Automated image alignment and segmentation to follow progression of geographic atrophy in age-related macular degeneration

Artificial Intelligence Approaches for Geographic Atrophy Segmentation: A Systematic Review and Meta-Analysis

Geographic atrophy (GA) is a progressive retinal disease associated with late-stage age-related macular degeneration (AMD), a significant cause of visual impairment in senior adults. GA lesion segmentation is important for disease monitoring in clinical trials and routine ophthalmic practice; however, its manual delineation is time-consuming, laborious, and subject to inter-grader variability. The use of artificial intelligence (AI) is rapidly expanding within the medical field and could potentially improve accuracy while reducing the workload by facilitating this task. This systematic review evaluates the performance of AI algorithms for GA segmentation and highlights their key limitations from the literature. Five databases and two registries were searched from inception until 23 March 2024, following the PRISMA methodology. Twenty-four studies met the prespecified eligibility criteria, and fifteen were included in this meta-analysis. The pooled Dice similarity coefficient (DSC) was 0.91 (95% CI 0.88-0.95), signifying a high agreement between the reference standards and model predictions. The risk of bias and reporting quality were assessed using QUADAS-2 and CLAIM tools. This review provides a comprehensive evaluation of AI applications for GA segmentation and identifies areas for improvement. The findings support the potential of AI to enhance clinical workflows and highlight pathways for improved future models that could bridge the gap between research settings and real-world clinical practice.

EyeLiner: A Deep Learning Pipeline for Longitudinal Image Registration Using Fundus Landmarks

Objective

Detecting and measuring changes in longitudinal fundus imaging is key to monitoring disease progression in chronic ophthalmic diseases, such as glaucoma and macular degeneration. Clinicians assess changes in disease status by either independently reviewing or manually juxtaposing longitudinally acquired color fundus photos (CFPs). Distinguishing variations in image acquisition due to camera orientation, zoom, and exposure from true disease-related changes can be challenging. This makes manual image evaluation variable and subjective, potentially impacting clinical decision-making. We introduce our deep learning (DL) pipeline, "EyeLiner," for registering, or aligning, 2-dimensional CFPs. Improved alignment of longitudinal image pairs may compensate for differences that are due to camera orientation while preserving pathological changes.

Design

EyeLiner registers a "moving" image to a "fixed" image using a DL-based keypoint matching algorithm.

Participants

We evaluate EyeLiner on 3 longitudinal data sets: Fundus Image REgistration (FIRE), sequential images for glaucoma forecast (SIGF), and our internal glaucoma data set from the Colorado Ophthalmology Research Information System (CORIS).

Methods

Anatomical keypoints along the retinal blood vessels were detected from the moving and fixed images using a convolutional neural network and subsequently matched using a transformer-based algorithm. Finally, transformation parameters were learned using the corresponding keypoints.

Main Outcome Measures

We computed the mean distance (MD) between manually annotated keypoints from the fixed and the registered moving image. For comparison to existing state-of-the-art retinal registration approaches, we used the mean area under the curve (AUC) metric introduced in the FIRE data set study.

Results

EyeLiner effectively aligns longitudinal image pairs from FIRE, SIGF, and CORIS, as qualitatively evaluated through registration checkerboards and flicker animations. Quantitative results show that the MD decreased for this model after alignment from 321.32 to 3.74 pixels for FIRE, 9.86 to 2.03 pixels for CORIS, and 25.23 to 5.94 pixels for SIGF. We also obtained an AUC of 0.85, 0.94, and 0.84 on FIRE, CORIS, and SIGF, respectively, beating the current state-of-the-art SuperRetina (AUCFIRE = 0.76, AUCCORIS = 0.83, AUCSIGF = 0.74).

Conclusions

Our pipeline demonstrates improved alignment of image pairs in comparison to the current state-of-the-art methods on 3 separate data sets. We envision that this method will enable clinicians to align image pairs and better visualize changes in disease over time.

Financial Disclosures

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

[Use of artificial intelligence in geographic atrophy in age-related macular degeneration]

The first regulatory approval of treatment for geographic atrophy (GA) secondary to age-related macular degeneration in the USA constitutes an important milestone; however, due to the nature of GA as a non-acute, insidiously progressing pathology, the ophthalmologist faces specific challenges concerning risk stratification, making treatment decisions, monitoring of treatment and patient education. Innovative retinal imaging modalities, such as fundus autofluorescence (FAF) and optical coherence tomography (OCT) have enabled identification of typical morphological alterations in relation to GA, which are also suitable for the quantitative characterization of GA. Solutions based on artificial intelligence (AI) enable automated detection and quantification of GA-specific biomarkers on retinal imaging data, also retrospectively and over time. Moreover, AI solutions can be used for the diagnosis and segmentation of GA as well as the prediction of structure and function without and under GA treatment, thereby making a valuable contribution to treatment monitoring and the identification of high-risk patients and patient education. The integration of AI solutions into existing clinical processes and software systems enables the broad implementation of informed and personalized treatment of GA secondary to AMD.

Recurrent and Concurrent Prediction of Longitudinal Progression of Stargardt Atrophy and Geographic Atrophy

Stargardt disease and age-related macular degeneration are the leading causes of blindness in the juvenile and geriatric populations, respectively. The formation of atrophic regions of the macula is a hallmark of the end-stages of both diseases. The progression of these diseases is tracked using various imaging modalities, two of the most common being fundus autofluorescence (FAF) imaging and spectral-domain optical coherence tomography (SD-OCT). This study seeks to investigate the use of longitudinal FAF and SD-OCT imaging (month 0, month 6, month 12, and month 18) data for the predictive modelling of future atrophy in Stargardt and geographic atrophy. To achieve such an objective, we develop a set of novel deep convolutional neural networks enhanced with recurrent network units for longitudinal prediction and concurrent learning of ensemble network units (termed ReConNet) which take advantage of improved retinal layer features beyond the mean intensity features. Using FAF images, the neural network presented in this paper achieved mean (± standard deviation, SD) and median Dice coefficients of 0.895 (± 0.086) and 0.922 for Stargardt atrophy, and 0.864 (± 0.113) and 0.893 for geographic atrophy. Using SD-OCT images for Stargardt atrophy, the neural network achieved mean and median Dice coefficients of 0.882 (± 0.101) and 0.906, respectively. When predicting only the interval growth of the atrophic lesions with FAF images, mean (± SD) and median Dice coefficients of 0.557 (± 0.094) and 0.559 were achieved for Stargardt atrophy, and 0.612 (± 0.089) and 0.601 for geographic atrophy. The prediction performance in OCT images is comparably good to that using FAF which opens a new, more efficient, and practical door in the assessment of atrophy progression for clinical trials and retina clinics, beyond widely used FAF. These results are highly encouraging for a high-performance interval growth prediction when more frequent or longer-term longitudinal data are available in our clinics. This is a pressing task for our next step in ongoing research.

