidates occurring frequently in the data and demonstrate that these markers show a predictive value in the task of detecting disease. A careful qualitative analysis of the identified data driven markers reveals how their quantifiable occurrence aligns with our current understanding of disease course, in early- and late age-related macular degeneration (AMD) patients. A multi-scale deep denoising autoencoder is trained on healthy images, and a one-class support vector machine identifies anomalies in new data. Clustering in the anomalies identifies stable categories. Using these markers to classify healthy-, early AMD- and late AMD cases yields an accuracy of 81.40%. In a second binary classification experiment on a publicly available data set (healthy versus intermediate AMD), the model achieves an area under the ROC curve of 0.944.

Layer boundary evolution method for macular OCT layer segmentation

Optical coherence tomography (OCT) is used to produce high resolution depth images of the retina and is now the standard of care for in-vivo ophthalmological assessment. It is also increasingly being used for evaluation of neurological disorders such as multiple sclerosis (MS). Automatic segmentation methods identify the retinal layers of the macular cube providing consistent results without intra- and inter-rater variation and is faster than manual segmentation. In this paper, we propose a fast multi-layer macular OCT segmentation method based on a fast level set method. Our framework uses contours in an optimized approach specifically for OCT layer segmentation over the whole macular cube. Our algorithm takes boundary probability maps from a trained random forest and iteratively refines the prediction to subvoxel precision. Evaluation on both healthy and multiple sclerosis subjects shows that our method is statistically better than a state-of-the-art graph-based method.

Robust segmentation of retinal layers in optical coherence tomography images based on a multistage active contour model

Optical Coherence Tomography (OCT) constitutes an imaging technique that is increasing its popularity in the ophthalmology field, since it offers a more complete set of information about the main retinal structures. Hence, it offers detailed information about the eye fundus morphology, allowing the identification of many intraretinal pathological signs. For that reason, over the recent years, Computer-Aided Diagnosis (CAD) systems have spread to work with this image modality and analyze its information. A crucial step for the analysis of the retinal tissues implies the identification and delimitation of the different retinal layers. In this context, we present in this work a fully automatic method for the identification of the main retinal layers that delimits the retinal region. Thus, an active contour-based model was completely adapted and optimized to segment these main retinal boundaries. This fully automatic method uses the information of the horizontal placement of these retinal layers and their relative location over the analyzed images to restrict the search space, considering the presence of shadows that are normally generated by pathological or non-pathological artifacts. The validation process was done using the groundtruth of an expert ophthalmologist analyzing healthy as well as unhealthy patients with different degrees of diabetic retinopathy (without macular edema, with macular edema and with lesions in the photoreceptor layers). Quantitative results are in line with the state of the art of this domain, providing accurate segmentations of the retinal layers even when significative pathological alterations are present in the eye fundus. Therefore, the proposed method is robust enough to be used in complex environments, making it feasible for the ophthalmologists in their routine clinical practice.

Serous Retinal Detachment Causes a Transient Reduction on Spectral Domain OCT Estimates of Ganglion Cell Layer Thickness

SIGNIFICANCE

During the acute stage of central serous chorioretinopathy (CSC) with retinal elevation, the spectral domain optical coherence tomography (SD-OCT) estimate of ganglion cell layer complex thickness is reduced. Thickness returns to normal after resolution of the event. Measurement error is at least partially responsible for this effect. The reduction in ganglion cell layer complex thickness does not represent atrophy and is not predictive of a poor outcome.

PURPOSE

We investigated the effects of serous retinal detachment on the ganglion cell layer complex analysis (GCA) by SD-OCT in CSC patients during the acute episodes and after resolution of fluid.

METHODS

We retrospectively reviewed medical records of 30 patients who visited the hospital with a first episode of CSC. We analyzed GCA maps using SD-OCT (Cirrus; Carl Zeiss Meditec, Dublin, CA) at the initial visit with serous retinal elevation and after the absorption of subretinal fluid. For repeatability analysis, we used the intraclass correlation and repeatability coefficient from two consecutive measurements 5 minutes apart in 12 patients.

RESULTS

At the initial visit, an average thickness of ganglion cell layer complex was thinner than that measured in the fellow eye (67.4 ± 27.4 μm), but after the absorption of subretinal fluid, it normalized to 87.0 ± 6.7 μm; the difference was statistically significant. The intraclass correlation and repeatability coefficient were low during the period of serous elevation but normalized after fluid resorption. Abnormalities of GCA resulted from the segmentation error of ganglion cell layer and inner plexiform layer during the acute phase of CSC.

