Search

PurposeTo test the hypothesis that visual field (VF) progression can be predicted from baseline and longitudinal optical coherence tomography (OCT) structural measurements.DesignProspective cohort study.MethodsA total of 104 eyes (104 patients) with ≥3 years of follow-up and ≥5 VF examinations were enrolled. We defined VF progression based on pointwise linear regression on 24-2 VF (≥3 locations with slope less than or equal to -1.0 dB/year and P < .01). We used elastic net logistic regression (ENR) and machine learning to predict VF progression with demographics, baseline circumpapillary retinal nerve fiber layer (RNFL), macular ganglion cell/inner plexiform layer (GCIPL) thickness, and RNFL and GCIPL change rates at central 24 superpixels and 3 eccentricities, 3.4°, 5.5°, and 6.8°, from fovea and hemimaculas. Areas-under-ROC curves (AUC) were used to compare models.ResultsAverage ± SD follow-up and VF examinations were 4.5 ± 0.9 years and 8.7 ± 1.6, respectively. VF progression was detected in 23 eyes (22%). ENR selected rates of change of superotemporal RNFL sector and GCIPL change rates in 5 central superpixels and at 3.4° and 5.6° eccentricities as the best predictor subset (AUC = 0.79 ± 0.12). Best machine learning predictors consisted of baseline superior hemimacular GCIPL thickness and GCIPL change rates at 3.4° eccentricity and 3 central superpixels (AUC = 0.81 ± 0.10). Models using GCIPL-only structural variables performed better than RNFL-only models.ConclusionsVF progression can be predicted with clinically relevant accuracy from baseline and longitudinal structural data. Further refinement of proposed models would assist clinicians with timely prediction of functional glaucoma progression and clinical decision making.

Goal: The purpose of this study was to identify clinically relevant patterns of glaucomatous vision loss through convex representation to predict glaucoma several years prior to disease onset. Methods: We developed a deep archetypal analysis to identify patterns of glaucomatous vision loss, and then projected visual fields over the identified patterns. Projections provided a representation that was more accurate in detecting glaucomatous vision loss, thus, more appropriate for recognizing preclinical signs of glaucoma prior to disease development. To overcome the class imbalance in prediction, we implemented a class-balanced bagging with neural networks. Results: Using original visual field as features of the class-balanced bagging classification provided an area under the receiver-operating characteristic curve (AUC) of 0.55 for predicting glaucoma approximately four years prior to disease development. Using convex representation of the visual fields as input features provided an AUC of 0.61 while using deep convex representation as input features improved the AUC to 0.71. Relevance vector machine (RVM) achieved an AUC of 0.64. Conclusion: Deep archetypal analysis representation of visual functional features with balanced bagging classification could serve as an automated tool for predicting glaucoma. Significance: Glaucoma is the second leading cause of worldwide blindness. Most people with glaucoma have no early symptoms or pain, delaying diagnosis in many patients until they reach late irreversible vision loss stages. In fact, about 50% of people with glaucoma are unaware they have the disease. Deep archetypal analysis models may impact clinical practice in effectively identifying at-risk glaucoma patients well prior to disease development.

PurposeTo evaluate the association between vessel density measurements using optical coherence tomography angiography (OCT-A) and severity of visual field loss in primary open-angle glaucoma.DesignObservational, cross-sectional study.ParticipantsA total of 153 eyes from 31 healthy participants, 48 glaucoma suspects, and 74 glaucoma patients enrolled in the Diagnostic Innovations in Glaucoma Study.MethodsAll eyes underwent imaging using OCT-A (Angiovue; Optovue, Fremont, CA), spectral-domain OCT (Avanti; Optovue), and standard automated perimetry (SAP). Retinal vasculature information was summarized as vessel density, the percentage of area occupied by flowing blood vessels in the selected region. Two measurements from the retinal nerve fiber layer (RNFL) were used: circumpapillary vessel density (cpVD) (750-μm-wide elliptical annulus around the optic disc) and whole-image vessel density (wiVD) (entire 4.5×4.5-mm scan field).Main outcome measuresAssociations between the severity of visual field loss, reported as SAP mean deviation (MD), and OCT-A vessel density.ResultsCompared with glaucoma eyes, normal eyes demonstrated a denser microvascular network within the RNFL. Vessel density was higher in normal eyes followed by glaucoma suspects, mild glaucoma, and moderate to severe glaucoma eyes for wiVD (55.5%, 51.3%, 48.3%, and 41.7%, respectively) and for cpVD (62.8%, 61.0%, 57.5%, 49.6%, respectively) (P < 0.001 for both). The association between SAP MD with cpVD and wiVD was stronger (R2 = 0.54 and R2 = 0.51, respectively) than the association between SAP MD with RNFL (R2 = 0.36) and rim area (R2 = 0.19) (P < 0.05 for all). Multivariate regression analysis showed that each 1% decrease in wiVD was associated with 0.66 decibel (dB) loss in MD and each 1% decrease in cpVD was associated with 0.64 dB loss in MD. In addition, the association between vessel density and severity of visual field damage was found to be significant even after controlling for the effect of structural loss.ConclusionsDecreased vessel density was significantly associated with the severity of visual field damage independent of the structural loss. Optical coherence tomography angiography is a promising technology in glaucoma management, potentially enhancing the understanding of the role of vasculature in the pathophysiology of the disease.

