The purpose of this study was to investigate the effect of image quality on retinal nerve fiber layer (RNFL) and retinal thickness measurements obtained using three commercially available spectral-domain optical coherence tomographers (SD-OCT). Subjectively determined good, medium and poor quality images were obtained from four healthy and one glaucoma suspect eyes. RNFL and retinal thickness measurements were compared as a function of image quality. Results indicate that when image quality is within the range specified as acceptable by SD-OCT manufacturers, RNFL and retinal thickness measurements are comparable.

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.

Purpose To determine the sensitivity and specificity of confocal scanning laser ophthalmoscope’s Topographic Change Analysis (TCA; Heidelberg Retina Tomograph [HRT]; Heidelberg Engineering, Heidelberg, Germany) parameters for discriminating between progressing glaucomatous and stable healthy eyes. Methods The 0.90, 0.95, and 0.99 specificity cutoffs for various (n = 70) TCA parameters were developed by using 1000 permuted topographic series derived from HRT images of 18 healthy eyes from Moorfields Eye Hospital, imaged at least four times. The cutoffs were then applied to topographic series from 36 eyes with known glaucomatous progression (by optic disc stereophotograph assessment and/or standard automated perimetry guided progression analysis, [GPA]) and 21 healthy eyes from the University of California, San Diego (UCSD) Diagnostic Innovations in Glaucoma Study (DIGS), all imaged at least four times, to determine TCA sensitivity and specificity. Cutoffs also were applied to 210 DIGS patients’ eyes imaged at least four times with no evidence of progression (nonprogressed) by stereophotography or GPA. Results The TCA parameter providing the best sensitivity/specificity tradeoff using the 0.90, 0.95, and 0.99 cutoffs was the largest clustered superpixel area within the optic disc margin (CAREAdisc mm2). Sensitivities/specificities for classifying progressing (by stereophotography and/or GPA) and healthy eyes were 0.778/0.809, 0.639/0.857, and 0.611/1.00, respectively. In nonprogressing eyes, specificities were 0.464, 0.570, and 0.647 (i.e., lower than in the healthy eyes). In addition, TCA parameter measurements of nonprogressing eyes were similar to those of progressing eyes. Conclusions TCA parameters can discriminate between progressing and longitudinally observed healthy eyes. Low specificity in apparently nonprogressing patients’ eyes suggests early progression detection using TCA.

This study examines the ability of RTVue, Cirrus and Spectralis OCT Spectral domain-optical coherence tomographs (SD-OCT) to detect localized retinal nerve fiber layer defects in glaucomatous eyes. In this observational case series, four glaucoma patients (8 eyes) were selected from the University of California, San Diego Shiley Eye Center and the Diagnostic Innovations in Glaucoma Study (DIGS) based on the presence of documented localized RNFL defects in at least one eye confirmed by masked stereophotograph assessment. One RTVue 3D Disc scan, one RTVue NHM4 scan, one Cirrus Optic Disk Cube 200×200 scan and one Spectralis scan centered on the optic disc (15×15 scan angle, 768 A-scans × 73 B-scans) were obtained on all undilated eyes within a single session. Results were compared with those obtained from stereophotographs. In 6 eyes the presence of localized RNFL defects was detected by stereophotography. In general, by qualitatively evaluating the retinal thickness maps generated, all SD-OCT instruments examined were able to confirm the presence of localized glaucomatous structural damage seen on stereophotographs. This study confirms SD-OCT is a promising technology for glaucoma detection as it may assist clinicians identify the presence of localized glaucomatous structural damage.

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.