Spectral-domain Optical Coherence Tomography Images Research Articles

Evaluation of choriocapillaris with high-speed spectral-domain optical coherence tomography angiography and image averaging

To study the capabilities of high-speed spectral-domain optical coherence tomography angiography (SD-OCTA) with image averaging in the evaluation of choriocapillaries in healthy individuals and patients with central serous chorioretinopathy (CSC). All participants underwent OCTA examination on the SOLIX tomograph (Optovue, USA) using quadruplicate 3×3 mm scan with image averaging. Image analysis was performed in 9-µm custom slab with Phansalkar auto local threshold algorithm and calculation of flow voids larger than 5000 µm2 and 10 000 µm2. The optimal slab depth was determined in healthy eyes by consecutive analysis of the slabs with a 3-µm shift from zero position to 33 µm below Bruch's membrane. The study included 18 eyes of 18 healthy volunteers (40.4±6.0 years old) and 18 fellow eyes of 18 unilateral CSC patients (37.4±10.7 years old). The slab 12 µm below the Bruch's membrane showed the minimal number of flow voids (23.1±7.0 of >5000 µm2 voids per scan) and was chosen for further analysis. The number of flow voids of >5000 µm2 in healthy paired eyes of CSC patients was statistically significantly higher than in the eyes of healthy individuals (32.7±10.7 and 25.3±8.1 voids/scan, respectively; p=0.022). The number of flow voids of >10000 µm2 was also statistically significantly higher in CSC eyes compared to healthy eyes (5.6±3.3 and 3.5±1.9 voids/scan, respectively; p=0.045). There was no statistically significant difference in total area of the voids between the eyes of healthy individuals and CSC patients (96406.1±3924.5 µm2 and 95395.7 ± 3615.1 µm2, respectively; p=0.42). The optimal settings for choriocapillaris imaging on the SOLIX tomograph include 9-µm slab 9 to 18 µm below the Bruch's membrane. Using a 9-µm slab 12 µm below the Bruch's membrane, a substantial difference was found in choriocapillaris perfusion between eyes of healthy individuals and CSC patients.

Visualization of Retinal Blood Vessels.

The retina offers a unique opportunity to directly visualize blood vessels in vivo noninvasively. Over the past few decades, several new imaging techniques have been adapted to study the retinal vasculature in the laboratory in animal models and in the clinic in human subjects. High-contrast, finely detailed fundus images can be acquired by confocal scanning laser ophthalmoscopy (cSLO). With fluorescein angiography (FA), the retinal microcirculation can be visualized. High-resolution spectral-domain optical coherence tomography (SD-OCT) is able to acquire cross-section images resolving the microarchitecture of the retina, similar to histology. The techniques and protocols for acquiring cSLO, FA, and SD-OCT imaging of the retinal vasculature and morphology in the rodent are described.

Reticular Pseudodrusen Characteristics and Associations in the Carotenoids in Age-Related Eye Disease Study 2 (CAREDS2), an Ancillary Study of the Women’s Health Initiative

