An Automated Method for Assessing Topographical Structure-Function Agreement in Abnormal Glaucomatous Regions

Retinal Nerve Fiber Layer Optical Texture Analysis and 10-2 Visual Field Assessment in Glaucoma

PURPOSE

To apply retinal nerve fiber layer (RNFL) optical texture analysis (ROTA) to 1) investigate the association between papillomacular and papillofoveal bundle defects with 10-2 visual field (VF) sensitivity abnormalities, and 2) integrate the information from RNFL bundle defect and 24-2 VF central test locations to determine the likelihood of 10-2 VF sensitivity abnormalities.

DESIGN

Cross-sectional.

METHODS

A total of 841 eyes (144 healthy, 317 glaucoma suspect, and 380 glaucoma) of 442 participants were included. Eyes underwent 24-2, and 10-2 VF testing and OCT for ROTA. The borders of RNFL defects were delineated from ROTA, and the involvement of the arcuate, papillomacular, and papillofoveal bundles was determined for each eye. Multilevel logistic regression analysis was applied to evaluate the structure-function association.

RESULTS

Papillomacular (92.1%) and papillofoveal (37.9%) RNFL bundle defects were prevalent in eyes with glaucoma. A 10-2 VF location that was projected onto a papillomacular or a papillofoveal RNFL bundle defect had a significantly increased likelihood of reduced sensitivity (ORs of 18.61 at PDP < 5%, and 20.17 at TDP < 5%, respectively, P < .001 for both). When predicting the likelihood of VF abnormality in a 10-2 test location, noticeably higher odds ratios were observed when overlapping with an RNFL bundle defect, compared to when an abnormal corresponding 24-2 central point was present.

CONCLUSIONS

Papillomacular and papillofoveal RNFL bundle defects are present in a considerable proportion of eyes with glaucoma. When detected, they significantly increase the likelihood of abnormality in the corresponding central VF test locations assessed by the 10-2 test.

Improving glaucoma staging in clinical practice by combining the ICD-10 glaucoma severity classification system and optical coherence tomography

OBJECTIVE

The International Classification of Disease, 10th revision (ICD-10) codes used for glaucoma severity classification are based on the 24-2 visual-field (VF) test. This study aim was to assess the added value of providing clinicians with optical coherence tomography (OCT) data, in addition to functional data, for glaucoma staging in clinical practice.

EXPOSURE

Disease classification was determined for 54 glaucoma eyes, according to the principles of the ICD-10 guidelines. Eyes were independently graded in a masked fashion using the 24-2 VF test and 10-2 VF test, with and without OCT information. The reference standard (RS) for severity was determined using a previously published automated structure-function topographic agreement for glaucomatous damage using all available information.

RESULTS

The RS classified eyes as mild, moderate and advanced in 3, 16 and 35 cases, respectively. Individual and combined 24-2 and 10-2 based gradings were significantly different from the RS (all P < 0.005), with Kappa agreements of 0.26, 0.45 and 0.42 respectively (P < 0.001). Classifications using OCT combined with either of the VF were not-significantly different from the RS (P > 0.3) with Kappa agreements of 0.56 and 0.57 respectively (P < 0.001). Combining 24-2 with OCT had less severity overestimations while 10-2 with OCT had fewer underestimations.

CONCLUSION

Combining OCT and VF data provides better staging of glaucoma severity than VF data alone. The 24-2 and OCT combination seems most appropriate given the high concordance with the RS and less overestimation of severity. Incorporating structural information into disease stages allows clinicians to set more appropriate severity-based treatment targets for individual patients.

Structure-function assessment in glaucoma based on perimetric sensitivity and en face optical coherence tomography images of retinal nerve fiber bundles

Many studies have assessed structure-function relations in glaucoma, but most without topographical comparison across the central 30°. We present a method for assessing structure-function relations with en face images of retinal nerve fiber layer (RNFL) bundles allowing topographical comparison across much of this retinal area. Forty-four patients with glaucoma (median age 61 years) were recruited and tested with Optical Coherence Tomography (OCT) and perimetry. Six rectangular volume scans were gathered, and then montaged to provide en face views of the RNFL bundles. We calculated the proportion of locations showing a perimetric defect that also showed an en face RNFL defect; and the proportion of locations falling on an RNFL defect that also showed a perimetric defect. A perimetric defect for a location was defined as a total deviation (TD) value equal to or deeper than -4 dB. We found that the median (IQR) number of locations with abnormal RNFL bundle reflectance that also had abnormal TD was 78% (60%) and for locations with abnormal TD that also had abnormal RNFL bundle reflectance was 75% (44%). We demonstrated a potential approach for structure-function assessment in glaucoma by presenting a topographic reflectance map, confirming results of previous studies and including larger retinal regions.