PurposeTo investigate whether vessel density assessed by optical coherence tomography angiography (OCT-A) is reduced in glaucomatous eyes with focal lamina cribrosa (LC) defects.DesignCross-sectional, case-control study.ParticipantsA total of 82 patients with primary open-angle glaucoma (POAG) from the Diagnostic Innovations in Glaucoma Study (DIGS) with and without focal LC defects (41 eyes of 41 patients in each group) matched by severity of visual field (VF) damage.MethodsOptical coherence tomography (OCT) angiography-derived circumpapillary vessel density (cpVD) was calculated as the percentage area occupied by vessels in the measured region extracted from the retinal nerve fiber layer (RNFL) in a 750-μm-wide elliptical annulus around the disc. Focal LC defects were detected using swept-source OCT images.Main outcome measuresComparison of global and sectoral (eight 45-degree sectors) cpVDs and circumpapillary RNFL (cpRNFL) thicknesses in eyes with and without LC defects.ResultsAge, global, and sectoral cpRNFL thicknesses, VF mean deviation (MD) and pattern standard deviation, presence of optic disc hemorrhage, and mean ocular perfusion pressure did not differ between patients with and without LC defects (P > 0.05 for all comparisons). Mean cpVDs of eyes with LC defects were significantly lower than in eyes without a defect globally (52.9%±5.6% vs. 56.8%±7.7%; P = 0.013) and in the inferotemporal (IT) (49.5%±10.3% vs. 56.8%±12.2%; P = 0.004), superotemporal (ST) (54.3%±8.8% vs. 58.8%±9.6%; P = 0.030), and inferonasal (IN) (52.4%±9.0% vs. 57.6%±9.1%; P = 0.009) sectors. Eyes with LC defects in the IT sector (n = 33) had significantly lower cpVDs than eyes without a defect in the corresponding IT and IN sectors (P < 0.05 for all). Eyes with LC defects in the ST sector (n = 19) had lower cpVDs in the ST, IT, and IN sectors (P < 0.05 for all).ConclusionsIn eyes with similar severity of glaucoma, OCT-A-measured vessel density was significantly lower in POAG eyes with focal LC defects than in eyes without an LC defect. Moreover, reduction of vessel density was spatially correlated with the location of the LC defect.

PurposeThe purpose of this study was to compare retinal nerve fiber layer (RNFL) thickness and optical coherence tomography angiography (OCT-A) retinal vasculature measurements in healthy, glaucoma suspect, and glaucoma patients.MethodsTwo hundred sixty-one eyes of 164 healthy, glaucoma suspect, and open-angle glaucoma (OAG) participants from the Diagnostic Innovations in Glaucoma Study with good quality OCT-A images were included. Retinal vasculature information was summarized as a vessel density map and as vessel density (%), which is the proportion of flowing vessel area over the total area evaluated. Two vessel density measurements extracted from the RNFL were analyzed: (1) circumpapillary vessel density (cpVD) measured in a 750-μm-wide elliptical annulus around the disc and (2) whole image vessel density (wiVD) measured over the entire image. Areas under the receiver operating characteristic curves (AUROC) were used to evaluate diagnostic accuracy.ResultsAge-adjusted mean vessel density was significantly lower in OAG eyes compared with glaucoma suspects and healthy eyes. (cpVD: 55.1 ± 7%, 60.3 ± 5%, and 64.2 ± 3%, respectively; P < 0.001; and wiVD: 46.2 ± 6%, 51.3 ± 5%, and 56.6 ± 3%, respectively; P < 0.001). For differentiating between glaucoma and healthy eyes, the age-adjusted AUROC was highest for wiVD (0.94), followed by RNFL thickness (0.92) and cpVD (0.83). The AUROCs for differentiating between healthy and glaucoma suspect eyes were highest for wiVD (0.70), followed by cpVD (0.65) and RNFL thickness (0.65).ConclusionsOptical coherence tomography angiography vessel density had similar diagnostic accuracy to RNFL thickness measurements for differentiating between healthy and glaucoma eyes. These results suggest that OCT-A measurements reflect damage to tissues relevant to the pathophysiology of OAG.