CONCLUSIONS

Serous retinal detachment can affect the GCA and repeatability measurements of the GCA. Clinicians should consider this finding when using the GCA measurement in the diagnosis and management of the patients with retinal contour changes such as retinal elevation including CSC.

Optimal surface segmentation with convex priors in irregularly sampled space

Optimal surface segmentation is a state-of-the-art method used for segmentation of multiple globally optimal surfaces in volumetric datasets. The method is widely used in numerous medical image segmentation applications. However, nodes in the graph based optimal surface segmentation method typically encode uniformly distributed orthogonal voxels of the volume. Thus the segmentation cannot attain an accuracy greater than a single unit voxel, i.e. the distance between two adjoining nodes in graph space. Segmentation accuracy higher than a unit voxel is achievable by exploiting partial volume information in the voxels which shall result in non-equidistant spacing between adjoining graph nodes. This paper reports a generalized graph based multiple surface segmentation method with convex priors which can optimally segment the target surfaces in an irregularly sampled space. The proposed method allows non-equidistant spacing between the adjoining graph nodes to achieve subvoxel segmentation accuracy by utilizing the partial volume information in the voxels. The partial volume information in the voxels is exploited by computing a displacement field from the original volume data to identify the subvoxel-accurate centers within each voxel resulting in non-equidistant spacing between the adjoining graph nodes. The smoothness of each surface modeled as a convex constraint governs the connectivity and regularity of the surface. We employ an edge-based graph representation to incorporate the necessary constraints and the globally optimal solution is obtained by computing a minimum s-t cut. The proposed method was validated on 10 intravascular multi-frame ultrasound image datasets for subvoxel segmentation accuracy. In all cases, the approach yielded highly accurate results. Our approach can be readily extended to higher-dimensional segmentations.

Automatic three-dimensional segmentation of endoscopic airway OCT images

Automatic delineation and segmentation of airway structures from endoscopic optical coherence tomography (OCT) images improve image analysis efficiency and thus has been of particular interest. Conventional two-dimensional automatic segmentation methods, such as the dynamic programming approach, ensures the edge-continuity in the xz-direction (intra-B-scan), but fails to preserve the surface-continuity when concerning the y-direction (inter-B-scan). To solve this, we present a novel automatic three-dimensional (3D) airway segmentation strategy. Our segmentation scheme includes an artifact-oriented pre-processing pipeline and a modified 3D optimal graph search algorithm incorporating adaptive tissue-curvature adjustment. The proposed algorithm is tested on endoscopic airway OCT image data sets acquired by different swept-source OCT platforms, and on different animal and human models. With our method, the results show continuous surface segmentation performance, which is both robust and accurate.

f-AnoGAN: Fast unsupervised anomaly detection with generative adversarial networks

Obtaining expert labels in clinical imaging is difficult since exhaustive annotation is time-consuming. Furthermore, not all possibly relevant markers may be known and sufficiently well described a priori to even guide annotation. While supervised learning yields good results if expert labeled training data is available, the visual variability, and thus the vocabulary of findings, we can detect and exploit, is limited to the annotated lesions. Here, we present fast AnoGAN (f-AnoGAN), a generative adversarial network (GAN) based unsupervised learning approach capable of identifying anomalous images and image segments, that can serve as imaging biomarker candidates. We build a generative model of healthy training data, and propose and evaluate a fast mapping technique of new data to the GAN's latent space. The mapping is based on a trained encoder, and anomalies are detected via a combined anomaly score based on the building blocks of the trained model - comprising a discriminator feature residual error and an image reconstruction error. In the experiments on optical coherence tomography data, we compare the proposed method with alternative approaches, and provide comprehensive empirical evidence that f-AnoGAN outperforms alternative approaches and yields high anomaly detection accuracy. In addition, a visual Turing test with two retina experts showed that the generated images are indistinguishable from real normal retinal OCT images. The f-AnoGAN code is available at https://github.com/tSchlegl/f-AnoGAN.

Inner macular layer thickness by spectral domain optical coherence tomography in children and adults: a hospital-based study

Purpose

To establish the normative ranges of macular ganglion cell layer (mGCL) and macular inner plexiform layer (mIPL) thickness using Spectralis spectral domain optical coherence tomography (SD-OCT) (Heidelberg Engineering, Inc., Heidelberg, Germany) in both Korean children and adults, and to determine factors associated with mGCL and mIPL thickness.