Diabetic microaneurysms detected by fluorescein angiography spatially correlate with regions of macular ischemia delineated by optical coherence tomography angiography

PURPOSE

To characterize the relationship between diabetic macular ischemia (DMI) delineated by optical coherence tomography angiography (OCTA) and microaneurysms (MAs) identified by fundus fluorescein angiography (FFA).

METHODS

Patients with diabetic retinopathy (DR) who underwent OCTA and FFA were retrospectively identified. FFA images were cropped and aligned with their respective OCTA images using i2k Align Retina software (Dual-Align, Clifton Park, NY, USA). Foveal avascular zone (FAZ) and ischemic areas were manually delineated on OCTA images, and MAs were marked on the corresponding FFA images before overlaying paired scans for analysis (ImageJ; National Institutes of Health, Bethesda, MD, USA).

RESULTS

Twenty-eight eyes of 20 patients were included. The average number of MAs identified in cropped FFA images was 127 ± 42. More DMI was noted in the superficial capillary plexus (SCP; 36 ± 13%) compared to the deep capillary plexus (DCP; 28 ± 14%, P < 0.001). Similarly, more MAs were associated with ischemic areas in SCP compared to DCP (92.0 ± 35.0 vs. 76.8 ± 36.5, P < 0.001). Most MAs bordered ischemic areas; fewer than 10% localized inside these regions. As DMI area increased, so did associated MAs (SCP: r = 0.695, P < 0.001; DCP: r = 0.726, P < 0.001). Density of MAs surrounding FAZ (7.7 ± 6.0 MAs/mm2) was similar to other DMI areas (SCP: 7.0 ± 4.0 MAs/mm2, P = 0.478; DCP: 9.2 ± 10.9 MAs/mm2, P = 0.394).

CONCLUSION

MAs identified in FFA strongly associate with, and border areas of, DMI delineated by OCTA. Although more MAs are localized to SCP ischemia, the concentration of MAs associated with DCP ischemia is greater. By contrast, few MAs are present inside low-flow regions, likely because capillary loss is associated with their regression.

Geographic Atrophy Segmentation Using Multimodal Deep Learning

Purpose

To examine deep learning (DL)-based methods for accurate segmentation of geographic atrophy (GA) lesions using fundus autofluorescence (FAF) and near-infrared (NIR) images.

Methods

This retrospective analysis utilized imaging data from study eyes of patients enrolled in Proxima A and B (NCT02479386; NCT02399072) natural history studies of GA. Two multimodal DL networks (UNet and YNet) were used to automatically segment GA lesions on FAF; segmentation accuracy was compared with annotations by experienced graders. The training data set comprised 940 image pairs (FAF and NIR) from 183 patients in Proxima B; the test data set comprised 497 image pairs from 154 patients in Proxima A. Dice coefficient scores, Bland-Altman plots, and Pearson correlation coefficient (r) were used to assess performance.

Results

On the test set, Dice scores for the DL network to grader comparison ranged from 0.89 to 0.92 for screening visit; Dice score between graders was 0.94. GA lesion area correlations (r) for YNet versus grader, UNet versus grader, and between graders were 0.981, 0.959, and 0.995, respectively. Longitudinal GA lesion area enlargement correlations (r) for screening to 12 months (n = 53) were lower (0.741, 0.622, and 0.890, respectively) compared with the cross-sectional results at screening. Longitudinal correlations (r) from screening to 6 months (n = 77) were even lower (0.294, 0.248, and 0.686, respectively).

Conclusions

Multimodal DL networks to segment GA lesions can produce accurate results comparable with expert graders.

Translational Relevance

DL-based tools may support efficient and individualized assessment of patients with GA in clinical research and practice.

Detection of macular atrophy in age-related macular degeneration aided by artificial intelligence

INTRODUCTION

Age-related macular degeneration (AMD) is a leading cause of irreversible visual impairment worldwide. The endpoint of AMD, both in its dry or wet form, is macular atrophy (MA) which is characterized by the permanent loss of the RPE and overlying photoreceptors either in dry AMD or in wet AMD. A recognized unmet need in AMD is the early detection of MA development.

AREAS COVERED

Artificial Intelligence (AI) has demonstrated great impact in detection of retinal diseases, especially with its robust ability to analyze big data afforded by ophthalmic imaging modalities, such as color fundus photography (CFP), fundus autofluorescence (FAF), near-infrared reflectance (NIR), and optical coherence tomography (OCT). Among these, OCT has been shown to have great promise in identifying early MA using the new criteria in 2018.

EXPERT OPINION

There are few studies in which AI-OCT methods have been used to identify MA; however, results are very promising when compared to other imaging modalities. In this paper, we review the development and advances of ophthalmic imaging modalities and their combination with AI technology to detect MA in AMD. In addition, we emphasize the application of AI-OCT as an objective, cost-effective tool for the early detection and monitoring of the progression of MA in AMD.