PurposeTo validate Gaussian mixture-model with expectation maximization (GEM) and variational Bayesian independent component analysis mixture-models (VIM) for detecting glaucomatous progression along visual field (VF) defect patterns (GEM-progression of patterns (POP) and VIM-POP). To compare GEM-POP and VIM-POP with other methods.MethodsGEM and VIM models separated cross-sectional abnormal VFs from 859 eyes and normal VFs from 1117 eyes into abnormal and normal clusters. Clusters were decomposed into independent axes. The confidence limit (CL) of stability was established for each axis with a set of 84 stable eyes. Sensitivity for detecting progression was assessed in a sample of 83 eyes with known progressive glaucomatous optic neuropathy (PGON). Eyes were classified as progressed if any defect pattern progressed beyond the CL of stability. Performance of GEM-POP and VIM-POP was compared to point-wise linear regression (PLR), permutation analysis of PLR (PoPLR), and linear regression (LR) of mean deviation (MD), and visual field index (VFI).ResultsSensitivity and specificity for detecting glaucomatous VFs were 89.9% and 93.8%, respectively, for GEM and 93.0% and 97.0%, respectively, for VIM. Receiver operating characteristic (ROC) curve areas for classifying progressed eyes were 0.82 for VIM-POP, 0.86 for GEM-POP, 0.81 for PoPLR, 0.69 for LR of MD, and 0.76 for LR of VFI.ConclusionsGEM-POP was significantly more sensitive to PGON than PoPLR and linear regression of MD and VFI in our sample, while providing localized progression information.Translational relevanceDetection of glaucomatous progression can be improved by assessing longitudinal changes in localized patterns of glaucomatous defect identified by unsupervised machine learning.

Detecting glaucomatous progression is an important aspect of glaucoma management. The assessment of longitudinal series of visual fields, measured using Standard Automated Perimetry (SAP), is considered the reference standard for this effort. We seek efficient techniques for determining progression from longitudinal visual fields by formulating the problem as an optimization framework, learned from a population of glaucoma data. The longitudinal data from each patient's eye were used in a convex optimization framework to find a vector that is representative of the progression direction of the sample population, as a whole. Post-hoc analysis of longitudinal visual fields across the derived vector led to optimal progression (change) detection. The proposed method was compared to recently described progression detection methods and to linear regression of instrument-defined global indices, and showed slightly higher sensitivities at the highest specificities than other methods (a clinically desirable result). The proposed approach is simpler, faster, and more efficient for detecting glaucomatous changes, compared to our previously proposed machine learning-based methods, although it provides somewhat less information. This approach has potential application in glaucoma clinics for patient monitoring and in research centers for classification of study participants.

Blood vessel segmentation is important for the analysis of ocular fundus images for diseases affecting vessel caliber, occlusion, leakage, inflammation, and proliferation. We introduce a novel supervised method to evaluate performance of Leung-Malik filters in delineating vessels. First, feature vectors are extracted for every pixel with respect to the response of Leung-Malik filters on green channel retinal images in different orientations and scales. A two level hierarchical learning framework is proposed to segment vessels in retinal images with confounding disease abnormalities. In the first level, three expert classifiers are trained to delineate 1) vessels, 2) background, and 3) retinal pathologies including abnormal pathologies such as lesions and anatomical structures such as optic disc. In the second level, a new classifier is trained to detect vessels and non-vessel pixels based on results of the expert classifiers. Qualitative evaluation shows the effectiveness of the proposed expert classifiers in modeling retinal pathologies. Quantitative results on two standard datasets STARE (AUC = 0.971, Acc=0.927) and DRIVE (AUC = 0.955, Acc =0.903) are comparable with other state-of-the-art vessel segmentation methods.

