Deep learning-based diagnostic classification of multiple sclerosis using multicenter optical coherence tomography data.

AimTo examine the relationship between retinal ganglion cell loss and changes in the inner nuclear layer (INL) in optic neuritis (ON).Methods36 multiple sclerosis (MS) patients with a history of ON and 36 age and sex-matched controls underwent Optical Coherence Tomography. The paramacular retinal nerve fiber layer (RNFL), combined ganglion cell and inner plexiform layers (GCL/IPL) and inner nuclear layer (INL) thickness were measured at 36 points around the fovea. To remove inter-subject variability, the difference in thickness of each layer between the ON and fellow eye of each patient was calculated. A topographic analysis was conducted.ResultsThe INL of the ON patients was thicker than the controls (42.9µm versus 39.6µm, p=0.002). ON patients also had a thinner RNFL (27.8µm versus 32.2µm, p<0.001) and GCL/IPL (69.3µm versus 98.1µm, p<0.001). Among the controls, there was no correlation between RNFL and GCL/IPL as well as RNFL and INL, but a positive correlation was seen between GCL/IPL and INL (r=0.65, p<0.001). In the ON group, there was a positive correlation between RNFL and GCL/IPL (r=0.80, p<0.001) but a negative correlation between RNFL and INL (r=-0.61, p<0.001) as well as GCL/IPL and INL (r=-0.44, p=0.007). The negative correlation between GCL/IPL and INL strengthened in the ON group when inter-subject variability was removed (r=-0.75, p<0.001). Microcysts within the INL were present in 5 ON patients, mainly in the superior and infero-nasal paramacular regions. While patients with microcysts lay at the far end of the correlation curve between GCL/IPL and INL (i.e. larger INL and smaller GCL/IPL compared to other patients), their exclusion did not affect the correlation (r= -0.76, p<0.001).ConclusionsINL enlargement in MS-related ON is associated with the severity of GCL loss. This is a continuous relationship and patients with INL microcysts may represent the extreme end of the scale.

Discrimination of multiple sclerosis using OCT images from two different centers.

Purpose Diabetes is known to cause alterations in retinal microvasculature and tissue that progressively lead to visual impairment. Optical coherence tomography (OCT) is useful for assessment of total retinal thickening due to diabetic macular edema (DME). In the current study, we determined associations between visual acuity (VA) and retinal layer thickness, reflectance, and interface disruption derived from enface OCT images in subjects with and without DME. Materials and Methods Best corrected VA was measured and high-density OCT volume scans were acquired in 149 diabetic subjects. A previously established image segmentation method identified retinal layer interfaces and locations of visually indiscernible (disrupted) interfaces. Enface thickness maps and reflectance images of the nerve fiber layer (NFL), combined ganglion cell and inner plexiform layer (GCLIPL), inner nuclear layer (INL), outer plexiform layer (OPL), outer nuclear layer (ONL), photoreceptor outer segment layer (OSL), and retinal pigment epithelium (RPE) were generated in the central macular subfield. The associations among VA and retinal layer metrics were determined by multivariate linear regressions after adjusting for covariates (age, sex, race, HbA1c, diabetes type, and duration) and correcting for multiple comparisons. Results In DME subjects, increased GCLIPL and OPL thickness and decreased OSL thickness were associated with reduced VA. Furthermore, increased NFL reflectance and decreased OSL reflectance were associated with reduced VA. Additionally, increased areas of INL and ONL interface disruptions were associated with reduced VA. In subjects without DME, increased INL thickness was associated with reduced VA, whereas in subjects without DME but with previous antivascular endothelium growth factor treatment, thickening of OPL was associated with reduced VA. Conclusions Alterations in retinal layer thickness and reflectance metrics derived from enface OCT images were associated with reduced VA with and without presence of DME, suggestive of their potential for monitoring development, progression, and treatment of DME.

