Quantification of retinal layer thickness changes in acute macular neuroretinopathy

Automated assessment of the smoothness of retinal layers in optical coherence tomography images using a machine learning algorithm

Quantifying the smoothness of different layers of the retina can potentially be an important and practical biomarker in various pathologic conditions like diabetic retinopathy. The purpose of this study is to develop an automated machine learning algorithm which uses support vector regression method with wavelet kernel and automatically segments two hyperreflective retinal layers (inner plexiform layer (IPL) and outer plexiform layer (OPL)) in 50 optical coherence tomography (OCT) slabs and calculates the smoothness index (SI). The Bland-Altman plots, mean absolute error, root mean square error and signed error calculations revealed a modest discrepancy between the manual approach, used as the ground truth, and the corresponding automated segmentation of IPL/ OPL, as well as SI measurements in OCT slabs. It was concluded that the constructed algorithm may be employed as a reliable, rapid and convenient approach for segmenting IPL/OPL and calculating SI in the appropriate layers.

Multimodal imaging in Schubert-Bornschein congenital stationary night blindness

BACKGROUND

Schubert-Bornschein (SB) is the most common type of people with congenital stationary night blindness (CSNB). The aim of the study is to describe the optical coherence tomography (OCT) and optical coherence tomography angiography (OCTA) findings in patients with SB CSNB.

METHODS

Prospective, observational case series including three patients with genetically confirmed CSNB along with matched controls, who underwent complete ophthalmic examination and multimodal imaging.

RESULTS

On SD-OCT, a significant focal outer plexiform layer (OPL) thickening and a corresponding focal outer nuclear layer (ONL) thinning were identified in the macular area (p < 0.001). OCTA analysis overall showed decreased density of macular deep capillary plexus (mDCP) and macular choriocapillaris (mCC) (p = 0.008 and p = 0.033, respectively). DCP vessel density in the area corresponding to OPL thickening was significantly increased compared to the remaining retina (p < 0.001).

CONCLUSION

SB CSNB is characterized by retinal vascular impairment, as detected on OCTA.

Acute macular neuroretinopathy in dengue virus serotype 1

Purpose

To report a case of acute macular neuroretinopathy (AMN) associated with dengue virus serotype 1 infection.

Observation

An 18-year-old Puerto Rican female was evaluated due to painless paracentral scotomas in each eye that developed after being hospitalized for dengue fever a week before. Clinical examination and multimodal imaging revealed bilateral hypopigmented macular lesions, hyperreflectivity at the outer nuclear and photoreceptor layer, and reduced flow signal in the deep capillary plexus. Additionally, hypoautofluorescent parafoveal lesions were found in the left eye. AMN was diagnosed. Two-month follow-up after the initial evaluation showed resolution of symptoms but persistence of some findings on optical coherence tomography.

Conclusions and importance

Patients with dengue virus serotype 1 may develop paracentral scotomas with classic AMN findings and obtain complete symptomatic recovery without treatment.

Assessment of patient specific information in the wild on fundus photography and optical coherence tomography

In this paper we analyse the performance of machine learning methods in predicting patient information such as age or sex solely from retinal imaging modalities in a heterogeneous clinical population. Our dataset consists of N = 135,667 fundus images and N = 85,536 volumetric OCT scans. Deep learning models were trained to predict the patient’s age and sex from fundus images, OCT cross sections and OCT volumes. For sex prediction, a ROC AUC of 0.80 was achieved for fundus images, 0.84 for OCT cross sections and 0.90 for OCT volumes. Age prediction mean absolute errors of 6.328 years for fundus, 5.625 years for OCT cross sections and 4.541 for OCT volumes were observed. We assess the performance of OCT scans containing different biomarkers and note a peak performance of AUC = 0.88 for OCT cross sections and 0.95 for volumes when there is no pathology on scans. Performance drops in case of drusen, fibrovascular pigment epitheliuum detachment and geographic atrophy present. We conclude that deep learning based methods are capable of classifying the patient’s sex and age from color fundus photography and OCT for a broad spectrum of patients irrespective of underlying disease or image quality. Non-random sex prediction using fundus images seems only possible if the eye fovea and optic disc are visible.

Longitudinal Retinal Layer Changes and Clinical Outcome in Patients with Multiple Evanescent White Dot Syndrome

Purpose: To evaluate longitudinal changes in retinal layer thickness and clinical outcome in patients with MEWDS.Methods: In 20 patients with MEWDS, SD-OCT images and BCVA were assessed at baseline, and at months 1, 3, and 12. SD-OCTs were segmented and measurements were performed within the fovea and a MEWDS lesion. Baseline and follow-up values in the affected eye were compared to measurements performed at the corresponding location in the fellow eye.Results: ONL thickness was 4.7% thicker in MEWDS-eyes compared with the baseline, with a significant decrease of 9% at 3 months. Within the lesion, INL thickness was 7.9% increased at baseline and decreased significantly over the follow-up of 12 months. BCVA was decreased at baseline (0.2 ± 0.18logMAR) and at the 3 months but after 12 months had increased to 0.01 ± 0.04 logMAR.Conclusion: MEWDS shows the involvement of different retinal layers and characteristic changes over the disease course.