Concordance of Objectively Detected Retinal Nerve Fiber Bundle Defects in En Face OCT Images with Conventional Structural and Functional Changes in Glaucoma

PURPOSE

To assess how objectively detected defects in retinal nerve fiber bundle (RNFB) reflectance on en face OCT images relate to circumpapillary retinal nerve fiber layer thickness (cpRNFLT) and visual field defects.

DESIGN

Cross-sectional study.

PARTICIPANTS

Sixteen participants with early glaucoma and 29 age-matched healthy controls, of whom 22 had usable en face images for the establishment of normative levels of RNFB reflectance.

METHODS

All the participants underwent cpRNFLT scans, visual field examination, and wide-field OCT. En face reflectivity was assessed objectively using the Summary of Multiple Anatomically Adjusted Slabs method. En face defects were deemed concordant with cpRNFLT when they had at least 1 cpRNFLT point with P < 0.01, within ± 15° of the predicted insertion on the optic disc. Visual fields were examined using custom suprathreshold perimetry and SITA Standard 24-2. For each visual field location, the corresponding reflectance was deemed abnormal if any en face superpixel within ± 1° was abnormal. The overall, positive, and negative agreements were measured in each participant.

MAIN OUTCOME MEASURES

Proportion of concordant defects between en face reflectance analysis and cpRNFLT (%) as well as overall, positive, and negative agreements between en face reflectance analysis and visual field results.

RESULTS

Most en face abnormalities had concordant cpRNFLT defects in the mapped sector (median proportion concordant, 0.85; interquartile range, 0.74-0.95). In eyes with glaucoma, a median of 8.1% (range, 2.4%-23.7%) and 14.9% (range, 3.5%-29.1%) locations showed corresponding en face and visual field defects using 24-2 and custom perimetry, respectively. Both the perimetric strategies had moderate-to-good raw agreement with en face analysis (0.66-0.68), with stronger agreement on normal findings than on defects (0.77-0.78 and 0.4-0.44).

CONCLUSIONS

Objectively extracted reflectance defects showed strong concordance with conventional cpRNFLT damage and good agreement with perimetry, which could be enhanced by further minimization of image artifacts.

The ICD-10 Glaucoma Severity Score Underestimates the Extent of Glaucomatous Optic Nerve Damage

PURPOSE

To evaluate the International Classification of Disease, Tenth Revision (ICD-10) codes used for glaucoma severity classification, which are based on the location of visual field (VF) defects; given the known poor sensitivity of the 24-2 visual field test to early disease and macular damage, we hypothesized that the ICD-10 codes would not accurately reflect the extent of glaucomatous damage.

DESIGN

Retrospective validity and reliability analysis.

METHODS

We evaluated 80 eyes with glaucomatous optic neuropathy (GON). Masked reviewers assigned an ICD-10 severity grade based on 24-2 VF. Two additional masked examiners determined the presence of optical coherence tomography (OCT) structural damage in each hemifield and/or central 5 degrees to define an OCT-based equivalent ICD-10 classification.

RESULTS

A total of 80 eyes with GON were classified as mild, moderate and advanced in 15, 23, and 42 cases, respectively, based on the 24-2 VF, and in 6, 7, and 67 cases, respectively, based on OCT. The OCT classifications were more severe in 29 of 80 cases (36%). In 33 cases (41.3%), macular damage detected by OCT was missed by the 24-2. In 4 of 80 cases (5%), the VF overestimated the severity, likely due to variability of the 24-2 test.

CONCLUSIONS

The ICD-10 system relies solely on damage seen on the 24-2 and as provides a 24-2 functional score rather than a "glaucoma" severity score. OCT revealed wide variation of damage across grades, with a significant proportion of the eyes showing macular structural damage missed with the 24-2 VF. Adding OCT information to the ICD-10 system would help it to more accurately reflect the extent of glaucomatous damage.

