Publications | Wilmer Glaucoma ML

2024

Assessment of linear regression of peripapillary optical coherence tomography retinal nerve fiber layer measurements to forecast glacuoma trajectory

Chris Bradley, Kaihua Hou, Patrick Herbert, and 5 more authors

PloS One 2024

Abs

Linear regression of optical coherence tomography measurements of peripapillary retinal nerve fiber layer thickness is often used to detect glaucoma progression and forecast future disease course. However, current measurement frequencies suggest that clinicians often apply linear regression to a relatively small number of measurements (e.g., less than a handful). In this study, we estimate the accuracy of linear regression in predicting the next reliable measurement of average retinal nerve fiber layer thickness using Zeiss Cirrus optical coherence tomography measurements of average retinal nerve fiber layer thickness from a sample of 6,471 eyes with glaucoma or glaucoma-suspect status. Linear regression is compared to two null models: no glaucoma worsening, and worsening due to aging. Linear regression on the first M ≥ 2 measurements was significantly worse at predicting a reliable M+1st measurement for 2 ≤ M ≤ 6. This range was reduced to 2 ≤ M ≤ 5 when retinal nerve fiber layer thickness measurements were first "corrected" for scan quality. Simulations based on measurement frequencies in our sample-on average 393 ± 190 days between consecutive measurements-show that linear regression outperforms both null models when M ≥ 5 and the goal is to forecast moderate (75th percentile) worsening, and when M ≥ 3 for rapid (90th percentile) worsening. If linear regression is used to assess disease trajectory with a small number of measurements over short time periods (e.g., 1-2 years), as is often the case in clinical practice, the number of optical coherence tomography examinations needs to be increased.
Deep learning-based identification of eyes at risk for glaucoma surgery

Ruolin Wang, Chris Bradley, Patrick Herbert, and 5 more authors

Scientific Reports Jan 2024

Abs

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 (\textgreater 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.
Opportunities for Improving Glaucoma Clinical Trials via Deep Learning-based Identification of Patients with Low Visual Field Variability

Wang Ruolin, Chris Bradley, Patrick Herbert, and 6 more authors

Ophthalmology. Glaucoma Jan 2024

Abs

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.

2023

The impact of achieving target intraocular pressure on glaucomatous retinal nerve fiber layer thinning in a treated clinical population

Alex T. Pham, Chris Bradley, Kaihua Hou, and 4 more authors

American Journal of Ophthalmology Nov 2023

Abs

Estimate the effect of being below and above the clinician-set target intraocular pressure (IOP) on rates of glaucomatous retinal nerve fiber layer (RNFL) thinning in a treated real-world clinical population.
Forecasting risk of future rapid glaucoma worsening using early visual field, Optical Coherence Tomography, and clinical data

Patrick Herbert, Kaihua Hou, Chris Bradley, and 5 more authors

Ophthalmol. Glaucoma Mar 2023

Abs

PURPOSE: 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: Retrospective cohort study. SUBJECTS: 4,536 eyes from 2,962 patients. 263 (5.80%) of eyes underwent rapid VF worsening (MD slope <-1dB/yr across all VFs). METHODS: We included eyes that met the following criteria: 1) followed for glaucoma or suspect status 2) had at least five longitudinal reliable VFs (VF1, VF2, VF3, VF4, VF5) 3) had one reliable baseline Optical Coherence Tomography (OCT) scan (OCT1) and one 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 that eye’s risk of rapid VF worsening across the five VFs. We compared the performance of models with differing inputs by computing area under receiver operating curve (AUC) in the test set. Specifically, we trained models with the following inputs: Model V: VF1; VC: VF1+ Clinical1; VO: VF1+ OCT1; VOC: VF1+ Clinical1+ OCT1; V2: VF1 + VF2; V2OC: VF1 + VF2 + Clinical1 + OCT1; V3: VF1 + VF2 + VF3; V3OC: VF1 + VF2 + VF3 + Clinical1 + OCT1. MAIN OUTCOME MEASURES: AUC of DLMs when forecasting rapidly worsening eyes. RESULTS: Model V3OC best forecasted rapid worsening with an AUC (95% CI) of 0.87 (0.77, 0.97). Remaining models in descending order of performance and their respective AUC [95% CI] were: Model V3 (0.84 [0.74 to 0.95]), Model V2OC (0.81 [0.70 to 0.92]), Model V2 (0.81 [0.70 to 0.82]), Model VOC (0.77 [0.65, 0.88]), Model VO [0.75 [0.64, 0.88], Model VC (0.75 [0.63, 0.87]), Model V (0.74 [0.62, 0.86]). CONCLUSION: DLMs 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.
Comparing the accuracy of peripapillary OCT scans and visual fields to detect glaucoma worsening

