Abstract
Retinal imaging has been proposed as a biomarker for neurological diseases such as multiple sclerosis (MS). Recently, a technique for non-invasive assessment of the retinal microvasculature ...called optical coherence tomography angiography (OCTA) was introduced. We investigated retinal microvasculature alterations in participants with relapsing–remitting MS (RRMS) without history of optic neuritis (ON) and compared them to a healthy control group. The study was performed in a prospective, case–control design, including 58 participants (n = 100 eyes) with RRMS without ON and 78 age- and sex-matched control participants (n = 136 eyes). OCTA images of the superficial capillary plexus (SCP), deep capillary plexus (DCP) and choriocapillaris (CC) were obtained using a commercial OCTA system (Zeiss Cirrus HD-5000 Spectral-Domain OCT with AngioPlex OCTA, Carl Zeiss Meditec, Dublin, CA). The outcome variables were perfusion density (PD) and foveal avascular zone (FAZ) features (area and circularity) in both the SCP and DCP, and flow deficit in the CC. MS group had on average higher intraocular pressure (IOP) than controls (
P
< 0.001). After adjusting for confounders, MS participants showed significantly increased PD in SCP (
P
= 0.003) and decreased PD in DCP (
P
< 0.001) as compared to controls. A significant difference was still noted when large vessels (LV) in the SCP were removed from the PD calculation (
P
= 0.004). Deep FAZ was significantly larger (
P
= 0.005) and less circular (
P
< 0.001) in the eyes of MS participants compared to the control ones. Neither LV, PD or FAZ features in the SCP, nor flow deficits in the CC showed any statistically significant differences between the MS group and control group (
P
> 0.186). Our study indicates that there are microvascular changes in the macular parafoveal retina of RRMS patients without ON, showing increased PD in SCP and decreased PD in DCP. Further studies with a larger cohort of MS patients and MRI correlations are necessary to validate retinal microvascular changes as imaging biomarkers for diagnosis and screening of MS.
Studies using machine learning (ML) approaches have reported high diagnostic accuracies for glaucoma detection. However, none assessed model performance across ethnicities. The aim of the study is to ...externally validate ML models for glaucoma detection from optical coherence tomography (OCT) data. We performed a prospective, cross-sectional study, where 514 Asians (257 glaucoma/257 controls) were enrolled to construct ML models for glaucoma detection, which was then tested on 356 Asians (183 glaucoma/173 controls) and 138 Caucasians (57 glaucoma/81 controls). We used the retinal nerve fibre layer (RNFL) thickness values produced by the compensation model, which is a multiple regression model fitted on healthy subjects that corrects the RNFL profile for anatomical factors and the original OCT data (measured) to build two classifiers, respectively. Both the ML models (area under the receiver operating AUC = 0.96 and accuracy = 92%) outperformed the measured data (AUC = 0.93; P < 0.001) for glaucoma detection in the Asian dataset. However, in the Caucasian dataset, the ML model trained with compensated data (AUC = 0.93 and accuracy = 84%) outperformed the ML model trained with original data (AUC = 0.83 and accuracy = 79%; P < 0.001) and measured data (AUC = 0.82; P < 0.001) for glaucoma detection. The performance with the ML model trained on measured data showed poor reproducibility across different datasets, whereas the performance of the compensated data was maintained. Care must be taken when ML models are applied to patient cohorts of different ethnicities.
•OCT diagnostics for MS improved after combining macular data with compensated peripapillary RNFL.
Optical coherence tomography (OCT) is a retinal imaging system that may improve the diagnosis of ...multiple sclerosis (MS) persons, but the evidence is currently equivocal. To assess whether compensating the peripapillary retinal nerve fiber layer (pRNFL) thickness for ocular anatomical features as well as the combination with macular layers can improve the capability of OCT in differentiating non-optic neuritis eyes of relapsing-remitting MS patients from healthy controls.
74 MS participants (n = 129 eyes) and 84 age- and sex-matched healthy controls (n = 149 eyes) were enrolled. Macular ganglion cell complex (mGCC) thickness was extracted and pRNFL measurement was compensated for ocular anatomical factors. Thickness measurements and their corresponding areas under the receiver operating characteristic curves (AUCs) were compared between groups.