Methods

We conducted a retrospective, observational study of 573 healthy subjects (5–70 years old) who underwent comprehensive ophthalmic examinations in a single institution. Each inner retinal layer thickness was measured using SD-OCT and automatic segmentation software. Cross-sectional analysis was used to evaluate the effect of gender, age and ocular parameters on mGCL and mIPL thickness. Normative ranges of mGCL and mIPL thickness according to age, gender and factors associated with mGCL and mIPL thickness were measured.

Results

The mean mGCL and mIPL thickness were 40.6±2.8 and 33.8±2.0 µm, respectively. Determinants of inner sector mGCL thickness were circumpapillary retinal nerve fibre layer (cpRNFL) thickness (β=1.172, p<0.001), age (β=−0.019, p=0.021) and male gender (β=1.452, p<0.001). Determinants of inner sector mIPL thickness were cpRNFL (β=0.952, p<0.001) and male gender (β=1.163, p<0.001). The inner sector mGCL and mIPL thickness increased significantly with age in children (β=0.174, p=0.009 and β=0.115, p=0.013), and then decreased in adults (β=−0.070, p<0.001 and β=−0.024, p=0.032). In the case of outer sectors, mGCL and mIPL thickness were not significantly related to age and gender.

Conclusions

This study ensured a normative range of the mGCL and mIPL thickness using Spectralis OCT. Gender, age and cpRNFL thickness significantly correlated with mGCL and mIPL thickness. This information should be considered in the interpretation of SD-OCT data.

Interplay Between Macular Retinal Changes and White Matter Integrity in Early Alzheimer's Disease

This study aims to investigate the relationship between structural changes in the retina and white matter in the brain, in early Alzheimer's disease (AD). Twenty-three healthy controls (mean age = 63.4±7.5 years) and seventeen AD patients (mean age = 66.5±6.6 years) were recruited for this study. By combining two imaging techniques-optical coherence tomography and diffusion tensor imaging (DTI)-the association between changes in the thickness of individual retinal layers and white matter dysfunction in early AD was assessed. Retinal layers were segmented, and thickness measurements were obtained for each layer. DTI images were analyzed with a quantitative data-driven approach to evaluating whole-brain diffusion metrics, using tract-based spatial statistics. Diffusion metrics, such as fractional anisotropy, are markers for white matter integrity. Multivariate and partial correlation analyses evaluating the association between individual retinal layers thickness and diffusion metrics were performed. We found that axial diffusivity, indexing axonal integrity, was significantly reduced in AD (p = 0.016, Cohen's d = 1.004) while in the retina, only a marginal trend for significance was found for the outer plexiform layer (p = 0.084, Cohen's d = 0.688). Furthermore, a positive association was found in the AD group between fractional anisotropy and the inner nuclear layer thickness (p < 0.05, r = 0.419, corrected for multiple comparisons by controlling family-wise error rate). Our findings suggest that axonal damage in the brain dominates early on in this condition and shows an association with retinal structural integrity already at initial stages of AD. These findings are consistent with an early axonal degeneration mechanism in AD.

Retinal optical coherence tomography image enhancement via deep learning

Optical coherence tomography (OCT) images of the retina are a powerful tool for diagnosing and monitoring eye disease. However, they are plagued by speckle noise, which reduces image quality and reliability of assessment. This paper introduces a novel speckle reduction method inspired by the recent successes of deep learning in medical imaging. We present two versions of the network to reflect the needs and preferences of different end-users. Specifically, we train a convolution neural network to denoise cross-sections from OCT volumes of healthy eyes using either (1) mean-squared error, or (2) a generative adversarial network (GAN) with Wasserstein distance and perceptual similarity. We then interrogate the success of both methods with extensive quantitative and qualitative metrics on cross-sections from both healthy and glaucomatous eyes. The results show that the former approach provides state-of-the-art improvement in quantitative metrics such as PSNR and SSIM, and aids layer segmentation. However, the latter approach, which puts more weight on visual perception, outperformed for qualitative comparisons based on accuracy, clarity, and personal preference. Overall, our results demonstrate the effectiveness and efficiency of a deep learning approach to denoising OCT images, while maintaining subtle details in the images.

