Faculty Bibliography

PURPOSE:To determine whether peripapillary atrophy (PPA) area is an indicator of glaucomatous structural and functional damage and progression. METHODS:In this retrospective longitudinal analysis from ongoing prospective study we qualified 71 eyes (50 subjects) with glaucoma. All subjects had a comprehensive ophthalmic examination, visual field (VF), and spectral-domain optical coherence tomography (OCT) testing in at least three visits. PPA was manually delineated on en face OCT optic nerve head scans, while observing the corresponding cross-sectional images, as the hyper-reflective area contiguous with the optic disc. RESULTS:The mean follow-up duration was 4.4 ± 1.4 years with an average of 6.8 ± 2.2 visits. At baseline, PPA area was significantly associated only with VF's mean deviation (MD; P = 0.041), visual field index (VFI; P = 0.041), superior ganglion cell inner plexiform layer (GCIPL; P = 0.011), and disc area (P = 0.011). Longitudinally, PPA area was negatively and significantly associated with MD (P = 0.015), VFI (P = 0.035), GCIPL (P = 0.009), superior GCIPL (P = 0.034), and disc area (P = 0.007, positive association). CONCLUSIONS:Longitudinal change in PPA area is an indicator of glaucomatous structural and functional progression but PPA area at baseline cannot predict future progression. TRANSLATIONAL RELEVANCE:Longitudinal changes in peripapillary atrophy area measured by OCT can be an indicator of structural and functional glaucoma progression.

PURPOSE/UNASSIGNED:To establish generalizable pointwise spatial relationship between structure and function through occlusion analysis of a deep-learning (DL) model for predicting the visual field (VF) sensitivities from 3-dimensional (3D) OCT scan. DESIGN/UNASSIGNED:Retrospective cross-sectional study. PARTICIPANTS/UNASSIGNED:A total of 2151 eyes from 1129 patients. METHODS/UNASSIGNED:A DL model was trained to predict 52 VF sensitivities of 24-2 standard automated perimetry from 3D spectral-domain OCT images of the optic nerve head (ONH) with 12 915 OCT-VF pairs. Using occlusion analysis, the contribution of each individual cube covering a 240 × 240 × 31.25 μm region of the ONH to the model's prediction was systematically evaluated for each OCT-VF pair in a separate test set that consisted of 996 OCT-VF pairs. After simple translation (shifting in x- and y-axes to match the ONH center), group t-statistic maps were derived to visualize statistically significant ONH regions for each VF test point within a group. This analysis allowed for understanding the importance of each super voxel (240 × 240 × 31.25 μm covering the entire 4.32 × 4.32 × 1.125 mm ONH cube) in predicting VF test points for specific patient groups. MAIN OUTCOME MEASURES/UNASSIGNED:The region at the ONH corresponding to each VF test point and the effect of the former on the latter. RESULTS/UNASSIGNED:The test set was divided to 2 groups, the healthy-to-early-glaucoma group (792 OCT-VF pairs, VF mean deviation [MD]: -1.32 ± 1.90 decibels [dB]) and the moderate-to-advanced-glaucoma group (204 OCT-VF pairs, VF MD: -17.93 ± 7.68 dB). Two-dimensional group t-statistic maps (x, y projection) were generated for both groups, assigning related ONH regions to visual field test points. The identified influential structural locations for VF sensitivity prediction at each test point aligned well with existing knowledge and understanding of structure-function spatial relationships. CONCLUSIONS/UNASSIGNED:This study successfully visualized the global trend of point-by-point spatial relationships between OCT-based structure and VF-based function without the need for prior knowledge or segmentation of OCTs. The revealed spatial correlations were consistent with previously published mappings. This presents possibilities of learning from trained machine learning models without applying any prior knowledge, potentially robust, and free from bias. FINANCIAL DISCLOSURES/UNASSIGNED:Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