To determine the prevalence and morphologic features of reticular pseudodrusen (RPD) and their association with participant demographics and age-related macular degeneration (AMD) status in the Carotenoids in Age-Related Eye Disease Study 2 (CAREDS2) sample, an ancillary study of the Women's Health Initiative Observational Study. Cross-sectional, multicenter, natural history study. Nine hundred and twenty-seven eyes from 466 postmenopausal women 69 to 101 years of age. Multimodal imaging, including spectral-domain (SD) OCT and infrared reflectance (IR), were used to identify RPD characteristics, including location (within or outside the 6-mm diameter circle centered at the macula), presence of peripapillary RPD, pattern of RPD, and RPD area. Age-related macular degeneration features from SD OCT, IR, and color photographs also were assessed and AMD severity was categorized. Reticular pseudodrusen prevalence using SD OCT and IR imaging and AMD status. Reticular pseudodrusen were present in 130 eyes (14% of eyes, 16% of participants), with increasing prevalence with age: 7% in those younger than 78 years, 14% in those 78 to 83 years of age, and 30% in those older than 83 years. Using clinical classification of AMD with color photography, RPD were seen in 2.4% of eyes with no AMD or aging changes, 11.5% in early AMD, 25.1% in intermediate AMD, and 51.1% in late AMD. Mean RPD area was 17.4 mm2 (standard deviation, 14.7 mm2). Ribbon morphologic RPD (53%) was more common than dot morphologic RPD (36%). Reticular pseudodrusen mostly were located both within and outside the 6-mm circle with primarily superior retinal distribution. Reticular pseudodrusen were visualized with corresponding color fundus photography in only 38 eyes (4% of total eyes). Participants with and without RPD had a visual acuity±standard error of 77.9 ± 1.4 letters and 81.3 ± 0.4 letters, respectively (P= 0.02). The prevalence of RPD in CAREDS2 increased with age and was associated with AMD severity. Reticular pseudodrusen were detected in eyes without other features of AMD and could represent an earlier disease state. Multimodal imaging with SD OCT and IR has significantly greater sensitivity for visualizing RPD than color fundus photography.

SiameseGAN: A Generative Model for Denoising of Spectral Domain Optical Coherence Tomography Images.

Optical coherence tomography (OCT) is a standard diagnostic imaging method for assessment of ophthalmic diseases. The speckle noise present in the high-speed OCT images hampers its clinical utility, especially in Spectral-Domain Optical Coherence Tomography (SDOCT). In this work, a new deep generative model, called as SiameseGAN, for denoising Low signal-to-noise ratio (LSNR) B-scans of SDOCT has been developed. SiameseGAN is a Generative Adversarial Network (GAN) equipped with a siamese twin network. The siamese network module of the proposed SiameseGAN model helps the generator to generate denoised images that are closer to groundtruth images in the feature space, while the discriminator helps in making sure they are realistic images. This approach, unlike baseline dictionary learning technique (MSBTD), does not require an apriori high-quality image from the target imaging subject for denoising and takes less time for denoising. Moreover, various deep learning models that have been shown to be effective in performing denoising task in the SDOCT imaging were also deployed in this work. A qualitative and quantitative comparison on the performance of proposed method with these state-of-the-art denoising algorithms has been performed. The experimental results show that the speckle noise can be effectively mitigated using the proposed SiameseGAN along with faster denoising unlike existing approaches.

Smartphone-based Ophthalmic Imaging Compared With Spectral-domain Optical Coherence Tomography Assessment of Vertical Cup-to-disc Ratio Among Adults in Southwestern Uganda.

Using optical coherence tomography (OCT) measurements as a reference standard for vertical cup-to-disc ratio (vCDR), a smartphone-based ophthalmic camera has a sensitivity of 67.7% and specificity of 96.7% to detect a vCDR>0.5. The purpose of this study was to assess the performance of a smartphone-based ophthalmic camera system using an Apple iPhone 6S and an adapter, Paxos Scope, to obtain adequate dilated fundus photos to measure clinically useful vCDR cutoffs. Adult patients from a government tertiary level eye hospital in Southwestern Uganda were prospectively recruited from January to April 2019. All patients experienced a comprehensive eye examination, dilated posterior segment indirect ophthalmoscope imaging with the Paxos Scope, and spectral-domain OCT imaging with a Cirrus HD-OCT to measure vCDR. Patients' eyes excluded had media opacities or existing disease precluding a view of the fundus. Fundus images underwent a single masked review to assign vCDR at increments of 0.1. Descriptive statistics, parametric and χ2 tests for significance, repeated measures correlation, κ, receiver operating characteristics curve, and Bland-Altman were used to assess the data. Among 467 (consecutive) individuals, fundus photographs acquired with the Paxos Scope demonstrated a 67.7% [95% confidence interval (CI), 63.0-72.0] sensitivity and 96.7% (95% CI, 94.2-98.3) specificity to detect a vCDR>0.5, using OCT as the reference standard. A total of 138 eyes were excluded due to poor imaging acquisition, such as dense cataract, rendering 796 eyes for analysis. The vCDR from graded Paxos Scope images and OCT correlated well with repeated measures correlation of 0.82 (95% CI, 0.77-0.86, P<0.001) and agreement, dichotomized as >0.5 or ≤0.5, was 80.9% (κ=0.63±0.034, P<0.001). Among glaucoma and glaucoma suspects (85 eyes), the sensitivity and specificity dichotomized using vCDR>0.5 were 97.5% (95% CI, 91.3-99.7) and 80.0% (95% CI, 28.4-99.5), respectively. The area under the receiver operating characteristics curve was 0.92 (95% CI, 0.89-0.94) for all eyes and 0.98 (95% CI, 0.78-1.0) for glaucoma and glaucoma suspects. The Paxos Scope produced images that can be reliably used to estimate vCDR, which is closely aligned with the automated algorithm from the OCT optic disc cube scan. The low-cost, ready-to-integrate adapter, and minimal training requirements make it a viable option for population-based screening in low-resource settings.