Automatic Segmentation of Retinal Layer in OCT Images With Choroidal Neovascularization

Age-related macular degeneration is one of the main causes of blindness. However, the internal structures of retinas are complex and difficult to be recognized due to the occurrence of neovascularization. Traditional surface detection methods may fail in the layer segmentation. In this paper, a supervised method is reported for simultaneously segmenting layers and neovascularization. Three spatial features, seven gray-level-based features, and 14 layer-like features are extracted for the neural network classifier. The coarse surfaces of different optical coherence tomography (OCT) images can thus be found. To describe and enhance retinal layers with different thicknesses and abnormalities, multi-scale bright and dark layer detection filters are introduced. A constrained graph search algorithm is also proposed to accurately detect retinal surfaces. The weights of nodes in the graph are computed based on these layer-like responses. The proposed method was evaluated on 42 spectral-domain OCT images with age-related macular degeneration. The experimental results show that the proposed method outperforms state-of-the-art methods.

Active contour method for ILM segmentation in ONH volume scans in retinal OCT

The optic nerve head (ONH) is affected by many neurodegenerative and autoimmune inflammatory conditions. Optical coherence tomography can acquire high-resolution 3D ONH scans. However, the ONH's complex anatomy and pathology make image segmentation challenging. This paper proposes a robust approach to segment the inner limiting membrane (ILM) in ONH volume scans based on an active contour method of Chan-Vese type, which can work in challenging topological structures. A local intensity fitting energy is added in order to handle very inhomogeneous image intensities. A suitable boundary potential is introduced to avoid structures belonging to outer retinal layers being detected as part of the segmentation. The average intensities in the inner and outer region are then rescaled locally to account for different brightness values occurring among the ONH center. The appropriate values for the parameters used in the complex computational model are found using an optimization based on the differential evolution algorithm. The evaluation of results showed that the proposed framework significantly improved segmentation results compared to the commercial solution.

Association between Visual Acuity and Retinal Layer Metrics in Diabetics with and without Macular Edema

Purpose

Diabetes is known to cause alterations in retinal microvasculature and tissue that progressively lead to visual impairment. Optical coherence tomography (OCT) is useful for assessment of total retinal thickening due to diabetic macular edema (DME). In the current study, we determined associations between visual acuity (VA) and retinal layer thickness, reflectance, and interface disruption derived from enface OCT images in subjects with and without DME.

Materials and Methods

Best corrected VA was measured and high-density OCT volume scans were acquired in 149 diabetic subjects. A previously established image segmentation method identified retinal layer interfaces and locations of visually indiscernible (disrupted) interfaces. Enface thickness maps and reflectance images of the nerve fiber layer (NFL), combined ganglion cell and inner plexiform layer (GCLIPL), inner nuclear layer (INL), outer plexiform layer (OPL), outer nuclear layer (ONL), photoreceptor outer segment layer (OSL), and retinal pigment epithelium (RPE) were generated in the central macular subfield. The associations among VA and retinal layer metrics were determined by multivariate linear regressions after adjusting for covariates (age, sex, race, HbA1c, diabetes type, and duration) and correcting for multiple comparisons.

Results

In DME subjects, increased GCLIPL and OPL thickness and decreased OSL thickness were associated with reduced VA. Furthermore, increased NFL reflectance and decreased OSL reflectance were associated with reduced VA. Additionally, increased areas of INL and ONL interface disruptions were associated with reduced VA. In subjects without DME, increased INL thickness was associated with reduced VA, whereas in subjects without DME but with previous antivascular endothelium growth factor treatment, thickening of OPL was associated with reduced VA.

Conclusions

Alterations in retinal layer thickness and reflectance metrics derived from enface OCT images were associated with reduced VA with and without presence of DME, suggestive of their potential for monitoring development, progression, and treatment of DME.

A supervised joint multi-layer segmentation framework for retinal optical coherence tomography images using conditional random field

BACKGROUND AND OBJECTIVE

Accurate segmentation of the intra-retinal tissue layers in Optical Coherence Tomography (OCT) images plays an important role in the diagnosis and treatment of ocular diseases such as Age-Related Macular Degeneration (AMD) and Diabetic Macular Edema (DME). The existing energy minimization based methods employ multiple, manually handcrafted cost terms and often fail in the presence of pathologies. In this work, we eliminate the need to handcraft the energy by learning it from training images in an end-to-end manner. Our method can be easily adapted to pathologies by re-training it on an appropriate dataset.

METHODS

We propose a Conditional Random Field (CRF) framework for the joint multi-layer segmentation of OCT B-scans. The appearance of each retinal layer and boundary is modeled by two convolutional filter banks and the shape priors are modeled using Gaussian distributions. The total CRF energy is linearly parameterized to allow a joint, end-to-end training by employing the Structured Support Vector Machine formulation.