The 24-2 Visual Field Guided Progression Analysis Can Miss the Progression of Glaucomatous Damage of the Macula Seen Using OCT

PURPOSE

To better understand the efficacy of the 24-2 guided progression analysis (GPA) in the detection of progression in eyes with early glaucoma (i.e., 24-2 mean deviation [MD] better than -6 dB) by comparing 24-2 GPA with a reference standard (RS) based on a combination of OCT and 24-2 and 10-2 visual field (VF) information.

DESIGN

Cross-sectional study.

PARTICIPANTS

Ninety-nine eyes from 99 individuals, including 70 suspected or early glaucomatous eyes (24-2 MD better than -6 dB) and 29 healthy controls (HCs).

METHODS

All the eyes had at least 4 OCT and VF test dates over a period that ranged from 12 to 59 months. The 24-2 VF tests included 2 baseline tests and at least 2 follow-up tests. The 2 baseline tests were performed within an average of 5.6 days (median, 7 days), and the last follow-up test was performed at least 1 year after the first baseline visit.

MAIN OUTCOME MEASURES

A commercial 24-2 GPA software, with default settings, characterized the eyes as having "likely progression" (LP) or "possible progression" (PP); both were considered "progressing" for this analysis. For RS, 3 authors graded progression using strict criteria and a combination of a custom OCT progression report and commercial 24-2 and 10-2 GPA reports for the same test dates as GPA.

RESULTS

The reference standard identified 10 (14%) of the 70 patient eyes and none of the HC eyes as having progression. The 24-2 guided progression analysis identified 13 of the 70 patient eyes as having progression (PP or LP). However, it correctly classified only 4 (40%) of the 10 RS progressors. All 6 of the RS progressors missed by the 24-2 GPA showed progression in the macula. In addition, the 24-2 GPA identified 2 of the 29 HC eyes as progressors and 9 patient eyes without progression based on the RS.

CONCLUSIONS

In eyes with early glaucoma (i.e., 24-2 MD, > -6 dB) in this study, the 24-2 GPA missed progression seen using OCT and exhibited a relatively high rate of false positives. Furthermore, the region progressing typically included the macula. The results suggest that including OCT and/or 10-2 VFs should improve the detection of progression.

Test of a Retinal Nerve Fiber Bundle Trajectory Model Using Eyes With Glaucomatous Optic Neuropathy

Purpose

To test a model of retinal nerve fiber bundle trajectories that predicts the arcuate-shaped patterns seen on optical coherence tomography (OCT) retinal nerve fiber layer (RNFL) probability/deviation maps (p-maps) in glaucomatous eyes.

Methods

Thirty-one glaucomatous eyes from a database of 250 eyes had clear arcuate-shaped patterns on RNFL p-maps derived from an OCT cube scan. The borders of the arcuate patterns were extracted from the RNFL p-maps. Next, the trajectories from an arcuate model were compared against these borders via a normalized root-mean-square difference analysis. The model's parameter, β, was varied, and the best-fitting, initial clock-hour position of the trajectory to the border was found for each β. Finally, the regions, as determined by the arcuate border's best-fit, initial clock-hour positions, were compared against the abnormal regions on the circumpapillary retinal nerve fiber layer (cpRNFL) profile.

Results

The arcuate model's mean β_Sup and β_Inf parameters minimized large differences between the trajectories and the arcuate borders on the RNFL p-maps. Furthermore, on average, 68% of the cpRNFL regions defined by the arcuate border's best-fit, initial clock-hour positions were abnormal (i.e., below the ≤5% threshold).

Conclusions

The arcuate model performed well in predicting the borders of arcuate patterns seen on RNFL p-maps. It also predicted the associated abnormal regions of the cpRNFL thickness plots.

Translational Relevance

This model should prove useful in helping clinicians understand topographical comparisons among different OCT representations and should improve structure-structure, as well as structure-function agreement analyses.

