Automated Segmentability Index for Layer Segmentation of Macular SD-OCT Images

Reduced Thickness of the Retina in de novo Parkinson's Disease Shows A Distinct Pattern, Different from Glaucoma

Background

Parkinson's disease (PD) patients experience visual symptoms and retinal degeneration. Studies using optical coherence tomography (OCT) have shown reduced thickness of the retina in PD, also a key characteristic of glaucoma.

Objective

To identify the presence and pattern of retinal changes in de novo, treatment-naive PD patients compared to healthy controls (HC) and early primary open angle glaucoma (POAG) patients.

Methods

Macular OCT data (10×10 mm) were collected from HC, PD, and early POAG patients, at the University Medical Center Groningen. Bayesian informative hypotheses statistical analyses were carried out comparing HC, PD-, and POAG patients, within each retinal cell layer.

Results

In total 100 HC, 121 PD, and 78 POAG patients were included. We showed significant reduced thickness of the inner plexiform layer and retinal pigment epithelium in PD compared to HC. POAG patients presented with a significantly thinner retinal nerve fiber layer, ganglion cell layer, inner plexiform layer, outer plexiform layer, and outer photoreceptor and subretinal virtual space compared to PD. Only the outer segment layer and retinal pigment epithelium were significantly thinner in PD compared to POAG.

Conclusions

De novo PD patients show reduced thickness of the retina compared to HC, especially of the inner plexiform layer, which differs significantly from POAG, showing a more extensive and widespread pattern of reduced thickness across layers. OCT is a useful tool to detect retinal changes in de novo PD, but its specificity versus other neurodegenerative disorders has to be established.

CLAHE-CapsNet: Efficient retina optical coherence tomography classification using capsule networks with contrast limited adaptive histogram equalization

Manual detection of eye diseases using retina Optical Coherence Tomography (OCT) images by Ophthalmologists is time consuming, prone to errors and tedious. Previous researchers have developed a computer aided system using deep learning-based convolutional neural networks (CNNs) to aid in faster detection of the retina diseases. However, these methods find it difficult to achieve better classification performance due to noise in the OCT image. Moreover, the pooling operations in CNN reduce resolution of the image that limits the performance of the model. The contributions of the paper are in two folds. Firstly, this paper makes a comprehensive literature review to establish current-state-of-act methods successfully implemented in retina OCT image classifications. Additionally, this paper proposes a capsule network coupled with contrast limited adaptive histogram equalization (CLAHE-CapsNet) for retina OCT image classification. The CLAHE was implemented as layers to minimize the noise in the retina image for better performance of the model. A three-layer convolutional capsule network was designed with carefully chosen hyperparameters. The dataset used for this study was presented by University of California San Diego (UCSD). The dataset consists of 84,495 X-Ray images (JPEG) and 4 categories (NORMAL, CNV, DME, and DRUSEN). The images went through a grading system consisting of multiple layers of trained graders of expertise for verification and correction of image labels. Evaluation experiments were conducted and comparison of results was done with state-of-the-art models to find out the best performing model. The evaluation metrics; accuracy, sensitivity, precision, specificity, and AUC are used to determine the performance of the models. The evaluation results show that the proposed model achieves the best performing model of accuracies of 97.7%, 99.5%, and 99.3% on overall accuracy (OA), overall sensitivity (OS), and overall precision (OP), respectively. The results obtained indicate that the proposed model can be adopted and implemented to help ophthalmologists in detecting retina OCT diseases.

Total retinal thickness: a neglected factor in the evaluation of inner retinal thickness

AIM

To determine whether macular retinal nerve fibre layer (mRNFL) and ganglion cell-inner plexiform layer (GC-IPL) thicknesses vary by ethnicity after accounting for total retinal thickness.

METHODS

We included healthy participants from the UK Biobank cohort who underwent macula-centred spectral domain-optical coherence tomography scans. mRNFL and GC-IPL thicknesses were determined for groups from different self-reported ethnic backgrounds. Multivariable regression models adjusting for covariables including age, gender, ethnicity and refractive error were built, with and without adjusting for total retinal thickness.

RESULTS

20237 participants were analysed. Prior to accounting for total retinal thickness, mRNFL thickness was on average 0.9 μm (-1.2, -0.6; p<0.001) lower among Asians and 1.5 μm (-2.3, -0.6; p<0.001) lower among black participants compared with white participants. Prior to accounting for total retinal thickness, the average GC-IPL thickness was 1.9 μm (-2.5, -1.4; p<0.001) lower among Asians compared with white participants, and 2.4 μm (-3.9, -1.0; p=0.001) lower among black participants compared with white participants. After accounting for total retinal thickness, the layer thicknesses were not significantly different among ethnic groups. When considered as a proportion of total retinal thickness, mRNFL thickness was ~0.1 and GC-IPL thickness was ~0.2 across age, gender and ethnic groups.

