Predicting eyes at risk for rapid glaucoma progression based on an initial visual field test using machine learning

Opportunities for Improving Glaucoma Clinical Trials via Deep Learning-Based Identification of Patients with Low Visual Field Variability

PURPOSE

Develop and evaluate the performance of a deep learning model (DLM) that forecasts eyes with low future visual field (VF) variability, and study the impact of using this DLM on sample size requirements for neuroprotective trials.

DESIGN

Retrospective cohort and simulation study.

METHODS

We included 1 eye per patient with baseline reliable VFs, OCT, clinical measures (demographics, intraocular pressure, and visual acuity), and 5 subsequent reliable VFs to forecast VF variability using DLMs and perform sample size estimates. We estimated sample size for 3 groups of eyes: all eyes (AE), low variability eyes (LVE: the subset of AE with a standard deviation of mean deviation [MD] slope residuals in the bottom 25th percentile), and DLM-predicted low variability eyes (DLPE: the subset of AE predicted to be low variability by the DLM). Deep learning models using only baseline VF/OCT/clinical data as input (DLM1), or also using a second VF (DLM2) were constructed to predict low VF variability (DLPE1 and DLPE2, respectively). Data were split 60/10/30 into train/val/test. Clinical trial simulations were performed only on the test set. We estimated the sample size necessary to detect treatment effects of 20% to 50% in MD slope with 80% power. Power was defined as the percentage of simulated clinical trials where the MD slope was significantly worse from the control. Clinical trials were simulated with visits every 3 months with a total of 10 visits.

RESULTS

A total of 2817 eyes were included in the analysis. Deep learning models 1 and 2 achieved an area under the receiver operating characteristic curve of 0.73 (95% confidence interval [CI]: 0.68, 0.76) and 0.82 (95% CI: 0.78, 0.85) in forecasting low VF variability. When compared with including AE, using DLPE1 and DLPE2 reduced sample size to achieve 80% power by 30% and 38% for 30% treatment effect, and 31% and 38% for 50% treatment effect.

CONCLUSIONS

Deep learning models can forecast eyes with low VF variability using data from a single baseline clinical visit. This can reduce sample size requirements, and potentially reduce the burden of future glaucoma clinical trials.

FINANCIAL DISCLOSURE(S)

Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Retinal Nerve Fiber Layer Damage Assessment in Glaucomatous Eyes Using Retinal Retardance Measured by Polarization-Sensitive Optical Coherence Tomography

Purpose

To assess the diagnostic performance and structure-function association of retinal retardance (RR), a customized metric measured by a prototype polarization-sensitive optical coherence tomography (PS-OCT), across various stages of glaucoma.

Methods

This cross-sectional pilot study analyzed 170 eyes from 49 healthy individuals and 68 patients with glaucoma. The patients underwent PS-OCT imaging and conventional spectral-domain optical coherence tomography (SD-OCT), as well as visual field (VF) tests. Parameters including RR and retinal nerve fiber layer thickness (RNFLT) were extracted from identical circumpapillary regions of the fundus. Glaucomatous eyes were categorized into early, moderate, or severe stages based on VF mean deviation (MD). The diagnostic performance of RR and RNFLT in discriminating glaucoma from controls was assessed using receiver operating characteristic (ROC) curves. Correlations among VF-MD, RR, and RNFLT were evaluated and compared within different groups of disease severity.

Results

The diagnostic performance of both RR and RNFLT was comparable for glaucoma detection (RR AUC = 0.98, RNFLT AUC = 0.97; P = 0.553). RR showed better structure-function association with VF-MD than RNFLT (RR VF-MD = 0.68, RNFLT VF-MD = 0.58; z = 1.99; P = 0.047) in glaucoma cases, especially in severe glaucoma, where the correlation between VF-MD and RR (r = 0.73) was significantly stronger than with RNFLT (r = 0.43, z = 1.96, P = 0.050). In eyes with early and moderate glaucoma, the structure-function association was similar when using RNFLT and RR.

Conclusions

RR and RNFLT have similar performance in glaucoma diagnosis. However, in patients with glaucoma especially severe glaucoma, RR showed a stronger correlation with VF test results. Further research is needed to validate RR as an indicator for severe glaucoma evaluation and to explore the benefits of using PS-OCT in clinical practice.