C. Bradley, P. Herbert, K. Hou, and 3 more authors

Ophthalmology Feb 2023

Abs

Compare the accuracy of detecting moderate and rapid rates of glaucoma worsening over a 2-year period with different numbers of OCT scans and VFs in a large sample of glaucoma and glaucoma-suspect eyes.
A deep learning model incorporating spatial and temporal information successfully detects visual field worsening using a consensus based approach

J. Sabharwal, K. Hou, P. Herbert, and 6 more authors

Sci Rep Jan 2023

Abs

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.

2022

Evidence based guidelines for the number of peripapillary OCT scans needed to detect glaucoma worsening

Chris Bradley, Kaihua Hou, Patrick Herbert, and 4 more authors

Ophthalmology Jan 2022

Abs

Estimate the number of OCT scans necessary to detect moderate and rapid rates of retinal nerve fiber layer (RNFL) thickness worsening at different levels of accuracy using a large sample of glaucoma and glaucoma-suspect eyes.
Improving visual field forecasting by correcting for the effects of poor visual field reliability

Gabriel A. Villasana, Chris Bradley, Tobias Elze, and 8 more authors

Translational Vision Science & Technology Jan 2022

Abs

The purpose of this study was to accurately forecast future reliable visual field (VF) mean deviation (MD) values by correcting for poor reliability.
Differences in visual field loss pattern when transitioning from SITA standard to Sita Faster

Christopher T. Le, Jacob Fiksel, Pradeep Ramulu, and 1 more author

Scientific Reports Jan 2022

Abs

Swedish Interactive Threshold Algorithm (SITA) Faster is the most recent and fastest testing algorithm for the evaluation of Humphrey visual fields (VF). However, existing evidence suggests that there are some differences in global measures of VF loss in eyes transitioning from SITA Standard to the newer SITA Faster. These differences may be relevant, especially in glaucoma, where VF changes over time influence clinical decisions around treatment. Furthermore, characterization of differences in localizable VF loss patterns between algorithms, rather than global summary measures, can be important for clinician interpretation when transitioning testing strategies. In this study, we determined the effect of transitioning from SITA Standard to SITA Faster on VF loss patterns in glaucomatous eyes undergoing longitudinal VF testing in a real-world clinical setting. Archetypal analysis was used to derive composition weights of 16 clinically relevant VF patterns (i.e., archetypes (AT)) from patient VFs. We found switching from SITA Standard to SITA Faster was associated with less preservation of VF loss (i.e., abnormal AT 2-4, 6-9, 11, 13, 14) relative to successive SITA Standard exams (P value < 0.01) and was associated with relatively greater preservation of AT 1, the normal VF (P value < 0.01). Eyes that transition from SITA Standard to SITA Faster in a real-world clinical setting have an increased likelihood of preserving patterns reflecting a normal VF and lower tendency to preserve patterns reflecting abnormal VF as compared to consecutive SITA Standard exams in the same eye.
The effect of achieving target intraocular pressure on visual field worsening

Gabriel A. Villasana, Chris Bradley, Pradeep Ramulu, and 2 more authors

Ophthalmology Sep 2022

Abs

To estimate the effect of achieving target intraocular pressure (IOP) values on visual field (VF) worsening in a treated clinical population.