RESULTS

The proposed method outperformed three benchmark algorithms on four public datasets. The NORMAL-1 and NORMAL-2 datasets contain healthy OCT B-scans while the AMD-1 and DME-1 dataset contain B-scans of AMD and DME cases respectively. The proposed method achieved an average unsigned boundary localization error (U-BLE) of 1.52 pixels on NORMAL-1, 1.11 pixels on NORMAL-2 and 2.04 pixels on the combined NORMAL-1 and DME-1 dataset across the eight layer boundaries, outperforming the three benchmark methods in each case. The Dice coefficient was 0.87 on NORMAL-1, 0.89 on NORMAL-2 and 0.84 on the combined NORMAL-1 and DME-1 dataset across the seven retinal layers. On the combined NORMAL-1 and AMD-1 dataset, we achieved an average U-BLE of 1.86 pixels on the ILM, inner and outer RPE boundaries and a Dice of 0.98 for the ILM-RPE_in region and 0.81 for the RPE layer.

CONCLUSION

We have proposed a supervised CRF based method to jointly segment multiple tissue layers in OCT images. It can aid the ophthalmologists in the quantitative analysis of structural changes in the retinal tissue layers for clinical practice and large-scale clinical studies.

Robust layer segmentation of esophageal OCT images based on graph search using edge-enhanced weights

Automatic segmentation of esophageal layers in OCT images is crucial for studying esophageal diseases and computer-assisted diagnosis. This work aims to improve the current techniques to increase the accuracy and robustness for esophageal OCT image segmentation. A two-step edge-enhanced graph search (EEGS) framework is proposed in this study. Firstly, a preprocessing scheme is applied to suppress speckle noise and remove the disturbance in the esophageal structure. Secondly, the image is formulated into a graph and layer boundaries are located by graph search. In this process, we propose an edge-enhanced weight matrix for the graph by combining the vertical gradients with a Canny edge map. Experiments on esophageal OCT images from guinea pigs demonstrate that the EEGS framework is more robust and more accurate than the current segmentation method. It can be potentially useful for the early detection of esophageal diseases.

Multiple surface segmentation using convolution neural nets: application to retinal layer segmentation in OCT images

Automated segmentation of object boundaries or surfaces is crucial for quantitative image analysis in numerous biomedical applications. For example, retinal surfaces in optical coherence tomography (OCT) images play a vital role in the diagnosis and management of retinal diseases. Recently, graph based surface segmentation and contour modeling have been developed and optimized for various surface segmentation tasks. These methods require expertly designed, application specific transforms, including cost functions, constraints and model parameters. However, deep learning based methods are able to directly learn the model and features from training data. In this paper, we propose a convolutional neural network (CNN) based framework to segment multiple surfaces simultaneously. We demonstrate the application of the proposed method by training a single CNN to segment three retinal surfaces in two types of OCT images - normal retinas and retinas affected by intermediate age-related macular degeneration (AMD). The trained network directly infers the segmentations for each B-scan in one pass. The proposed method was validated on 50 retinal OCT volumes (3000 B-scans) including 25 normal and 25 intermediate AMD subjects. Our experiment demonstrated statistically significant improvement of segmentation accuracy compared to the optimal surface segmentation method with convex priors (OSCS) and two deep learning based UNET methods for both types of data. The average computation time for segmenting an entire OCT volume (consisting of 60 B-scans each) for the proposed method was 12.3 seconds, demonstrating low computation costs and higher performance compared to the graph based optimal surface segmentation and UNET based methods.

Automated segmentation of retinal layer boundaries and capillary plexuses in wide-field optical coherence tomographic angiography

Advances in the retinal layer segmentation of structural optical coherence tomography (OCT) images have allowed the separation of capillary plexuses in OCT angiography (OCTA). With the increased scanning speeds of OCT devices and wider field images (≥10 mm on fast-axis), greater retinal curvature and anatomic variations have introduced new challenges. In this study, we developed a novel automated method to segment seven retinal layer boundaries and two retinal plexuses in wide-field OCTA images. The algorithm was initialized by a series of points forming a guidance point array that estimates the location of retinal layer boundaries. A guided bidirectional graph search method consisting of an improvement of our previous segmentation algorithm was used to search for the precise boundaries. We validated the method on normal and diseased eyes, demonstrating subpixel accuracy for all groups. By allowing independent visualization of the superficial and deep plexuses, this method shows potential for the detection of plexus-specific peripheral vascular abnormalities.