Determination of Neurodegeneration in Polycystic Ovary Syndrome with Retinal Segmentation Analysis

ABSTRACT Purpose: To compare the thickness of each retinal layer in patients with polycystic ovary syndrome (PCOS) versus healthy, age-matched controls by using retinal segmentation analysis. Methods: In our cross-sectional study, 37 patients with PCOS (i.e., patient group) and 35 healthy individuals (i.e., control group) underwent spectral-domain optical coherence tomography imaging. Using built-in automatic retinal segmentation software to analyze the images collected, we compared the thickness of the retinal nerve fiber layer (RNFL), ganglion cell layer (GCL), inner plexiform layer (IPL), inner nuclear layer, outer plexiform layer, outer nuclear layer, photoreceptor layer (PRL), retinal pigment epithelium (RPE), inner retinal layers, and outer retinal layers between the groups. To analyze the measurements, we used a traditional Early Treatment Diabetic Retinopathy Study (ETDRS) grid. Results: In ETDRS subfields, 6-mm nasal RNFL thickness; 3- and 6-mm nasal GCL thickness; 3-mm superior and 6-mm nasal IPL thickness; 1-mm central, 3-mm nasal, superior, and inferior, and 6-mm nasal and inferior PRL thickness; and 6-mm inferior RPE thickness were significantly thinner in patients with PCOS than that of healthy controls. Conclusion: The results of our retinal segmentation analysis indicate that patients with PCOS tend to have thinner GCL, IPL, and PRL than healthy, age-matched controls due to neurodegeneration likely caused by insulin resistance, or subclinical retinal inflammation.

Local Anatomic Precursors to New-Onset Geographic Atrophy in Age-Related Macular Degeneration as Defined on OCT

