Comparing automated retinal layer segmentation using proprietary software (Heidelberg Spectralis HRA + OCT) and cross-platform Optical Coherence Tomography (OCT) segmentation software (Orion). Image segmentations of normal and diseased (iAMD, DME) eyes were performed using both softwares and then compared to the 'gold standard' of manual segmentation. A qualitative assessment and quantitative (layer volume) comparison of segmentations were performed. Segmented images from the two softwares were graded by two masked graders and in cases with difference, a senior retina specialist made a final independent decisive grading. Cross-platform software was significantly better than the proprietary software in the segmentation of NFL and INL layers in Normal eyes. It generated significantly better segmentation only for NFL in iAMD and for INL and OPL layers in DME eyes. In normal eyes, all retinal layer volumes calculated by the two softwares were moderate-strongly correlated except OUTLY. In iAMD eyes, GCIPL, INL, ONL, INLY, TRV layer volumes were moderate-strongly correlated between softwares. In eyes with DME, all layer volume values were moderate-strongly correlated between softwares. Cross-platform software can be used reliably in research settings to study the retinal layers as it compares well against manual segmentation and the commonly used proprietary software for both normal and diseased eyes.

Multi-modal retinal image registration between 2D Ultra-Widefield (UWF) and narrow-angle (NA) images has not been well-studied, since most existing methods mainly focus on NA image alignment. The stereographic projection model used in UWF imaging causes strong distortions in peripheral areas, which leads to inferior alignment quality. We propose a distortion correction method that remaps the UWF images based on estimated camera view points of NA images. In addition, we set up a CNN-based registration pipeline for UWF and NA images, which consists of the distortion correction method and three networks for vessel segmentation, feature detection and matching, and outlier rejection. Experimental results on our collected dataset shows the effectiveness of the proposed pipeline and the distortion correction method.

Purpose

Macular pigment, composed of lutein, zeaxanthin, and meso-zeaxanthin, is postulated to protect against age-related macular degeneration, likely because of filtering blue light and its antioxidant properties. Macular pigment optical density (MPOD) is reported to be associated with macular function evaluated by visual acuity and multifocal electroretinogram. Given the importance of macular pigment, reliable and accurate measurement methods are important. The main purpose of this study is to determine the reproducibility of MPOD measurement by two-wavelength autofluorescence method using scanning laser ophthalmoscopy.

Methods

Sixty-eight eyes of 39 persons were enrolled in the study, including 11 normal eyes, 16 eyes with wet age-related macular degeneration, 16 eyes with dry age-related macular degeneration, 11 eyes with macular edema due to diabetic mellitus, branch retinal vein occlusion or macular telangiectasia, and 14 eyes with tractional maculopathy, including vitreomacular traction, epiretinal membrane, or macular hole. MPOD was measured with a two-wavelength (488 and 514 nm) autofluorescence method with the Spectralis HRA + OCT after pupil dilation. The measurement was repeated for each eye 10 minutes later. The analysis of variance and Bland-Altman plot were used to assess the reproducibility between the two measurements.

Results

The mean MPOD at eccentricities of 1° and 2° was 0.36 ± 0.17 (range: 0.04-0.69) and 0.15 ± 0.08 (range: -0.03 to 0.35) for the first measurement and 0.35 ± 0.17 (range: 0.02-0.68) and 0.15 ± 0.08 (range: -0.01 to 0.33) for the second measurement, respectively. The difference between the 2 measurements was not statistically significant, and the Bland-Altman plot showed 7.4% and 5.9% points outside the 95% limits of agreement, indicating an overall excellent reproducibility. Similarly, there is no significant difference between the first and second measurements of MPOD volume within eccentricities of 1°, 2°, and 6° radius, and the Bland-Altman plot showed 8.8%, 2.9%, and 4.4% points outside the 95% limits of agreement, respectively. The data for the reproducibility did not differ significantly among the various disease and normal eyes.

Conclusion

Under routine examination conditions with pupil dilation, MPOD measurement by two-wavelength autofluorescence method showed a high reproducibility.