Automated segmentation and quantification of calcified drusen in 3D swept source OCT imaging

Qualitative and quantitative assessments of calcified drusen are clinically important for determining the risk of disease progression in age-related macular degeneration (AMD). This paper reports the development of an automated algorithm to segment and quantify calcified drusen on swept-source optical coherence tomography (SS-OCT) images. The algorithm leverages the higher scattering property of calcified drusen compared with soft drusen. Calcified drusen have a higher optical attenuation coefficient (OAC), which results in a choroidal hypotransmission defect (hypoTD) below the calcified drusen. We show that it is possible to automatically segment calcified drusen from 3D SS-OCT scans by combining the OAC within drusen and the hypoTDs under drusen. We also propose a correction method for the segmentation of the retina pigment epithelium (RPE) overlying calcified drusen by automatically correcting the RPE by an amount of the OAC peak width along each A-line, leading to more accurate segmentation and quantification of drusen in general, and the calcified drusen in particular. A total of 29 eyes with nonexudative AMD and calcified drusen imaged with SS-OCT using the 6 × 6 mm² scanning pattern were used in this study to test the performance of the proposed automated method. We demonstrated that the method achieved good agreement with the human expert graders in identifying the area of calcified drusen (Dice similarity coefficient: 68.27 ± 11.09%, correlation coefficient of the area measurements: r = 0.9422, the mean bias of the area measurements = 0.04781 mm²).

Multimodal image alignment aids in the evaluation and monitoring of sector retinitis pigmentosa

PURPOSE

To present a semi-automated method of image alignment to aid in monitoring the progression of inherited retinal degenerations (IRDs).

RESULTS

A 22-year-old woman presented with nyctalopia and a family history of retinitis pigmentosa (RP), but with no prior genetic testing. Fundus examination showed a sectoral retinal degeneration involving the inferior and nasal retina with rare, pigmented deposits. Goldmann kinetic perimetry demonstrated corresponding superotemporal visual field defects. The best-corrected visual acuity was 20/20 in both eyes. Multimodal imaging delineated geographically restricted peripheral retinal degeneration extending to the inferior edge of the macula. Central visual function remained intact with normal multifocal electroretinography findings. Optical coherence tomography (OCT) through the leading edge of the retinal degeneration confirmed loss of the photoreceptor layer and associated retinal pigment epithelium. In the region of retinal degeneration, loss of vascular flow density was noted on optical coherence tomography angiography (OCTA). Genetic testing identified a pathologic sequence variant in RHO (c.68C>A, p.Pro23His), confirming autosomal dominant sector retinitis pigmentosa (SRP). Image alignment allowed for precise measurement of the progression of SRP over a period of 18 months.

CONCLUSION

SRP is a rare subtype of RP characterized by focal, typically inferior and nasal, retinal degeneration of the peripheral retina. Although the onset and extent of peripheral retinal degeneration varies, compared with RP, SRP typically progresses more slowly to involve the macula. In this report, we highlight the utility of image registration and alignment to aid in monitoring disease progression in IRDs by means of multimodal imaging.

Automated segmentation and feature discovery of age-related macular degeneration and Stargardt disease via self-attended neural networks

Age-related macular degeneration (AMD) and Stargardt disease are the leading causes of blindness for the elderly and young adults respectively. Geographic atrophy (GA) of AMD and Stargardt atrophy are their end-stage outcomes. Efficient methods for segmentation and quantification of these atrophic lesions are critical for clinical research. In this study, we developed a deep convolutional neural network (CNN) with a trainable self-attended mechanism for accurate GA and Stargardt atrophy segmentation. Compared with traditional post-hoc attention mechanisms which can only visualize CNN features, our self-attended mechanism is embedded in a fully convolutional network and directly involved in training the CNN to actively attend key features for enhanced algorithm performance. We applied the self-attended CNN on the segmentation of AMD and Stargardt atrophic lesions on fundus autofluorescence (FAF) images. Compared with a preexisting regular fully convolutional network (the U-Net), our self-attended CNN achieved 10.6% higher Dice coefficient and 17% higher IoU (intersection over union) for AMD GA segmentation, and a 22% higher Dice coefficient and a 32% higher IoU for Stargardt atrophy segmentation. With longitudinal image data having over a longer time, the developed self-attended mechanism can also be applied on the visual discovery of early AMD and Stargardt features.

Automatic geographic atrophy segmentation using optical attenuation in OCT scans with deep learning

A deep learning algorithm was developed to automatically identify, segment, and quantify geographic atrophy (GA) based on optical attenuation coefficients (OACs) calculated from optical coherence tomography (OCT) datasets. Normal eyes and eyes with GA secondary to age-related macular degeneration were imaged with swept-source OCT using 6 × 6 mm scanning patterns. OACs calculated from OCT scans were used to generate customized composite en face OAC images. GA lesions were identified and measured using customized en face sub-retinal pigment epithelium (subRPE) OCT images. Two deep learning models with the same U-Net architecture were trained using OAC images and subRPE OCT images. Model performance was evaluated using DICE similarity coefficients (DSCs). The GA areas were calculated and compared with manual segmentations using Pearson's correlation and Bland-Altman plots. In total, 80 GA eyes and 60 normal eyes were included in this study, out of which, 16 GA eyes and 12 normal eyes were used to test the models. Both models identified GA with 100% sensitivity and specificity on the subject level. With the GA eyes, the model trained with OAC images achieved significantly higher DSCs, stronger correlation to manual results and smaller mean bias than the model trained with subRPE OCT images (0.940 ± 0.032 vs 0.889 ± 0.056, p = 0.03, paired t-test, r = 0.995 vs r = 0.959, mean bias = 0.011 mm vs mean bias = 0.117 mm). In summary, the proposed deep learning model using composite OAC images effectively and accurately identified, segmented, and quantified GA using OCT scans.

Automated Image Alignment for Comparing Microvascular Changes Detected by Fluorescein Angiography and Optical Coherence Tomography Angiography in Diabetic Retinopathy

PURPOSE

To quantitatively compare microvascular features in the macula of patients with diabetic retinopathy (DR) using fluorescein angiography (FA) and optical coherence tomography angiography (OCTA).

METHODS

Patients with DR were recruited from the Cairo University Hospital. FA was performed using a Topcon TRC-50DX or Heidelberg Spectralis HRA+OCT. OCTA was performed using an Optovue RTVue-XR Avanti. FA images were cropped and aligned to the corresponding OCTA images using i2k Align Retina software. The foveal avascular zone (FAZ), area of ischemia, and microaneurysms (MAs) were manually quantified using ImageJ. The fractal dimension (FD) was calculated from each skeletonized image using the FracLac plugin of ImageJ after retinal vascular segmentation.