Translational Relevance

We demonstrated that PS-OCT has the potential to evaluate the status of RNFL structural damage in eyes with severe glaucoma, which is currently challenging in clinics.

The Impact of Social Vulnerability on Structural and Functional Glaucoma Severity, Worsening, and Variability

PURPOSE

To determine the association between social vulnerability index (SVI) and baseline severity, worsening, and variability of glaucoma, as assessed by visual field (VF) and optical coherence tomography (OCT).

DESIGN

Retrospective longitudinal cohort study.

PARTICIPANTS

Adults with glaucoma or glaucoma suspect status in one or both eyes. VFs were derived from 7,897 eyes from 4,482 patients, while OCTs were derived from 6,271 eyes from 3,976 patients. All eyes had a minimum of 5 tests over follow-up using either the Humphrey Field Analyzer or the Cirrus HD-OCT.

METHODS

SVI, which measures neighborhood-level environmental factors, was linked to patient addresses at the census tract level. Rates of change in mean deviation (MD) and retinal nerve fiber layer (RNFL) thickness were computed using linear regression. The slope of the regression line was used to assess worsening, while the standard deviation (SD) of residuals was used as a measure of variability. Multivariable linear mixed-effects models were used to investigate the impact of SVI on baseline, worsening, and variability in both MD and RNFL. We further explored the interaction effect of mean IOP and SVI on worsening in MD and RNFL.

MAIN OUTCOME MEASURES

Glaucoma severity defined based on baseline MD and RNFL thickness. Worsening defined as MD and RNFL slope. Variability defined as the SD of the residuals obtained from MD and RNFL slopes.

RESULTS

Increased (worse) SVI was significantly associated with worse baseline MD (β = -1.07 dB, 95% CI: [-1.54, -0.60]), thicker baseline RNFL (β = 2.46 μm, 95% CI: [0.75, 4.17]), greater rates of RNFL loss (β = -0.12 μm, 95% CI: [-0.23, -0.02]), and greater VF variability (β = 0.16 dB, 95% CI: [0.07, 0.24]). Having worse SVI was associated with worse RNFL loss with increases in IOP (βinteraction = -0.07, [95% CI: -0.12, -0.02]).

CONCLUSIONS

Increased social vulnerability index score is associated with worse functional (VF) loss at baseline, higher rates of structural (OCT) worsening over time, higher VF variability, and a greater effect of IOP on RNFL loss. Further studies are needed to enhance our understanding of these relationships and establish their cause.

In vivo identification of angle dysgenesis and its relation to genetic markers associated with glaucoma using artificial intelligence

PURPOSE

To predict the presence of angle dysgenesis on anterior-segment optical coherence tomography (ADoA) by using deep learning (DL) and to correlate ADoA with mutations in known glaucoma genes.

PARTICIPANTS

In total, 800 high-definition anterior-segment optical coherence tomography (AS-OCT) images were included, of which 340 images were used to build the machine learning (ML) model. Images used to build the ML model included 170 scans of primary congenital glaucoma (16 patients), juvenile-onset open-angle glaucoma (62 patients), and adult-onset primary open-angle glaucoma eyes (37 patients); the rest were controls (n = 85). The genetic validation dataset consisted of another 393 images of patients with known mutations that were compared with 320 images of healthy controls.

METHODS

ADoA was defined as the absence of Schlemm's canal, the presence of hyperreflectivity over the region of the trabecular meshwork, or a hyperreflective membrane. DL was used to classify a given AS-OCT image as either having angle dysgenesis or not. ADoA was then specifically looked for on AS-OCT images of patients with mutations in the known genes for glaucoma.

RESULTS

The final prediction, which was a consensus-based outcome from the three optimized DL models, had an accuracy of >95%, a specificity of >97%, and a sensitivity of >96% in detecting ADoA in the internal test dataset. Among the patients with known gene mutations, ( MYOC, CYP1B1, FOXC1, and LTBP2 ) ADoA was observed among all the patients in the majority of the images, compared to only 5% of the healthy controls.

CONCLUSION

ADoA can be objectively identified using models built with DL.

