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.

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.

Uveal melanoma, the most common intraocular primary cancer in adults, is characterized by striking variability in metastatic tendencies. BAP1 deletion in the primary tumor is associated with uveal melanoma metastasis, but it cannot always be resolved by bulk DNA sequencing of heterogeneous tumors. Here, we show that assessment of BAP1 methylation is an accurate and readily clinically actionable assay to accurately identify high-risk uveal melanoma patients.

Purpose

To apply computational techniques to wide-angle swept-source optical coherence tomography (SS-OCT) images to identify novel, glaucoma-related structural features and improve detection of glaucoma and prediction of future glaucomatous progression.

Methods

Wide-angle SS-OCT, OCT circumpapillary retinal nerve fiber layer (cpRNFL) circle scans spectral-domain (SD)-OCT, standard automated perimetry (SAP), and frequency doubling technology (FDT) visual field tests were completed every 3 months for 2 years from a cohort of 28 healthy participants (56 eyes) and 93 glaucoma participants (179 eyes). RNFL thickness maps were extracted from segmented SS-OCT images and an unsupervised machine learning approach based on principal component analysis (PCA) was used to identify novel structural features. Area under the receiver operating characteristic curve (AUC) was used to assess diagnostic accuracy of RNFL PCA for detecting glaucoma and progression compared to SAP, FDT, and cpRNFL measures.

Results

The RNFL PCA features were significantly associated with mean deviation (MD) in both SAP (R2 = 0.49, P < 0.0001) and FDT visual field testing (R2 = 0.48, P < 0.0001), and with mean circumpapillary RNFL thickness (cpRNFLt) from SD-OCT (R2 = 0.58, P < 0.0001). The identified features outperformed each of these measures in detecting glaucoma with an AUC of 0.95 for RNFL PCA compared to an 0.90 for mean cpRNFLt (P = 0.09), 0.86 for SAP MD (P = 0.034), and 0.83 for FDT MD (P = 0.021). Accuracy in predicting progression was also significantly higher for RNFL PCA compared to SAP MD, FDT MD, and mean cpRNFLt (P = 0.046, P = 0.007, and P = 0.044, respectively).

Conclusions

A computational approach can identify structural features that improve glaucoma detection and progression prediction.

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 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.

Methods

GEM 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).

Results

Sensitivity 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.

Conclusions

GEM-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 relevance

Detection of glaucomatous progression can be improved by assessing longitudinal changes in localized patterns of glaucomatous defect identified by unsupervised machine learning.

Purpose

To develop and evaluate a deep learning system for differentiating between eyes with and without glaucomatous visual field damage (GVFD) and predicting the severity of GFVD from spectral domain OCT (SD OCT) optic nerve head images.

Design

Evaluation of a diagnostic technology.

Participants

A total of 9765 visual field (VF) SD OCT pairs collected from 1194 participants with and without GVFD (1909 eyes).

Methods

Deep learning models were trained to use SD OCT retinal nerve fiber layer (RNFL) thickness maps, RNFL en face images, and confocal scanning laser ophthalmoscopy (CSLO) images to identify eyes with GVFD and predict quantitative VF mean deviation (MD), pattern standard deviation (PSD), and mean VF sectoral pattern deviation (PD) from SD OCT data.

Main outcome measures

Deep learning models were compared with mean RNFL thickness for identifying GVFD using area under the curve (AUC), sensitivity, and specificity. For predicting MD, PSD, and mean sectoral PD, models were evaluated using R² and mean absolute error (MAE).

Results

In the independent test dataset, the deep learning models based on RNFL en face images achieved an AUC of 0.88 for identifying eyes with GVFD and 0.82 for detecting mild GVFD significantly (P < 0.001) better than using mean RNFL thickness measurements (AUC = 0.82 and 0.73, respectively). Deep learning models outperformed standard RNFL thickness measurements in predicting all quantitative VF metrics. In predicting MD, deep learning models based on RNFL en face images achieved an R² of 0.70 and MAE of 2.5 decibels (dB) compared with 0.45 and 3.7 dB for RNFL thickness measurements. In predicting mean VF sectoral PD, deep learning models achieved high accuracy in the inferior nasal (R² = 0.60) and superior nasal (R² = 0.67) sectors, moderate accuracy in inferior (R² = 0.26) and superior (R² = 0.35) sectors, and lower accuracy in the central (R² = 0.15) and temporal (R² = 0.12) sectors.

Conclusions

Deep learning models had high accuracy in identifying eyes with GFVD and predicting the severity of functional loss from SD OCT images. Accurately predicting the severity of GFVD from SD OCT imaging can help clinicians more effectively individualize the frequency of VF testing to the individual patient.

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.

Purpose

To compare change over time in eye-specific optical coherence tomography (OCT) retinal nerve fiber layer (RNFL)-based region-of-interest (ROI) maps developed using unsupervised deep-learning auto-encoders (DL-AE) to circumpapillary RNFL (cpRNFL) thickness for the detection of glaucomatous progression.

Methods

Forty-four progressing glaucoma eyes (by stereophotograph assessment), 189 nonprogressing glaucoma eyes (by stereophotograph assessment), and 109 healthy eyes were followed for ≥3 years with ≥4 visits using OCT. The San Diego Automated Layer Segmentation Algorithm was used to automatically segment the RNFL layer from raw three-dimensional OCT images. For each longitudinal series, DL-AEs were used to generate individualized eye-based ROI maps by identifying RNFL regions of likely progression and no change. Sensitivities and specificities for detecting change over time and rates of change over time were compared for the DL-AE ROI and global cpRNFL thickness measurements derived from a 2.22-mm to 3.45-mm annulus centered on the optic disc.

Results

The sensitivity for detecting change in progressing eyes was greater for DL-AE ROIs than for global cpRNFL annulus thicknesses (0.90 and 0.63, respectively). The specificity for detecting not likely progression in nonprogressing eyes was similar (0.92 and 0.93, respectively). The mean rates of change in DL-AE ROI were significantly faster than for cpRNFL annulus thickness in progressing eyes (-1.28 µm/y vs. -0.83 µm/y) and nonprogressing eyes (-1.03 µm/y vs. -0.78 µm/y).

Conclusions

Eye-specific ROIs identified using DL-AE analysis of OCT images show promise for improving assessment of glaucomatous progression.

Translational relevance

The detection and monitoring of structural glaucomatous progression can be improved by considering eye-specific regions of likely progression identified using deep learning.

Purpose

The 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.

Methods

FDT 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.

Results

FDT 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<0.001). VIM identified meaningful clusters of FDT data and positioned a set of statistically independent axes through the mean of each cluster. The optimal VIM model separated the FDT fields into 3 clusters. Cluster N contained primarily normal fields (1109/1190, specificity 93.1%) and clusters G1 and G2 combined, contained primarily abnormal fields (651/786, sensitivity 82.8%). For clusters G1 and G2 the optimal number of axes were 2 and 5, respectively. Patterns automatically generated along axes within the glaucoma clusters were similar to those known to be indicative of glaucoma. Fields located farther from the normal mean on each glaucoma axis showed increasing field defect severity.

Conclusions

VIM successfully separated FDT fields from healthy and glaucoma eyes without a priori information about class membership, and identified familiar glaucomatous patterns of loss.