Development of an efficient algorithm for the detection of macular edema from optical coherence tomography images

PURPOSE

Detection of eye diseases and their treatment is a key to reduce blindness, which impacts human daily needs like driving, reading, writing, etc. Several methods based on image processing have been used to monitor the presence of macular diseases. Optical coherence tomography (OCT) imaging is the most efficient technique used to observe eye diseases. This paper proposes an efficient algorithm to automatically classify normal as well as disease-affected (macular edema) retinal OCT images by using segmentation of Inner Limiting Membrane and the Choroid Layer.

METHODS

In the proposed method, preprocessing of the input image is done to improve the quality and reduce the speckle noise. The layer segmentation is done on the gradient image, and graph theory and dynamic programming algorithm is performed. The feature vectors from segmented image are in terms of thickness profile and cyst fluid parameter, and these features are applied to various classifiers.

RESULTS

The proposed method was tested with the standard dataset collected from the Department of Ophthalmology, Duke University, and achieved a high accuracy rate of 99.4975%, sensitivity of 100%, and specificity of 99% for the SVM classifier.

CONCLUSIONS

An efficient algorithm is proposed for macular edema detection from OCT images using segmentation based on graph theory and dynamic programming algorithm. The comparison with alternative methods yielded results that demonstrate the superiority of the proposed algorithm for macular edema detection.

Three-dimensional graph-based skin layer segmentation in optical coherence tomography images for roughness estimation

Automatic skin layer segmentation in optical coherence tomography (OCT) images is important for a topographic assessment of skin or skin disease detection. However, existing methods cannot deal with the problem of shadowing in OCT images due to the presence of hair, scales, etc. In this work, we propose a method to segment the topmost layer of the skin (or the skin surface) using 3D graphs with a novel cost function to deal with shadowing in OCT images. 3D graph cuts use context information across B-scans when segmenting the skin surface, which improves the segmentation as compared to segmenting each B-scan separately. The proposed method reduces the segmentation error by more than 20% as compared to the best performing related work. The method has been applied to roughness estimation and shows a high correlation with a manual assessment. Promising results demonstrate the usefulness of the proposed method for skin layer segmentation and roughness estimation in both normal OCT images and OCT images with shadowing.

Automated Layer Segmentation of Retinal Optical Coherence Tomography Images Using a Deep Feature Enhanced Structured Random Forests Classifier

Optical coherence tomography (OCT) is a high-resolution and noninvasive imaging modality that has become one of the most prevalent techniques for ophthalmic diagnosis. Retinal layer segmentation is very crucial for doctors to diagnose and study retinal diseases. However, manual segmentation is often a time-consuming and subjective process. In this work, we propose a new method for automatically segmenting retinal OCT images, which integrates deep features and hand-designed features to train a structured random forests classifier. The deep convolutional features are learned from deep residual network. With the trained classifier, we can get the contour probability graph of each layer; finally, the shortest path is employed to achieve the final layer segmentation. The experimental results show that our method achieves good results with the mean layer contour error of 1.215 pixels, whereas that of the state of the art was 1.464 pixels, and achieves an F1-score of 0.885, which is also better than 0.863 that is obtained by the state of the art method.

Effect of patch size and network architecture on a convolutional neural network approach for automatic segmentation of OCT retinal layers

Deep learning strategies, particularly convolutional neural networks (CNNs), are especially suited to finding patterns in images and using those patterns for image classification. The method is normally applied to an image patch and assigns a class weight to the patch; this method has recently been used to detect the probability of retinal boundary locations in OCT images, which is subsequently used to segment the OCT image using a graph-search approach. This paper examines the effects of a number of modifications to the CNN architecture with the aim of optimizing retinal layer segmentation, specifically the effect of patch size as well as the network architecture design on CNN performance and subsequent layer segmentation. The results demonstrate that increasing patch size can improve the performance of the classification and provides a more reliable segmentation in the analysis of retinal layer characteristics in OCT imaging. Similarly, this work shows that changing aspects of the CNN network design can also significantly improve the segmentation results. This work also demonstrates that the performance of the method can change depending on the number of classes (i.e. boundaries) used to train the CNN, with fewer classes showing an inferior performance due to the presence of similar image features between classes that can trigger false positives. Changes in the network (patch size and or architecture) can be applied to provide a superior segmentation performance, which is robust to the class effect. The findings from this work may inform future CNN development in OCT retinal image analysis.