Deep learning-based identification of eyes at risk for glaucoma surgery

To develop and evaluate the performance of a deep learning model (DLM) that predicts eyes at high risk of surgical intervention for uncontrolled glaucoma based on multimodal data from an initial ophthalmology visit. Longitudinal, observational, retrospective study. 4898 unique eyes from 4038 adult glaucoma or glaucoma-suspect patients who underwent surgery for uncontrolled glaucoma (trabeculectomy, tube shunt, xen, or diode surgery) between 2013 and 2021, or did not undergo glaucoma surgery but had 3 or more ophthalmology visits. We constructed a DLM to predict the occurrence of glaucoma surgery within various time horizons from a baseline visit. Model inputs included spatially oriented visual field (VF) and optical coherence tomography (OCT) data as well as clinical and demographic features. Separate DLMs with the same architecture were trained to predict the occurrence of surgery within 3 months, within 3-6 months, within 6 months-1 year, within 1-2 years, within 2-3 years, within 3-4 years, and within 4-5 years from the baseline visit. Included eyes were randomly split into 60%, 20%, and 20% for training, validation, and testing. DLM performance was measured using area under the receiver operating characteristic curve (AUC) and precision-recall curve (PRC). Shapley additive explanations (SHAP) were utilized to assess the importance of different features. Model prediction of surgery for uncontrolled glaucoma within 3 months had the best AUC of 0.92 (95% CI 0.88, 0.96). DLMs achieved clinically useful AUC values (> 0.8) for all models that predicted the occurrence of surgery within 3 years. According to SHAP analysis, all 7 models placed intraocular pressure (IOP) within the five most important features in predicting the occurrence of glaucoma surgery. Mean deviation (MD) and average retinal nerve fiber layer (RNFL) thickness were listed among the top 5 most important features by 6 of the 7 models. DLMs can successfully identify eyes requiring surgery for uncontrolled glaucoma within specific time horizons. Predictive performance decreases as the time horizon for forecasting surgery increases. Implementing prediction models in a clinical setting may help identify patients that should be referred to a glaucoma specialist for surgical evaluation.

Identifying central 10° visual subfield associated with future worsening of visual acuity in eyes with advanced glaucoma

BACKGROUND/AIMS

To determine a cluster of test points: visual subfield (VSF) of Humphrey Field Analyzer 10-2 test (HFA 10-2) of which baseline sensitivities were associated with future worsening of visual acuity (VA) in eyes with advanced glaucoma.

METHODS

A total of 175 advanced glaucoma eyes of 175 advanced glaucoma patients with well controlled intraocular pressure (IOP), a mean deviation of the Humphrey Field Analyzer 24-2 (HFA 24-2) test ≤ -20 decibels and best corrected VA ≥20/40, were included. At baseline, HFA 24-2 and HFA 10-2 tests were performed along with VA measurements. All patients underwent prospective follow-up of 5 years, and VA was measured every 6 months. The Cox proportional hazards model was used to identify visual field sensitivities associated with deterioration of VA and also blindness.

RESULTS

Deterioration of VA and blindness were observed in 15.4% and 3.4% of the eyes, respectively. More negative total deviation (TD) values in the temporal papillomacular bundle VSF were significantly associated with deterioration in VA. Averages of the TD values in this area of the HFA 10-2 test had the most predictive power of future VA deterioration (OR: 0.92, p<0.001). A very similar tendency was observed for blindness.

CONCLUSION

In advanced glaucoma eyes with well-controlled IOP, careful attention is needed when the mean TD values in the temporal papillomacular bundle VSF, measured with a HFA 10-2 test is deteriorated. TD values of this VSF indicate higher risks for future deterioration of VA and also blindness.

Artificial intelligence in glaucoma: opportunities, challenges, and future directions

Artificial intelligence (AI) has shown excellent diagnostic performance in detecting various complex problems related to many areas of healthcare including ophthalmology. AI diagnostic systems developed from fundus images have become state-of-the-art tools in diagnosing retinal conditions and glaucoma as well as other ocular diseases. However, designing and implementing AI models using large imaging data is challenging. In this study, we review different machine learning (ML) and deep learning (DL) techniques applied to multiple modalities of retinal data, such as fundus images and visual fields for glaucoma detection, progression assessment, staging and so on. We summarize findings and provide several taxonomies to help the reader understand the evolution of conventional and emerging AI models in glaucoma. We discuss opportunities and challenges facing AI application in glaucoma and highlight some key themes from the existing literature that may help to explore future studies. Our goal in this systematic review is to help readers and researchers to understand critical aspects of AI related to glaucoma as well as determine the necessary steps and requirements for the successful development of AI models in glaucoma.

