Search

Introduction: Detection of neurological conditions is of high importance in the current context of increasingly ageing populations. Imaging of the retina and the optic nerve head represents a unique opportunity to detect brain diseases, but requires specific human expertise. We review the current outcomes of artificial intelligence (AI) methods applied to retinal imaging for the detection of neurological and neuro-ophthalmic conditions. Method: Current and emerging concepts related to the detection of neurological conditions, using AI-based investigations of the retina in patients with brain disease were examined and summarised. Results: Papilloedema due to intracranial hypertension can be accurately identified with deep learning on standard retinal imaging at a human expert level. Emerging studies suggest that patients with Alzheimer’s disease can be discriminated from cognitively normal individuals, using AI applied to retinal images. Conclusion: Recent AI-based systems dedicated to scalable retinal imaging have opened new perspectives for the detection of brain conditions directly or indirectly affecting retinal structures. However, further validation and implementation studies are required to better understand their potential value in clinical practice. Keywords: Alzheimer’s disease, deep learning, dementia, optic neuropathy, papilloedema

Age-related macular degeneration, a prevalent degenerative retinal disease, is associated with non-visual and psychosocial impairments that may affect sleep. In this systematic review, we evaluated associations between age-related macular degeneration (AMD) and sleep, highlighted knowledge gaps and provided evidence-based recommendations to clinicians to enable holistic management of AMD patients.We searched PubMed, Embase and the Cochrane Central registries for papers published before May 2022. Non-English, qualitative studies and grey literature were excluded. Studies evaluating the association between AMD and sleep (including sleep disorders like insomnia and sleep apnea), and vice versa, were included. The quality of shortlisted studies was evaluated using the Newcastle Ottawa Scale.Six (two case-control studies, three longitudinal cohort studies and one cross-sectional study) of 551 studies were included in this review. Four studies found that AMD was associated with increased rates of sleep apnea and poorer reported sleep quality, while five studies showed that patients with sleep apnea or insomnia were at higher risk of developing AMD. Associations between self-reported sleep quantity and AMD were conflicting. No study evaluated the relationship between AMD and sleep using objective sleep assessment tools.Only a limited number of studies investigated associations between AMD and sleep. These studies suggest a bidirectional relationship between AMD and sleep dysfunction yet disagree on the relationship between sleep quantity and the likelihood of AMD. Additional studies, using objective characterisation of sleep in patients with AMD are required to confirm these findings.

To evaluate the ability of handheld chromatic pupillometry to reveal and localise retinal neural dysfunction in diabetic patients with and without diabetic retinopathy (DR).This cross-sectional study included 82 diabetics (DM) and 93 controls (60.4 ± 8.4 years, 44.1% males). DM patients included those without (n = 25, 64.7 ± 6.3 years, 44.0% males) and with DR (n = 57, 60.3 ± 8.5 years, 64.9% males). Changes in horizontal pupil radius in response to blue (469 nm) and red (640 nm) light stimuli were assessed monocularly, in clinics, using a custom-built handheld pupillometer. Pupillometric parameters (phasic constriction amplitudes [predominantly from the outer retina], maximal constriction amplitudes [from the inner and outer retina] and post-illumination pupillary responses [PIPRs; predominantly from the inner retina]) were extracted from baseline-adjusted pupillary light response traces and compared between controls, DM without DR, and DR. Net PIPR was defined as the difference between blue and red PIPRs.Phasic constriction amplitudes to blue and red lights were decreased in DR compared to controls (p 0.001; p 0.001). Maximal constriction amplitudes to blue and red lights were decreased in DR compared to DM without DR (p 0.001; p = 0.02), and in DM without DR compared to controls (p 0.001; p = 0.005). Net PIPR was decreased in both DR and DM without DR compared to controls (p = 0.02; p = 0.03), suggesting a wavelength-dependent (and hence retinal) pupillometric dysfunction in diabetic patients with or without DR.Handheld chromatic pupillometry can reveal retinal neural dysfunction in diabetes, even without DR. Patients with DM but no DR displayed primarily inner retinal dysfunction, while patients with DR showed both inner and outer retinal dysfunction.