A hierarchical approach to learn from visual field data was adopted to identify glaucomatous visual field defect patterns and to detect glaucomatous progression. The analysis pipeline included three stages, namely, clustering, glaucoma boundary limit detection, and glaucoma progression detection testing. First, cross-sectional visual field tests collected from each subject were clustered using a mixture of Gaussians and model parameters were estimated using expectation maximization. The visual field clusters were further estimated to recognize glaucomatous visual field defect patterns by decomposing each cluster into several axes. The glaucoma visual field defect patterns along each axis then were identified. To derive a definition of progression, the longitudinal visual fields of stable glaucoma eyes on the abnormal cluster axes were projected and the slope was approximated using linear regression (LR) to determine the confidence limit of each axis. For glaucoma progression detection, the longitudinal visual fields of each eye on the abnormal cluster axes were projected and the slope was approximated by LR. Progression was assigned if the progression rate was greater than the boundary limit of the stable eyes; otherwise, stability was assumed. The proposed method was compared to a recently developed progression detection method and to clinically available glaucoma progression detection software. The clinical accuracy of the proposed pipeline was as good as or better than the currently available methods.

Machine learning classifiers were employed to detect glaucomatous progression using longitudinal series of structural data extracted from retinal nerve fiber layer thickness measurements and visual functional data recorded from standard automated perimetry tests. Using the collected data, a longitudinal feature vector was created for each patient's eye by computing the norm 1 difference vector of the data at the baseline and at each follow-up visit. The longitudinal features from each patient's eye were then fed to the machine learning classifier to classify each eye as stable or progressed over time. This study was performed using several machine learning classifiers including Bayesian, Lazy, Meta, and Tree, composing different families. Combinations of structural and functional features were selected and ranked to determine the relative effectiveness of each feature. Finally, the outcomes of the classifiers were assessed by several performance metrics and the effectiveness of structural and functional features were analyzed.

Glaucoma is a potentially blinding optic neuropathy that results in a decrease in visual sensitivity. Visual field abnormalities (decreased visual sensitivity on psychophysical tests) are the primary means of glaucoma diagnosis. One form of visual field testing is Frequency Doubling Technology (FDT) that tests sensitivity at 52 points within the visual field. Like other psychophysical tests used in clinical practice, FDT results yield specific patterns of defect indicative of the disease. We used Gaussian Mixture Model with Expectation Maximization (GEM), (EM is used to estimate the model parameters) to automatically separate FDT data into clusters of normal and abnormal eyes. Principal component analysis (PCA) was used to decompose each cluster into different axes (patterns). FDT measurements were obtained from 1,190 eyes with normal FDT results and 786 eyes with abnormal (i.e., glaucomatous) FDT results, recruited from a university-based, longitudinal, multi-center, clinical study on glaucoma. The GEM input was the 52-point FDT threshold sensitivities for all eyes. The optimal GEM model separated the FDT fields into 3 clusters. Cluster 1 contained 94% normal fields (94% specificity) and clusters 2 and 3 combined, contained 77% abnormal fields (77% sensitivity). For clusters 1, 2 and 3 the optimal number of PCA-identified axes were 2, 2 and 5, respectively. GEM with PCA successfully separated FDT fields from healthy and glaucoma eyes and identified familiar glaucomatous patterns of loss.

PurposeThe variational Bayesian independent component analysis-mixture model (VIM), an unsupervised machine-learning classifier, was used to automatically separate Matrix Frequency Doubling Technology (FDT) perimetry data into clusters of healthy and glaucomatous eyes, and to identify axes representing statistically independent patterns of defect in the glaucoma clusters.MethodsFDT measurements were obtained from 1,190 eyes with normal FDT results and 786 eyes with abnormal FDT results from the UCSD-based Diagnostic Innovations in Glaucoma Study (DIGS) and African Descent and Glaucoma Evaluation Study (ADAGES). For all eyes, VIM input was 52 threshold test points from the 24-2 test pattern, plus age.ResultsFDT mean deviation was -1.00 dB (S.D. = 2.80 dB) and -5.57 dB (S.D. = 5.09 dB) in FDT-normal eyes and FDT-abnormal eyes, respectively (p