RESULTS

Twenty-four eyes of 17 patients were evaluated, but only 18 eyes were successfully aligned. There was no difference in FAZ area measured for FA and OCTA images. Compared with OCTA images, FD was significantly less for FA images (1.66 ± 0.048 versus 1.72 ± 0.023, p < .001). Significantly more MAs were identified on FA images (102 ± 27.5) compared with OCTA (47.5 ± 11.7, p < .0001). The number of MAs on FA correlated with decreasing best corrected visual acuity (r2 = 0.315, p = .015) and increasing central macular thickness (r2 = 0.492, p = .001). No such associations were found with MAs detected on OCTA. Nevertheless, the area of ischemia in the FA images (8.5 ± 4.1%) was significantly smaller compared with the area measured in both the superficial (30.7 ± 9.5%) and deep capillary plexus (21.6 ± 10.9%) of the OCTA (p < .001). Interestingly, number of MAs in the FA images correlated with increasing area of ischemia in the FA (r2 = 0.568, p < .001) but only the superficial segment of the depth-resolved OCTA scans (r2 = 0.539, p < .001).

CONCLUSIONS

OCTA is a non-invasive tool capable of resolving the retinal vasculature in greater detail when compared with FA but detects significantly fewer MAs. Automatic alignment facilitates quantitative comparison of the microvascular features in DR.

Normalization of series of fundus images to monitor the geographic atrophy growth in dry age-related macular degeneration

BACKGROUND AND OBJECTIVE

Age-related macular degeneration (ARMD) is a degenerative disease that affects the retina, and the leading cause of visual loss. In its dry form, the pathology is characterized by the progressive, centrifugal expansion of retinal lesions, called geographic atrophy (GA). In infrared eye fundus images, the GA appears as localized bright areas and its growth can be observed in series of images acquired at regular time intervals. However, illumination distortions between the images make impossible the direct comparison of intensities in order to study the GA progress. Here, we propose a new method to compensate for illumination distortion between images.

METHODS

We process all images of the series so that any two images have comparable gray levels. Our approach relies on an illumination/reflectance model. We first estimate the pixel-wise illumination ratio between any two images of the series, in a recursive way; then we correct each image against all the others, based on those estimates. The algorithm is applied on a sliding temporal window to cope with large changes in reflectance. We also propose morphological processing to suppress illumination artefacts.

RESULTS

The corrected illumination function is homogeneous in the series, enabling the direct comparison of grey-levels intensities in each pixel, and so the detection of the GA growth between any two images. To demonstrate that, we present numerous experiments performed on a dataset of 18 series (328 images), manually segmented by an ophthalmologist. First, we show that the normalization preprocessing dramatically increases the contrast of the GA growth areas. Secondly, we apply segmentation algorithms derived from Otsu's thresholding to detect automatically the GA total growth and the GA progress between consecutive images. We demonstrate qualitatively and quantitatively that these algorithms, although fully automatic, unsupervised and basic, already lead to interesting segmentation results when applied to the normalized images. Colored maps representing the GA evolution can be derived from the segmentations.

CONCLUSION

To our knowledge, the proposed method is the first one which corrects automatically and jointly the illumination inhomogeneity in a series of fundus images, regardless of the number of images, the size, shape and progression of lesion areas. This algorithm greatly facilitates the visual interpretation by the medical expert. It opens up the possibility of treating automatically each series as a whole (not just in pairs of images) to model the GA growth.

Unsupervised Approaches for the Segmentation of Dry ARMD Lesions in Eye Fundus cSLO Images

Age-related macular degeneration (ARMD), a major cause of sight impairment for elderly people, is still not well understood despite intensive research. Measuring the size of the lesions in the fundus is the main biomarker of the severity of the disease and as such is widely used in clinical trials yet only relies on manual segmentation. Artificial intelligence, in particular automatic image analysis based on neural networks, has a major role to play in better understanding the disease, by analyzing the intrinsic optical properties of dry ARMD lesions from patient images. In this paper, we propose a comparison of automatic segmentation methods (classical computer vision method, machine learning method and deep learning method) in an unsupervised context applied on cSLO IR images. Among the methods compared, we propose an adaptation of a fully convolutional network, called W-net, as an efficient method for the segmentation of ARMD lesions. Unlike supervised segmentation methods, our algorithm does not require annotated data which are very difficult to obtain in this application. Our method was tested on a dataset of 328 images and has shown to reach higher quality results than other compared unsupervised methods with a F1 score of 0.87, while having a more stable model, even though in some specific cases, texture/edges-based methods can produce relevant results.

Deep Learning Applied to Automated Segmentation of Geographic Atrophy in Fundus Autofluorescence Images

Purpose

This study describes the development of a deep learning algorithm based on the U-Net architecture for automated segmentation of geographic atrophy (GA) lesions in fundus autofluorescence (FAF) images.

Methods

Image preprocessing and normalization by modified adaptive histogram equalization were used for image standardization to improve effectiveness of deep learning. A U-Net-based deep learning algorithm was developed and trained and tested by fivefold cross-validation using FAF images from clinical datasets. The following metrics were used for evaluating the performance for lesion segmentation in GA: dice similarity coefficient (DSC), DSC loss, sensitivity, specificity, mean absolute error (MAE), accuracy, recall, and precision.

Results

In total, 702 FAF images from 51 patients were analyzed. After fivefold cross-validation for lesion segmentation, the average training and validation scores were found for the most important metric, DSC (0.9874 and 0.9779), for accuracy (0.9912 and 0.9815), for sensitivity (0.9955 and 0.9928), and for specificity (0.8686 and 0.7261). Scores for testing were all similar to the validation scores. The algorithm segmented GA lesions six times more quickly than human performance.

Conclusions

The deep learning algorithm can be implemented using clinical data with a very high level of performance for lesion segmentation. Automation of diagnostics for GA assessment has the potential to provide savings with respect to patient visit duration, operational cost and measurement reliability in routine GA assessments.

Translational Relevance

A deep learning algorithm based on the U-Net architecture and image preprocessing appears to be suitable for automated segmentation of GA lesions on clinical data, producing fast and accurate results.