In macula-wide analyses, spectral-domain (SD) optical coherence tomography (OCT) features including drusen volume, hyperreflective foci, and OCT-reflective drusen substructures independently predict geographic atrophy (GA) onset secondary to age-related macular degeneration (AMD). We sought to identify SD OCT features in the location of new GA before its onset. Retrospective study. Age-Related Eye Disease Study 2 Ancillary SD OCT Study participants. We analyzed longitudinally captured SD OCT images and color photographs from 488 eyes of 488 participants with intermediate AMD at baseline. Sixty-two eyes with sufficient image quality demonstrated new-onset GA on color photographs during study years 2 through 7. The area of new-onset GA and one size-matched control region in the same eye were segmented separately, and corresponding spatial volumes on registered SD OCT images at the GA incident year and at 2, 3, and 4 years previously were defined. Differences in SD OCT features between paired precursor regions were evaluated through matched-pairs analyses. Localized SD OCT features 2 years before GA onset. Compared with paired control regions, GA precursor regions at 2, 3, and 4 years before (n= 54, 33, and 25, respectively) showed greater drusen volume (P= 0.01, P= 0.003, and P= 0.003, respectively). At 2 and 3 years before GA onset, they were associated with the presence of hypertransmission (P < 0.001 and P= 0.03, respectively), hyperreflective foci (P < 0.001 and P= 0.045, respectively), OCT-reflective drusen substructures (P= 0.004 and P= 0.03, respectively), and loss or disruption of the photoreceptor zone, ellipsoid zone, and retinal pigment epithelium (RPE, P < 0.001 and P= 0.005-0.045, respectively). At 4 years before GA onset, precursor regions were associated with photoreceptor zone thinning (P= 0.007) and interdigitation zone loss (P= 0.045). Evolution to GA is heralded by early local photoreceptor changes and drusen accumulation, detectable4 years before GA onset. These precede other anatomic heralds such as RPE changes and drusen substructure emergence detectable1 to 2 years before GA. This study thus identified earlier end points for GA as potential therapeutic targets in clinical trials.

Pigmented Paravenous Chorioretinal Atrophy: Clinical Spectrum and Multimodal Imaging Characteristics

To investigate the clinical findings and natural course of patients with pigmented paravenous chorioretinal atrophy (PPCRA) using multimodal imaging. Retrospective, observational case series. We reviewed the records of consecutive patients diagnosed with PPCRA at a single center and assessed serial fundus photographs, fundus autofluorescence (FAF), and spectral-domain optical coherence tomography images. Electrophysiological findings and visual field analysis were also reviewed. The study included 50 eyes in 25 patients. The mean age of the population was 51.6 ± 14.6 years. Nine patients (36.0%) were asymptomatic and 9 (36.0%) complained of nyctalopia. We divided fundus appearance into one of 3 groups: paravenous (58.0%), focal (16.0%), and confluent (26.0%). Of the 50 eyes, macular involvement was present in 13 eyes (26.0%). Fifteen patients (60.0%) demonstrated a symmetric fundus appearance, whereas 10 (40.0%) had marked asymmetry. Eight eyes (16.0%) exhibited apparent changes in fundus findings, over a mean follow-up period of 8.8 years. FAF imaging was most sensitive to evaluate the extent of lesions. Sixteen eyes (44.4%) showed progressive visual field loss during the follow-up period. Most patients maintained stable vision, and 36 eyes (72.0%) had a final visual acuity of 20/50 or better. Nevertheless, some eyes with macular involvement experienced severe deterioration in vision. Electrophysiological data were variable, and interocular asymmetry was common (45.8%). PPCRA can present with a more variable expressivity than previously described. Multimodal imaging can provide insights into its clinical characteristics to facilitate the diagnosis, classification, and follow-up of these patients.

Optical coherence tomography biomarkers for visual acuity in patients with idiopathic epiretinal membrane.

To investigate the correlation between outer and inner retina optical coherence tomography (OCT) biomarkers and visual acuity in patients with idiopathic epiretinal membrane (iERM) and identify which of them may be predictive of visual function. A retrospective cross-sectional single-center study was conducted that included patients diagnosed with iERM. Spectral domain OCT images were obtained and assessed qualitatively and quantitatively. The association of OCT parameters with best corrected visual acuity was analyzed. Charts of 97 eyes of 97 patients were reviewed. Central foveal thickness, maximal retinal thickness (MRT), photoreceptor outer segment length, outer foveal thickness, ganglion cell-inner plexiform layer complex thickening, inner retinal thickness and inner retinal layer irregularity index were among the major outcome measures. OCT scans were also assessed for the presence of cotton ball sign, ellipsoid zone disruption, ectopic inner foveal layer, disorganization of retinal inner layers (DRIL), intraretinal fluid, subretinal fluid (SRF) and epimacular membrane rip. Univariate analysis showed statistically significant association between all the aforementioned parameters with worse vision, except for cotton ball sign and SRF. Multivariate analysis found that MRT and severe DRIL were strongly correlated with worse vision (p < 0.001). MRT and severe DRIL should be considered as negative prognostic factors for visual acuity.