2021

The association between intraocular pressure and visual field worsening in treated glaucoma patients

Jithin Yohannan, Michael V. Boland, and Pradeep Ramulu

Journal of Glaucoma Sep 2021

Abs

The purpose of this study was to describe the relationship between mean treated IOP and VF worsening and understand how this relationship is affected by glaucoma severity and IOP range.
Predicting eyes at risk for rapid glaucoma progression based on an initial visual field test using machine learning

Scott R. Shuldiner, Michael V. Boland, Pradeep Y. Ramulu, and 6 more authors

PLOS ONE Sep 2021

Abs

To assess whether machine learning algorithms (MLA) can predict eyes that will undergo rapid glaucoma progression based on an initial visual field (VF) test.
The effect of transitioning from SITA standard to Sita Faster on visual field performance

Alex T. Pham, Pradeep Y. Ramulu, Michael V. Boland, and 1 more author

Ophthalmology Mar 2021

Abs

To determine the effect of transitioning from Swedish Interactive Thresholding Algorithm (SITA) Standard to SITA Faster on visual field (VF) performance in glaucomatous eyes with a broad spectrum of disease severity undergoing longitudinal VF testing in a real-world clinical setting.

2020

Assessing glaucoma progression using machine learning trained on longitudinal visual field and clinical data

Avyuk Dixit, Jithin Yohannan, and Michael V. Boland

Ophthalmology Dec 2020

Abs

Rule-based approaches to determining glaucoma progression from visual fields (VFs) alone are discordant and have tradeoffs. To detect better when glaucoma progression is occurring, we used a longitudinal data set of merged VF and clinical data to assess the performance of a convolutional long short-term memory (LSTM) neural network.
Factors predicting a greater likelihood of poor visual field reliability in glaucoma patients and suspects

Inas F. Aboobakar, Jiangxia Wang, Balwantray C. Chauhan, and 4 more authors

Translational Vision Science & Technology Jan 2020

Abs

Identify factors predicting worse or better than expected visual field (VF) performance.

2019

Evidence-based criteria for determining peripapillary oct reliability

Jithin Yohannan, Michael Cheng, Joseph Da, and 6 more authors

Ophthalmology Aug 2019

Abs

To assess the impact of OCT signal strength (SS) and artifact on retinal nerve fiber layer (RNFL) measurement reliability and to understand whether glaucoma severity modifies this relationship.

2017

The evolving role of the relationship between optic nerve structure and function in glaucoma

Jithin Yohannan, and Michael V. Boland

Ophthalmology Dec 2017

Abs

The relationship between functional vision loss and structural changes of the optic nerve head and retinal ganglion cells is the hallmark of glaucoma diagnosis. Understanding and measuring this relationship has been the focus of numerous studies, the goal of which have been to improve glaucoma diagnosis and detection of glaucoma worsening. In this review, a historical perspective is used to understand structure–function relationships in glaucoma and their application to improve glaucoma diagnosis and monitoring. Initially, histologic studies that link visual field sensitivity to retinal ganglion cell count are discussed. Additionally, those studies that determined the mathematical relationship between visual field sensitivity and ganglion cell number are reviewed. Next, those studies that attempt to create a map of the structure–function relationship using fundus photography and visual field sensitivity are examined. Subsequently, studies that use more recent imaging technology, such as optical coherence tomography, confocal scanning laser ophthalmoscopy, or scanning laser perimetry, to measure structure quantitatively in vivo and to correlate these measures with automated perimetry are explored. Among these studies that use advanced imaging, those that use cross-sectional data to explore structure–function relationships to improve glaucoma diagnosis first are discussed. Second, those studies that use longitudinal data to improve detection of worsening are reviewed. Finally, areas of further research and steps needed to implement structure–function relationships clinically are explored.
Evidence-based criteria for assessment of Visual Field Reliability

Jithin Yohannan, Jiangxia Wang, Jamie Brown, and 4 more authors

Ophthalmology Jul 2017

Abs

Assess the impact of false-positives (FP), false-negatives (FN), fixation losses (FL), and test duration (TD) on visual field (VF) reliability at different stages of glaucoma severity.