Fiber-based visible and near infrared optical coherence tomography (vnOCT) enables quantitative elastic light scattering spectroscopy in human retina

Elastic light scattering spectroscopy (ELSS) has been proven a powerful method in measuring tissue structures with exquisite nanoscale sensitivity. However, ELSS contrast in the living human retina has been relatively underexplored, primarily due to the lack of imaging tools with a large spectral bandwidth. Here, we report a simple all fiber-based setup to implement dual-channel visible and near infrared (NIR) optical coherence tomography (vnOCT) for human retinal imaging, bridging over a 300nm spectral gap. Remarkably, the fiber components in our vnOCT system support single-mode propagation for both visible and NIR light, both of which maintain excellent interference efficiencies with fringe visibility of 97% and 90%, respectively. The longitudinal chromatic aberration from the eye is corrected by a custom-designed achromatizing lens. The elegant fiber-based design enables simultaneous imaging for both channels and allows comprehensive ELSS analysis on several important anatomical layers, including nerve fiber layer, outer segment of the photoreceptors and retinal pigment epithelium. This vnOCT platform and method of ELSS analysis open new opportunities in understanding structure-function relationship in the human retina and in exploring new biomarkers for retinal diseases.

The retinal ganglion cell layer predicts normal-appearing white matter tract integrity in multiple sclerosis: A combined diffusion tensor imaging and optical coherence tomography approach

We investigated the relationship between retinal layers and normal-appearing white matter (WM) integrity in the brain of patients with relapsing-remitting multiple sclerosis (MS), using a combined diffusion tensor imaging and high resolution optical coherence tomography approach. Fifty patients and 62 controls were recruited. The patients were divided into two groups according to presence (n = 18) or absence (n = 32) of optic neuritis. Diffusion tensor data were analyzed with a voxel-wise whole brain analysis of diffusion metrics in WM with tract-based spatial statistics. Thickness measurements were obtained for each individual retinal layer. Partial correlation and multivariate regression analyses were performed, assessing the association between individual retinal layers and diffusion metrics across all groups. Region-based analysis was performed, by focusing on tracts associated with the visual system. Receiver operating characteristic (ROC) curves were computed to compare the biomarker potential for the diagnosis of MS, using the thickness of each retinal layer and diffusion metrics. In patients without optic neuritis, both ganglion cell layer (GCL) and inner plexiform layer thickness correlated with the diffusion metrics within and outside the visual system. GCL thickness was a significant predictor of diffusion metrics in the whole WM skeleton, unlike other layers. No association was observed for either controls or patients with a history of optic neuritis. ROC analysis showed that the biomarker potential for the diagnosis of MS based on the GCL was high when compared to other layers. We conclude that GCL integrity is a predictor of whole-brain WM disruption in MS patients without optic neuritis.

Multi-layer Fast Level Set Segmentation for Macular OCT

Segmenting optical coherence tomography (OCT) images of the retina is important in the diagnosis, staging, and tracking of ophthalmological diseases. Whereas automatic segmentation methods are typically much faster than manual segmentation, they may still take several minutes to segment a three-dimensional macular scan, and this can be prohibitive for routine clinical application. In this paper, we propose a fast, multi-layer macular OCT segmentation method based on a fast level set method. In our framework, the boundary evolution operations are computationally fast, are specific to each boundary between retinal layers, guarantee proper layer ordering, and avoid level set computation during evolution. Subvoxel resolution is achieved by reconstructing the level set functions after convergence. Experiments demonstrate that our method reduces the computation expense by 90% compared to graph-based methods and produces comparable accuracy to both graph-based and level set retinal OCT segmentation methods.

Sparse high order potentials for extending multi-surface segmentation of OCT images with drusen

Drusen quantification is important for evaluating age-related macular degeneration (AMD) progress. Most methods for retinal layers segmentation in optical coherence tomography (OCT) depend heavily on prior information. This improves robustness, but also has the downside of increasing surface rigidity. Hence, those algorithms normally smooth drusen borders, as significant local variations are not expected. In this work, we propose to integrate sparse higher order potentials (SHOPs) into a multi-surface segmentation framework to cope with local boundary variations caused by drusen. The algorithm was evaluated in a database of 20 patients with AMD. The mean unsigned error for the inner retinal pigment epithelium (IRPE) and Bruch's membrane (BM) was 5.65±6.26 and 4.37±5.25 μm, respectively. These results are relative to the average of two experts, whose inter-observer variability was 7.30±6.87 μm for IRPE and 5.03±4.37 μm for BM. The use SHOPs resulted in a successful segmentation of the IRPE. The remaining boundaries were also successfully segmented.