Effect of macular edema in the premature period on refraction in infants screened and treated for retinopathy of prematurity.

To determine the effect of macular edema on refraction in infants with retinopathy of prematurity (ROP) at the ages of one, two and three years using Optical Coherence Tomography (OCT). Optical coherence tomography (OCT) was performed to detect and categorize edema in 280 eyes of 280 premature infants. The cross-sectional spectral domain OCT imaging of the macular region was performed between weeks 36 and 42 to assess macular development. Refraction measurement was performed during follow-up at the ages of one, two and three years. Macular thickness, choroidal thickness, macular edema severity, and the spherical, cylindrical and spherical equivalent (SE) values were measured. A positive correlation was found between macular thickness in the premature period and the SE value at age one and a negative correlation at age three. No correlation was found between gestational age and the SE value, but there was a positive correlation between birth weight and the SE value at ages one and two. Although no correlation was determined between edema and gestational age or between edema and birth weight, the prevalence of macular edema in infants with ROP was significantly higher than that of infants without this disease. Macular edema during the premature period can have an impact on refraction at the ages one, two and three years by effecting the emmetropization process.

Análisis de la capa de células ganglionares con deep learning en el diagnóstico de glaucoma

ObjectiveTo determine and compare the diagnostic precision in glaucoma of two deep learning models using infrared images of the optic nerve, eye fundus, and the ganglion cell layer (GCL). MethodsWe have selected a sample of normal and glaucoma patients. Three infrared images were registered with a spectral-domain optical coherence tomography (SD-OCT). The first corresponds to the confocal scan image of the fundus, the second is a cut-out of the first centered on the optic nerve, and the third was the SD-OCT image of the GCL. Our deep learning models are developed on the MatLab platform with the ResNet50 and VGG19 pre-trained neural networks. Results498 eyes of 298 patients were collected. Of the 498 eyes, 312 are glaucoma and 186 are normal. In the test, the precision of the models was 96% (ResNet50) and 96% (VGG19) for the GCL images, 90% (ResNet50) and 90% (VGG19) for the optic nerve images and 82% (ResNet50) and 84% (VGG19) for the fundus images. The ROC area in the test was 0.96 (ResNet50) and 0.97 (VGG19) for the GCL images, 0.87 (ResNet50) and 0.88 (VGG19) for the optic nerve images, and 0.79 (ResNet50) and 0.81 (VGG19) for the fundus images. ConclusionsBoth deep learning models, applied to the GCL images, achieve high diagnostic precision, sensitivity and specificity in the diagnosis of glaucoma.

MS-CAM: Multi-Scale Class Activation Maps for Weakly-Supervised Segmentation of Geographic Atrophy Lesions in SD-OCT Images.

As one of the most critical characteristics in advanced stage of non-exudative Age-related Macular Degeneration (AMD), Geographic Atrophy (GA) is one of the significant causes of sustained visual acuity loss. Automatic localization of retinal regions affected by GA is a fundamental step for clinical diagnosis. In this paper, we present a novel weakly supervised model for GA segmentation in Spectral-Domain Optical Coherence Tomography (SD-OCT) images. A novel Multi-Scale Class Activation Map (MS-CAM) is proposed to highlight the discriminatory significance regions in localization and detail descriptions. To extract available multi-scale features, we design a Scaling and UpSampling (SUS) module to balance the information content between features of different scales. To capture more discriminative features, an Attentional Fully Connected (AFC) module is proposed by introducing the attention mechanism into the fully connected operations to enhance the significant informative features and suppress less useful ones. Based on the location cues, the final GA region prediction is obtained by the projection segmentation of MS-CAM. The experimental results on two independent datasets demonstrate that the proposed weakly supervised model outperforms the conventional GA segmentation methods and can produce similar or superior accuracy comparing with fully supervised approaches. The source code has been released and is available on GitHub: https://github.com/ jizexuan/Multi-Scale-Class-Activation-Map-Tensorflow.