PurposeThis article reports a method for en face optical coherence tomography (OCT) imaging and quantitative assessment of alterations in both thickness and reflectance of individual retinal layers at different stages of diabetic retinopathy (DR).MethodsHigh-density OCT raster volume scans were acquired in 29 diabetic subjects divided into no DR (NDR) or non-proliferative DR (NPDR) groups and 22 control subjects (CNTL). A customized image segmentation method identified eight retinal layer interfaces and generated en face thickness maps and reflectance images for nerve fiber layer (NFL), ganglion cell and inner plexiform layers (GCLIPL), inner nuclear layer (INL), outer plexiform layer (OPL), outer nuclear layer (ONL), photoreceptor outer segment layer (OSL), and retinal pigment epithelium (RPE). Mean thickness and intensity values were calculated in nine macular subfields for each retinal layer.ResultsEn face thickness maps and reflectance images of retinal layers in CNTL subjects corresponded to normal retinal anatomy. Total retinal thickness correlated negatively with age in nasal subfields (R ≤−0.31; P ≤ 0.03, N = 51). In NDR subjects, NFL and OPL thickness were decreased (P = 0.05), and ONL thickness was increased (P = 0.04) compared to CNTL. In NPDR subjects, GCLIPL thickness was increased in perifoveal subfields (P < 0.05) and INL intensity was higher in all macular subfields (P = 0.04) compared to CNTL.ConclusionsDepth and spatially resolved retinal thickness and reflectance measurements are potential biomarkers for assessment and monitoring of DR.

To study the relationship between retinal nerve fiber layer (RNFL) thickness and brain atrophy using magnetic resonance imaging (MRI) with bicaudate ratio (BCR) in patients with multiple sclerosis (MS) with different levels of disease severity. We also assessed whether RNFL thickness correlated with Expanded Disability Status Scale (EDSS) score. The participants consisted of 88 patients with MS and 59 age- and sex-matched healthy control subjects. Eleven patients had clinically isolated syndrome (CIS), 68 patients had relapsing-remitting MS (RR-MS), and 9 patients had secondary progressive MS. Patients and controls were evaluated using optical coherence tomography (OCT, Cirrus) and scanning laser polarimetry with variable corneal compensation (GDx VCC). Patients underwent the same brain MRI scanning protocol. Disability was evaluated according to the EDSS. The BCR was calculated by dividing the minimum intercaudate distance by brain width along the same level. The BCR was higher in patients with MS (0.12 ± 0.03) than in controls (0.08 ± 0.009) (P < 0.001). OCT average RNFL thickness in patients with MS was significantly lower (84.51 ± 14.27 μm) than in control subjects (98.44 ± 6.83 μm). BCR was correlated with OCT average RNFL thickness (r = -0.48, P = 0.002) in patients with MS without optic neuritis. Significant correlations were found between average RNFL thickness and EDSS (r = -0.43, P = 0.003). Additionally, there were correlations between BCR with GDx parameters in patients with MS without optic neuritis. This study shows that RNFL thickness correlates with BCR and with MS subtypes. Additionally, our study indicates that OCT is better suited for MS assessment than GDx. We conclude that the damage of retinal axons appears related to brain damage in patients with MS.

PurposeTo evaluate retinal layer thickness (RLT) after retinal detachment (RD) using high-resolution (HighRes) optical coherence tomography (OCT).MethodsThis prospective, cross-sectional, observational study included 44 (out of 53; 14 [31.8%] female; mean age = 62 ± 10 years) patients with successfully treated macula-off RD. HighRes OCT and Spectralis OCT volume scans (97 scans; both Heidelberg Engineering GmbH, Heidelberg, Germany) were generated for study eyes (SEs) and healthy fellow eyes (FEs), followed by semiautomatic layer segmentation analysis.ResultsThe thicknesses of the ganglion cell layer and inner plexiform layer (GCL-IPL) and inner nuclear layer (INL) were higher in SEs than in FEs (GCL-IPL: 64.4 ± 5.4 µm vs. 61.2 ± 5.2 µm, P < 0.001; INL: 35.1 ± 2.9 µm vs. 32.9 ± 2.7 µm, P < 0.001). HighRes OCT revealed thinner GCL-IPL and ellipsoid zone (EZ) (GCL-IPL: 64.2 ± 5.5 µm vs. 65.2 ± 5.7 µm, P < 0.001; EZ: 8.9 ± 1.4 µm vs. 13.7 ± 1.0 µm, P < 0.001) and thicker INL, outer plexiform layer to myoid zone (OPL-MZ), and combined outer segments of photoreceptors to interdigitation zone (PROS-IZ) than Spectralis OCT (INL: 35.2 ± 3.2 µm vs. 33.9 ± 2.8 µm, P < 0.001; OPL-MZ: 102.7 ± 8.1 µm vs. 99.8 ± 7.6 µm, P < 0.001; PROS-IZ: 19.4 ± 4.9 µm vs. 8.4 ± 2.1 µm, P < 0.001). HighRes OCT revealed two distinct hyperreflective bands between the retinal pigment epithelium and EZ in all SEs, which were not visible in Spectralis OCT.ConclusionsGCL-IPL and INL were thicker in SEs, with RLT differing between imaging devices. Compared with Spectralis OCT, HighRes OCT provides enhanced images of outer retinal layers, which are most relevant for visual outcome in patients with macula-off RD.Translational RelevanceHighRes OCT may improve the diagnostic accuracy, potentially enhancing clinical management of RD.