PURPOSE/UNASSIGNED:Prior evidence suggests racial disparities in the utilization of visual field testing (VFT) for the diagnosis and monitoring of glaucoma. In this study, we considered the effect of baseline glaucoma severity and socioeconomic disadvantage along with other potential confounders such as test reliability, ancillary tests, and glaucoma surgeries on racial disparity in the frequency of VFT. METHODS/UNASSIGNED:The records of all subjects with a diagnosis of glaucoma who received VFT at an academic, tertiary care facility from January 2018 to December 2021 were accessed. Analysis was performed to compare VFT frequency, the total number of office visits (DoS), and the ratio of VFT frequency to DoS (VFT/DoS) across self-reported races while controlling for sex, age, socioeconomic disadvantage (Area Deprivation Index), VF reliability indicators and baseline mean deviation, optical coherence tomography frequency, and glaucoma surgeries. RESULTS/UNASSIGNED:Among the 2654 subjects (1515 White, 782 Black, and 357 Asian) included in this study, Black subjects had the worst socioeconomic status and disease severity at baseline. They also experienced a 3% lower VFT/DoS ratio compared to White subjects (P = 0.031). Asian subjects had a 5% lower VFT/DoS ratio compared to White subjects (P = 0.015). DISCUSSION/UNASSIGNED:We identified racial disparity in performing VFT in subjects with glaucoma even when multiple confounders were considered. Further investigation is necessary to identify other race-associated factors to work toward reducing racial disparities in VFT. TRANSLATIONAL RELEVANCE/UNASSIGNED:Black and Asian subjects with glaucoma receive fewer VFT per visit compared to White subjects even when considering socioeconomic disadvantage and disease severity.

Excitotoxicity from the impairment of glutamate uptake constitutes an important mechanism in neurodegenerative diseases such as Alzheimer's, multiple sclerosis, and Parkinson's disease. Within the eye, excitotoxicity is thought to play a critical role in retinal ganglion cell death in glaucoma, diabetic retinopathy, retinal ischemia, and optic nerve injury, yet how excitotoxic injury impacts different retinal layers is not well understood. Here, we investigated the longitudinal effects of N-methyl-D-aspartate (NMDA)-induced excitotoxic retinal injury in a rat model using deep learning-assisted retinal layer thickness estimation. Before and after unilateral intravitreal NMDA injection in nine adult Long Evans rats, spectral-domain optical coherence tomography (OCT) was used to acquire volumetric retinal images in both eyes over 4 weeks. Ten retinal layers were automatically segmented from the OCT data using our deep learning-based algorithm. Retinal degeneration was evaluated using layer-specific retinal thickness changes at each time point (before, and at 3, 7, and 28 days after NMDA injection). Within the inner retina, our OCT results showed that retinal thinning occurred first in the inner plexiform layer at 3 days after NMDA injection, followed by the inner nuclear layer at 7 days post-injury. In contrast, the retinal nerve fiber layer exhibited an initial thickening 3 days after NMDA injection, followed by normalization and thinning up to 4 weeks post-injury. Our results demonstrated the pathological cascades of NMDA-induced neurotoxicity across different layers of the retina. The early inner plexiform layer thinning suggests early dendritic shrinkage, whereas the initial retinal nerve fiber layer thickening before subsequent normalization and thinning indicates early inflammation before axonal loss and cell death. These findings implicate the inner plexiform layer as an early imaging biomarker of excitotoxic retinal degeneration, whereas caution is warranted when interpreting the ganglion cell complex combining retinal nerve fiber layer, ganglion cell layer, and inner plexiform layer thicknesses in conventional OCT measures. Deep learning-assisted retinal layer segmentation and longitudinal OCT monitoring can help evaluate the different phases of retinal layer damage upon excitotoxicity.

