Search

We reconstruct the three-dimensional shape and location of the retinal vascular network from commercial spectral-domain (SD) optical coherence tomography (OCT) data. The two-dimensional location of retinal vascular network on the eye fundus is obtained through support vector machines classification of properly defined fundus images from OCT data, taking advantage of the fact that on standard SD-OCT, the incident light beam is absorbed by hemoglobin, creating a shadow on the OCT signal below each perfused vessel. The depth-wise location of the vessel is obtained as the beginning of the shadow. The classification of crossovers and bifurcations within the vascular network is also addressed. We illustrate the feasibility of the method in terms of vessel caliber estimation and the accuracy of bifurcations and crossovers classification.

Two-dimensional images computed from three-dimensional optical coherence tomography (OCT) data are intrinsically aligned with it, allowing to accurately position a retinal OCT scan within the ocular fundus. In this work, we aim to compute an OCT fundus reference image with improved retinal vasculature extension and contrast over traditional approaches. Based on the shadow casted by hemoglobin on the outer layers of the retina, we compute three independent images from the OCT volumetric data (including the traditional fundus reference image). Combining these images, a fourth one is created that is able to outperform the other three, both quantitatively and qualitatively (as evaluated by retina specialists). The vascular network extension provided by this method was also compared with widely used fundus imaging modalities, showing that it is similar to that achieved with color fundus photography. In this way, the proposed method is an important starting point to the segmentation of the vascular tree and provides users with a detailed fundus reference image., (Copyright © 2014 Elsevier Ltd. All rights reserved.)

The automatic segmentation of the retinal vascular network from ocular fundus images has been performed by several research groups. Although different approaches have been proposed for traditional imaging modalities, only a few have addressed this problem for optical coherence tomography (OCT). Furthermore, these approaches were focused on the optic nerve head region. Compared to color fundus photography and fluorescein angiography, two-dimensional ocular fundus reference images computed from three-dimensional OCT data present additional problems related to system lateral resolution, image contrast, and noise. Specifically, the combination of system lateral resolution and vessel diameter in the macular region renders the process particularly complex, which might partly explain the focus on the optic disc region. In this report, we describe a set of features computed from standard OCT data of the human macula that are used by a supervised-learning process (support vector machines) to automatically segment the vascular network. For a set of macular OCT scans of healthy subjects and diabetic patients, the proposed method achieves 98% accuracy, 99% specificity, and 83% sensitivity. This method was also tested on OCT data of the optic nerve head region achieving similar results.

Introduction: The presence of large-sized drusen (≥125 μm), soft indistinct drusen, pigmentary changes, a large area of drusen and a choroidal neovascular membrane in one eye have been found to be predictive risk factors of late exudative age-related macular degeneration (AMD). Multimodal imaging potentially increases the possibility of indentifying further potential risk factors of developing wet AMD., Purpose: To identify morphological and/or functional baseline risk factors for the development of choroidal neovascularization (CNV) in a multimodal set of images from fellow eyes of patients with exudative AMD., Methods: Single-center, prospective, observational, longitudinal 2-year plus 1-year extension study of 62 patients with neovascular AMD in one eye (the nonstudy eye) and early age-related maculopathy (ARM) in the fellow eye (study eye). Best-corrected ETDRS visual acuity, fluorescein angiography (FA) and indocyanine green angiography (ICG), fundus photography, retinal leakage analysis, fundus autofluorescence imaging and optical coherence tomography (OCT Stratus 4.0.2, Carl Zeiss Meditech Inc.) were performed at baseline and every 6 months in order to identify both conversion to CNV as well as possible predictive features present before conversion. A semiautomated computer-assisted grading system was used for classifying fundus color images. Only eyes with 3 years of follow-up were considered for statistical analysis., Results: Fifty-two patients completed the 3-year study follow-up: 26 men and 26 women aged from 56 to 92 years (mean ± SD: 76 ± 6 years). CNV confirmed with FA developed in 46% of the 52 study eyes during the 3-year follow-up (24 converted eyes: 7 in the first year, 11 in the second and 6 in the third). A significantly higher risk for conversion to wet AMD was found only for leakage on a retinal leakage analyzer (odds ratio, OR = 5.0; 95% confidence interval, CI = 1.5-16.4; p = 0.006) detected at least in one visit before the onset of exudative lesions, for baseline ICG hot spots (OR = 7.2; 95% CI = 2.0-25.7; p = 0.002), baseline late ICG hot spots (OR = 4.7; 95% CI = 1.4-15.4; p = 0.009) and baseline early ICG hypofluorescent spots (OR = 3.7; 95% CI = 1.2-12.1; p = 0.025). The total area of drusen, the area of drusen in subfield 1, inner circle or outer circle, the total number of drusen and the number of drusen ≥125 μm, fundus autofluorescence patterns, OCT findings and the severity of ARM at baseline did not show any correlation with an increased risk of conversion to wet AMD., Conclusion: At 3 years, progression from early to late exudative AMD was superior to the expected rate (44%). ICG early and late hyperfluorescent spots or areas, ICG early hypofluorescent spots or areas and early leakage detected with the retinal leakage analyzer, but not pigmentary changes, large drusen, number and area of drusen at any location or a greater severity of ARM at baseline, showed to be a predictive parameter of conversion to wet AMD., (Copyright © 2011 S. Karger AG, Basel.)

