Automatic Three-dimensional Detection of Photoreceptor Ellipsoid Zone Disruption Caused by Trauma in the OCT

Morphological Biomarkers Related to Visual Acuity in Patients With Radiation Retinopathy Treated With Intravitreal Ranibizumab

BACKGROUND AND OBJECTIVE

Our objective was to monitor variables via spectral-domain optical coherence tomography (SD-OCT) and identify the most relevant biomarkers related to best-corrected visual acuity (BCVA) in radiation retinopathy (RR).

PATIENTS AND METHODS

A post-hoc analysis of the two-year Ranibizumab for Radiation Retinopathy (RRR) trial analyzed vision and OCT parameters including intraretinal fluid, ellipsoid zone (EZ) disruption, retinal pigment epithelium atrophy, hard exudates, retinal hemorrhage, retinal neovascularization, and subfoveal fluid. BCVA and SD-OCT parameters were evaluated by univariate analysis and a mixed-effects model.

RESULTS

Forty eyes from the RRR trial were included. Intraretinal cyst vertical size (week 24: P = 0.032; week 48: P = 0.021), neovascularization (week 48: P = 0.028; week 72: P = 0.025), and EZ disruption (week 72: P = 0.029; week 104: P = 0.019) were the clinical parameters most relevant to BCVA by univariate analysis in at least two time points. The mixed-effects model confirmed the relevance of intraretinal cyst vertical size (P = 0.001) and neovascularization (P = 0.001) but not EZ disruption (P = 0.119) over the course of the study.

CONCLUSIONS

This study characterizes the course of visual loss in RR by identifying intraretinal cyst vertical size, neovascularization, and EZ disruption as biomarkers of poor BCVA over a span of two years. Larger multicenter studies are needed to confirm these findings. [Ophthalmic Surg Lasers Imaging Retina 2024;55:255-262.].

Automated Identification and Segmentation of Ellipsoid Zone At-Risk Using Deep Learning on SD-OCT for Predicting Progression in Dry AMD

BACKGROUND

The development and testing of a deep learning (DL)-based approach for detection and measurement of regions of Ellipsoid Zone (EZ) At-Risk to study progression in nonexudative age-related macular degeneration (AMD).

METHODS

Used in DL model training and testing were 341 subjects with nonexudative AMD with or without geographic atrophy (GA). An independent dataset of 120 subjects were used for testing model performance for prediction of GA progression. Accuracy, specificity, sensitivity, and intraclass correlation coefficient (ICC) for DL-based EZ At-Risk percentage area measurement was calculated. Random forest-based feature ranking of EZ At-Risk was compared to previously validated quantitative OCT-based biomarkers.

RESULTS

The model achieved a detection accuracy of 99% (sensitivity = 99%; specificity = 100%) for EZ At-Risk. Automatic EZ At-Risk measurement achieved an accuracy of 90% (sensitivity = 90%; specificity = 84%) and the ICC compared to ground truth was high (0.83). In the independent dataset, higher baseline mean EZ At-Risk correlated with higher progression to GA at year 5 (p < 0.001). EZ At-Risk was a top ranked feature in the random forest assessment for GA prediction.

CONCLUSIONS

This report describes a novel high performance DL-based model for the detection and measurement of EZ At-Risk. This biomarker showed promising results in predicting progression in nonexudative AMD patients.

Impact of Baseline Quantitative OCT Features on Response to Risuteganib for the Treatment of Dry Age-Related Macular Degeneration: The Importance of Outer Retinal Integrity

OBJECTIVE

The purpose of the study was to perform a post hoc analysis to explore the effect of baseline anatomic characteristics identified on OCT on best-corrected visual acuity (BCVA) responses to risuteganib from the completed phase II study in subjects with dry age-related macular degeneration (AMD).

DESIGN

Post hoc analysis of a randomized, double-masked, placebo-controlled, phase II study.

SUBJECTS

Eyes with intermediate dry AMD with BCVA between 20/40 and 20/200. Patients with concurrent vision-influencing or macula-obscuring ocular pathologies were excluded.

METHODS

Patients were randomized to receive a 1-mg intravitreal risuteganib injection or a sham injection at baseline. A second 1-mg intravitreal injection of risuteganib was given at week 16 to those in the treatment arm. Two independent, masked reading centers evaluated the baseline anatomic characteristics on OCT to explore features associated with positive responses to risuteganib.

MAIN OUTCOME MEASURES

Treatment response was defined as a gain of ≥ 8 letters in BCVA from baseline to week 28 in the treatment arm, compared with baseline to week 12 in the sham group. Anatomic parameters, measured by retinal segmentation platforms, including measures of retinal thickness were compared between the responders and nonresponders to risuteganib.