Artificial Intelligence Algorithms for Analysis of Geographic Atrophy: A Review and Evaluation

Purpose

The purpose of this study was to summarize and evaluate artificial intelligence (AI) algorithms used in geographic atrophy (GA) diagnostic processes (e.g. isolating lesions or disease progression).

Methods

The search strategy and selection of publications were both conducted in accordance with the Preferred of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. PubMed and Web of Science were used to extract literary data. The algorithms were summarized by objective, performance, and scope of coverage of GA diagnosis (e.g. lesion automation and GA progression).

Results

Twenty-seven studies were identified for this review. A total of 18 publications focused on lesion segmentation only, 2 were designed to detect and classify GA, 2 were designed to predict future overall GA progression, 3 focused on prediction of future spatial GA progression, and 2 focused on prediction of visual function in GA. GA-related algorithms reported sensitivities from 0.47 to 0.98, specificities from 0.73 to 0.99, accuracies from 0.42 to 0.995, and Dice coefficients from 0.66 to 0.89.

Conclusions

Current GA-AI publications have a predominant focus on lesion segmentation and a minor focus on classification and progression analysis. AI could be applied to other facets of GA diagnoses, such as understanding the role of hyperfluorescent areas in GA. Using AI for GA has several advantages, including improved diagnostic accuracy and faster processing speeds.

Translational Relevance

AI can be used to quantify GA lesions and therefore allows one to impute visual function and quality-of-life. However, there is a need for the development of reliable and objective models and software to predict the rate of GA progression and to quantify improvements due to interventions.

Automatic Quantification Software for Geographic Atrophy Associated with Age-Related Macular Degeneration: A Validation Study

Aims

To determine the accuracy and repeatability of new software to automatically quantify GA areas associated to age-related macular degeneration (AMD) by swept-source optical coherence tomography (SS-OCT). Settings and Design. Tertiary referral hospital in Spain. Cross-sectional and noninterventional. Methods and Material. Forty-six eyes from 33 AMD patients with GA, without previous choroidal neovascularization, were scanned using a SS-OCT (Topcon Corporation, Japan), including three consecutive 7 × 7 mm OCT scans. Three independent masked observers manually measured the GA area using FAF images. These measures were compared to the three automatic determinations of the GA. Lesions were classified according to their morphology and number as regular/irregular and single/multiple. Statistical Analysis Used. Intraclass correlation coefficients (ICCs) were estimated to study the agreement between the three physicians in manual measurements. ICC through a two-way mixed effects model was used for the software measures, and Lin's concordance correlation coefficient (CCC) was used to analyse the agreement between the physicians and the software.

Results

The mean age was 76.3 ± 11.7 years. Eighteen cases showed regular lesions, and 30 showed single lesions. The CCC between manual and automatic measures was 0.95 for the whole sample. The CCC for the area according to the lesion type was 0.92 and 0.97; it was 0.99 for single lesions and 0.89 for multiple lesions. The ICC between the three physicians was 0.94 for the whole sample and 0.88 in multiple lesions. The ICC between the three automatic measures for the area was 0.98 for the whole sample, regular or irregular lesions, and single or multiple lesions.

Conclusions

The accuracy of this new software is substantial for the area with a high degree of repeatability agreement, being very precise in single lesions.

MS-CAM: Multi-Scale Class Activation Maps for Weakly-Supervised Segmentation of Geographic Atrophy Lesions in SD-OCT Images

As one of the most critical characteristics in advanced stage of non-exudative Age-related Macular Degeneration (AMD), Geographic Atrophy (GA) is one of the significant causes of sustained visual acuity loss. Automatic localization of retinal regions affected by GA is a fundamental step for clinical diagnosis. In this paper, we present a novel weakly supervised model for GA segmentation in Spectral-Domain Optical Coherence Tomography (SD-OCT) images. A novel Multi-Scale Class Activation Map (MS-CAM) is proposed to highlight the discriminatory significance regions in localization and detail descriptions. To extract available multi-scale features, we design a Scaling and UpSampling (SUS) module to balance the information content between features of different scales. To capture more discriminative features, an Attentional Fully Connected (AFC) module is proposed by introducing the attention mechanism into the fully connected operations to enhance the significant informative features and suppress less useful ones. Based on the location cues, the final GA region prediction is obtained by the projection segmentation of MS-CAM. The experimental results on two independent datasets demonstrate that the proposed weakly supervised model outperforms the conventional GA segmentation methods and can produce similar or superior accuracy comparing with fully supervised approaches. The source code has been released and is available on GitHub: https://github.com/ jizexuan/Multi-Scale-Class-Activation-Map-Tensorflow.

Analyzing Age-Related Macular Degeneration Progression in Patients with Geographic Atrophy Using Joint Autoencoders for Unsupervised Change Detection

Age-Related Macular Degeneration (ARMD) is a progressive eye disease that slowly causes patients to go blind. For several years now, it has been an important research field to try to understand how the disease progresses and find effective medical treatments. Researchers have been mostly interested in studying the evolution of the lesions using different techniques ranging from manual annotation to mathematical models of the disease. However, artificial intelligence for ARMD image analysis has become one of the main research focuses to study the progression of the disease, as accurate manual annotation of its evolution has proved difficult using traditional methods even for experienced practicians. In this paper, we propose a deep learning architecture that can detect changes in the eye fundus images and assess the progression of the disease. Our method is based on joint autoencoders and is fully unsupervised. Our algorithm has been applied to pairs of images from different eye fundus images time series of 24 ARMD patients. Our method has been shown to be quite effective when compared with other methods from the literature, including non-neural network based algorithms that still are the current standard to follow the disease progression and change detection methods from other fields.

Evaluation of Changes in Macular Perfusion Detected by Optical Coherence Tomography Angiography following 3 Intravitreal Monthly Bevacizumab Injections for Diabetic Macular Edema in the IMPACT Study

Objective

To evaluate macular perfusion changes following intravitreal bevacizumab injections for diabetic macular edema (DME) using spectral domain optical coherence tomography angiography (SD-OCTA).