Multi-modal retinal image registration plays an important role in the ophthalmological diagnosis process. The conventional methods lack robustness in aligning multi-modal images of various imaging qualities. Deep-learning methods have not been widely developed for this task, especially for the coarse-to-fine registration pipeline. To handle this task, we propose a two-step method based on deep convolutional networks, including a coarse alignment step and a fine alignment step. In the coarse alignment step, a global registration matrix is estimated by three sequentially connected networks for vessel segmentation, feature detection and description, and outlier rejection, respectively. In the fine alignment step, a deformable registration network is set up to find pixel-wise correspondence between a target image and a coarsely aligned image from the previous step to further improve the alignment accuracy. Particularly, an unsupervised learning framework is proposed to handle the difficulties of inconsistent modalities and lack of labeled training data for the fine alignment step. The proposed framework first changes multi-modal images into a same modality through modality transformers, and then adopts photometric consistency loss and smoothness loss to train the deformable registration network. The experimental results show that the proposed method achieves state-of-the-art results in Dice metrics and is more robust in challenging cases.

Optical Coherence Tomography (OCT) is a powerful technique for non-invasive 3D imaging of biological tissues at high resolution that has revolutionized retinal imaging. A major challenge in OCT imaging is the motion artifacts introduced by involuntary eye movements. In this paper, we propose a convolutional neural network that learns to correct axial motion in OCT based on a single volumetric scan. The proposed method is able to correct large motion, while preserving the overall curvature of the retina. The experimental results show significant improvements in visual quality as well as overall error compared to the conventional methods in both normal and disease cases.

We previously identified a homozygous G178R mutation in human ASRGL1 (hASRGL1) through whole-exome analysis responsible for early onset retinal degeneration (RD) in patients with cone-rod dystrophy. The mutant G178R ASRGL1 expressed in Cos-7 cells showed altered localization, while the mutant ASRGL1 in E. coli lacked the autocatalytic activity needed to generate the active protein. To evaluate the effect of impaired ASRGL1 function on the retina in vivo, we generated a mouse model with c.578_579insAGAAA (NM_001083926.2) mutation (Asrgl1^mut/mut) through the CRISPR/Cas9 methodology. The expression of ASGRL1 and its asparaginase activity were undetectable in the retina of Asrgl1^mut/mut mice. The ophthalmic evaluation of Asrgl1^mut/mut mice showed a significant and progressive decrease in scotopic electroretinographic (ERG) response observed at an early age of 3 months followed by a decrease in photopic response around 5 months compared with age-matched wildtype mice. Immunostaining and RT-PCR analyses with rod and cone cell markers revealed a loss of cone outer segments and a significant decrease in the expression of Rhodopsin, Opn1sw, and Opn1mw at 3 months in Asrgl1^mut/mut mice compared with age-matched wildtype mice. Importantly, the retinal phenotype of Asrgl1^mut/mut mice is consistent with the phenotype observed in patients harboring the G178R mutation in ASRGL1 confirming a critical role of ASRGL1 in the retina and the contribution of ASRGL1 mutations in retinal degeneration.

Objective

To develop automated algorithms for the detection of posterior vitreous detachment (PVD) using OCT imaging.

Design

Evaluation of a diagnostic test or technology.

Subjects

Overall, 42 385 consecutive OCT images (865 volumetric OCT scans) obtained with Heidelberg Spectralis from 865 eyes from 464 patients at an academic retina clinic between October 2020 and December 2021 were retrospectively reviewed.

Methods

We developed a customized computer vision algorithm based on image filtering and edge detection to detect the posterior vitreous cortex for the determination of PVD status. A second deep learning (DL) image classification model based on convolutional neural networks and ResNet-50 architecture was also trained to identify PVD status from OCT images. The training dataset consisted of 674 OCT volume scans (33 026 OCT images), while the validation testing set consisted of 73 OCT volume scans (3577 OCT images). Overall, 118 OCT volume scans (5782 OCT images) were used as a separate external testing dataset.