Distinguishing Healthy From Glaucomatous Eyes With Optical Coherence Tomography Global Circumpapillary Retinal Nerve Fiber Thickness in the Bottom 5th Percentile

PRCIS

Two novel, quantitative metrics, and 1 traditional metric were able to distinguish between many, but not all healthy and glaucomatous eyes in the bottom 5th percentile of global circumpapillary retinal nerve fiber layer (cpRNFL) thickness.

PURPOSE

To test the hypothesis that objective optical coherence tomography measures can distinguish between a healthy control with global cpRNFL thickness within the lower 5% of normal and a glaucoma patient with an equivalent cpRNFL thickness.

PATIENTS AND METHODS

A total of 37 healthy eyes from over 700 normative eyes fell within the bottom 5th percentile in global cpRNFL thickness. The global cpRNFL thickness of 35 glaucomatous eyes from 188 patients fell within the same range. For the traditional methods, the global cpRNFL thickness percentile and the global ganglion cell layer (GCL) thickness percentile for the central ±8 degrees, were calculated for all 72 eyes. For the novel cpRNFL method, the normalized root mean square (RMS) difference between the cpRNFL thickness profile and the global thickness-matched normative thickness profile was calculated. For the superior-inferior (SI) GCL method, the normalized mean difference in superior and inferior GCL thickness was calculated for the central ±8 degrees.

RESULTS

The best quantitative metric, the RMS cpRNFL method, had an accuracy of 90% compared with 81% for the SI GCL and 81% for the global GCL methods. As expected, the global cpRNFL had the worst accuracy, 72%. Similarly, the RMS cpRNFL method had an area under the curve of 0.93 compared with 0.83 and 0.84 for the SI GCL and global GCL methods, respectively. The global cpRNFL method had the worst area under the curve, 0.75.

CONCLUSION

Quantitative metrics can distinguish between most of the healthy and glaucomatous eyes with low global cpRNFL thickness. However, even the most successful metric, RMS cpRNFL, missed some glaucomatous eyes.

Rationale and Development of an OCT-Based Method for Detection of Glaucomatous Optic Neuropathy

A specific, sensitive, and intersubjectively verifiable definition of disease for clinical care and research remains an important unmet need in the field of glaucoma. Using an iterative, consensus-building approach and employing pilot data, an optical coherence tomography (OCT)-based method to aid in the detection of glaucomatous optic neuropathy was sought to address this challenge. To maximize the chance of success, we utilized all available information from the OCT circle and cube scans, applied both quantitative and semiquantitative data analysis methods, and aimed to limit the use of perimetry to cases where it is absolutely necessary. The outcome of this approach was an OCT-based method for the diagnosis of glaucomatous optic neuropathy that did not require the use of perimetry for initial diagnosis. A decision tree was devised for testing and implementation in clinical practice and research that can be used by reading centers, researchers, and clinicians. While initial pilot data were encouraging, future testing and validation will be needed to establish its utility in clinical practice, as well as for research.

Detection of Early Glaucomatous Damage: Performance of Summary Statistics From Optical Coherence Tomography and Perimetry

Purpose

To evaluate the diagnostic performance of optical coherence tomography (OCT) and visual field (VF) summary statistics (metrics) that are available in OCT and VF reports.

Methods

OCT disc and macular scans and 24-2 and 10-2 VFs were obtained from 56 healthy control (HC) eyes/participants and 61 eyes/patients with 24-2 mean deviation of better than –6 dB. All metrics were obtained from OCT radial, circle, and posterior pole cube scans and 24-2 and 10-2 VFs. Their diagnostic performances were evaluated, in isolation and in combinations. For specificity, the 56 HC eyes were used. For sensitivity, 40 of the 61 patient eyes were deemed likely glaucomatous based on an automated topographic method that evaluates structure–function (S–F) agreement. Any 1 of these 40 eyes not judged as abnormal by any given metric was considered a false negative.

Results

All single OCT and VF metrics misclassified HCs as glaucomatous and missed likely glaucomatous eyes. The best performing single metric was the temporal inferior thickness of the 3.5-mm circle scan, with 96% specificity and 83% sensitivity. Combinations of OCT–OCT and OCT–VF metrics markedly improved specificity. A newly proposed metric that evaluates structure–structure (S–S) agreement at a hemifield level had the highest accuracy. This S–S metric had 98% specificity and 80% sensitivity.

Conclusions

OCT and VF metrics, single or in combinations, have only moderate sensitivity for eyes with early glaucoma.