Xian Hui Lim,1 Monisha Esther Nongpiur,1– 3 Raymond P Najjar,2,3 Quan V Hoang,1– 4 Dan Milea,1– 3 Chee Wai Wong,1– 3 Rahat Husain,1,3 Hla Myint Htoon,2,3 Tin Aung,1– 3,5 Shamira Perera,1– 3 Tina Tzee Ling Wong1– 3,5 1Singapore National Eye Centre, Singapore; 2Singapore Eye Research Institute, Singapore; 3Duke-NUS Medical School, Singapore; 4Department of Ophthalmology, Edward S. Harkness Eye Institute, Columbia College of Physicians and Surgeons, New York, NY, USA; 5Department of Ophthalmology, Yong Loo Lin School of Medicine, National University of Singapore, SingaporeCorrespondence: Tina Tzee Ling WongSingapore National Eye Centre, 11 Third Hospital Avenue, 168751, SingaporeEmail tina.wong.t.l@singhealth.com.sgPurpose: To investigate features of the steady-state pattern electroretinogram (ssPERG) in subjects with glaucoma (G), high myopia (HM; spherical equivalent ≤-6D) and glaucoma with high myopia (GHM).Patients and Methods: Our study included 48 participants divided into 3 groups (G, HM, and GHM) who each underwent monocular ssPERG testing with Diopsys NOVA PERG protocols. The ConStim protocol detects distinct topographic patterns of dysfunction 16° and 24° around the central macula. MagD is the amplitude of the average signal and MagD/Mag ratio indicates the consistency of the response. ssPERG indices were compared between groups and correlated with functional (ie, visual field mean deviation (VFMD)) and structural (ie, average retinal nerve fibre layer (RNFL) thickness; Cirrus optical coherence tomography) features.Results: Participants had an average age of 59.4± 7.6 years. Mean Humphrey VFMD was − 14.22 ± 2.88dB, − 2.62 ± 1.18dB and − 12.80 ± 2.60dB for G, HM and GHM groups, respectively. Mean RNFL thickness was 63.0 ± 8.20μm, 69.5 ± 15.7μm and 60.6 ± 5.0μm for G, HM and GHM groups, respectively. For the 24° setting, no significant differences were noted for any of the parameters. For the 16° setting, MagD was lower in the GHM group compared to the HM group (0.29μV vs 0.52μV; p = 0.02). Significant differences were noted for the MagD/Mag ratio between HM and G groups (0.58 vs 0.40; p = 0.02) and between HM and GHM groups (0.58 vs 0.35; p = 0.002). There were positive correlations between both MagD 16° and MagD/Mag ratio 16° with VFMD (correlation coefficient [r]=0.37, p = 0.009; and r = 0.44, p = 0.002, respectively) and RNFL (r = 0.43, p = 0.002; and r = 0.48, p = 0.001, respectively).Conclusion: MagD/Mag ratio at 16° was significantly lower in glaucomatous eyes (with or without high myopia) compared to those with high myopia without glaucoma, suggesting that glaucoma has a distinct impact on MagD/Mag ratio at 16° irrespective of the presence of myopia.Keywords: pattern electroretinogram, visual field mean deviation, optical coherence tomography, retinal nerve fibre layer

PURPOSETo evaluate the duration-dependent and synergetic impact of high-intensity light (HL) and optical refocus (RF) on lens-induced myopia (LIM) development in chickens.METHODSMyopia was induced in one eye in chicks (10 groups, n=126) from day 1 post- hatching (D1) until D8 using -10D lenses. Fellow eyes remained uncovered as controls. Nine groups were exposed daily to continuous 2 hours (h), 4h, or 6h of either HL (15,000 lux); RF (removal of -10D lens); or both (HL+RF). One group served as the LIM group without any interventions. Ocular axial length (AL), refractive error, and choroidal thickness were measured on D1, D4, and D8. Outcome measures are expressed as inter-ocular difference (IOD= experimental - control eye) ±SEM.RESULTSBy D8, LIM increased AL (0.36±0.04 mm), myopic refraction (-9.02±0.37D), and choroidal thinning (-90.27±16.44 µm) in the LIM group (all, PCONCLUSIONDaily exposure to 2h, 4h, or 6h of HL, RF, or HL+RF reduced lens-induced myopic refraction in a duration-dependent manner in chickens. Only 6h of HL+RF completely stopped LIM development. The synergetic effect of HL and RF is dependent on the duration of the interventions.