The cellular-level visualization of retinal microstructures such as blood vessel wall components, not available with other imaging modalities, is provided with unprecedented details by dark-field imaging configurations; however, the interpretation of such images alone is sometimes difficult since multiple structural disturbances may be present in the same time. Particularly in eyes with retinal pathology, microstructures may appear in high-resolution retinal images with a wide range of sizes, sharpnesses, and brightnesses. In this paper we show that motion contrast and phase gradient imaging modalities, as well as the simultaneous acquisition of depth-resolved optical coherence tomography (OCT) images, provide additional insight to help understand the retinal neural and vascular structures seen in dark-field images and may enable improved diagnostic and treatment plans.

PURPOSE/UNASSIGNED:Broken stick analysis is a widely used approach for detecting unknown breakpoints where the association between measurements is nonlinear. We propose LIMBARE, an advanced linear mixed-effects breakpoint analysis with robust estimation, especially designed for longitudinal ophthalmic studies. LIMBARE accommodates repeated measurements from both eyes and over time, and it effectively addresses the presence of outliers. METHODS/UNASSIGNED:The model setup of LIMBARE and the computing algorithm for point and confidence interval estimates of the breakpoint were introduced. The performance of LIMBARE and other competing methods was assessed via comprehensive simulation studies and application to a longitudinal ophthalmic study with 216 eyes (145 subjects) followed for an average of 3.7 ± 1.3 years to examine the longitudinal association between structural and functional measurements. RESULTS/UNASSIGNED:In simulation studies, LIMBARE showed the smallest bias and mean squared error for estimating the breakpoint, with an empirical coverage probability of corresponding confidence interval estimates closest to the nominal level for scenarios with and without outlier data points. In the application to the longitudinal ophthalmic study, LIMBARE detected two breakpoints between visual field mean deviation (MD) and retinal nerve fiber layer thickness and one breakpoint between MD and cup-to-disc ratio, whereas the cross-sectional analysis approach detected only one and none, respectively. CONCLUSIONS/UNASSIGNED:LIMBARE enhances breakpoint estimation accuracy in longitudinal ophthalmic studies, and the cross-sectional analysis approach is not recommended for future studies. TRANSLATIONAL RELEVANCE/UNASSIGNED:Our proposed method and companion R package provide a valuable computational tool for advancing longitudinal ophthalmology research and exploring the association relationships among ophthalmic variables.

PURPOSE/UNASSIGNED:stimation, especially designed for longitudinal ophthalmic studies. LIMBARE accommodates repeated measurements from both eyes and overtime, and effectively address the presence of outliers. METHODS/UNASSIGNED:The model setup of LIMBARE and computing algorithm for point and confidence interval estimates of the breakpoint was introduced. The performance of LIMBARE and other competing methods was assessed via comprehensive simulation studies and application to a longitudinal ophthalmic study with 216 eyes (145 subjects) followed for an average of 3.7±1.3 years to examine the longitudinal association between structural and functional measurements. RESULTS/UNASSIGNED:In simulation studies, LIMBARE showed the smallest bias and mean squared error (MSE) for estimating the breakpoint, with empirical coverage probability of corresponding CI estimate closest to the nominal level for scenarios with and without outlier data points. In the application to the longitudinal ophthalmic study, LIMBARE detected two breakpoints between visual field mean deviation (MD) and retinal nerve fiber layer thickness (RNFL) and one breakpoint between MD and cup to disc ratio (CDR), while the cross-sectional analysis approach only detected one and none, respectively. CONCLUSIONS/UNASSIGNED:LIMBARE enhances breakpoint estimation accuracy in longitudinal ophthalmic studies, while cross-sectional analysis approach is not recommended for future studies. TRANSLATIONAL RELEVANCE/UNASSIGNED:Our proposed method and companion software R package provides a valuable computational tool for advancing longitudinal ophthalmology research and exploring the association relationships between ophthalmic variables.