RESULTS

Thirty-nine patients completed the study and underwent analysis. In the treatment arm, 48% of eyes demonstrated treatment responses, compared with 7% in the sham group. In the quantitative anatomic assessment, enhanced ellipsoid integrity, greater outer retinal thickness, and decreased geographic atrophy were associated with increased BCVA gains to risuteganib.

CONCLUSIONS

This post hoc analysis demonstrated that baseline OCT features may help determine the likelihood of a functional response to risuteganib. The characterization of higher-order OCT features may provide important information regarding biomarkers for treatment response and could facilitate optimized clinical trial enrollment and enrichment.

VALIDATION OF A DEEP LEARNING-BASED ALGORITHM FOR SEGMENTATION OF THE ELLIPSOID ZONE ON OPTICAL COHERENCE TOMOGRAPHY IMAGES OF AN USH2A-RELATED RETINAL DEGENERATION CLINICAL TRIAL

PURPOSE

To assess the generalizability of a deep learning-based algorithm to segment the ellipsoid zone (EZ).

METHODS

The dataset consisted of 127 spectral-domain optical coherence tomography volumes from eyes of participants with USH2A-related retinal degeneration enrolled in the RUSH2A clinical trial (NCT03146078). The EZ was segmented manually by trained readers and automatically by deep OCT atrophy detection, a deep learning-based algorithm originally developed for macular telangiectasia Type 2. Performance was evaluated using the Dice similarity coefficient between the segmentations, and the absolute difference and Pearson's correlation of measurements of interest obtained from the segmentations.

RESULTS

With deep OCT atrophy detection, the average (mean ± SD, median) Dice similarity coefficient was 0.79 ± 0.27, 0.90. The average absolute difference in total EZ area was 0.62 ± 1.41, 0.22 mm2 with a correlation of 0.97. The average absolute difference in the maximum EZ length was 222 ± 288, 126 µm with a correlation of 0.97.

CONCLUSION

Deep OCT atrophy detection segmented EZ in USH2A-related retinal degeneration with good performance. The algorithm is potentially generalizable to other diseases and other biomarkers of interest as well, which is an important aspect of clinical applicability.

Retrospective Study of Ellipsoid Zone Integrity Following Treatment with Intravitreal Ocriplasmin (OZONE Study)

Purpose

To assess generalized (GD) and focal ellipsoid zone disruption (FD) in patients with symptomatic vitreomacular adhesion (sVMA) using spectral domain optical coherence tomography (SD-OCT) following ocriplasmin.

Patients and methods

OZONE was a Phase 4, retrospective study of patients with sVMA treated with a single intravitreal injection of ocriplasmin (0.125 mg). Data from adult patients with at least 6-month follow-up after ocriplasmin were included. SD-OCT was performed at baseline (within 30 days before ocriplasmin), before Day 21 post-injection (early observation, EO), and by last observation (LO) which was maximally 6 months post-injection. The main outcome measure was the development of new and the evolution of existing FD/GD at EO and LO.

Results

The study enrolled 134 eyes/patients from 22 sites in the USA. At baseline, 87 eyes (64.9%) had FD, 21 eyes (15.7%) had GD and 26 eyes (19.4%) had no FD/GD. Among the eyes without FD/GD at baseline, 13 (50%) and 8 (30.8%) developed FD or GD, respectively, by EO. By LO, FD/GD improvement or resolution was seen in >80% of these eyes. Among the eyes with FD/GD at baseline, <40% had improving/resolving EZ integrity at LO. The absence of FD/GD at baseline was associated with less persistent FD/GD at LO (P<0.0005). The presence of FD with MH at baseline was associated with persistent FD at LO (P=0.027).

Conclusion

The fact that a large majority of eyes had FD/GD prior to ocriplasmin was unexpected and demonstrates that EZ disruptions are common in sVMA. This suggests that loss of EZ integrity may be part of the natural history of this disorder. It is hypothesized that the status of the EZ at baseline is a contributing, ocriplasmin independent modulator of subsequent EZ changes after ocriplasmin. Prospective analyses which include a sham control group would be required to test this hypothesis.