CONCLUSIONS

The previously reported ethnic differences in layer thickness among groups are likely driven by differences in total retinal thickness. Our results suggest using layer thickness ratio (retinal layer thicknesses/total retinal thickness) rather than absolute thickness values when comparing retinal layer thicknesses across groups.

Detecting retinal neurodegeneration in people with diabetes: Findings from the UK Biobank

Importance

Efforts are underway to incorporate retinal neurodegeneration in the diabetic retinopathy severity scale. However, there is no established measure to quantify diabetic retinal neurodegeneration (DRN).

Objective

We compared total retinal, macular retinal nerve fiber layer (mRNFL) and ganglion cell-inner plexiform layer (GC-IPL) thickness among participants with and without diabetes (DM) in a population-based cohort.

Design/setting/participants

Cross-sectional analysis, using the UK Biobank data resource. Separate general linear mixed models (GLMM) were created using DM and glycated hemoglobin as predictor variables for retinal thickness. Sub-analyses included comparing thickness measurements for patients with no/mild diabetic retinopathy (DR) and evaluating factors associated with retinal thickness in participants with and without diabetes. Factors found to be significantly associated with DM or thickness were included in a multiple GLMM.

Exposure

Diagnosis of DM was determined via self-report of diagnosis, medication use, DM-related complications or glycated hemoglobin level of ≥ 6.5%.

Main outcomes and measures

Total retinal, mRNFL and GC-IPL thickness.

Results

74,422 participants (69,985 with no DM; 4,437 with DM) were included. Median age was 59 years, 46% were men and 92% were white. Participants with DM had lower total retinal thickness (-4.57 μm, 95% CI: -5.00, -4.14; p<0.001), GC-IPL thickness (-1.73 μm, 95% CI: -1.86, -1.59; p<0.001) and mRNFL thickness (-0.68 μm, 95% CI: -0.81, -0.54; p<0.001) compared to those without DM. After adjusting for co-variates, in the GLMM, total retinal thickness was 1.99 um lower (95% CI: -2.47, -1.50; p<0.001) and GC-IPL was 1.02 μm lower (95% CI: -1.18, -0.87; p<0.001) among those with DM compared to without. mRNFL was no longer significantly different (p = 0.369). GC-IPL remained significantly lower, after adjusting for co-variates, among those with DM compared to those without DM when including only participants with no/mild DR (-0.80 μm, 95% CI: -0.98, -0.62; p<0.001). Total retinal thickness decreased 0.40 μm (95% CI: -0.61, -0.20; p<0.001), mRNFL thickness increased 0.20 μm (95% CI: 0.14, 0.27; p<0.001) and GC-IPL decreased 0.26 μm (95% CI: -0.33, -0.20; p<0.001) per unit increase in A1c after adjusting for co-variates. Among participants with diabetes, age, DR grade, ethnicity, body mass index, glaucoma, spherical equivalent, and visual acuity were significantly associated with GC-IPL thickness.

Conclusion

GC-IPL was thinner among participants with DM, compared to without DM. This difference persisted after adjusting for confounding variables and when considering only those with no/mild DR. This confirms that GC-IPL thinning occurs early in DM and can serve as a useful marker of DRN.

Modular deep neural networks for automatic quality control of retinal optical coherence tomography scans

Retinal optical coherence tomography (OCT) with intraretinal layer segmentation is increasingly used not only in ophthalmology but also for neurological diseases such as multiple sclerosis (MS). Signal quality influences segmentation results, and high-quality OCT images are needed for accurate segmentation and quantification of subtle intraretinal layer changes. Among others, OCT image quality depends on the ability to focus, patient compliance and operator skills. Current criteria for OCT quality define acceptable image quality, but depend on manual rating by experienced graders and are time consuming and subjective. In this paper, we propose and validate a standardized, grader-independent, real-time feedback system for automatic quality assessment of retinal OCT images. We defined image quality criteria for scan centering, signal quality and image completeness based on published quality criteria and typical artifacts identified by experienced graders when inspecting OCT images. We then trained modular neural networks on OCT data with manual quality grading to analyze image quality features. Quality analysis by a combination of these trained networks generates a comprehensive quality report containing quantitative results. We validated the approach against quality assessment according to the OSCAR-IB criteria by an experienced grader. Here, 100 OCT files with volume, circular and radial scans, centered on optic nerve head and macula, were analyzed and classified. A specificity of 0.96, a sensitivity of 0.97 and an accuracy of 0.97 as well as a Matthews correlation coefficient of 0.93 indicate a high rate of correct classification. Our method shows promising results in comparison to manual OCT grading and may be useful for real-time image quality analysis or analysis of large data sets, supporting standardized application of image quality criteria.