Optical coherence tomography angiography for identifying choroidal neovascular membranes: a masked study in clinical practice.

Optical coherence tomography angiography (OCT-A) allows non-invasive imaging of chorio-retinal vasculature, and is a potential alternative to fluorescein angiography (FA). Sensitivity and specificity of OCT-A for detecting choroidal neovascularisation (CNV) in treatment-naïve neovascular age-related macular degeneration (nAMD) patients is examined, using the Heidelberg Spectralis in a 'real world' setting. Overall, 43 eyes from 26 patients were included in the study. Spectral domain OCT (SD-OCT), OCT-A and FA images were obtained at baseline. Each of the three retinal image modalities was systematically assessed by three masked clinicians. Decisions about the presence/absence of CNV were recorded using an automated segmentation for OCT-A, a manual method, and using both OCT-A and SD-OCT in conjunction. Additional information about the presence of sub-retinal hyper-reflective material (SHRM) and the 'double layer sign' (DLS) were recorded. The average sensitivity and specificity of the OCT-A for the detection of CNV in treatment naïve AMD was 89% and 87% for the combined SD-OCT and OCT-A, 76% and 91% for the automated segmentation and 84% and 85% for the manual segmentation, respectively. Inter-clinician agreement was 0.59-65 kappa. In patients without CNV, SHRM was present in only 6% while DLS was present in 28%. Sensitivity and specificity was >78% for both SHRM and DLS. OCT-A provides a reliable tool for detecting CNV in treatment naïve nAMD patients, with high sensitivity and specificity. Combined use of SD-OCT images and SHRM as an additional bio-marker, OCT-A could become an alternative to FA in routine clinical practice.

Disorganisation of retinal inner layers is associated with reduced contrast sensitivity in retinal vein occlusion

Background/AimsTo determine if disorganisation of retinal inner layers (DRIL) is associated with reduced contrast sensitivity (CS) in patients with retinal vein occlusion (RVO) with a history of macular oedema (ME).MethodsProspective,...

Effects of age and binarising area on choroidal vascularity index in healthy eyes: an optical coherence tomography study.

To investigate the changes in the choroidal vascularity index (CVI) with age and to compare the effect of the binarised area on CVI in healthy eyes using spectral-domain optical coherence tomography (SD-OCT). Two hundred and twenty-four eyes of 224 healthy subjects were included in this prospective cross-sectional study.The eyes were divided into different age groups to analyse the possible age-related choroidal structural changes. Subfoveal choroidal thickness (SFCT), CVI, total choroidal area (TCA), stromal area (SA), luminal area (LA), and CVI within the central 1500µm of the macula were analysed using enhanced depth imaging SD-OCT. The CVI was defined as the proportion of the LA to the TCA, and its values for the two binarised areas were compared (CVItotal vs. CVI1500). The mean age was 34.77 ± 20.97 (range: 5-70) years. The mean CVItotal was statistically lower (66.71 ± 2.58%) than the mean CVI1500 (67.54 ± 3.13%, p = 0.008) among all the healthy participants. TCA, LA, CVItotal, and CVI1500 were statistically higher in the ≤ 18-year-old group compared to the > 18-year-old group (p < 0.05), but SA was not significantly different between the groups (p = 0.327). Similarly, TCA, LA, CVItotal, and CVI1500 between the five studied age groups were statistically different (p < 0.001), showing larger figures in the 0-10-year-old group. However, this was not true for the stromal region (p = 0.139). CVItotal exhibited a very strong positive correlation with CVI1500. No significant gender-related difference was observed in CVI. Decreased LA, TCA, and CVI were observed in healthy eyes with increasing age. CVI1500 was higher than CVItotal in a single B scan OCT. This result may provide valuable information about the choroid under different conditions, such as its physiological changes and disease pathophysiology.