Participants with MS showed significantly thinner mGCC, measured and compensated pRNFL (p ≤ 0.026). Compensated pRNFL achieved better performance than measured pRNFL for MS differentiation (AUC, 0.75 vs 0.80; p = 0.020). Combining macular and compensated pRNFL parameters provided the best discrimination of MS (AUC = 0.85 vs 0.75; p < 0.001), translating to an average improvement in sensitivity of 24 percent for differentiation of MS individuals.
The capability of OCT in MS differentiation is made more robust by accounting OCT scans for individual anatomical differences and incorporating information from both optic disc and macular regions, representing markers of axonal damage and neuronal injury, respectively.
There is a clear evidence that pregnancy is associated with high production of sex hormones. During the first, second and third trimester of pregnancy, blood hormones levels increase gradually. Cells ...with affinity for sex hormones have been identified in different ocular tissues, such as: lid, lacrimal gland, meibomian gland, bulbar and palpebral conjunctivae, cornea, iris, ciliary body, lens, retina (retinal pigment epithelium) and choroid. This is why pregnancy is associated with changes at ocular level, involving anterior and posterior segments. Several clinical trials have been made trying to highlight changes in corneal biomechanics during pregnancy. By conducting this review, we want to evaluate both the changes in parameters that define corneal biomechanics and intraocular pressure values in pregnant.
Following a systematic search in the literature related mainly to changes in corneal biomechanics during pregnancy, focusing on the paper published in the last decade, we included in a meta-analysis the cumulative results of three prospective comparative studies.
Important changes in corneal biomechanics (corneal hysteresis and corneal resistance factor) parameters were observed in women in the third trimester of pregnancy, but these variations were not statistically significant. Also, a decrease in intraocular pressure was mentioned in these women, but only the corneal compensation intraocular pressure showed a decrease with statistical significance.
A decrease in corneal compensatory intraocular pressure was observed in pregnant women in the third trimester of pregnancy, but without other statistically significant changes resulting from the analysis of the other three parameters (corneal hysteresis, corneal resistance factor and Goldmann-correlated intraocular pressure).
Abstract
Data on how retinal structural and vascular parameters jointly influence the diagnostic performance of detection of multiple sclerosis (MS) patients without optic neuritis (MSNON) are ...lacking. To investigate the diagnostic performance of structural and vascular changes to detect MSNON from controls, we performed a cross‐sectional study of 76 eyes from 51 MS participants and 117 eyes from 71 healthy controls. Retinal macular ganglion cell complex (GCC), retinal nerve fiber layer (RNFL) thicknesses, and capillary densities from the superficial (SCP) and deep capillary plexuses (DCP) were obtained from the Cirrus AngioPlex. The best structural parameter for detecting MS was compensated RNFL from the optic nerve head (AUC = 0.85), followed by GCC from the macula (AUC = 0.79), while the best vascular parameter was the SCP (AUC = 0.66). Combining structural and vascular parameters improved the diagnostic performance for MS detection (AUC = 0.90;
p
<0.001). Including both structure and vasculature in the joint model considerably improved the discrimination between MSNON and normal controls compared to each parameter separately (
p
= 0.027). Combining optical coherence tomography (OCT)‐derived structural metrics and vascular measurements from optical coherence tomography angiography (OCTA) improved the detection of MSNON. Further studies may be warranted to evaluate the clinical utility of OCT and OCTA parameters in the prediction of disease progression.
To evaluate machine learning (ML) approaches for structure–function modeling to estimate visual field (VF) loss in glaucoma, models from different ML approaches were trained on optical coherence ...tomography thickness measurements to estimate global VF mean deviation (VF MD) and focal VF loss from 24‐2 standard automated perimetry. The models were compared using mean absolute errors (MAEs). Baseline MAEs were obtained from the VF values and their means. Data of 832 eyes from 569 participants were included, with 537 Asian eyes for training, and 148 Asian and 111 Caucasian eyes set aside as the respective test sets. All ML models performed significantly better than baseline. Gradient‐boosted trees (XGB) achieved the lowest MAE of 3.01 (95% CI: 2.57, 3.48) dB and 3.04 (95% CI: 2.59, 3.99) dB for VF MD estimation in the Asian and Caucasian test sets, although difference between models was not significant. In focal VF estimation, XGB achieved median MAEs of 4.44 IQR 3.45–5.17 dB and 3.87 IQR 3.64–4.22 dB across the 24‐2 VF for the Asian and Caucasian test sets and was comparable to VF estimates from support vector regression (SVR) models. VF estimates from both XGB and SVR were significantly better than the other models. These results show that XGB and SVR could potentially be used for both global and focal structure–function modeling in glaucoma.