Semi-Supervised Capsule cGAN for Speckle Noise Reduction in Retinal OCT Images

Speckle noise is the main cause of poor optical coherence tomography (OCT) image quality. Convolutional neural networks (CNNs) have shown remarkable performances for speckle noise reduction. However, speckle noise denoising still meets great challenges because the deep learning-based methods need a large amount of labeled data whose acquisition is time-consuming or expensive. Besides, many CNNs-based methods design complex structure based networks with lots of parameters to improve the denoising performance, which consume hardware resources severely and are prone to overfitting. To solve these problems, we propose a novel semi-supervised learning based method for speckle noise denoising in retinal OCT images. First, to improve the model's ability to capture complex and sparse features in OCT images, and avoid the problem of a great increase of parameters, a novel capsule conditional generative adversarial network (Caps-cGAN) with small number of parameters is proposed to construct the semi-supervised learning system. Then, to tackle the problem of retinal structure information loss in OCT images caused by lack of detailed guidance during unsupervised learning, a novel joint semi-supervised loss function composed of unsupervised loss and supervised loss is proposed to train the model. Compared with other state-of-the-art methods, the proposed semi-supervised method is suitable for retinal OCT images collected from different OCT devices and can achieve better performance even only using half of the training data.

TOPOGRAPHIC OPTICAL COHERENCE TOMOGRAPHY SEGMENTATION SHOWS LIMITED ELLIPSOID ZONE RECOVERY IN MILD HYDROXYCHLOROQUINE RETINOPATHY

PURPOSE

Optical coherence tomography (OCT) cross-sections have shown limited ellipsoid zone (EZ) improvement in mild hydroxychloroquine (HCQ) retinopathy within a few years after drug cessation. However, the extent, functional significance, and stability of such changes over time remain unclear.

METHODS

We created en face EZ maps using automated pixel-by-pixel segmentation for four patients with early-moderate HCQ toxicity followed for 6-8 years after drug cessation. These maps were compared with OCT cross-sections, fundus autofluorescence, and automated 10-2 visual fields.

RESULTS

One patient had no EZ line loss; one had stable EZ loss throughout follow-up; two showed 30 to 40% reduction in the area of loss, largely in the first 2 years. This limited recovery mostly occurred in regions where the EZ line was only thinned or fragmented; other similar areas did not improve. Fundus autofluorescence hyperfluorescence and visual fields did not show consistent correlation with topography.

CONCLUSION

Anatomic EZ recovery, when present, was restricted to regions of mild damage and did not correlate with fundus autofluorescence or improvement in visual fields. Topographic mapping seemed no more sensitive locally than cross-sectional OCT but may aid detection and longitudinal follow-up of toxicity by showing early damage or changes in the macula that could be missed with individual cross-sections.

Beyond Performance Metrics: Automatic Deep Learning Retinal OCT Analysis Reproduces Clinical Trial Outcome

PURPOSE

To validate the efficacy of a fully automatic, deep learning-based segmentation algorithm beyond conventional performance metrics by measuring the primary outcome of a clinical trial for macular telangiectasia type 2 (MacTel2).

DESIGN

Evaluation of diagnostic test or technology.

PARTICIPANTS

A total of 92 eyes from 62 participants with MacTel2 from a phase 2 clinical trial (NCT01949324) randomized to 1 of 2 treatment groups METHODS: The ellipsoid zone (EZ) defect areas were measured on spectral domain OCT images of each eye at 2 time points (baseline and month 24) by a fully automatic, deep learning-based segmentation algorithm. The change in EZ defect area from baseline to month 24 was calculated and analyzed according to the clinical trial protocol.

MAIN OUTCOME MEASURE

Difference in the change in EZ defect area from baseline to month 24 between the 2 treatment groups.

RESULTS

The difference in the change in EZ defect area from baseline to month 24 between the 2 treatment groups measured by the fully automatic segmentation algorithm was 0.072±0.035 mm² (P = 0.021). This was comparable to the outcome of the clinical trial using semiautomatic measurements by expert readers, 0.065±0.033 mm² (P = 0.025).

CONCLUSIONS

The fully automatic segmentation algorithm was as accurate as semiautomatic expert segmentation to assess EZ defect areas and was able to reliably reproduce the statistically significant primary outcome measure of the clinical trial. This approach, to validate the performance of an automatic segmentation algorithm on the primary clinical trial end point, provides a robust gauge of its clinical applicability.

IDENTIFICATION AND CLASSIFICATION OF MACULAR MORPHOLOGIC BIOMARKERS RELATED TO VISUAL ACUITY IN RADIATION MACULOPATHY: A Multimodal Imaging Study

PURPOSE

To identify and classify, by a multimodal imaging approach, the most relevant macular morphologic biomarkers related to visual acuity in patients affected by radiation maculopathy secondary to brachytherapy.

METHODS

Fifty-one consecutive patients previously treated with Iodine-125 brachytherapy because of uveal melanoma were enrolled. Each patient underwent full ophthalmologic examination including best-corrected visual acuity and multimodal macular imaging analysis. Macular morphological parameters were processed by a stepwise selection analysis.