Ocular fundus imaging plays a key role in monitoring the health status of the human eye. Currently, a large number of imaging modalities allow the assessment and/or quantification of ocular changes from a healthy status. This review focuses on the main digital fundus imaging modality, color fundus photography, with a brief overview of complementary techniques, such as fluorescein angiography. While focusing on two-dimensional color fundus photography, the authors address the evolution from nondigital to digital imaging and its impact on diagnosis. They also compare several studies performed along the transitional path of this technology. Retinal image processing and analysis, automated disease detection and identification of the stage of diabetic retinopathy (DR) are addressed as well. The authors emphasize the problems of image segmentation, focusing on the major landmark structures of the ocular fundus: the vascular network, optic disk and the fovea. Several proposed approaches for the automatic detection of signs of disease onset and progression, such as microaneurysms, are surveyed. A thorough comparison is conducted among different studies with regard to the number of eyes/subjects, imaging modality, fundus camera used, field of view and image resolution to identify the large variation in characteristics from one study to another. Similarly, the main features of the proposed classifications and algorithms for the automatic detection of DR are compared, thereby addressing computer-aided diagnosis and computer-aided detection for use in screening programs., (Copyright © 2011 S. Karger AG, Basel.)

Purpose: To demonstrate the association between changes in the blood-retinal barrier (BRB) identified by fluorescein leakage and those in the optical properties of the human retina determined by optical coherence tomography (OCT) and show how these changes can be quantified and their location identified within the retina., Methods: Two imaging techniques were applied: the retinal leakage analyzer, to map BRB function into intact or disrupted regions, and OCT, to measure refractive index changes along the light path within the human ocular fundus., Results: A total of 140 comparisons were made, 77 between areas of regions receiving the same classification (intact or disrupted BRB) and 63 between areas of regions receiving distinct classifications, from 4 pathological cases: 2 eyes with nonproliferative diabetic retinopathy and 2 eyes with wet age-related macular degeneration. In all cases, the distribution of OCT data between regions of intact and regions of disrupted BRB, identified by the retinal leakage analyzer, was quantified and was statistically significantly different (p < 0.001). In addition, it was found that the differences could be localized in the retina to specific structural sequences., Conclusions: Using a novel method to analyze OCT data, we showed that it may be possible to quantify differences in the extracellular compartment in eyes with retinal disease and alterations of the BRB. Based on quantitative techniques, our findings demonstrate the presence of indirect information on the BRB status within noninvasive OCT data., (Copyright © 2011 S. Karger AG, Basel.)

Purpose: To examine the relationship between microaneurysm turnover (formation rate), using a new semi-automatic method (MA-Tracker) based on color fundus photographs, and diabetic retinopathy (DR) progression to clinically significant macular edema (CSME)., Methods: In total, 113 patients/eyes with nonproliferative DR (NPDR) were followed up every 6 months for 2 years as controls of the DR clinical trials, and by conventional general and ophthalmological care for the next 8 years (over a total of 10 years' follow-up). Microaneurysm turnover for the 2 first years was computed using the MA-Tracker., Results: The 17 patients that developed CSME over the 10 years of follow-up presented a microaneurysm formation rate of 9.2 +/- 18.2 microaneurysms/year (mean +/- SD) during the first 2 years, which was statistically higher than the eyes that did not develop CSME (0.5 +/- 1.2 microaneurysms/year, p < 0.001). These 17 patients also presented higher HbA(1C) levels at baseline (8.5 +/- 1.2%) compared to the patients who did not develop CSME (7.3 +/- 1.2%, p = 0.001)., Conclusions: A high microaneurysm formation rate on color fundus photographs appears to be a good biomarker for DR progression to CSME in type 2 diabetic patients with NPDR., ((c) 2009 S. Karger AG, Basel.)