Different machine learning models for global and focal visual field estimation were compared using independent internal and external test sets with different demographics. We compared several models and show that those based on gradient‐boosted trees generally performed well in both internal and external sets and thus may be a useful approach for future structure–function studies. We also found that all models had difficulties to varying degrees in more severe visual fields.
Purpose: Retinal imaging has attracted much interest as a non‐invasive low‐budget biomarker for neurological diseases such as multiple sclerosis (MS). Optical coherence tomography angiography (OCTA) ...is a functional extension of OCT and allows for the non‐invasive visualization of the retinal and choroidal microvasculature. We investigated retinal microvasculature changes in patients with relapsing–remitting MS (RRMS) without history of optic neuritis (ON) and compared them to a healthy control group.
Methods: The study was performed in a prospective, case–control design, including 58 participants (n = 100 eyes) with RRMS without ON and 78 age‐ and sex‐matched control participants (n = 136 eyes). OCTA images of the superficial capillary plexus (SCP), deep capillary plexus (DCP) and choriocapillaris (CC) were obtained using a commercial OCTA system (Zeiss Cirrus HD‐5000 Spectral‐Domain OCT with AngioPlex OCTA, Carl Zeiss Meditec, Dublin, CA). Perfusion density (PD) and foveal avascular zone (FAZ) features (area and circularity) in both the SCP and DCP, as well as flow deficit in the CC were used as outcome variables.
Results: MS patients showed significantly increased PD in SCP (p = 0.003) and decreased PD in DCP (p < 0.001) as compared to controls when data were corrected for confounders. A significant difference was also noted when large vessels (LV) in the SCP were removed from the PD calculation (p = 0.004). Deep FAZ was significantly larger (p = 0.005) and less circular (p < 0.001) in the eyes of MS patients compared to healthy controls. Neither LV, PD or FAZ features in the SCP, nor flow deficits in the CC showed any statistically significant differences between the MS patients and the controls (p > 0.186).
Conclusions: Our study shows that MS patients have microvascular changes in the macular parafoveal retina even without ON. They show increased PD in SCP and decreased PD in DCP. To which degree retinal biomarkers in MS are associated with the progression of MS remains to be studied.
IMPORTANCE: Deep learning (DL) networks require large data sets for training, which can be challenging to collect clinically. Generative models could be used to generate large numbers of synthetic ...optical coherence tomography (OCT) images to train such DL networks for glaucoma detection. OBJECTIVE: To assess whether generative models can synthesize circumpapillary optic nerve head OCT images of normal and glaucomatous eyes and determine the usability of synthetic images for training DL models for glaucoma detection. DESIGN, SETTING, AND PARTICIPANTS: Progressively growing generative adversarial network models were trained to generate circumpapillary OCT scans. Image gradeability and authenticity were evaluated on a clinical set of 100 real and 100 synthetic images by 2 clinical experts. DL networks for glaucoma detection were trained with real or synthetic images and evaluated on independent internal and external test data sets of 140 and 300 real images, respectively. MAIN OUTCOMES AND MEASURES: Evaluations of the clinical set between the experts were compared. Glaucoma detection performance of the DL networks was assessed using area under the curve (AUC) analysis. Class activation maps provided visualizations of the regions contributing to the respective classifications. RESULTS: A total of 990 normal and 862 glaucomatous eyes were analyzed. Evaluations of the clinical set were similar for gradeability (expert 1: 92.0%; expert 2: 93.0%) and authenticity (expert 1: 51.8%; expert 2: 51.3%). The best-performing DL network trained on synthetic images had AUC scores of 0.97 (95% CI, 0.95-0.99) on the internal test data set and 0.90 (95% CI, 0.87-0.93) on the external test data set, compared with AUCs of 0.96 (95% CI, 0.94-0.99) on the internal test data set and 0.84 (95% CI, 0.80-0.87) on the external test data set for the network trained with real images. An increase in the AUC for the synthetic DL network was observed with the use of larger synthetic data set sizes. Class activation maps showed that the regions of the synthetic images contributing to glaucoma detection were generally similar to that of real images. CONCLUSIONS AND RELEVANCE: DL networks trained with synthetic OCT images for glaucoma detection were comparable with networks trained with real images. These results suggest potential use of generative models in the training of DL networks and as a means of data sharing across institutions without patient information confidentiality issues.