Methods

This study was a prospective noncomparative interventional case series. Treatment naïve patients with DME underwent full ophthalmological examination and SD-OCTA scanning at baseline and after 3 intravitreal bevacizumab injections. Both the 6 × 6 and 3 × 3 mm macular scan protocols were used. Pretreatment and posttreatment OCTA images were automatically aligned using a commercially available retina alignment software (i2k Align Retina software); then the fractal dimension (FD), vascular density (VD), and skeleton VD changes were obtained at the full retinal thickness (Full) and superficial (SCP) and deep (DCP) capillary plexuses after processing images using a semiautomated program. The foveal avascular zone (FAZ) was manually measured and FD was calculated using the FracLac plugin of ImageJ.

Results

Forty eyes of 26 patients were included. Following injections, there were an 8.1% increase in FAZ, 1.3% decrease in FD-Full and FD-SCP, 1.9% decrease in FD-DCP, 8% decrease in VD-Full, 9.1% decrease in VD-SCP, 10.6% decrease in VD-DCP, 13.3% decrease in skeleton VD-Full, 12.5% decrease in skeleton VD-SCP, and 16.3% decrease in skeleton VD-DCP in the 6 × 6 mm macular area and a 2.6% decrease in FD-Full, 3.4% decrease in FD-SCP, 11.5% decrease in VD-Full, 14.3% decrease in VD-SCP, and 25.1% decrease in skeleton VD-SCP in the 3 × 3 mm macular area which were all statistically significant (p < 0.05). Using univariate and multivariate analysis, the pretreatment FD, VD, and skeleton VD at each capillary layer significantly negatively correlated with the change in FD, VD, and skeleton VD at the corresponding capillary layer, respectively (p < 0.05).

Conclusion

OCTA is a useful noninvasive tool for quantitative evaluation of macular perfusion changes following DME treatment. This trial is registered with NCT03246152.

Image registration of the human accommodating eye demonstrates equivalent increases in lens equatorial radius and central thickness

AIM

To compare the results of in vivo human high resolution image registration studies of the eye during accommodation to the predictions of mathematical and finite element models of accommodation.

METHODS

Data from published high quality image registration studies of pilocarpine induced accommodative changes of equatorial lens radius (ELR) and central lens thickness (CLT) were statistically analyzed.

RESULTS

The mean changes in ELR and CLT were 6.76 µm/diopter and 6.51 µm/diopter, respectively. The linear regressions, reflecting the association between ELR and accommodative amplitude (AAELR) was: slope=6.58 µm/diopter, r2 =0.98, P<0.0001 and between CLT and AACLT was: slope=6.75 µm/diopter, r2 =0.83, P<0.001. On the basis of these relationships, the CLT slope and the AAELR were used to predict the measured change in ELR (ELRpredicted). There was no statistical difference between ELRpredicted and the measured ELR as demonstrated by a Student's paired t-test: P=0.96 and linear regression analysis: slope=0.97, r2 =0.98, P<0.00001.

CONCLUSION

Image registration with invariant positional references demonstrates that ELR and CLT equivalently minimally increase ∼7.0 µm/diopter during accommodation. The small equivalent increases in ELR and CLT are associated with a large accommodative amplitude. These findings are consistent with the predictions of mathematical and finite element models that specified the stiffness of the lens nucleus is the same or greater than the lens cortex and that accommodation involves a small force (<5 g).

A Deep Learning Approach for Automated Detection of Geographic Atrophy from Color Fundus Photographs

PURPOSE

To assess the utility of deep learning in the detection of geographic atrophy (GA) from color fundus photographs and to explore potential utility in detecting central GA (CGA).

DESIGN

A deep learning model was developed to detect the presence of GA in color fundus photographs, and 2 additional models were developed to detect CGA in different scenarios.

PARTICIPANTS

A total of 59 812 color fundus photographs from longitudinal follow-up of 4582 participants in the Age-Related Eye Disease Study (AREDS) dataset. Gold standard labels were from human expert reading center graders using a standardized protocol.

METHODS

A deep learning model was trained to use color fundus photographs to predict GA presence from a population of eyes with no AMD to advanced AMD. A second model was trained to predict CGA presence from the same population. A third model was trained to predict CGA presence from the subset of eyes with GA. For training and testing, 5-fold cross-validation was used. For comparison with human clinician performance, model performance was compared with that of 88 retinal specialists.

MAIN OUTCOME MEASURES

Area under the curve (AUC), accuracy, sensitivity, specificity, and precision.

RESULTS

The deep learning models (GA detection, CGA detection from all eyes, and centrality detection from GA eyes) had AUCs of 0.933-0.976, 0.939-0.976, and 0.827-0.888, respectively. The GA detection model had accuracy, sensitivity, specificity, and precision of 0.965 (95% confidence interval [CI], 0.959-0.971), 0.692 (0.560-0.825), 0.978 (0.970-0.985), and 0.584 (0.491-0.676), respectively, compared with 0.975 (0.971-0.980), 0.588 (0.468-0.707), 0.982 (0.978-0.985), and 0.368 (0.230-0.505) for the retinal specialists. The CGA detection model had values of 0.966 (0.957-0.975), 0.763 (0.641-0.885), 0.971 (0.960-0.982), and 0.394 (0.341-0.448). The centrality detection model had values of 0.762 (0.725-0.799), 0.782 (0.618-0.945), 0.729 (0.543-0.916), and 0.799 (0.710-0.888).

CONCLUSIONS

A deep learning model demonstrated high accuracy for the automated detection of GA. The AUC was noninferior to that of human retinal specialists. Deep learning approaches may also be applied to the identification of CGA. The code and pretrained models are publicly available at https://github.com/ncbi-nlp/DeepSeeNet.

Two-Year Results After AAV2-Mediated Gene Therapy for Choroideremia: The Alberta Experience

PURPOSE

To assess the safety of a recombinant adeno-associated viral vector expressing REP1 (rAAV2.REP1) in choroideremia subjects.

METHODS

Design: Phase I clinical trial.