The quality of ocular fundus photographs can affect the accuracy of the morphologic assessment of the optic nerve head (ONH), either by humans or by deep learning systems (DLS). In order to automatically identify ONH photographs of optimal quality, we have developed, trained, and tested a DLS, using an international, multicentre, multi-ethnic dataset of 5015 ocular fundus photographs from 31 centres in 20 countries participating to the Brain and Optic Nerve Study with Artificial Intelligence (BONSAI). The reference standard in image quality was established by three experts who independently classified photographs as of “good”, “borderline”, or “poor” quality. The DLS was trained on 4208 fundus photographs and tested on an independent external dataset of 807 photographs, using a multi-class model, evaluated with a one-vs-rest classification strategy. In the external-testing dataset, the DLS could identify with excellent performance “good” quality photographs (AUC = 0.93 (95% CI, 0.91–0.95), accuracy = 91.4% (95% CI, 90.0–92.9%), sensitivity = 93.8% (95% CI, 92.5–95.2%), specificity = 75.9% (95% CI, 69.7–82.1%) and “poor” quality photographs (AUC = 1.00 (95% CI, 0.99–1.00), accuracy = 99.1% (95% CI, 98.6–99.6%), sensitivity = 81.5% (95% CI, 70.6–93.8%), specificity = 99.7% (95% CI, 99.6–100.0%). “Borderline” quality images were also accurately classified (AUC = 0.90 (95% CI, 0.88–0.93), accuracy = 90.6% (95% CI, 89.1–92.2%), sensitivity = 65.4% (95% CI, 56.6–72.9%), specificity = 93.4% (95% CI, 92.1–94.8%). The overall accuracy to distinguish among the three classes was 90.6% (95% CI, 89.1–92.1%), suggesting that this DLS could select optimal quality fundus photographs in patients with neuro-ophthalmic and neurological disorders affecting the ONH.

We are developing an age-related macular degeneration (AMD) health-related quality of life (HRQoL) item bank, applicable to Western and Asian populations. We report primarily on content generation and refinement, but also compare the HRQoL issues reported in our study with Western studies and current AMD-HRQoL questionnaires. In this cross-sectional, qualitative study of AMD patients attending the Singapore National Eye Centre (May–December 2019), items/domains were generated from: (1) AMD-specific questionnaires; (2) published articles; (3) focus groups/semi-structured interviews with AMD patients (n = 27); and (4) written feedback from retinal experts. Following thematic analysis, items were systematically refined to a minimally representative set and pre-tested using cognitive interviews with 16 AMD patients. Of the 27 patients (mean ± standard deviation age 67.9 ± 7.0; 59.2% male), 18 (66.7%), two (7.4%), and seven (25.9%) had no, early-intermediate, and late/advanced AMD (better eye), respectively. Whilst some HRQoL issues, e.g. activity limitation, mobility, lighting, and concerns were similarly reported by Western patients and covered by other questionnaires, others like anxiety about intravitreal injections, work tasks, and financial dependency were novel. Overall, 462 items within seven independent HRQoL domains were identified: Activity limitation, Lighting, Mobility, Emotional, Concerns, AMD management, and Work. Following item refinement, items were reduced to 219, with 31 items undergoing amendment. Our 7-domain, 219-item AMD-specific HRQoL instrument will undergo psychometric testing and calibration for computerized adaptive testing. The future instrument will enable users to precisely, rapidly, and comprehensively quantify the HRQoL impact of AMD and associated treatments, with item coverage relevant across several populations.

Recent advances in artificial intelligence have provided ophthalmologists with fast, accurate, and automated means for diagnosing and treating ocular conditions, paving the way to a modern and scalable eye care system. Compared to other ophthalmic disciplines, neuro-ophthalmology has, until recently, not benefitted from significant advances in the area of artificial intelligence. In this narrative review, we summarize and discuss recent advancements utilizing artificial intelligence for the detection of structural and functional optic nerve head abnormalities, and ocular movement disorders in neuro-ophthalmology.

The purpose of this study is to characterize the retinal and choroidal vascular networks in some of the most common animal species using swept-source optical coherence tomography angiography (SS-OCTA). Retinal angiographic images were acquired from healthy, anesthetized animals of seven species (mouse, rat, pig, rabbit, guinea pig, chicken, and non-human primate). We generated the enface angiograms to visualize the different retinal vascular plexuses and the choroidal vascular plexus. Quantitative OCTA metrics, including perfusion density, vessel density, and fractal dimension, were compared amongst the different species. There was a noticeable difference in the OCTA enface maps of the distinct vascular layers amongst the various species. Specifically, the non-human primate retina has the highest level of perfusion density and vessel density, whereas the rabbit retina exhibited the lowest level of vessel density. The mouse and the rat retina shared similar vascular patterns, and there was no difference in the OCTA metrics. Using one specific SS-OCTA system for all experiments eliminated multiple instrument-dependent factors, but the lateral resolution was still affected by eye size. Effects of varying lateral resolution on vascular metrics were investigated via a simulation. Overall, we achieved a large field of view of the distinct retinal and choroidal vascular plexuses and quantified the vascular metrics in multiple species, which could serve as protocol guidance and atlas to study the retinal and choroidal vascular abnormalities and their roles in ocular diseases resembling in-vivo histology.