Retinal neuronal and vascular changes have been observed in multiple sclerosis (MS) patients. The aim of this review was to highlight the most current optical coherence tomography (OCT) and optical ...coherence tomography angiography (OCT-A) data in MS and to provide information about the possibility of using OCT / OCT-A parameters as biomarkers for screening, diagnosis and monitoring of MS.
To carry out this review, a meticulous literature search was undergone on PubMed between 2014 and the present day, using the following terms: "multiple", "sclerosis", "optical", "coherence", "tomography" and "angiography". Additional studies were found via references, being chosen according to relevance.
Retinal nerve fiber layer (RNFL) and ganglion cell layer (GCL) were significantly lower in MS patients compared to controls, and correlated with clinical and paraclinical variables, such as visual function, disability, and magnetic resonance imaging (MRI). Retinal capillary plexuses could be higher, lower or the same, and the best OCT-A microvasculature parameter for the detection of MS was the superficial capillary plexus (SCP). The reduced retinal vessel density (VD) was correlated with the disability in MS.
OCT and OCT-A parameters could improve the development of retinal biomarkers for screening, early diagnosis and monitoring the disease progression of MS, and they could improve the development of potential future therapies that could slow or stop the course of this incurable disease.
DCP = deep capillary plexus; EDSS = Expanded Disability Status Scale; GCC = ganglion cell complex; GCL = ganglion cell layer; MRI = magnetic resonance imaging; MS = Multiple sclerosis; OCT = optical coherence tomography; OCT-A = optical coherence tomography angiography; ON = optic neuritis; RNFL = retinal nerve fiber layer; SCP = superficial capillary plexus; VD = vessel density.
Purpose: To evaluate the use of synthetically generated OCT images for the development of deep‐learning models in glaucoma detection.
Methods: Progressively Grown Generative Adversarial Network ...(PGGAN) models for glaucoma and healthy eyes were developed with data from 862 Asian glaucoma eyes and 990 Asian normal eyes to generate synthetic circumpapillary OCT images. Glaucoma detection deep‐learning models were trained using 1200, 10 000, 60 000 or 200 000 of the generated images, equally split between glaucoma and normal. Detection performance was evaluated on real images from an Asian dataset of 140 eyes from 112 subjects and a Caucasian dataset of 300 eyes from 160 subjects, with half of each being glaucoma. Results were compared with a glaucoma detection model trained with real images from 600 glaucoma and 600 healthy eyes, and with global retinal nerve fibre layer (RNFL) measurements using Area Under the Curve (AUC) analysis.
Results: Glaucoma detection performance improved with increasing synthetic dataset size, from an AUC of 0.945 95% CI: 0.917–0.974 and 0.856 95% CI: 0.819–0.889 on the Asian and Caucasian test data respectively when 1200 synthetic images were used, to AUCs of 0.969 95% CI: 0.949–0.987 and 0.897 95% CI: 0.867–0.927 when 200 000 synthetic images were used. The model trained on 200 000 synthetic images was significantly better (p < 0.05) than the model trained on real images on the Caucasian test data (0.838 95% CI: 0.800–0.874) but not on the Asian test data (0.962 95% CI: 0.939–0.985). Both detection models showed significantly better AUCs than that of global RNFL for both Asian (0.920 95% CI: 0.884–0.957) and Caucasian (0.833 95% CI: 0.793–0.869) test datasets.
Conclusions: Synthetically circumpapillary OCT images used to train deep‐learning glaucoma detection models showed similar performance to a model trained with real data, and could potentially be used to facilitate sharing of imaging data while addressing privacy considerations.