Predicting glaucoma progression using deep learning framework guided by generative algorithm

Glaucoma is a slowly progressing optic neuropathy that may eventually lead to blindness. To help patients receive customized treatment, predicting how quickly the disease will progress is important. Structural assessment using optical coherence tomography (OCT) can be used to visualize glaucomatous optic nerve and retinal damage, while functional visual field (VF) tests can be used to measure the extent of vision loss. However, VF testing is patient-dependent and highly inconsistent, making it difficult to track glaucoma progression. In this work, we developed a multimodal deep learning model comprising a convolutional neural network (CNN) and a long short-term memory (LSTM) network, for glaucoma progression prediction. We used OCT images, VF values, demographic and clinical data of 86 glaucoma patients with five visits over 12 months. The proposed method was used to predict VF changes 12 months after the first visit by combining past multimodal inputs with synthesized future images generated using generative adversarial network (GAN). The patients were classified into two classes based on their VF mean deviation (MD) decline: slow progressors (< 3 dB) and fast progressors (> 3 dB). We showed that our generative model-based novel approach can achieve the best AUC of 0.83 for predicting the progression 6 months earlier. Further, the use of synthetic future images enabled the model to accurately predict the vision loss even earlier (9 months earlier) with an AUC of 0.81, compared to using only structural (AUC = 0.68) or only functional measures (AUC = 0.72). This study provides valuable insights into the potential of using synthetic follow-up OCT images for early detection of glaucoma progression.

Forecasting Risk of Future Rapid Glaucoma Worsening Using Early Visual Field, OCT, and Clinical Data

PURPOSE

To assess whether we can forecast future rapid visual field (VF) worsening using deep learning models (DLMs) trained on early VF, OCT, and clinical data.

DESIGN

A retrospective cohort study.

SUBJECTS

In total, 4536 eyes from 2962 patients. Overall, 263 (5.80%) eyes underwent rapid VF worsening (mean deviation slope less than -1 dB/year across all VFs).

METHODS

We included eyes that met the following criteria: (1) followed for glaucoma or suspect status; (2) had at least 5 longitudinal reliable VFs (VF1, VF2, VF3, VF4, and VF5); and (3) had 1 reliable baseline OCT scan (OCT1) and 1 set of baseline clinical measurements (clinical1) at the time of VF1. We designed a DLM to forecast future rapid VF worsening. The input consisted of spatially oriented total deviation values from VF1 (including or not including VF2 and VF3 in some models) and retinal nerve fiber layer thickness values from the baseline OCT. We passed this VF/OCT stack into a vision transformer feature extractor, the output of which was concatenated with baseline clinical data before putting it through a linear classifier to predict the eye's risk of rapid VF worsening across the 5 VFs. We compared the performance of models with differing inputs by computing area under the curve (AUC) in the test set. Specifically, we trained models with the following inputs: (1) model V: VF1; (2) VC: VF1+ Clinical1; (3) VO: VF1+ OCT1; (4) VOC: VF1+ Clinical1+ OCT1; (5) V2: VF1 + VF2; (6) V2OC: VF1 + VF2 + Clinical1 + OCT1; (7) V3: VF1 + VF2 + VF3; and (8) V3OC: VF1 + VF2 + VF3 + Clinical1 + OCT1.

MAIN OUTCOME MEASURES

The AUC of DLMs when forecasting rapidly worsening eyes.

RESULTS

Model V3OC best forecasted rapid worsening with an AUC (95% confidence interval [CI]) of 0.87 (0.77-0.97). Remaining models in descending order of performance and their respective AUC (95% CI) were as follows: (1) model V3 (0.84 [0.74-0.95]), (2) model V2OC (0.81 [0.70-0.92]), (3) model V2 (0.81 [0.70-0.82]), (4) model VOC (0.77 [0.65-0.88]), (5) model VO (0.75 [0.64-0.88]), (6) model VC (0.75 [0.63-0.87]), and (7) model V (0.74 [0.62-0.86]).

