Predictors of Glaucomatous Progression in Individuals with Small and Large Optic Discs.

PURPOSE: To identify factors associated with glaucomatous progression in individuals with small and large optic discs. DESIGN: Retrospective review. SUBJECTS: 4505 individuals with glaucoma at UCLA; 233 (59.7%) with small discs, 157 (40.3%) with large discs. METHODS: Small and large disc sizes were defined by OCT or Heidelberg Retinal Tomography as disc area ≤ 5% (≤ 1.3 mm2) and ≥ 95% (≥ 2.9 mm2), respectively. Medical records were reviewed for demographics, systemic comorbidities, glaucoma type, ocular comorbidities, and ocular surgery. Logistic regression was used to identify predictors of visual field (VF) progression in individuals with small and large discs and predictors of large versus small discs. MAIN OUTCOME MEASURES: The VF deterioration with mean deviation, pointwise linear regression, and glaucoma rate index (GRI); large vs. small disc. RESULTS: In individuals with small discs, Asian versus non-Hispanic White ethnicity was associated with increased progression (adjusted odds ratio [aOR] = 4.05; 95% confidence interval [CI] = 1.12-14.59 for GRI). Higher intraocular pressure (IOP) range and peak were associated with increased progression in individuals with both small discs (aOR = 1.12; 95% CI = 1.00-1.27 and aOR = 1.05; 95% CI = 1.00-1.10 per 1 mmHg for range and peak with GRI) and large discs (aOR = 1.35; 95% CI = 1.12-1.66 and aOR = 1.11; 95% CI = 1.03-1.20 per 1 mmHg for range and peak with GRI). Multivariable predictors of having large vs. small discs included vasospastic phenotype (aOR = 2.58; 95% CI = 1.35-5.19) and Black (aOR = 20.46; 95% CI = 8.33-61.84), Hispanic/Latino (aOR = 9.65; 95% CI = 4.14-25.39), Asian (aOR = 4.87; 95% CI = 2.96-8.1), and other (aOR = 2.79; 95% CI = 1.69-4.63) versus non-Hispanic White ethnicity. CONCLUSIONS: Increased odds of glaucomatous progression were associated with Asian vs. non-Hispanic White ethnicity in glaucoma patients with small optic discs, as well as with increased IOP range and peak in those with small and large discs. Individuals with a vasospastic phenotype and those from racial and ethnic minority backgrounds had increased odds of having large vs. small optic discs. Further characterization of discernible phenotypes would improve disease prognostication and help individualize glaucoma treatment. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Phenotypic expressions of the optic disc in primary open-angle glaucoma.

BACKGROUND: Which phenotypes are we able to recognize in the optic nerve of patients with primary open angle glaucoma? METHODS: Retrospective interventional case series. 885 eyes from 885 patients at an outpatient tertiary care centre who met specified criteria for POAG were included. Disc photographs were classified by three glaucoma specialists into the following phenotypes according to their predominant characteristics: (1) concentric rim thinning, (2) focal rim thinning, (3) acquired pit of the optic nerve (APON), (4) tilted, (5) extensive peripapillary atrophy (PPA), and (6) broad rim thinning. Demographic, medical, and ocular data were collected. Kruskal-Wallis was used as a non-parametric test and pairwise comparison was performed by using Wilcoxon rank sum test corrected. RESULTS: Phenotypic distribution was as follows: 398(45%) focal thinning, 153(18%) concentric thinning, 153(17%) broad thinning, 109(12%) tilted, 47(5%) extensive PPA and 25(3%) APON. Phenotypic traits of interest included a higher proportion of female patients with the focal thinning phenotype (p = 0.015); myopia (p = 0.000), Asian race (OR: 8.8, p = 0.000), and younger age (p = 0.000) were associated with the tilted phenotype; the concentric thinning patients had thicker RNFL (p = 0.000), higher MD (p = 0.008) and lower PSD (p = 0.043) than broad thinning, despite no difference in disc sizes (p = 0.849). The focal thinning group had a localized VF pattern with high PSD compared to concentric thinning (p = 0.005). CONCLUSION: We report six phenotypic classifications of POAG patients with demographic and ocular differences between phenotypes. Future refinement of phenotypes should allow enhanced identification of genetic associations and improved individualization of patient care.

A Neural Network for Automated Image Quality Assessment of Optic Disc Photographs.

This study describes the development of a convolutional neural network (CNN) for automated assessment of optic disc photograph quality. Using a code-free deep learning platform, a total of 2377 optic disc photographs were used to develop a deep CNN capable of determining optic disc photograph quality. Of these, 1002 were good-quality images, 609 were acceptable-quality, and 766 were poor-quality images. The dataset was split 80/10/10 into training, validation, and test sets and balanced for quality. A ternary classification model (good, acceptable, and poor quality) and a binary model (usable, unusable) were developed. In the ternary classification system, the model had an overall accuracy of 91% and an AUC of 0.98. The model had higher predictive accuracy for images of good (93%) and poor quality (96%) than for images of acceptable quality (91%). The binary model performed with an overall accuracy of 98% and an AUC of 0.99. When validated on 292 images not included in the original training/validation/test dataset, the model's accuracy was 85% on the three-class classification task and 97% on the binary classification task. The proposed system for automated image-quality assessment for optic disc photographs achieves high accuracy in both ternary and binary classification systems, and highlights the success achievable with a code-free platform. There is wide clinical and research potential for such a model, with potential applications ranging from integration into fundus camera software to provide immediate feedback to ophthalmic photographers, to prescreening large databases before their use in research.