Livelayer: a semi-automatic software program for segmentation of layers and diabetic macular edema in optical coherence tomography images

Given the capacity of Optical Coherence Tomography (OCT) imaging to display structural changes in a wide variety of eye diseases and neurological disorders, the need for OCT image segmentation and the corresponding data interpretation is latterly felt more than ever before. In this paper, we wish to address this need by designing a semi-automatic software program for applying reliable segmentation of 8 different macular layers as well as outlining retinal pathologies such as diabetic macular edema. The software accommodates a novel graph-based semi-automatic method, called "Livelayer" which is designed for straightforward segmentation of retinal layers and fluids. This method is chiefly based on Dijkstra's Shortest Path First (SPF) algorithm and the Live-wire function together with some preprocessing operations on the to-be-segmented images. The software is indeed suitable for obtaining detailed segmentation of layers, exact localization of clear or unclear fluid objects and the ground truth, demanding far less endeavor in comparison to a common manual segmentation method. It is also valuable as a tool for calculating the irregularity index in deformed OCT images. The amount of time (seconds) that Livelayer required for segmentation of Inner Limiting Membrane, Inner Plexiform Layer-Inner Nuclear Layer, Outer Plexiform Layer-Outer Nuclear Layer was much less than that for the manual segmentation, 5 s for the ILM (minimum) and 15.57 s for the OPL-ONL (maximum). The unsigned errors (pixels) between the semi-automatically labeled and gold standard data was on average 2.7, 1.9, 2.1 for ILM, IPL-INL, OPL-ONL, respectively. The Bland-Altman plots indicated perfect concordance between the Livelayer and the manual algorithm and that they could be used interchangeably. The repeatability error was around one pixel for the OPL-ONL and < 1 for the other two. The unsigned errors between the Livelayer and the manual algorithm was 1.33 for ILM and 1.53 for Nerve Fiber Layer-Ganglion Cell Layer in peripapillary B-Scans. The Dice scores for comparing the two algorithms and for obtaining the repeatability on segmentation of fluid objects were at acceptable levels.

Association of Retinal Neurodegeneration on Optical Coherence Tomography With Dementia: A Population-Based Study

Importance

Retinal structures may serve as a biomarker for dementia, but longitudinal studies examining this link are lacking.

Objective

To investigate the association of inner retinal layer thickness with prevalent and incident dementia in a general population of Dutch adults.

Design, Setting, and Participants

From September 2007 to June 2012, participants from the prospective population-based Rotterdam Study who were 45 years and older and had gradable retinal optical coherence tomography images and at baseline were free from stroke, Parkinson disease, multiple sclerosis, glaucoma, macular degeneration, retinopathy, myopia, hyperopia, and optic disc pathology were included. They were followed up until January 1, 2015, for the onset of dementia.

Exposures

Inner retinal layer thicknesses (ie, retinal nerve fiber layer [RNFL]) and ganglion cell-inner plexiform layer (GC-IPL) thicknesses measured on optical coherence tomography images.

Main Outcomes and Measures

Odds ratios and hazard ratios for incident dementia per SD decrease in retinal layer thickness adjusted for age, sex, education, and cardiovascular risk factors.

Results

Of 5065 individuals eligible for optical coherence tomography scanning, 3289 (64.9%) (mean [SD] age 68.9 [9.9] years, 1879 [57%] women) were included in the analysis. Of these 3289 individuals, 41 (1.2%) already had dementia. Thinner GC-IPL was associated with prevalent dementia (odds ratio per SD decrease in GC-IPL, 1.37 [95% CI, 0.99-1.90]). No association was found of RNFL with prevalent dementia. During 14 674 person-years of follow-up (mean [SD], 4.5 [1.6] years), 86 individuals (2.6%) developed dementia of whom 68 (2.1%) had Alzheimer disease. Thinner RNFL at baseline was associated with an increased risk of developing dementia (hazard ratio per SD decrease in RNFL, 1.44 [95% CI, 1.19-1.75]), which was similar for Alzheimer disease (hazard ratio, 1.43 [95% CI, 1.15-1.78]). No association was found between GC-IPL thickness and incident dementia (hazard ratio, 1.13 [95% CI, 0.90-1.43]).