AUTOMATED RETINAL LAYER SEGMENTATION AND THEIR THICKNESS PROFILES IN HEALTHY SUBJECTS: A Comparison of 55° Wide-field and Conventional 30° Spectral Domain-Optical Coherence Tomography

Fifty-five degree wide-field spectral domain optical coherence tomography provides an alternative to standard 30° optical coherence tomography lens with comparable thickness values with a greater field of view.

Purpose:

To assess whether retinal thickness measurements with a standard 30° spectral domain optical coherence tomography (SD-OCT) are comparable with wide-field 55° SD-OCT.

Methods:

Thirty-three healthy individuals were scanned using 55° as well as 30° SD-OCT according to a standardized protocol. Automated retinal layer segmentation of standard and wide-field SD-OCTs was assessed using customized software.

Results:

Both lenses showed a high correlation when analyzing total retinal thickness within the central, the inner, and the outer retinal ring (r = > 0.9). Automated thickness measurements with the 55° system were marginally higher compared with the 30° lens. The thickness of each separate retinal layer using automated segmentation showed excellent correlations within the inner and outer rings (range: r = 0.6—r = 0.9 for the inner ring and range: r = 0.9—r = 1.0 for the outer ring).

Conclusion:

Fifty-five degree wide-field SD-OCT provides a good overview of the posterior pole and presents similar quantitative values as a standard 30° OCT lens. Therefore, thickness values are comparable when switching between these two lenses.

DIRECTIONAL VARIABILITY OF FUNDUS REFLECTANCE IN ACUTE MACULAR NEURORETINOPATHY: EVIDENCE FOR A CONTRIBUTION OF THE STILES-CRAWFORD EFFECT

PURPOSE

To document directional reflectivity of fundus lesions in a case of acute macular neuroretinopathy.

METHODS

Case report. Clinical and imaging data from a patient with acute macular neuroretinopathy were reviewed. Imaging comprised infrared scanning laser ophthalmoscopy, optical coherence tomography and flood-illumination adaptive optics images acquired through different entry pupils in the cardinal directions (approximately 2° eccentricity).

RESULTS

The patient reported acute bilateral paracentral scotoma revealing dark, wedge-shaped macular plaques which by optical coherence tomography were associated with focal loss of the visibility of the cone outer segment tip and inner/outer segment lines. Comparing scanning laser ophthalmoscopy images taken at different entry points in the pupil showed that macular plaques varied from hyporeflectance to isoreflectance. Cone counts by flood-illumination adaptive optics within plaques and optical coherence tomography features of the cone outer segment tip showed also a strong directional variability, peaking at near-normal values. Within each modality, fusion images showed that directional variability covered most of macular plaques.

CONCLUSION

The characteristic fundus abnormalities of acute macular neuroretinopathy may show a strong directional variability. Our findings suggest that the Stiles-Crawford effect may be an important factor in signs and symptoms of acute macular neuroretinopathy.

Acute macular neuroretinopathy: pathogenetic insights from optical coherence tomography angiography

Aim

To describe the optical coherence tomography angiography (OCT-A) features of patients affected by acute macular neuroretinopathy (AMN).

Methods

This is a prospective, observational, cross-sectional study. Multimodal imaging, including spectral domain OCT (Spectralis HRA+OCT; Heidelberg Engineering, Heidelberg, Germany) and 6×6 mm swept source OCT-A (ZEISS PLEX Elite 9000; ZEISS, Dublin, California), was performed on all patients. The primary outcome measure was assessment of the vascular alterations of the superficial (SCP) and deep (DCP) capillary plexuses and choriocapillaris (CC). A segmentation of the AMN lesions using enface OCT images of photoreceptors-retinal pigment epithelium complex was used to discriminate intralesional and extralesional regions on the OCT-A enface slabs of the DCP and CC reconstructions. Each OCT-A slab was imported into ImageJ V.1.50 and digitally binarised for quantitative analyses.

Results

Overall, seven patients (mean age 19.4±3.2 years, six women) affected by AMN were included. The mean best-corrected visual acuity was 0.00±0.00 logarithm of the minimum angle of resolution. Twelve healthy age-matched subjects (mean age 22±3.4 years, 10 women) represented the control group. The quantitative analysis of global vessel densities showed that the CC vessel density was significantly lower in patients with AMN (0.495±0.03) compared with the age-matched controls (0.545±0.02) (corrected p=0.0003). The intralesional vessel density of the DCP was 0.349±0.04 in AMN and vs 0.497±0.02 in the controls (corrected p=0.0002).

Conclusions

Our study confirmed inner choroidal vascular flow void as a possible pathogenetic mechanism of AMN. We also found a focal impairment of the DCP within the AMN lesions. Future studies are needed to clarify which is the primary location of the vascular insult in this condition.