Association between central retinal thickness and low luminance visual acuity in early age-related macular degeneration.

To examine whether central retinal thickness (CRT) is related to mesopic visual acuity (VA) and low luminance deficit (LLD, difference between photopic and mesopic VA) in eyes with early and intermediate age-related macular degeneration (AMD). In a cross-sectional study, 50 pseudophakic subjects older than 63 years were divided into three groups (no AMD, early AMD and intermediate AMD). Spectral domain optical coherence tomography (SD-OCT) was used to measure CRT in the 1 mm-central-area. Best-corrected distance VA was measured under photopic or mesopic luminance conditions and LLD calculated. Subjects were stratified by VA impairment to compare CRTs across these groups. Relationships were examined by stepwise multiple linear regression. No significant differences in mean CRT, photopic and mesopic VA or LLD were detected between the groups no AMD, early AMD and intermediate AMD. However, mean CRTs were 20 microns and 18 microns thicker in the eyes with impaired mesopic VA (> 0.3 logMAR) and impaired LLD (⩾ 0.3 logMAR) compared to the eyes with non-impaired VA or LLD respectively (both p < 0.01). CRT and mesopic pupil size were independent predictors of mesopic VA (p = 0.001). CRT emerged as the only independent predictor of LLD (p = 0.004). Increased CRT was linked to worse retinal function when measured under mesopic conditions in eyes without AMD and eyes with early to intermediate AMD. SD-OCT imaging combined with VA measurements under low luminance conditions could be a useful tool to detect early AMD.

RPE layer detection and baseline estimation using statistical methods and randomization for classification of AMD from retinal OCT

Background and Objective: Age-related macular degeneration (AMD) is a condition of the eye that affects the aged people. Optical coherence tomography (OCT) is a diagnostic tool capable of analyzing and identifying the disease affected retinal layers with high resolution. The objective of this work is to extract the retinal pigment epithelium (RPE) layer and the baseline (natural eye curvature, particular to every patient) from retinal spectral-domain OCT (SD-OCT) images. It uses them to find the height of drusen (abnormalities) in the RPE layer and classify it as AMD or normal.Methods: In the proposed work, the contrast enhancement based adaptive denoising technique is used for speckle elimination. Pixel grouping and iterative elimination based on the knowledge of typical layer intensities and positions are used to obtain the RPE layer. Using this estimate, randomization techniques are employed, followed by polynomial fitting and drusen removal to arrive at a baseline estimate. The classification is based on the drusen height obtained by taking the difference between the RPE and baseline levels. We have used a patient, wise classification approach where a patient is classified diseased if more than a threshold number of patient images have drusen of more than a certain height. Since all slices of an affected patient will not show drusen, we are justified in adopting this technique.Results: The proposed method is tested on a public data set of 2130 images/slices, which belonged to 30 patient volumes (15 AMD and 15 Normal) and achieved an overall accuracy of 96.66%, with no false positives. In comparison with existing works, the proposed method achieved higher overall accuracy and a better baseline estimate.Conclusions: The proposed work focuses on AMD/normal classification using a statistical approach. It does not require any training. The proposed method modifies the motion restoration paradigm to obtain an application-specific denoising algorithm. The existing RPE detection algorithm is modified significantly to make it robust and applicable even to images where the RPE is not very evident/there is a significant amount of perforations (drusen). The baseline estimation algorithm employs a powerful combination of randomization, iterative polynomial fitting, and pixel elimination in contrast to mere fitting techniques. The main highlight of this work is, it achieved an exact estimation of the baseline in the retinal image compared to the existing methods.

CHARACTERIZING THE NATURAL HISTORY OF FOVEAL-SPARING ATROPHIC LATE-ONSET RETINAL DEGENERATION.