BackgroundMultiple sclerosis cortical lesions are areas of demyelination and neuroaxonal loss. Retinal layer thickness, measured with optical coherence tomography (OCT), is an emerging biomarker of neuroaxonal loss. Studies have reported correlations between cortical lesions and retinal layer thinning in established multiple sclerosis, suggesting a shared pathophysiological process. Here, we assessed the correlation between cortical lesions and OCT metrics at the onset of multiple sclerosis, examining, for the first time, associations with physical or cognitive disability. ObjectiveTo examine the relationship between cortical lesions, optic nerve and retinal layer thicknesses, and physical and cognitive disability at the first demyelinating event. MethodsThirty-nine patients and 22 controls underwent 3T-MRI, optical coherence tomography, and clinical tests. We identified cortical lesions on phase-sensitive inversion recovery sequences, including occipital cortex lesions. We measured the estimated total intracranial volume and the white matter lesion volume. OCT metrics included peripapillary retinal nerve fibre layer (pRNFL), ganglion cell and inner plexiform layer (GCIPL) and inner nuclear layer (INL) thicknesses. ResultsHigher total cortical and leukocortical lesion volumes correlated with thinner pRNFL (B = -0.0005, 95 % CI -0.0008 to -0.0001, p = 0.01; B = -0.0005, 95 % CI -0.0008 to -0.0001, p = 0.01, respectively). Leukocortical lesion number correlated with colour vision deficits (B = 0.58, 95 %CI 0.039 to 1,11, p = 0.036). Thinner GCIPL correlated with a higher Expanded Disability Status Scale (B = -0.06, 95 % CI -1.1 to -0.008, p = 0.026). MS diagnosis (n = 18) correlated with higher cortical and leukocortical lesion numbers (p = 0.004 and p = 0.003), thinner GCIPL (p = 0.029) and INL (p = 0.041). ConclusionThe association between cortical lesions and axonal damage in the optic nerve reinforces the role of neurodegenerative processes in MS pathogenesis at onset.

Multiple sclerosis (MS) is associated with progressive brain atrophy, which in turn correlates with disability, depression, and cognitive impairment. Relapsing-remitting multiple sclerosis (RRMS) is a type of MS in which relapses of the disease are followed by remission periods. This is the most common type of the disease. There is a significant need for easy and low-cost methods to these cerebral changes. Changes in retinal layer thickness may reflect alterations in brain white and gray matter volumes. Therefore, this paper aims to determine whether retinal layer thickness, measured using optical coherence tomography (OCT), correlates with volumetric brain assessments obtained by magnetic resonance imaging (MRI). This retrospective cohort study recruited 53 patients with relapsing-remitting MS who underwent MRI and OCT examinations for evaluation of brain compartment volumes and thickness of retinal layers, respectively. OCT parameters, including central retinal thickness; retinal nerve fiber layer thickness (RNFL, peripapillary thickness); ganglion cell complex thickness (GCC, macular thickness); and Expanded Disability Status Scale (EDSS) results were compared with MRI parameters (cerebral cortex; cerebral cortex and basal ganglia combined; brain hemispheres without the ventricular system; and white matter plaques). We also checked whether there is a correlation between the number of RRMS and OCT parameters. Our primary objective was to identify whether these patients had retinal thickness changes, and our secondary objective was to check if those changes correlated with the MRI brain anatomical changes. RNFL and GCC thicknesses were strongly (p-value < 0.05) associated with (i) cerebral cortex volume, (ii) combination of brain cortex and basal ganglia volumes, and (iii) the hemispheres but without the ventricular system. White matter plaques (combined) showed only weak or no correlation with RNFL and GCC. There was no correlation between central retinal thickness and brain compartment volumes, and there were weak or no correlations between the summary EDSS scores and OCT results. Retinal layer thickness measured by OCT correlates with select volumetric brain assessments on MRI. During the course of RRMS, the anatomo-pathological structure of the retina might serve as a surrogate marker of brain atrophy and clinical progression within selected domains.