Translational Relevance

OCT and VF metrics combinations evaluating S–S or S–F agreement can be highly specific, which is an important implication for clinical and research purposes.

The OCT RNFL Probability Map and Artifacts Resembling Glaucomatous Damage

Purpose

The purpose of this study was to improve the diagnostic ability of the optical coherence tomography (OCT) retinal nerve fiber layer (RNFL) probability (p-) map by understanding the frequency and pattern of artifacts seen on the p-maps of healthy control (HC) eyes resembling glaucomatous damage.

Methods

RNFL p-maps were generated from wide-field OCT cube scans of 2 groups of HC eyes, 200 from a commercial normative group (HC-norm) and 54 from a prospective study group, as well as from 62 patient eyes, which included 32 with early glaucoma (EG). These 32 EG eyes had 24-2 mean deviation (MD) better than -6 dB and perimetric glaucoma as defined by 24-2 and 10-2 criteria. For the HC groups, "glaucoma-like" arcuates were defined as any red region near the temporal half of the disc.

Results

Seven percent of the 200 HC-norm and 11% of the 54 HC RNFL p-maps satisfied the definition of "glaucoma-like," as did all the patients' p-maps. The HC p-maps showed two general patterns of abnormal regions, "arcuate" and "temporal quadrant," and these patterns resembled those seen on some of the RNFL p-maps of the EG eyes. A "vertical midline" rule, which required the abnormal region to cross the vertical midline through the fovea, had a specificity of >99%, and a sensitivity of 75% for EG and 93% for moderate to advanced eyes.

Conclusions

Glaucoma-like artifacts on RNFL p-maps are relatively common and can masquerade as arcuate and/or widespread/temporal damage.

Translational Relevance

A vertical midline rule had excellent specificity. However, other OCT information is necessary to obtain high sensitivity, especially in eyes with early glaucoma.

Detecting glaucoma with only OCT: Implications for the clinic, research, screening, and AI development

A method for detecting glaucoma based only on optical coherence tomography (OCT) is of potential value for routine clinical decisions, for inclusion criteria for research studies and trials, for large-scale clinical screening, as well as for the development of artificial intelligence (AI) decision models. Recent work suggests that the OCT probability (p-) maps, also known as deviation maps, can play a key role in an OCT-based method. However, artifacts seen on the p-maps of healthy control eyes can resemble patterns of damage due to glaucoma. We document in section 2 that these glaucoma-like artifacts are relatively common and are probably due to normal anatomical variations in healthy eyes. We also introduce a simple anatomical artifact model based upon known anatomical variations to help distinguish these artifacts from actual glaucomatous damage. In section 3, we apply this model to an OCT-based method for detecting glaucoma that starts with an examination of the retinal nerve fiber layer (RNFL) p-map. While this method requires a judgment by the clinician, sections 4 and 5 describe automated methods that do not. In section 4, the simple model helps explain the relatively poor performance of commonly employed summary statistics, including circumpapillary RNFL thickness. In section 5, the model helps account for the success of an AI deep learning model, which in turn validates our focus on the RNFL p-map. Finally, in section 6 we consider the implications of OCT-based methods for the clinic, research, screening, and the development of AI models.

A Case for The Use of Artificial Intelligence in Glaucoma Assessment

We hypothesize that artificial intelligence applied to relevant clinical testing in glaucoma has the potential to enhance the ability to detect glaucoma. This premise was discussed at the recent Collaborative Community for Ophthalmic Imaging meeting, "The Future of Artificial Intelligence-Enabled Ophthalmic Image Interpretation: Accelerating Innovation and Implementation Pathways," held virtually September 3-4, 2020. The Collaborative Community in Ophthalmic Imaging (CCOI) is an independent self-governing consortium of stakeholders with broad international representation from academic institutions, government agencies, and the private sector whose mission is to act as a forum for the purpose of helping speed innovation in healthcare technology. It was one of the first two such organizations officially designated by the FDA in September 2019 in response to their announcement of the collaborative community program as a strategic priority for 2018-2020. Further information on the CCOI can be found online at their website (https://www.cc-oi.org/about). Artificial intelligence for glaucoma diagnosis would have high utility globally, as access to care is limited in many parts of the world and half of all people with glaucoma are unaware of their illness. The application of artificial intelligence technology to glaucoma diagnosis has the potential to broadly increase access to care worldwide, in essence flattening the Earth by providing expert level evaluation to individuals even in the most remote regions of the planet.