CONCLUSIONS

Deep learning models can forecast future rapid glaucoma worsening with modest to high performance when trained using data from early in the disease course. Including baseline data from multiple modalities and subsequent visits improves performance beyond using VF data alone.

FINANCIAL DISCLOSURE(S)

Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Use of artificial intelligence in forecasting glaucoma progression

Artificial intelligence (AI) has been widely used in ophthalmology for disease detection and monitoring progression. For glaucoma research, AI has been used to understand progression patterns and forecast disease trajectory based on analysis of clinical and imaging data. Techniques such as machine learning, natural language processing, and deep learning have been employed for this purpose. The results from studies using AI for forecasting glaucoma progression however vary considerably due to dataset constraints, lack of a standard progression definition and differences in methodology and approach. While glaucoma detection and screening have been the focus of most research that has been published in the last few years, in this narrative review we focus on studies that specifically address glaucoma progression. We also summarize the current evidence, highlight studies that have translational potential, and provide suggestions on how future research that addresses glaucoma progression can be improved.

Detecting disease progression in mild, moderate and severe glaucoma

PURPOSE OF REVIEW

The purpose of this review is to examine contemporary techniques for detecting the progression of glaucoma. We provide a general overview of detection principles and review evidence-based diagnostic strategies and specific considerations for detecting glaucomatous progression in patients with mild, moderate and severe disease.

RECENT FINDINGS

Diagnostic techniques and technologies for glaucoma have dramatically evolved in recent years, affording clinicians an expansive toolkit with which to detect glaucoma progression. Each stage of glaucoma, however, presents unique diagnostic challenges. In mild disease, either structural or functional changes can develop first in disease progression. In moderate disease, structural or functional changes can occur either in tandem or in isolation. In severe disease, standard techniques may fail to detect further disease progression, but such detection can still be measured using other modalities.

SUMMARY

Detecting disease progression is central to the management of glaucoma. Glaucomatous progression has both structural and functional elements, both of which must be carefully monitored at all disease stages to determine when interventions are warranted.

A deep learning model incorporating spatial and temporal information successfully detects visual field worsening using a consensus based approach

Glaucoma is a leading cause of irreversible blindness, and its worsening is most often monitored with visual field (VF) testing. Deep learning models (DLM) may help identify VF worsening consistently and reproducibly. In this study, we developed and investigated the performance of a DLM on a large population of glaucoma patients. We included 5099 patients (8705 eyes) seen at one institute from June 1990 to June 2020 that had VF testing as well as clinician assessment of VF worsening. Since there is no gold standard to identify VF worsening, we used a consensus of six commonly used algorithmic methods which include global regressions as well as point-wise change in the VFs. We used the consensus decision as a reference standard to train/test the DLM and evaluate clinician performance. 80%, 10%, and 10% of patients were included in training, validation, and test sets, respectively. Of the 873 eyes in the test set, 309 [60.6%] were from females and the median age was 62.4; (IQR 54.8-68.9). The DLM achieved an AUC of 0.94 (95% CI 0.93-0.99). Even after removing the 6 most recent VFs, providing fewer data points to the model, the DLM successfully identified worsening with an AUC of 0.78 (95% CI 0.72-0.84). Clinician assessment of worsening (based on documentation from the health record at the time of the final VF in each eye) had an AUC of 0.64 (95% CI 0.63-0.66). Both the DLM and clinician performed worse when the initial disease was more severe. This data shows that a DLM trained on a consensus of methods to define worsening successfully identified VF worsening and could help guide clinicians during routine clinical care.