PurposeRetinal layer thickness parameters measured by optical coherence tomography (OCT) are emerging biomarkers of neuroaxonal degeneration and inflammation in multiple sclerosis (MS). We aimed to evaluate the value of retinal layer thickness for prediction of disability worsening and relapse in a real-world MS cohort.Patients and MethodsFor this longitudinal observational study, we included MS patients with spectral-domain OCT scans available and ≥1 year of clinical follow-up. The value of peripapillary retinal nerve fiber layer (pRNFL), macular ganglion-cell-and-inner-plexiform-layer (GCIPL) and inner nuclear layer (INL) thickness for prediction of disability worsening and relapse during the observation period was tested by multivariate models.ResultsWe analyzed 60 MS patients during a mean observation period of 2.9 years (SD 1.8). Lower baseline thickness of GCIPL (cut-off <77µm; HR 4.1, p=0.001) and pRNFL (cut-off ≤88µm; HR 3.1, p=0.019) were associated with an increased risk of disability worsening. Longitudinally, mean thinning rates were −0.8µm/year (SD 1.6) for GCIPL, −0.6µm/year (SD 3.5) for pRNFL. GCIPL thinning ≥1.0µm/year and pRNFL >1.5µm/year is associated with higher likelihood of disability worsening (HR 5.7, p=0.009 and HR 6.8, p=0.003, respectively). INL thickened in patients with relapse by a mean 0.9µm while thinning by 0.3µm in patients without relapse (p=0.04). In multivariate analyses, INL thickening was associated with an increased probability of relapse (OR 17.8, p=0.023).ConclusionCross-sectional and longitudinal measurement of GCIPL and pRNFL thinning is reliable as a biomarker of disability worsening in a real-world setting. Change of INL thickness is a promising marker of relapse, i.e. inflammatory activity.

Hyperreflective retinal foci (HRF) visualized by optical coherence tomography (OCT) potentially represent clusters of microglia. We compared HRF frequencies and their association with retinal neurodegeneration between people with clinically isolated syndrome (pwCIS), multiple sclerosis (pwMS), aquaporin 4-IgG positive neuromyelitis optica spectrum disorder (pwNMOSD), and healthy controls (HC)-as well as between eyes with (ON+eyes) and without a history of optic neuritis (ON-eyes). Cross-sectional data of pwCIS, pwMS, and pwNMOSD with previous ON and HC were acquired at Charité-Universitätsmedizin Berlin. HRF analysis was performed manually on the central macular OCT scan. Semi-manual OCT segmentation was performed to acquire the combined ganglion cell and inner plexiform layer (GCIPL), inner nuclear layer (INL), and peripapillary retinal nerve fiber layer (pRNFL) thickness. Group comparisons were performed by linear mixed models. In total, 227 eyes from 88 patients (21 pwCIS, 32 pwMS, and 35 pwNMOSD) and 35 HCs were included. HRF in GCIPL and INL were more frequently detected in pwCIS, pwMS, and pwNMOSD than HCs (p < 0.001 for all comparisons) with pwCIS exhibiting the greatest numbers. ON+eyes of pwMS had less HRF in GCIPL than ON-eyes (p = 0.036), but no difference was seen in pwCIS and pwNMOSD. HRF GCIPL were correlated to GCIPL thickness in ON+eyes in pwMS (p = 0.040) and pwNMOSD (p = 0.031). HRF occur in ON+eyes and ON-eyes across neuroinflammatory diseases. In pwMS and pwNMOSD, HRF frequency was positively associated with GCIPL thickness indicating that HRF formation might be dependent on retinal ganglion cells.