Automated quantification of Haller's layer in choroid using swept-source optical coherence tomography

OBJECTIVE

To develop an algorithm for automated quantification of Haller's layer in choroid using swept-source optical coherence tomography (OCT).

BACKGROUND

So far, to understand the association of various diseases with structural changes of choroid, only gross indicators such as thickness, volume and vascularity index have been examined. However, certain diseases affect specific sublayers of the choroid. Accordingly, a need for targeted quantitation arises. In particular, there is significant interest in understanding Haller's layer, a choroidal sublayer comprising relatively large blood vessels. Unfortunately, its intricate vasculature makes, manual quantitation difficult, tedious, and error-prone. To surmount this difficulty, it is imperative to develop an algorithmic method.

METHODOLOGY

The primary contribution of this work consists in developing an approach for detecting the boundary between Haller's and Sattler's layers, the latter comprising medium-sized vessels. The proposed algorithm estimates vessel cross-sections using exponentiation-based binarization, and labels a vessel large if its cross-section exceeds certain statistically determined threshold. Finally, the desired boundary is obtained as a smooth curve joining the innermost points of such large vessels. On 50 OCT B-scans (of 50 healthy eyes), our algorithm was validated both qualitatively and quantitatively, by comparing with intra-observer variability. Extensive statistical analysis was performed using metrics including Dice coefficient (DC), correlation coefficient (CC) and absolute difference (AD).

RESULTS

The proposed algorithm achieved a mean DC of 89.48% (SD:5.03%) in close agreement with the intra-observer repeatability of 89.12% (SD:5.68%). Corresponding mean AD and mean CC were of 17.54 μm (SD:16.45μm) and 98.10% (SD:1.60%) which too approximate the respective intra-observer repeatability values 19.19 μm (SD:17.69 μm) and 98.58% (SD:1.12%).

CONCLUSION

High correlation between algorithmic and manual delineations indicates suitability of our algorithm for clinically analyzing choroid in greater finer details, especially, in diseased eyes.

Eye-motion-corrected optical coherence tomography angiography using Lissajous scanning

To correct eye motion artifacts in en face optical coherence tomography angiography (OCT-A) images, a Lissajous scanning method with subsequent software-based motion correction is proposed. The standard Lissajous scanning pattern is modified to be compatible with OCT-A and a corresponding motion correction algorithm is designed. The effectiveness of our method was demonstrated by comparing en face OCT-A images with and without motion correction. The method was further validated by comparing motion-corrected images with scanning laser ophthalmoscopy images, and the repeatability of the method was evaluated using a checkerboard image. A motion-corrected en face OCT-A image from a blinking case is presented to demonstrate the ability of the method to deal with eye blinking. Results show that the method can produce accurate motion-free en face OCT-A images of the posterior segment of the eye in vivo.

EFFECT OF OPTIC DISK-FOVEA DISTANCE ON MEASUREMENTS OF INDIVIDUAL MACULAR INTRARETINAL LAYERS IN NORMAL SUBJECTS

PURPOSE

To investigate the effect of optic disk-fovea distance (DFD) on measurements of macular intraretinal layers using spectral domain optical coherence tomography in normal subjects.

METHODS

One hundred and eighty-two eyes from 182 normal subjects were imaged using spectral domain optical coherence tomography. The average thicknesses of eight macular intraretinal layers were measured using an automatic segmentation algorithm. Partial correlation test and multiple regression analysis were used to determine the effect of DFD on thicknesses of intraretinal layers.

RESULTS

Disk-fovea distance correlated negatively with the overall average thickness in all the intraretinal layers (r ≤ -0.17, all P ≤ 0.025) except the ganglion cell layer and photoreceptor. In multiple regression analysis, greater DFD was associated with thinner nerve fiber layer (6.78 μm decrease per each millimeter increase in DFD, P < 0.001), thinner ganglion cell-inner plexiform layer (2.16 μm decrease per each millimeter increase in DFD, P = 0.039), thinner ganglion cell complex (8.94 μm decrease per each millimeter increase in DFD, P < 0.001), thinner central macular thickness (18.16 μm decrease per each millimeter increase in DFD, P < 0.001), and thinner total macular thickness (15.94 μm decrease per each millimeter increase in DFD, P < 0.001).

CONCLUSION

Thinner measurements of macular intraretinal layers were significantly associated with greater DFD. A clinical assessment of macular intraretinal layers in the evaluation of various macular diseases should always be interpreted in the context of DFD.