To identify quantifiable markers of disease progression in patients with foveal-sparing atrophic late-onset retinal degeneration using fundus autofluorescence and spectral-domain optical coherence tomography imaging. Natural history study evaluating patients within a 3-year interval. Disease progression was assessed based on the area of retinal atrophy, macular topographic distribution of lesions, retinal and choroidal thickness and volume, and choroidal vascularity index. Twenty-four eyes (12 individuals) were included for fundus autofluorescence, and 31 eyes (16 individuals) for spectral-domain optical coherence tomography studies. Measurements were symmetrical between eyes of the same patient. The area of atrophy significantly enlarged (P = 0.002), with a growth rate of 2.67 mm2/year (SD: 2.13; square rooted: 0.57 mm/year, SD = 0.34). Baseline area of atrophy and progression both correlated with age. Most atrophic lesions were found in the temporal macula and progressed nasally at follow-up. Central choroidal and retinal thicknesses and volume in late-onset retinal degeneration cases were significantly reduced compared with controls, but only central retinal thickness decreased significantly at follow-up. This study identifies the area of atrophy and central retinal thickness, but not chorioretinal volume or choroidal thickness, as markers of short-term progression in late-onset retinal degeneration. These findings may be useful for disease monitoring and late-onset retinal degeneration interventional studies.

Optical coherence tomography-based deep-learning model for detecting central serous chorioretinopathy

Central serous chorioretinopathy (CSC) is a common condition characterized by serous detachment of the neurosensory retina at the posterior pole. We built a deep learning system model to diagnose CSC, and distinguish chronic from acute CSC using spectral domain optical coherence tomography (SD-OCT) images. Data from SD-OCT images of patients with CSC and a control group were analyzed with a convolutional neural network. Sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUROC) were used to evaluate the model. For CSC diagnosis, our model showed an accuracy, sensitivity, and specificity of 93.8%, 90.0%, and 99.1%, respectively; AUROC was 98.9% (95% CI, 0.983–0.995); and its diagnostic performance was comparable with VGG-16, Resnet-50, and the diagnoses of five different ophthalmologists. For distinguishing chronic from acute cases, the accuracy, sensitivity, and specificity were 97.6%, 100.0%, and 92.6%, respectively; AUROC was 99.4% (95% CI, 0.985–1.000); performance was better than VGG-16 and Resnet-50, and was as good as the ophthalmologists. Our model performed well when diagnosing CSC and yielded highly accurate results when distinguishing between acute and chronic cases. Thus, automated deep learning system algorithms could play a role independent of human experts in the diagnosis of CSC.

Adaptive-Guided-Coupling-Probability Level Set for Retinal Layer Segmentation.

Quantitative assessment of retinal layer thickness in spectral domain-optical coherence tomography (SD-OCT) images is vital for clinicians to determine the degree of ophthalmic lesions. However, due to the complex retinal tissues, high-level speckle noises and low intensity constraint, how to accurately recognize the retinal layer structure still remains a challenge. To overcome this problem, this paper proposes an adaptive-guided-coupling-probability level set method for retinal layer segmentation in SD-OCT images. Specifically, based on Bayes's theorem, each voxel probability representation is composed of two probability terms in our method. The first term is constructed as neighborhood Gaussian fitting distribution to characterize intensity information for each intra-retinal layer. The second one is boundary probability map generated by combining anatomical priors and adaptive thickness information to ensure surfaces evolve within a proper range. Then, the voxel probability representation is introduced into the proposed segmentation framework based on coupling probability level set to detect layer boundaries. A total of 1792 retinal B-scan images from 4 SD-OCT cubes in healthy eyes, 5 cubes in abnormal eyes with central serous chorioretinaopathy and 5 SD-OCT cubes in abnormal eyes with age-related macular disease are used to evaluate the proposed method. The experiment demonstrates that the segmentation results obtained by the proposed method have a good consistency with ground truth, and the proposed method outperforms six methods in the layer segmentation of uneven retinal SD-OCT images.