BackgroundThe novel optic neuritis (ON) diagnostic criteria include intereye differences (IED) of optical coherence tomography (OCT) parameters. IED has proven valuable for ON diagnosis in multiple sclerosis but has not...

To examine structural changes in retinal layers over time in patients with non-arteritic anterior ischaemic optic neuropathy (NAION) and determine the layers that predict visual outcomes. The optical coherence tomography parameters in NAION eyes at <2 months, 2-5 months, and 6-18 months from the onset were compared to age-matched normal controls. Generalised estimating equation analysis was used to analyse the changes over time and regression analysis was performed to identify the layer that could predict visual field outcomes. Less than 2 months from the onset, the peripapillary retinal nerve fibre layer (RNFL) (p = 0.001) and macular outer nuclear layer (ONL) (p = 0.024) were significantly thicker in the NAION eyes than in the control eyes. The average peripapillary RNFL, macular RNFL, and ganglion cell layer and inner plexiform layer (GCIPL) showed reductions in thickness within 2-5 months (peripapillary RNFL: -19.8 μm/month, p < 0.001, macular RNFL: -14.5 μm/month, p < 0.001, GCIPL: -26.8 μm/month, p < 0.001). The change of thickness in temporal and superior peripapillary RNFL, GCIPL, inner nuclear layer (INL), and ONL by 2-5 months was associated with the final visual field results (p = 0.018, p < 0.001, p = 0.040, p = 0.020, and p = 0.002, respectively). The peripapillary RNFL swelling initially observed started to decrease within 2-5 months along with macular RNFL and GCIPL thinning. The rate of thickness changes in the peripapillary RNFL, GCIPL, INL, and ONL by 2-5 months was associated with visual field outcomes.

This study was undertaken to investigate baseline peripapillary retinal nerve fiber layer (pRNFL) and macular ganglion cell and inner plexiform layer (GCIPL) thickness for prediction of disability accumulation in early relapsing multiple sclerosis (RMS). From a prospective observational study, we included patients with newly diagnosed RMS and obtained spectral-domain optical coherence tomography scan within 90 days after RMS diagnosis. Impact of pRNFL and GCIPL thickness for prediction of disability accumulation (confirmed Expanded Disability Status Scale [EDSS] score ≥ 3.0) was tested by multivariate (adjusted hazard ratio [HR] with 95% confidence interval [CI]) Cox regression models. We analyzed 231 MS patients (mean age=30.3 years, SD=8.1, 74% female) during a median observation period of 61 months (range=12-93). Mean pRNFL thickness was 92.6 μm (SD = 12.1), and mean GCIPL thickness was 81.4 μm (SD=11.8). EDSS ≥ 3 was reached by 28 patients (12.1%) after a median 49 months (range=9-92). EDSS ≥ 3 was predicted with GCIPL < 77 μm (HR=2.7, 95% CI=1.6-4.2, p< 0.001) and pRNFL thickness ≤ 88 μm (HR=2.0, 95% CI=1.4-3.3, p< 0.001). Higher age (HR=1.4 per 10 years, p< 0.001), incomplete remission of first clinical attack (HR=2.2, p< 0.001), ≥10 magnetic resonance imaging (MRI) lesions (HR=2.0, p< 0.001), and infratentorial MRI lesions (HR=1.9, p< 0.001) were associated with increased risk of disability accumulation, whereas highly effective disease-modifying treatment was protective (HR=0.6, p< 0.001). Type of first clinical attack and presence of oligoclonal bands were not significantly associated. Retinal layer thickness (GCIPL more than pRNFL) is a useful predictor of future disability accumulation in RMS, independently adding to established markers.