PARTICIPANTS

Six adult male subjects, 30-42 years of age, with genetically confirmed choroideremia (CHM) were enrolled. The eye with the worse vision, for all subjects, received a single subfoveal injection of 0.1 mL rAAV2.REP1 containing 10¹¹ genome particles. Subjects were followed up for 2 years thereafter.

OUTCOME MEASURES

The primary outcome measure was safety, determined by the number of ocular and systemic adverse events assessed by ophthalmic examination, spectral-domain optical coherence tomography (SD-OCT), and short-wavelength autofluorescence (FAF). Secondary outcome measures were the change from baseline in best-corrected visual acuity (BCVA) in the treated eye compared to the untreated eye, changes in visual function using microperimetry, and the area of retinal pigment epithelium (RPE) preservation by FAF.

RESULTS

One subject had an 8-ETDRS-letter BCVA loss from baseline measured at 24 months, while 1 subject had a ≥15-letter BCVA gain. A similar improvement was noted in the untreated eye of another subject throughout the follow-up period. Microperimetry sensitivity showed no improvement or significant change up to 2 years after vector administration. The area of preserved RPE as measured by FAF was noted to decline at a similar rate between the treated and untreated eyes. One subject experienced a serious adverse event: a localized intraretinal immune response, resulting in marked decline in visual function and loss of SD-OCT outer retinal structures.

CONCLUSIONS

One serious adverse event was experienced in 6 subjects treated with a subfoveal injection of AAV2.REP1. The area of remaining functional RPE in the treated eye and untreated eye declined at the same rate over a 2-year period. Fundus autofluorescence area is a remarkably predictive biomarker and objective outcome measure for future studies of ocular gene therapy in CHM subjects.

Beyond Retinal Layers: A Deep Voting Model for Automated Geographic Atrophy Segmentation in SD-OCT Images

PURPOSE

To automatically and accurately segment geographic atrophy (GA) in spectral-domain optical coherence tomography (SD-OCT) images by constructing a voting system with deep neural networks without the use of retinal layer segmentation.

METHODS

An automatic GA segmentation method for SD-OCT images based on the deep network was constructed. The structure of the deep network was composed of five layers, including one input layer, three hidden layers, and one output layer. During the training phase, the labeled A-scans with 1024 features were directly fed into the network as the input layer to obtain the deep representations. Then a soft-max classifier was trained to determine the label of each individual pixel. Finally, a voting decision strategy was used to refine the segmentation results among 10 trained models.

RESULTS

Two image data sets with GA were used to evaluate the model. For the first dataset, our algorithm obtained a mean overlap ratio (OR) 86.94% ± 8.75%, absolute area difference (AAD) 11.49% ± 11.50%, and correlation coefficients (CC) 0.9857; for the second dataset, the mean OR, AAD, and CC of the proposed method were 81.66% ± 10.93%, 8.30% ± 9.09%, and 0.9952, respectively. The proposed algorithm was capable of improving over 5% and 10% segmentation accuracy, respectively, when compared with several state-of-the-art algorithms on two data sets.

CONCLUSIONS

Without retinal layer segmentation, the proposed algorithm could produce higher segmentation accuracy and was more stable when compared with state-of-the-art methods that relied on retinal layer segmentation results. Our model may provide reliable GA segmentations from SD-OCT images and be useful in the clinical diagnosis of advanced nonexudative AMD.

TRANSLATIONAL RELEVANCE

Based on the deep neural networks, this study presents an accurate GA segmentation method for SD-OCT images without using any retinal layer segmentation results, and may contribute to improved understanding of advanced nonexudative AMD.

Mechanism of accommodation assessed by change in precisely registered ocular images associated with concurrent change in auto-refraction

PURPOSE

Our purpose was to determine the changes in anterior chamber depth (ACD) and central lens thickness (CLT) during pharmacologically induced accommodation.

METHODS

Following pupillary dilation with phenylephrine 10%, baseline auto-refractions and swept-source optical coherence tomographic biometric images (Zeiss IOLMaster 700) were obtained from the right eyes of 25 subjects aged 19 to 24 years. Pilocarpine 4% and phenylephrine 10% were then instilled into these right eyes. One hour later, auto-refractions and biometric imaging were repeated. Only data from eight of 25 subjects met the following stringent criteria to be included in the study analysis: pre and post-pilocarpine biometric foveal images were registerable, the images of the corneal centers were shifted by ≤100 μm, pupils >5 mm and the pharmacologically induced refractive change was ≥ -7 diopters.

RESULTS

The mean auto-refractive accommodative change for the eight included subjects was -12.45 diopters (± 3.45 diopters). The mean change in CLT was 81 μm (± 54 μm) and the mean change in ACD was -145 μm (± 86 μm). Superimposition of the registered pre and post-pilocarpine biometric images of the sagittal sections of the whole eye from each subject demonstrated that the position of the whole lens did not shift either anteriorly, posteriorly or vertically during pharmacologically induced accommodation.

CONCLUSIONS

A small increase in lens thickness was associated with a large change in accommodative amplitude and no significant change in lens position as predicted by the Schachar theory.

Image registration reveals central lens thickness minimally increases during accommodation

Purpose

To evaluate anterior chamber depth, central crystalline lens thickness and lens curvature during accommodation.

Setting

California Retina Associates, El Centro, CA, USA.

Design

Healthy volunteer, prospective, clinical research swept-source optical coherence biometric image registration study of accommodation.

Methods

Ten subjects (4 females and 6 males) with an average age of 22.5 years (range: 20–26 years) participated in the study. A 45° beam splitter attached to a Zeiss IOLMaster 700 (Carl Zeiss Meditec Inc., Jena, Germany) biometer enabled simultaneous imaging of the cornea, anterior chamber, entire central crystalline lens and fovea in the dilated right eyes of subjects before, and during focus on a target 11 cm from the cornea. Images with superimposable foveal images, obtained before and during accommodation, that met all of the predetermined alignment criteria were selected for comparison. This registration requirement assured that changes in anterior chamber depth and central lens thickness could be accurately and reliably measured. The lens radii of curvatures were measured with a pixel stick circle.