In this population-based, cross-sectional study of Indian and Malay adults in Singapore aged 40 years or above, intermediate or high risk of obstructive sleep apnea (OSA) was associated with 50% higher odds of having glaucoma.The relationship between OSA and glaucoma is unclear. We assessed the association between the risk of OSA and glaucoma in an Asian population.In this population-based, cross-sectional study, we included Indian and Malay adults aged 40 years or above recruited between 2011 and 2015. Glaucoma was assessed by trained ophthalmologists and classified into primary open angle glaucoma (POAG) and primary angle closure glaucoma (PACG). OSA risk was assessed with the Snoring, Tiredness, Observed apnea, High blood pressure, Body mass index, Age, Neck circumference, and male Gender (STOP-Bang) questionnaire and categorized as low risk (3) or intermediate/higher risk (≥3). We used multivariable logistic regression models to evaluate the relationship between risk of OSA and glaucoma adjusted for key variables, and further stratified for subtype and ethnicity.Of the 3126 participants (mean age: 63.1±9.6 y; 52.5% female), 134 (4.3%) had glaucoma, comprising 86 (2.8%) POAG, 22 (0.7%) PACG and 26 (0.8%) secondary glaucomas, and 1182 (37.8%) had an intermediate/higher risk of OSA. Compared with individuals with a low risk of OSA, individuals with intermediate/higher risk had 50% greater odds of having glaucoma (odds ratio: 1.55, 95% confidence interval: 1.03-2.33; P =0.035). We observed a nonsignificant increase in likelihood of having POAG in those with intermediate/higher risk of OSA compared with those with low risk. The OSA-glaucoma relationship was modified by ethnicity, with Malays with intermediate/higher risk of OSA having a 2-fold risk of having any glaucoma (odds ratio: 2.01, 95% confidence interval: 1.12-3.59 P =0.019); while the same elevated risk was not observed for Indians.Intermediate or high risk of OSA is associated with 50% higher odds of having glaucoma in our Singaporean population, with a 2-fold higher risk of glaucoma observed in Malays (but not Indians); however a conformational sleep study is needed.

BackgroundTo evaluate the impact of full-spectrum light-emitting diodes (LEDs) mimicking sunlight on ocular axial elongation and refractive error development in a chicken model of myopia.MethodsA total of 39 chicks (Lohmann brown), 1 day-old, were randomly distributed into 3 groups. Animals were housed for 28 days in a temperature-controlled enclosure, under a 12/12h light/dark cycle of isoluminant (∼285 Lux) fluorescent [n = 18, (4000K, FL-4000)] or Sunlike-LED [n=12, (4000K, SL-4000); n = 9, (6500K, SL-6500)] white lights. Myopia was induced monocularly in all chicks by random occlusion of one eye with a frosted diffuser, from day 1 post-hatching (D1) until D14. On D14, diffusers were removed, and recovery from myopia was monitored under the same experimental light condition. Axial length (AL), refractive status, choroidal thickness and anterior chamber depth were recorded on days 1, 7, 14, 22 and 28.Ex vivoscleral collagen fibre thicknesses were measured from scanning electron microscopy images. Differences in outcome measures between eyes and groups were compared using 2-way repeated-measures ANOVA.ResultsThere was no significant difference between groups in the AL and refraction of form-deprived (FD) eyes during form-deprivation (D1 to D14). FD eyes of animals raised under SL-4000 and SL-6500 recovered more rapidly from excessive axial elongation than those of animals raised under FL-4000, by D22 and D28. Correspondingly, the refractive status of FD eyes exposed to SL-4000 and SL-6500 was close to that of control eyes by D28. The choroid became thicker during recovery in FD eyes compared to control eyes, in all groups. Choroidal thickness was significantly greater in FD eyes of chickens raised under SL-6500 than in animals raised under FL-4000 (P< 0.01). The diameter of scleral collagen fibrils was significantly greater in recovering FD eyes of chickens raised under SL-6500, than in those raised under FL-4000 (P= 0.04) and SL-4000 (P= 0.002).ConclusionsCompared to fluorescent light, moderate intensities of full-spectrum Sunlike-LEDs can accelerate recovery from form-deprivation myopia in chickens, potentially through choroid-mediated pathways increasing the diameter of scleral collagen fibrils. This study highlights an important implication of the spectral content of white light on ocular growth and emmetropization.