Purpose: To assess the reliability of microaneurysm turnover, computed from color fundus photographs, in evaluating diabetic retinopathy in patients with type 2 diabetes and nonproliferative retinopathy., Methods: A new method (MA-Tracker) was developed to count microaneurysms by mapping their locations through image co-registration. To compute the reliability of microaneurysm turnover, 3 different graders were asked to earmark microaneurysms on the same set of color fundus photographs., Results: The total numbers of microaneurysms earmarked in each of 5 visits suggest that microaneurysms remain stable over time (p >or= 0.138). However, an analysis of each microaneurysm showed that only 29.4% remained at the same location. By computing the formation and disappearance rates of microaneurysms (2.3 and 1.7 microaneurysms/year, respectively), a significant turnover of microaneurysms was found., Conclusions: The formation and disappearance rates of microaneurysms obtained from color fundus photographs using MA-Tracker show very good agreement between different graders, and can be used as indicators of microaneurysm turnover in the initial stages of diabetic retinopathy., ((c) 2009 S. Karger AG, Basel.)

The aim of the work herein presented is to map blood-retinal barrier function by measuring retinal fluorescein leakage from the blood stream into the human vitreous using a confocal scanning laser ophthalmoscope (CSLO). Existing methods for the assessment of fluorescein leakage into the human vitreous are based on the qualitative evaluation of fluorescein angiographies (FA) and on volume measurements, as performed by the Fluorotron Master. A new procedure is presented capable of measuring fluorescein leakage into the vitreous while simultaneously imaging the retina. The present methodology computes the fluorescein leakage in a fully automated way, based on the three-dimensional fluorescence distribution in the human eye by using a single data acquisition. The processing includes signal filtering, volume alignment and profile deconvolution. The deconvolved profile obeys the established physical model. Representative cases shown are: a healthy eye; an eye with drusen from a nondiabetic person; a photocoagulated eye; and an eye with nonproliferative diabetic retinopathy. The results are in agreement with previous findings and go a step further by making possible its daily usage in a clinical setup based on currently available instrumentation.

Background: To identify alterations of retinal capillary blood flow in the papillomacular area in preclinical diabetic retinopathy using the Heidelberg scanning laser Doppler flowmeter., Methods: Ten eyes from ten patients with type 2 diabetes and no lesions visible on fundus photography (level 10 of Wisconsin grading) and ten eyes from ten healthy subjects of similar age range were examined with the HRF. Intravisit reproducibility of retinal capillary blood flow measurements was assessed in normal subjects and in type 2 diabetic patients, comparing different measurement areas and different analysis procedures: (a) 10x10 pixel box with original software, (b) 10x10 pixel box with SLDF software, and (c) whole-scan analysis with SLDF software (automatic full-field perfusion image analysis)., Results: Intravisit reproducibility for the whole-scan analysis in the papillomacular area was 3.52%, 4.81% and 4.60% for volume (VOL), flow (FLW) and velocity (VEL) respectively. Using this method, mean and SD values for retinal capillary blood-flow are 13.25+/-2.87, 214.58+/-55.30 and 0.74+/-0.17, for VOL, FLW and VEL for healthy eyes, comparing with 19.85+/-6.22, 360.87+/-158.70 and 1.20+/-0.48 in eyes with preclinical diabetic retinopathy (P<0.010, P<0.019 and P<0.015 respectively)., Conclusions: The HRF shows acceptable reproducibility when using whole-scan analysis in the papillomacular area. Retinal capillary blood VOL, FLW and VEL were particularly increased in five of the ten diabetic eyes examined, with values over the mean + 2SD of the control population, suggesting that eyes showing increased retinal capillary blood flow may indicate risk of progression.