RESULTS

Three macular parameters were identified as the most relevant macular morphologic biomarkers of poor visual acuity: the vertical thickness of the thickest macular cyst (P = 0.0001), the presence of foveal inner segment/outer segment (IS/OS) layer disruption (P = 0.0054), and the presence of foveal retinal pigment epithelium atrophy (0.0884). The intergrader agreement for these morphologic biomarkers was 0.98, 0.92, and 0.92, respectively (interclass correlation coefficient).

CONCLUSION

The vertical thickness of the thickest macular cyst, the presence of foveal retinal pigment epithelium atrophy, and IS/OS layer disruption can be used to clinically characterize radiation maculopathy. These parameters allow for separation of the edematous component of radiation maculopathy, which is potentially treatable in early disease stages, from late onset atrophic components, which are theoretically irreversible.

OIPAV: an Integrated Software System for Ophthalmic Image Processing, Analysis, and Visualization

Ophthalmic medical images, such as optical coherence tomography (OCT) images and color photo of fundus, provide valuable information for clinical diagnosis and treatment of ophthalmic diseases. In this paper, we introduce a software system specially oriented to ophthalmic images processing, analysis, and visualization (OIPAV) to assist users. OIPAV is a cross-platform system built on a set of powerful and widely used toolkit libraries. Based on the plugin mechanism, the system has an extensible framework. It provides rich functionalities including data I/O, image processing, interaction, ophthalmic diseases detection, data analysis, and visualization. By using OIPAV, users can easily access to the ophthalmic image data manufactured from different imaging devices, facilitate workflows of processing ophthalmic images, and improve quantitative evaluations. With a satisfying function scalability and expandability, the software is applicable for both ophthalmic researchers and clinicians.

Deep learning for the segmentation of preserved photoreceptors on en face optical coherence tomography in two inherited retinal diseases

The objective quantification of photoreceptor loss in inherited retinal degenerations (IRD) is essential for measuring disease progression, and is now especially important with the growing number of clinical trials. Optical coherence tomography (OCT) is a non-invasive imaging technology widely used to recognize and quantify such anomalies. Here, we implement a versatile method based on a convolutional neural network to segment the regions of preserved photoreceptors in two different IRDs (choroideremia and retinitis pigmentosa) from OCT images. An excellent segmentation accuracy (~90%) was achieved for both IRDs. Due to the flexibility of this technique, it has potential to be extended to additional IRDs in the future.

Identification of biomarker sets for predicting the efficacy of sublingual immunotherapy against pollen-induced allergic rhinitis

Sublingual immunotherapy (SLIT) is effective against allergic rhinitis, although a substantial proportion of individuals is refractory. Herein, we describe a predictive modality to reliably identify SLIT non-responders (NRs). We conducted a 2-year clinical study in 193 adult patients with Japanese cedar pollinosis, with biweekly administration of 2000 Japanese allergy units of cedar pollen extract as the maintenance dose. After identifying high-responder (HR) patients with improved severity scores and NR patients with unchanged or exacerbated symptoms, differences in 33 HR and 34 NR patients were evaluated in terms of peripheral blood cellular profiles by flow cytometry and serum factors by ELISA and cytokine bead array, both pre- and post-SLIT. Improved clinical responses were seen in 72% of the treated patients. Pre-therapy IL-12p70 and post-therapy IgG1 serum levels were significantly different between HR and NR patients, although these parameters alone failed to distinguish NR from HR patients. However, the analysis of serum parameters in the pre-therapy samples with the Adaptive Boosting (AdaBoost) algorithm distinguished NR patients with high probability within the training data set. Cluster analysis revealed a positive correlation between serum Th1/Th2 cytokines and other cytokines/chemokines in HR patients after SLIT. Thus, processing of pre-therapy serum parameters with AdaBoost and cluster analysis can be reliably used to develop a prediction method for HR/NR patients.

Automated detection of photoreceptor disruption in mild diabetic retinopathy on volumetric optical coherence tomography

Diabetic retinopathy is a pathology where microvascular circulation abnormalities ultimately result in photoreceptor disruption and, consequently, permanent loss of vision. Here, we developed a method that automatically detects photoreceptor disruption in mild diabetic retinopathy by mapping ellipsoid zone reflectance abnormalities from en face optical coherence tomography images. The algorithm uses a fuzzy c-means scheme with a redefined membership function to assign a defect severity level on each pixel and generate a probability map of defect category affiliation. A novel scheme of unsupervised clustering optimization allows accurate detection of the affected area. The achieved accuracy, sensitivity and specificity were about 90% on a population of thirteen diseased subjects. This method shows potential for accurate and fast detection of early biomarkers in diabetic retinopathy evolution.