DDLSNet: A Novel Deep Learning-Based System for Grading Funduscopic Images for Glaucomatous Damage.

Purpose: To report an image analysis pipeline, DDLSNet, consisting of a rim segmentation (RimNet) branch and a disc size classification (DiscNet) branch to automate estimation of the disc damage likelihood scale (DDLS). Design: Retrospective observational. Participants: RimNet and DiscNet were developed with 1208 and 11 536 optic disc photographs (ODPs), respectively. DDLSNet performance was evaluated on 120 ODPs from the RimNet test set, for which the DDLS scores were graded by clinicians. Reproducibility was evaluated on a group of 781 eyes, each with 2 ODPs taken within 4 years apart. Methods: Disc damage likelihood scale calculation requires estimation of optic disc size, provided by DiscNet (VGG19 network), and the minimum rim-to-disc ratio (mRDR) or absent rim width (ARW), provided by RimNet (InceptionV3/LinkNet segmentation model). To build RimNet's dataset, glaucoma specialists marked optic disc rim and cup boundaries on ODPs. The "ground truth" mRDR or ARW was calculated. For DiscNet's dataset, corresponding OCT images provided "ground truth" disc size. Optic disc photographs were split into 80/10/10 for training, validation, and testing, respectively, for RimNet and DiscNet. DDLSNet estimation was tested against manual grading of DDLS by clinicians with the average score used as "ground truth." Reproducibility of DDLSNet grading was evaluated by repeating DDLS estimation on a dataset of nonprogressing paired ODPs taken at separate times. Main Outcome Measures: The main outcome measure was a weighted kappa score between clinicians and the DDLSNet pipeline with agreement defined as ± 1 DDLS score difference. Results: RimNet achieved an mRDR mean absolute error (MAE) of 0.04 (± 0.03) and an ARW MAE of 48.9 (± 35.9) degrees when compared to clinician segmentations. DiscNet achieved 73% (95% confidence interval [CI]: 70%, 75%) classification accuracy. DDLSNet achieved an average weighted kappa agreement of 0.54 (95% CI: 0.40, 0.68) compared to clinicians. Average interclinician agreement was 0.52 (95% CI: 0.49, 0.56). Reproducibility testing demonstrated that 96% of ODP pairs had a difference of ≤ 1 DDLS score. Conclusions: DDLSNet achieved moderate agreement with clinicians for DDLS grading. This novel approach illustrates the feasibility of automated ODP grading for assessing glaucoma severity. Further improvements may be achieved by increasing the number of incomplete rims sample size, expanding the hyperparameter search, and increasing the agreement of clinicians grading ODPs.

RimNet: A Deep Neural Network Pipeline for Automated Identification of the Optic Disc Rim.

Purpose: Accurate neural rim measurement based on optic disc imaging is important to glaucoma severity grading and often performed by trained glaucoma specialists. We aim to improve upon existing automated tools by building a fully automated system (RimNet) for direct rim identification in glaucomatous eyes and measurement of the minimum rim-to-disc ratio (mRDR) in intact rims, the angle of absent rim width (ARW) in incomplete rims, and the rim-to-disc-area ratio (RDAR) with the goal of optic disc damage grading. Design: Retrospective cross sectional study. Participants: One thousand and twenty-eight optic disc photographs with evidence of glaucomatous optic nerve damage from 1021 eyes of 903 patients with any form of primary glaucoma were included. The mean age was 63.7 (± 14.9) yrs. The average mean deviation of visual fields was -8.03 (± 8.59). Methods: The images were required to be of adequate quality, have signs of glaucomatous damage, and be free of significant concurrent pathology as independently determined by glaucoma specialists. Rim and optic cup masks for each image were manually delineated by glaucoma specialists. The database was randomly split into 80/10/10 for training, validation, and testing, respectively. RimNet consists of a deep learning rim and cup segmentation model, a computer vision mRDR measurement tool for intact rims, and an ARW measurement tool for incomplete rims. The mRDR is calculated at the thinnest rim section while ARW is calculated in regions of total rim loss. The RDAR was also calculated. Evaluation on the Drishti-GS dataset provided external validation (Sivaswamy 2015). Main Outcome Measures: Median Absolute Error (MAE) between glaucoma specialists and RimNet for mRDR and ARW. Results: On the test set, RimNet achieved a mRDR MAE of 0.03 (0.05), ARW MAE of 31 (89)°, and an RDAR MAE of 0.09 (0.10). On the Drishti-GS dataset, an mRDR MAE of 0.03 (0.04) and an mRDAR MAE of 0.09 (0.10) was observed. Conclusions: RimNet demonstrated acceptably accurate rim segmentation and mRDR and ARW measurements. The fully automated algorithm presented here would be a valuable component in an automated mRDR-based glaucoma grading system. Further improvements could be made by improving identification and segmentation performance on incomplete rims and expanding the number and variety of glaucomatous training images.