Objective: To identify alterations of the blood-retinal barrier by mapping retinal fluorescein leakage into the vitreous and changes in retinal thickness occurring in the macular region in preclinical diabetic retinopathy., Methods: Ten eyes from 10 patients with type 2 diabetes and no lesions visible on fundus photography (level 10 of Wisconsin grading) were examined with the retinal leakage analyzer (RLA) (Confocal Scanning Laser Ophthalmoscope [modified]; Carl Zeiss Inc, Thornwood, NY) and the retinal thickness analyzer (RTA) (Talia Technology, Mevaseret Zion, Israel). The maps of retinal leakage and retinal thickness were aligned and integrated in the same image to correlate leakage with thickness. Data from the group of individuals with diabetes were compared with those of a healthy control population (N = 14; mean age, 48 years; range, 42-55 years) and used to establish reference maps for the RLA and RTA., Results: Areas of abnormally increased fluorescein leakage were detected in 9 of 10 eyes examined. The increased leakage in 6 (67%) of 9 eyes reached values higher than 40% more than the mean +2 SD RLA control value. Areas of abnormally increased thickness were found in 7 of 10 eyes examined. For the most part, the increases in retinal thickness were not severe (ie, <15% increase in 5 eyes and an 18% increase in 1 eye). The eyes with the most extensive leakage (cases 1, 3, and 9) showed relatively good coincidence between the location of the areas of increased leakage and the location of the areas of increased thickness. In 4 eyes (cases 2, 5, 7, and 8), no such correlation was apparent. The 3 remaining eyes showed little coincidence between these locations. Characteristically, the latter 3 eyes had areas of abnormally increased thickness that were much larger than the areas of increased fluorescein leakage, which were relatively moderate or absent of any leakage., Conclusions: Localized sites of increased fluorescein leakage and zones of increased retinal thickness were found in most eyes in a series of 10 eyes in the preretinopathy stage from 10 patients with type 2 diabetes. Increases in retinal thickness may be observed that do not coincide with sites of retinal leakage. Two types of increased retinal thickness may, therefore, be present in the preretinopathy stage of diabetic retinopathy, one directly associated with an alteration of the blood-retinal barrier, and another occurring without apparent breakdown of blood-retinal barrier.

Objective: To demonstrate an objective, quantitative, and sensitive method of mapping retinal fluorescein leakage into the vitreous while simultaneously imaging the retina., Methods: A prototype Zeiss confocal scanning laser ophthalmoscope was modified to obtain fluorometric measurements from 18 optical planes across the retina and cortical vitreous, separated from each other by 150 microm, and parallel to the retinal surface. After intravenous administration of fluorescein, an axial graphic of equivalent fluorescein concentration in the vitreous may be obtained from any region of interest. After correcting for fluorescence levels in the retina and choroid and plasma levels of free fluorescein, permeability values of the blood-retinal barrier to fluorescein were obtained from 1512 regions measuring 75 x 75 microm, from a total 3150 x 2700-microm area of the fundus, generating a detailed map of retinal fluorescein leakage. The method was assessed in vitro and in 7 healthy subjects who underwent scans during separate visits. Depth resolution and influence of chorioretinal fluorescence were further tested in 2 patients with multiple drusen and in 2 eyes after vitrectomy. Fourteen eyes from 7 patients with diabetes and nonproliferative retinopathy were also examined. Lateral resolution was tested in 3 diabetic eyes that underwent focal photocoagulation. Four eyes from 2 patients with diabetes and minimal retinopathy were examined at 3-month intervals. All eyes examined had less than 2 diopters of astigmatism., Results: Characteristics of the modified confocal scanning laser fluorometer included a lower limit of detection equal to 0.40 Eq ng/mL and depth precision of +/-15 microm. Values for the blood-retinal barrier permeability index in healthy subjects, measured 30 minutes after a single intravenous pulse of fluorescein (14 mg/kg), ranged from 1.3 +/- 0.4 x 10(-6) cm/s over the foveal avascular zone to 2.2 +/- 0.6 x 10(-6) cm/s over vessels in the retina. Diabetic eyes with retinopathy showed higher values, ranging from 1.4 to 15.0 x 10(-6) cm/s. Vitrectomized eyes and eyes with multiple drusen showed the validity of the correction algorithm demonstrating that measurements of fluorescence in the vitreous are not influenced by the chorioretinal fluorescence level. Argon laser photocoagulation burns placed in the diabetic retina demonstrated a lateral resolution on the order of 75 to 100 microm. Intravisit and intervisit reproducibility was +/-10.2% and +/-13%, respectively., Conclusions: This new method measures localized alterations of the blood-retinal barrier and allows for direct correlation with retinal anatomy. Its most interesting feature is the ability to map retinal fluorescein leakage while simultaneously imaging the retina. This capability is expected to improve our understanding and management of retinal disease.