Reverse translation of artificial intelligence in glaucoma: Connecting basic science with clinical applications

Artificial intelligence (AI) has been approved for biomedical research in diverse areas from bedside clinical studies to benchtop basic scientific research. For ophthalmic research, in particular glaucoma, AI applications are rapidly growing for potential clinical translation given the vast data available and the introduction of federated learning. Conversely, AI for basic science remains limited despite its useful power in providing mechanistic insight. In this perspective, we discuss recent progress, opportunities, and challenges in the application of AI in glaucoma for scientific discoveries. Specifically, we focus on the research paradigm of reverse translation, in which clinical data are first used for patient-centered hypothesis generation followed by transitioning into basic science studies for hypothesis validation. We elaborate on several distinctive areas of research opportunities for reverse translation of AI in glaucoma including disease risk and progression prediction, pathology characterization, and sub-phenotype identification. We conclude with current challenges and future opportunities for AI research in basic science for glaucoma such as inter-species diversity, AI model generalizability and explainability, as well as AI applications using advanced ocular imaging and genomic data.

Predictors of limited early response to anti-vascular endothelial growth factor therapy in neovascular age-related macular degeneration with machine learning feature importance

PURPOSE

Patients with neovascular age-related macular degeneration (nAMD) have varying responses to anti-vascular endothelial growth factor injections. Limited early response (LER) after three monthly loading doses is associated with poor long-term vision outcomes. This study predicts LER in nAMD and uses feature importance analysis to explain how baseline variables influence predicted LER risk.

METHODS

Baseline age, best visual acuity (BVA), central subfield thickness (CST), and baseline and 3 months intraretinal fluid (IRF) and subretinal fluid (SRF) for 286 eyes were collected in a retrospective clinical chart review. At month 3, LER was defined as the presence of fluid, while early response (ER) was the absence thereof. Decision tree classification and feature importance methods determined the influence of baseline age, BVA, CST, IRF, and SRF, on predicted LER risk.

RESULTS

One hundred and sixty-seven eyes were LERs and 119 were ERs. The algorithm achieved area under the curve = 0.66 in predicting LER. Baseline SRF was most important for predicting LER while age, BVA, CST, and IRF were somewhat less important. Nonlinear trends were observed between baseline variables and predicted LER risk. Zones of increased predicted LER risk were identified, including age <74 years, and CST <290 or >350 μm, IRF >750 nL, and SRF >150 nL.

CONCLUSION

These findings explain baseline variable importance for predicting LER and show SRF to be the most important. The nonlinear impact of baseline variables on predicted risk is shown, increasing understanding of LER and aiding clinicians in assessing personalized LER risk.

A deep-learning system predicts glaucoma incidence and progression using retinal photographs

BackgroundDeep learning has been widely used for glaucoma diagnosis. However, there is no clinically validated algorithm for glaucoma incidence and progression prediction. This study aims to develop a clinically feasible deep-learning system for predicting and stratifying the risk of glaucoma onset and progression based on color fundus photographs (CFPs), with clinical validation of performance in external population cohorts.MethodsWe established data sets of CFPs and visual fields collected from longitudinal cohorts. The mean follow-up duration was 3 to 5 years across the data sets. Artificial intelligence (AI) models were developed to predict future glaucoma incidence and progression based on the CFPs of 17,497 eyes in 9346 patients. The area under the receiver operating characteristic (AUROC) curve, sensitivity, and specificity of the AI models were calculated with reference to the labels provided by experienced ophthalmologists. Incidence and progression of glaucoma were determined based on longitudinal CFP images or visual fields, respectively.ResultsThe AI model to predict glaucoma incidence achieved an AUROC of 0.90 (0.81-0.99) in the validation set and demonstrated good generalizability, with AUROCs of 0.89 (0.83-0.95) and 0.88 (0.79-0.97) in external test sets 1 and 2, respectively. The AI model to predict glaucoma progression achieved an AUROC of 0.91 (0.88-0.94) in the validation set, and also demonstrated outstanding predictive performance with AUROCs of 0.87 (0.81-0.92) and 0.88 (0.83-0.94) in external test sets 1 and 2, respectively.ConclusionOur study demonstrates the feasibility of deep-learning algorithms in the early detection and prediction of glaucoma progression.FUNDINGNational Natural Science Foundation of China (NSFC); the High-level Hospital Construction Project, Zhongshan Ophthalmic Center, Sun Yat-sen University; the Science and Technology Program of Guangzhou, China (2021), the Science and Technology Development Fund (FDCT) of Macau, and FDCT-NSFC.