Did the OCT Show Progression Since the Last Visit?

Identifying progression is of fundamental importance to the management of glaucoma. It is also a challenge. The most sophisticated, and probably the most useful, commercially available clinical tool for identifying progression is the Guided Progression Analysis (GPA), which was initially developed to identify progression using 24-2 visual field tests. More recently, it has been extended to retinal nerve fiber layer (RNFL) and ganglion cell+inner plexiform layer thicknesses measured with optical coherence tomography (OCT). However, the OCT GPA requires a minimum of 3 tests to determine "possible loss (progression)" and a minimum of 4 tests to determine if the patient shows "likely loss (progression)." Thus, it is not designed to answer a fundamental question asked by both the clinician and the patient, namely: Did damage progress since the last visit? Some clinicians use changes in summary statistics, such as global/average circumpapillary RNFL thickness. However, these statistics have poor sensitivity and specificity due to segmentation and alignment errors. Instead of relying on the GPA analysis or summary statistics, one needs to evaluate RNFL and ganglion cell+inner plexiform layer probability maps and circumpapillary OCT B-scan images. In addition, we argue that the clinician can make a better decision about suspected progression between 2 test days by topographically comparing the changes in the different OCT maps and images, in addition to topographically comparing the changes in the visual field with the changes in OCT probability maps.

Strategies to Improve Convolutional Neural Network Generalizability and Reference Standards for Glaucoma Detection From OCT Scans

Purpose

To develop and evaluate methods to improve the generalizability of convolutional neural networks (CNNs) trained to detect glaucoma from optical coherence tomography retinal nerve fiber layer probability maps, as well as optical coherence tomography circumpapillary disc (circle) b-scans, and to explore impact of reference standard (RS) on CNN accuracy.

Methods

CNNs previously optimized for glaucoma detection from retinal nerve fiber layer probability maps, and newly developed CNNs adapted for glaucoma detection from optical coherence tomography b-scans, were evaluated on an unseen dataset (i.e., data collected at a different site). Multiple techniques were used to enhance CNN generalizability, including augmenting the training dataset, using multimodal input, and training with confidently rated images. Model performance was evaluated with different RS.

Results

Training with data augmentation and training on confident images enhanced the accuracy of the CNNs for glaucoma detection on a new dataset by 5% to 9%. CNN performance was optimal when a similar RS was used to establish labels both for the training and the testing sets. However, interestingly, the CNNs described here were robust to variation in the RS.

Conclusions

CNN generalizability can be improved with data augmentation, multiple input image modalities, and training on images with confident ratings. CNNs trained and tested with the same RS achieved best accuracy, suggesting that choosing a thorough and consistent RS for training and testing improves generalization to new datasets.

Translational Relevance

Strategies for enhancing CNN generalizability and for choosing optimal RS should be standard practice for CNNs before their deployment for glaucoma detection.

An Evaluation of a New 24-2 Metric for Detecting Early Central Glaucomatous Damage

PURPOSE

We sought to test the hypothesis that a recently proposed pattern standard deviation (PSD) metric, based upon the 24-2 visual field (VF) test, as well as the PSD of the 10-2 VF, will miss central glaucomatous damage confirmed with an objective structure-function method.

DESIGN

Cross-sectional study.

METHODS

A glaucoma (G) group (70 eyes/patients) diagnosed with glaucoma and a 24-2 mean deviation better than -6 dB and a healthy (H) group (45 eyes/patients) had 24-2 and 10-2 VFs and optical coherence tomography (OCT) scans twice within 4 weeks. The PSD(C24-2), based upon the central 12 points of the 24-2, was compared with the PSD(10-2). To evaluate central damage (CD) in G eyes with normal PSD(C24-2) values, a post hoc analysis was combined with a CD reference standard (CD-RS), which was based upon an objective, topographic comparison between abnormal points on the 10-2 VF and OCT probability maps.