Main outcome measures

Accuracy, sensitivity, specificity, F1-scores, and area under the receiver operator characteristic curves (AUROCs) were measured to assess the performance of the automated algorithms.

Results

Both the customized computer vision algorithm and DL model results were largely in agreement with the PVD status labeled by trained graders. The DL approach achieved an accuracy of 90.7% and an F1-score of 0.932 with a sensitivity of 100% and a specificity of 74.5% for PVD detection from an OCT volume scan. The AUROC was 89% at the image level and 96% at the volume level for the DL model. The customized computer vision algorithm attained an accuracy of 89.5% and an F1-score of 0.912 with a sensitivity of 91.9% and a specificity of 86.1% on the same task.

Conclusions

Both the computer vision algorithm and the DL model applied on OCT imaging enabled reliable detection of PVD status, demonstrating the potential for OCT-based automated PVD status classification to assist with vitreoretinal surgical planning.

Financial disclosures

Proprietary or commercial disclosure may be found after the references.

Purpose

This study aimed to compare a new Artificial Intelligence (AI) method to conventional mathematical warping in accurately overlaying peripheral retinal vessels from two different imaging devices: confocal scanning laser ophthalmoscope (cSLO) wide-field images and SLO ultra-wide field images.

Methods

Images were captured using the Heidelberg Spectralis 55-degree field-of-view and Optos ultra-wide field. The conventional mathematical warping was performed using Random Sample Consensus-Sample and Consensus sets (RANSAC-SC). This was compared to an AI alignment algorithm based on a one-way forward registration procedure consisting of full Convolutional Neural Networks (CNNs) with Outlier Rejection (OR CNN), as well as an iterative 3D camera pose optimization process (OR CNN + Distortion Correction [DC]). Images were provided in a checkerboard pattern, and peripheral vessels were graded in four quadrants based on alignment to the adjacent box.

Results

A total of 660 boxes were analysed from 55 eyes. Dice scores were compared between the three methods (RANSAC-SC/OR CNN/OR CNN + DC): 0.3341/0.4665/4784 for fold 1-2 and 0.3315/0.4494/4596 for fold 2-1 in composite images. The images composed using the OR CNN + DC have a median rating of 4 (out of 5) versus 2 using RANSAC-SC. The odds of getting a higher grading level are 4.8 times higher using our OR CNN + DC than RANSAC-SC (p < 0.0001).

Conclusion

Peripheral retinal vessel alignment performed better using our AI algorithm than RANSAC-SC. This may help improve co-localizing retinal anatomy and pathology with our algorithm.

Background

Retinitis pigmentosa (RP) represents a group of progressive, genetically heterogenous blinding diseases. Recently, relationships between measures of retinal function and structure are needed to help identify outcome measures or biomarkers for clinical trials. The ability to align retinal multimodal images, taken on different platforms, will allow better understanding of this relationship. We investigate the efficacy of artificial intelligence (AI) in overlaying different multimodal retinal images in RP patients.

Methods

We overlayed infrared images from microperimetry on near-infra-red images from scanning laser ophthalmoscope and spectral domain optical coherence tomography in RP patients using manual alignment and AI. The AI adopted a two-step framework and was trained on a separate dataset. Manual alignment was performed using in-house software that allowed labelling of six key points located at vessel bifurcations. Manual overlay was considered successful if the distance between same key points on the overlayed images was ≤1/2°.

Results

Fifty-seven eyes of 32 patients were included in the analysis. AI was significantly more accurate and successful in aligning images compared to manual alignment as confirmed by linear mixed-effects modelling (p < 0.001). A receiver operating characteristic analysis, used to compute the area under the curve of the AI (0.991) and manual (0.835) Dice coefficients in relation to their respective 'truth' values, found AI significantly more accurate in the overlay (p < 0.001).

Conclusion

AI was significantly more accurate than manual alignment in overlaying multimodal retinal imaging in RP patients and showed the potential to use AI algorithms for future multimodal clinical and research applications.