Glaucoma causes progressive visual field deterioration and is the leading cause of blindness worldwide. Glaucomatous damage is irreversible and greatly impacts quality of life. Therefore, it is critically important to detect glaucoma early and closely monitor progression to preserve functional vision. Glaucoma is routinely monitored in the clinical setting using optical coherence tomography (OCT) for derived measures such as the thickness of important visual structures. There is not a consensus of what measures represent the most relevant biomarkers of glaucoma progression. Further, despite the increasing availability of longitudinal OCT data, a quantitative model of 3D structural change over time associated with glaucoma does not exist. In this paper we present an algorithm that will perform hierarchical geodesic modeling at the imaging level, considering 3D OCT images as observations of structural change over time. Hierarchical modeling includes subject-wise trajectories as geodesics in the space of diffeomorphisms and population level (glaucoma vs control) trajectories are also geodesics which explain subject-wise trajectories as deviations from the mean. Our preliminary experiments demonstrate a greater magnitude of structural change associated with glaucoma compared to normal aging. Our algorithm has the potential application in patient-specific monitoring and analysis of glaucoma progression as well as a statistical model of population trends and population variability.

PURPOSE/UNASSIGNED:Race disparities in the healthcare system and the resulting inequality in clinical data among different races hinder the ability to generate equitable prediction results. This study aims to reduce healthcare disparities arising from data imbalance by leveraging advanced transfer learning (TL) methods. METHOD/UNASSIGNED:We examined the ophthalmic healthcare disparities at a population level using electronic medical records data from a study cohort (N = 785) receiving care at an academic institute. Regression-based TL models were usesd, transferring valuable information from the dominant racial group (White) to improve visual field mean deviation (MD) rate of change prediction particularly for data-disadvantaged African American (AA) and Asian racial groups. Prediction results of TL models were compared with two conventional approaches. RESULTS/UNASSIGNED:Disparities in socioeconomic status and baseline disease severity were observed among the AA and Asian racial groups. The TL approach achieved marked to comparable improvement in prediction accuracy compared to the two conventional approaches as evident by smaller mean absolute errors or mean square errors. TL identified distinct key features of visual field MD rate of change for each racial group. CONCLUSIONS/UNASSIGNED:The study introduces a novel application of TL that improved reliability of the analysis in comparison with conventional methods, especially in small sample size groups. This can improve assessment of healthcare disparity and subsequent remedy approach. TRANSLATIONAL RELEVANCE/UNASSIGNED:TL offers an equitable and efficient approach to mitigate healthcare disparities analysis by enhancing prediction performance for data-disadvantaged group.

PURPOSE:The presence of imbalanced datasets in medical applications can negatively affect deep learning methods. This study aims to investigate how the performance of convolutional neural networks (CNNs) for glaucoma diagnosis can be improved by addressing imbalanced learning issues through utilizing glaucoma suspect samples, which are often excluded from studies because they are a mixture of healthy and preperimetric glaucomatous eyes, in a semi-supervised learning approach. METHODS:A baseline 3D CNN was developed and trained on a real-world glaucoma dataset, which is naturally imbalanced (like many other real-world medical datasets). Then, three methods, including reweighting samples, data resampling to form balanced batches, and semi-supervised learning on glaucoma suspect data were applied to practically assess their impacts on the performances of the trained methods. RESULTS:The proposed method achieved a mean accuracy of 95.24%, an F1 score of 97.42%, and an area under the curve of receiver operating characteristic (AUC ROC) of 95.64%, whereas the corresponding results for the traditional supervised training using weighted cross-entropy loss were 92.88%, 96.12%, and 92.72%, respectively. The obtained results show statistically significant improvements in all metrics. CONCLUSIONS:Exploiting glaucoma suspect eyes in a semi-supervised learning method coupled with resampling can improve glaucoma diagnosis performance by mitigating imbalanced learning issues. TRANSLATIONAL RELEVANCE:Clinical imbalanced datasets may negatively affect medical applications of deep learning. Utilizing data with uncertain diagnosis, such as glaucoma suspects, through a combination of semi-supervised learning and class-imbalanced learning strategies can partially address the problems of having limited data and learning on imbalanced datasets.