Conclusions and Relevance

Thinner RNFL is associated with an increased risk of dementia, including Alzheimer disease, suggesting that retinal neurodegeneration may serve as a preclinical biomarker for dementia.

Thinner retinal layers are associated with changes in the visual pathway: A population-based study

Increasing evidence shows that thinner retinal nerve fiber layer (RNFL) and ganglion cell layer (GCL), assessed on optical coherence tomography (OCT), are reflecting global brain atrophy. Yet, little is known on the relation of these layers with specific brain regions. Using voxel-based analysis, we aimed to unravel specific brain regions associated with these retinal layers. We included 2,235 persons (mean age: 67.3 years, 55% women) from the Rotterdam Study (2007-2012) who had gradable retinal OCT images and brain magnetic resonance imaging (MRI) scans, including diffusion tensor (DT) imaging. Thicknesses of peripapillary RNFL and perimacular GCL were measured using an automated segmentation algorithm. Voxel-based morphometry protocols were applied to process DT-MRI data. We investigated the association between retinal layer thickness with voxel-wise gray matter density and white matter microstructure by performing linear regression models. We found that thinner RNFL and GCL were associated with lower gray matter density in the visual cortex, and with lower fractional anisotropy and higher mean diffusivity in white matter tracts that are part of the optic radiation. Furthermore, thinner GCL was associated with lower gray matter density of the thalamus. Thinner RNFL and GCL are associated with gray and white matter changes in the visual pathway suggesting that retinal thinning on OCT may be specifically associated with changes in the visual pathway rather than with changes in the global brain. These findings may serve as a basis for understanding visual symptoms in elderly patients, patients with Alzheimer's disease, or patients with posterior cortical atrophy.

Automated drusen detection in dry age-related macular degeneration by multiple-depth, en face optical coherence tomography

We introduce a method to automatically detect drusen in dry age-related macular degeneration (AMD) from optical coherence tomography with minimum need for layer segmentation. The method is based on the en face detection of drusen areas in C-scans at certain distances above the Bruch's membrane, circumventing the difficult task of pathologic retinal pigment epithelium segmentation. All types of drusen can be detected, including the challenging subretinal drusenoid deposits (pseudodrusen). The high sensitivity and accuracy demonstrated here shows its potential for detection of drusen onset in early AMD.

Optical Coherence Tomography Analysis Based Prediction of Humphrey 24-2 Visual Field Thresholds in Patients With Glaucoma

Purpose

A pilot study showed that prediction of individual Humphrey 24-2 visual field (HVF 24-2) sensitivity thresholds from optical coherence tomography (OCT) image analysis is possible. We evaluate performance of an improved approach as well as 3 other predictive algorithms on a new, fully independent set of glaucoma subjects.

Methods

Subjects underwent HVF 24-2 and 9-field OCT (Heidelberg Spectralis) testing. Nerve fiber (NFL), and ganglion cell and inner plexiform (GCL+IPL) layers were cosegmented and partitioned into 52 sectors matching HVF 24-2 test locations. The Wilcoxon rank sum test was applied to test correlation R, root mean square error (RMSE), and limits of agreement (LoA) between actual and predicted thresholds for four prediction models. The training data consisted of the 9-field OCT and HVF 24-2 thresholds of 111 glaucoma patients from our pilot study.

Results

We studied 112 subjects (112 eyes) with early, moderate, or advanced primary and secondary open angle glaucoma. Subjects with less than 9 scans (15/112) or insufficient quality segmentations (11/97) were excluded. Retinal ganglion cell axonal complex (RGC-AC) optimized had superior average R = 0.74 (95% confidence interval [CI], 0.67-0.76) and RMSE = 5.42 (95% CI, 5.1-5.7) dB, which was significantly better (P < 0.05/3) than the other three models: Naïve (R = 0.49; 95% CI, 0.44-0.54; RMSE = 7.24 dB; 95% CI, 6.6-7.8 dB), Garway-Heath (R = 0.66; 95% CI, 0.60-0.68; RMSE = 6.07 dB; 95% CI, 5.7-6.5 dB), and Donut (R = 0.67; 95% CI, 0.61-0.69; RMSE = 6.08 dB, 95% CI, 5.8-6.4 dB).

Conclusions

The proposed RGC-AC optimized predictive algorithm based on 9-field OCT image analysis and the RGC-AC concept is superior to previous methods and its performance is close to the reproducibility of HVF 24-2.