Results

Images from only 3 of 10 subjects met the predetermined criteria for registration. Mean anterior chamber depth decreased, −67 μm (range: −0.40 to −110 μm), and mean central lens thickness increased, 117 μm (range: 100–130 μm). The lens surfaces steepened, anterior greater than posterior, while the lens, itself, did not move or shift its position as appeared from the lack of movement of the lens nucleus, during 7.8 diopters of accommodation, (range: 6.6–9.7 diopters).

Conclusion

Image registration, with stable invariant references for image correspondence, reveals that during accommodation a large increase in lens surface curvatures is associated with only a small increase in central lens thickness and no change in lens position.

CLINICAL ENDPOINTS FOR THE STUDY OF GEOGRAPHIC ATROPHY SECONDARY TO AGE-RELATED MACULAR DEGENERATION

PURPOSE

To summarize the recent literature describing the application of modern technologies in the study of patients with geographic atrophy (GA) secondary to age-related macular degeneration.

METHODS

Review of the literature describing the terms and definitions used to describe GA, imaging modalities used to capture and measure GA, and the tests of visual function and functional deficits that occur in patients with GA.

RESULTS

In this paper, we describe the evolution of the definitions used to describe GA. We compare imaging modalities used in the characterization of GA, report on the sensitivity and specificity of the techniques where data exist, and describe the correlations between these various modes of capturing the presence of GA. We review the functional tests that have been used in patients with GA, and critically examine their ability to detect and quantify visual deficits.

CONCLUSION

Ophthalmologists and retina specialists now have a wide range of assessments available for the functional and anatomic characterization of GA in patients with age-related macular degeneration. To date, studies have been limited by their unimodal approach, and we recommend that future studies of GA use multimodal imaging. We also suggest strategies for the optimal functional testing of patients with GA.

Geographic atrophy in patients with advanced dry age-related macular degeneration: current challenges and future prospects

Geographic atrophy (GA) of the retinal pigment epithelium (RPE) is a devastating complication of age-related macular degeneration (AMD). GA may be classified as drusen-related (drusen-associated GA) or neovascularization-related (neovascular-associated GA). Drusen-related GA remains a large public health concern due to the burden of blindness it produces, but pathophysiology of the condition is obscure and there are no proven treatment options. Genotyping, cell biology, and clinical imaging point to upregulation of parainflammatory pathways, oxidative stress, and choroidal sclerosis as contributors, among other factors. Onset and monitoring of progression is accomplished through clinical imaging instrumentation such as optical coherence tomography, photography, and autofluorescence, which are the tools most helpful in determining end points for clinical trials at present. A number of treatment approaches with diverse targets are in development at this time, some of which are in human clinical trials. Neovascular-associated GA is a consequence of RPE loss after development of neovascular AMD. The neovascular process leads to a plethora of cellular stresses such as ischemia, inflammation, and dramatic changes in cell environment that further taxes RPE cells already dysfunctional from drusen-associated changes. GA may therefore develop secondary to the neovascular process de novo or preexisting drusen-associated GA may continue to worsen with the development of neovascular AMD. Neovascular-associated GA is a prominent cause of continued vision loss in patients with otherwise successfully treated neovascular AMD. Clearly, treatment with vascular endothelial growth factor (VEGF) inhibitors early in the course of the neovascular disease is of great clinical benefit. However, there is a rationale and some suggestive evidence that anti-VEGF agents themselves could be toxic to RPE and enhance neovascular-associated GA. The increasing prevalence of legal blindness from this condition due to the aging of the general population lends urgency to the search for a therapy to ameliorate GA.

Automated segmentation of geographic atrophy of the retinal epithelium via random forests in AREDS color fundus images

BACKGROUND

Age-related macular degeneration (AMD), left untreated, is the leading cause of vision loss in people older than 55. Severe central vision loss occurs in the advanced stage of the disease, characterized by either the in growth of choroidal neovascularization (CNV), termed the "wet" form, or by geographic atrophy (GA) of the retinal pigment epithelium (RPE) involving the center of the macula, termed the "dry" form. Tracking the change in GA area over time is important since it allows for the characterization of the effectiveness of GA treatments. Tracking GA evolution can be achieved by physicians performing manual delineation of GA area on retinal fundus images. However, manual GA delineation is time-consuming and subject to inter-and intra-observer variability.

METHODS

We have developed a fully automated GA segmentation algorithm in color fundus images that uses a supervised machine learning approach employing a random forest classifier. This algorithm is developed and tested using a dataset of images from the NIH-sponsored Age Related Eye Disease Study (AREDS). GA segmentation output was compared against a manual delineation by a retina specialist.

RESULTS

Using 143 color fundus images from 55 different patient eyes, our algorithm achieved PPV of 0.82±0.19, and NPV of 0:95±0.07.

DISCUSSION

This is the first study, to our knowledge, applying machine learning methods to GA segmentation on color fundus images and using AREDS imagery for testing. These preliminary results show promising evidence that machine learning methods may have utility in automated characterization of GA from color fundus images.

Automated segmentation of geographic atrophy in fundus autofluorescence images using supervised pixel classification

Geographic atrophy (GA) is a manifestation of the advanced or late stage of age-related macular degeneration (AMD). AMD is the leading cause of blindness in people over the age of 65 in the western world. The purpose of this study is to develop a fully automated supervised pixel classification approach for segmenting GA, including uni- and multifocal patches in fundus autofluorescene (FAF) images. The image features include region-wise intensity measures, gray-level co-occurrence matrix measures, and Gaussian filter banks. A [Formula: see text]-nearest-neighbor pixel classifier is applied to obtain a GA probability map, representing the likelihood that the image pixel belongs to GA. Sixteen randomly chosen FAF images were obtained from 16 subjects with GA. The algorithm-defined GA regions are compared with manual delineation performed by a certified image reading center grader. Eight-fold cross-validation is applied to evaluate the algorithm performance. The mean overlap ratio (OR), area correlation (Pearson's [Formula: see text]), accuracy (ACC), true positive rate (TPR), specificity (SPC), positive predictive value (PPV), and false discovery rate (FDR) between the algorithm- and manually defined GA regions are [Formula: see text], [Formula: see text], [Formula: see text], [Formula: see text], [Formula: see text], [Formula: see text], and [Formula: see text], respectively.