RESULTS

The 115 PSD(C24-2) and PSD(10-2) values were significantly correlated (Spearman correclation coefficient: rho = 0.55; P < .001) and the number of G eyes (19) identified as abnormal by the PSD(C24-2) was not significantly different from the number (22) identified by the PSD(10-2) (P = .15). However, based upon the CD-RS, 44 of 70 G eyes were classified as abnormal. The PSD(C24-2) missed 27 (61%) of these 44 eyes, and the PSD(10-2) missed 23 (52%) of these eyes. Post hoc analysis revealed clear CD in most of these eyes.

CONCLUSION

Neither the PSD(C24-2) nor the PSD(10-2) metric is good measure of early CD. Instead we recommend a topographic approach based upon OCT probability maps and a 10-2 VF.

Local Glaucomatous Defects of the Circumpapillary Retinal Nerve Fiber Layer Show a Variety of Patterns of Progression

PRECIS

The region of glaucomatous progression, seen on optical coherence tomography (OCT) images of the circumpapillary retinal nerve fiber layer (cRNFL), increases in width and depth in all eyes, but shows a variety of different patterns of loss across eyes.

PURPOSE

The purpose of this study was to examine the patterns of cRNFL loss secondary to glaucomatous progression in a region associated with the superior hemifield of the 24-2/30-2 visual field (VF).

METHODS

Twenty-four eyes (20 patients) with a diagnosis of glaucoma and evidence of progression on OCT had OCT disc cube scans on at least 3 separate visits (mean follow-up 7.4 y; range: 3.9 to 11.4). Circumpapillary b-scans were derived after enface images were aligned to assure that the study region (ie, 0 to -135 degrees, where 0 degree is 9 o'clock, on a right eye) coincided. Within this region, a region of progression (ROP) was defined based on the loss in cRNFL thickness between the first and subsequent visits. The width of the ROP was determined, along with the locations of its leading (close to fixation) and trailing edges. In addition, for each ROP, the location and depth at the point of maximal loss, total loss, and average remaining retinal nerve fiber layer were measured.

RESULTS

The ROP proceeded both toward and away from fixation. Across eyes, the ROP varied widely in width (32 to 131 degrees, mean 82.7 degrees), location, and loss at point of deepest loss (22 to 99 μm, mean 52.9 μm), as well as total cRNFL loss.

CONCLUSIONS

All eyes showed a widening and deepening of the ROP, but a variety of different patterns of progressive cRNFL loss. Thus, one should expect considerable variation in patterns of VF loss. Furthermore, conventional metrics (global or quadrant cRNFL thickness) do not fully depict the progressive changes that can be appreciated by inspecting OCT images.

Global optical coherence tomography measures for detecting the progression of glaucoma have fundamental flaws

Objective

To understand the problems involved in using global OCT measures for detecting progression in early glaucoma.

Subjects/Methods

Eyes from 76 patients and 28 healthy controls (HC) had a least two OCT scans at least 1 year apart. To determine the 95% confidence intervals (CI), 151 eyes (49 HC and 102 patients) had at least two scans within 6 months. All eyes had 24-2 mean deviation ≥-6dB. The average (global) thicknesses of the circumpapillary retinal nerve fibre layer (cRNFL), G_ONH, and of the retinal ganglion cell layer plus inner plexiform layer (RGCLP), G_mac, were calculated. Using quantile regression, the 95% CI intervals were determined. Eyes outside the CIs were classified as “progressors.” For a reference standard (RS), four experts evaluated OCT and VF information.

Results

Compared to the RS, 31 of the 76 (40.8%) patient eyes were identified as progressors (RS-P), and 45 patient, and all 28 HC, eyes as nonprogressors (RS-NP). The metrics missed (false negative, FN) 15 (48%) (G_ONH) and 9 (29%) (G_mac) of the 31 RS-P. Further, G_ONH and/or G_mac falsely identified (false positive, FP) 10 (22.2%) of 45 patient RS-NP eyes and 7 (25%) of the 28 HC eyes as progressing. Post-hoc analysis identified three reasons (segmentation, centring, and local damage) for these errors.

Conclusions

Global metrics lead to FPs and FNs because of problems inherent in OCT scanning (segmentation and centring), and to FNs because they can miss local damage. These problems are difficult, if not impossible, to correct, and raise concerns about the advisability of using G_ONH and G_mac for detecting progression.