Spectral domain-optical coherence tomography plays a crucial role in the early detection and monitoring of multiple sclerosis (MS) pathophysiology. We aimed to quantify differences in retinal layer measures among different groups of MS and explored different variables that correlate with retinal measures. This study was reported according PRISMA guidelines. A comprehensive search was done across PubMed, Embase, and Google Scholar. The mean difference in thickness of retinal layers and macular volume was assessed. Meta-regression was done to assess the sources of heterogeneity. A total of 100 articles were included in the meta-analyses. The peripapillary retinal nerve fiber layer (pRNFL) thickness significantly decreased in the MSON (MD: -16.44, P < 0.001), MSNON (MD: -6.97, P < 0.001), and PMS (MD: -11.35, P < 0.001) versus HC. The macular RNFL was lower among the MSON (MD: -6.24, P = 0.013) and MSNON (MD: -3.84, P <0.001) versus HC. Macular ganglion cell layer and inner plexiform layer (GCIPL) was thinner among MSON (MD: -14.83, P <0.001), MSNON (MD: -6.38, P < 0.001), and PMS (MD: -11.52, P < 0.001) compared with control eyes. Inner nuclear layer (INL) was higher in the MSON (MD: 0.49, P < 0.001) versus HC. Outer nuclear layer (ONL) thickness significantly lower in the MSNON (MD: -1.15, P = 0.019) versus HC. Meta-regression showed that disease duration, age, EDSS score, and percentage of patients taking DMT are all negatively correlated with pRNFL and GCIPL thickness; however, female gender was correlated with less atrophy. As conclusion, the study highlights substantial thinning in the pRNFL and macular GCIPL between MS versus controls. INL as valuable parameter for capturing inflammatory disease activity.

Objective To study the relationship between retinal nerve fiber layer (RNFL) thickness measured by optical coherence tomography (OCT) and visual function, disease course and system dysfunction in patients with multiple sclerosis (MS); to evaluate the significance of RNFL thickness as a biological marker for axonal loss in the course of MS disease. Methods This was a cross-sectional study. Seventeen MS patients (32 eyes) and 17 healthy people (17 eyes) were matched by age and sex. All participants received a neurological evaluation and a complete ophthalmological examination, including visual acuity, OCT measurement of RNFL thickness, pattern visual evoked potentials (PVEP) and electroretinography (ERG) examinations. Patients were divided into three groups: MS patients with a history of optic neuritis (MS-ON group), MS patients without a history of optic neuritis (MS-NON group) and the control group. The results of OCT measurement of RNFL thickness, and PVEP and ERG examinations of the three groups were compared by ANOVA. The relationship between RNFL thickness and the score on the expanded disability status scale (EDSS), and disease course, and electrophysiology results were analyzed by Spearman correlation analysis. Results The MS-ON eyes showed RNFL thinning in the average, superior, inferior, nasal, and temporal quadrants [(72.4±16.6), (80.5±26.3), (84.2±29.5), (65.8±14.0), (54.2±16.4)μm] compared to the control eyes [(105.6±10.8), (119.7±18.2), (123.5±17.9), (91.1 ±21.2), (88.2±13.0)μm] (P=0.000 each). The differences in RNFL thickness in the average, superior, inferior, and temporal quadrants between MS-ON eyes and MS-NON eyes [(98.3±12.0), (115.5±18.2), (117.7±21.5), (72.0±15.0), (68.1 ±16.1 )μm] were significant (P＜0.05 each), while the nasal quadrant did not show significant differences. Compared to control eyes, RNFL in the nasal and temporal quadrants in MS-NON eyes showed a significant reduction (P＜0.05 each), while the average, superior and inferior quadrants did not show significant thinning. RNFL thickness in MS-ON eyes was significantly correlated with the disease course of MS [(8.8±9.8)year] (r=-0.659, P=0.028) and EDSS scores (2.9±2.2) (r=-0.741, P=0.046). In this research, no significant relationships were found between RNFL thickness and disease course or EDSS scores in the MS-NON group. RNFL thickness in MS eyes showed a significant relationship to the latency and amplitude of P100 wave of both the 60' and 15' angles (r=-0.416, 0.332,-0.317, 0.265, P＜0.05 each), and the latency and amplitude of the a wave and latency of the b wave in the ERG maximum response (r=-0.471, 0.415,-0.360, P＜0.05 each), while no relationship was found between RNFL thickness and the amplitude of the b wave. Conclusion RNFL thickness measured by OCT can be used for clinical application as a structural biomarker of axonal loss in MS patients. Key words: Multiple sclerosis; Nerve fiber layer,retinal; Tomography,optical coherence; Evoked potential,visual; Electroretinography; EDSS score