Robust and Interpretable Convolutional Neural Networks to Detect Glaucoma in Optical Coherence Tomography Images

Recent studies suggest that deep learning systems can now achieve performance on par with medical experts in diagnosis of disease. A prime example is in the field of ophthalmology, where convolutional neural networks (CNNs) have been used to detect retinal and ocular diseases. However, this type of artificial intelligence (AI) has yet to be adopted clinically due to questions regarding robustness of the algorithms to datasets collected at new clinical sites and a lack of explainability of AI-based predictions, especially relative to those of human expert counterparts. In this work, we develop CNN architectures that demonstrate robust detection of glaucoma in optical coherence tomography (OCT) images and test with concept activation vectors (TCAVs) to infer what image concepts CNNs use to generate predictions. Furthermore, we compare TCAV results to eye fixations of clinicians, to identify common decision-making features used by both AI and human experts. We find that employing fine-tuned transfer learning and CNN ensemble learning create end-to-end deep learning models with superior robustness compared to previously reported hybrid deep-learning/machine-learning models, and TCAV/eye-fixation comparison suggests the importance of three OCT report sub-images that are consistent with areas of interest fixated upon by OCT experts to detect glaucoma. The pipeline described here for evaluating CNN robustness and validating interpretable image concepts used by CNNs with eye movements of experts has the potential to help standardize the acceptance of new AI tools for use in the clinic.

Improving the Detection of Glaucoma and Its Progression: A Topographical Approach

Glaucoma is typically defined as a progressive optic neuropathy characterized by a specific (arcuate) pattern of visual field (VF) and anatomic changes. Therefore, we should be comparing arcuate patterns of damage seen on VFs with those seen on optical coherence tomography (OCT) maps. Instead, clinicians often use summary metrics such as VF pattern standard deviation, OCT retinal nerve fiber (RNF) global thickness, etc. There are 2 major impediments to topographically comparing patterns of damage on VF and OCT maps. First, until recently, it was not easy to make these comparisons with commercial reports. While recent reports do make it easier to compare VF and OCT maps, they have shortcomings. In particular, the 24-2 VF covers a larger retinal region than the commercial OCT scans, and, further, it is not easy to understand the topographical relationship among the different maps/plots within the current OCT reports. Here we show how a model of RNF bundles can overcome these problems. The second major impediment is the lack of a quantitative, and automated, method for comparing patterns of damage seen on VF and OCT maps. However, it is now possible to objectively and automatically quantify this agreement. Together, the RNF bundle model and the automated structure-function method should improve the power of topographical methods for detecting glaucoma and its progression. This should prove useful in clinical studies and trials, as well as for training and validating artificial intelligence/deep learning approaches for these purposes.

Glaucomatous Maculopathy: Thickness Differences on Inner and Outer Macular Layers between Ocular Hypertension and Early Primary Open-Angle Glaucoma Using 8 × 8 Posterior Pole Algorithm of SD-OCT

The purpose of this study was to compare the thickness of all inner and outer macular layers between ocular hypertension (OHT) and early primary open-angle glaucoma (POAG) using spectral domain optical coherence tomography (SD-OCT) 8 × 8 posterior pole algorithm (8 × 8 PPA). Fifty-seven eyes of 57 OHT individuals and fifty-seven eyes of 57 early POAG patients were included. The thickness of macular retinal nerve fiber layer (mRNFL), ganglion cell layer (GCL), inner plexiform layer (IPL), inner nuclear layer (INL), outer plexiform and nuclear layer, photoreceptor layer (PRL) and retinal pigment epithelium were obtained in 64 cells for each macular layer and mean thickness of superior and inferior hemispheres was also calculated. Thinning of superior and inferior hemisphere mean thickness in mRNFL, GCL and IPL and thickening of superior and inferior hemisphere mean thickness in PRL and inferior hemisphere in INL were found in early GPAA group. Otherwise, heatmaps representing cell-to-cell comparisons showed thinning patterns in inner retinal layers (except for INL) and thickening patterns in outer retinal layers in GPAA group. We found that 8 × 8 PPA not only allows the detection of significant thinning patterns in inner retinal layers, but also thickening patterns in outer retinal layers when comparing early POAG eyes to OHT eyes.