Background and ObjectivesThe distinction of papilledema from other optic nerve head (ONH) lesions mimicking papilledema, such as optic disc drusen (ODD), can be difficult in clinical practice. We aimed the following: (1) to develop a deep learning algorithm to automatically identify major structures of the ONH in 3-dimensional (3D) optical coherence tomography (OCT) scans and (2) to exploit such information to robustly differentiate among ODD, papilledema, and healthy ONHs.MethodsThis was a cross-sectional comparative study of patients from 3 sites (Singapore, Denmark, and Australia) with confirmed ODD, those with papilledema due to raised intracranial pressure, and healthy controls. Raster scans of the ONH were acquired using OCT imaging and then processed to improve deep-tissue visibility. First, a deep learning algorithm was developed to identify major ONH tissues and ODD regions. The performance of our algorithm was assessed using the Dice coefficient. Second, a classification algorithm (random forest) was designed to perform 3-class classifications (1: ODD, 2: papilledema, and 3: healthy ONHs) strictly from their drusen and prelamina swelling scores (calculated from the segmentations). To assess performance, we reported the area under the receiver operating characteristic curve for each class.ResultsA total of 241 patients (256 imaged ONHs, including 105 ODD, 51 papilledema, and 100 healthy ONHs) were retrospectively included in this study. Using OCT images of the ONH, our segmentation algorithm was able to isolate neural and connective tissues and ODD regions/conglomerates whenever present. This was confirmed by an averaged Dice coefficient of 0.93 ± 0.03 on the test set, corresponding to good segmentation performance. Classification was achieved with high AUCs, that is, 0.99 ± 0.001 for the detection of ODD, 0.99 ± 0.005 for the detection of papilledema, and 0.98 ± 0.01 for the detection of healthy ONHs.DiscussionOur artificial intelligence approach can discriminate ODD from papilledema, strictly using a single OCT scan of the ONH. Our classification performance was very good in the studied population, with the caveat that validation in a much larger population is warranted. Our approach may have the potential to establish OCT imaging as one of the mainstays of diagnostic imaging for ONH disorders in neuro-ophthalmology, in addition to fundus photography.

Deep learning (DL)-based retinal image quality assessment (RIQA) algorithms have been gaining popularity, as a solution to reduce the frequency of diagnostically unusable images. Most existing RIQA tools target retinal conditions, with a dearth of studies looking into RIQA models for optic nerve head (ONH) disorders. The recent success of DL systems in detecting ONH abnormalities on color fundus images prompts the development of tailored RIQA algorithms for these specific conditions. In this review, we discuss recent progress in DL-based RIQA models in general and the need for RIQA models tailored for ONH disorders. Finally, we propose suggestions for such models in the future.

BACKGROUND To assess pointwise corneal elevation changes after forced eyelid closure test (FECT) in the eyes of healthy subjects and in eyes with keratoconus. METHODS Twenty-nine subjects with keratoconus and 31 healthy volunteers were evaluated. Patients with keratoconus who had corneal hydrops, apical scarring, corneal thickness ≤ 400 μm, ocular surface disease, contact lens wear on the examination day and a history of corneal cross-linking were excluded. Exclusion criteria for healthy participants were spherical error > +3.00 D and 1.50 D, corneal curvature > 47 D, ocular allergy, clinical findings and family history of keratoconus. Pentacam was performed before and after 20 seconds of FECT and raw data were extracted from the built-in software. Pointwise anterior and posterior elevation changes in the central 8 mm cornea were assessed using paired samples t-test and heat maps were constructed to reflect mean changes and statistically significant data points. Statistical significance was assumed at p

Chromatic pupillometry is an emerging modality in the assessment of retinal and optic nerve disorders. Herein, we evaluate the effect of low and moderate refractive errors on pupillary responses to blue- and red-light stimuli in a healthy older population. This study included 139 participants (≥50 years) grouped by refractive error: moderate myopes (>−6.0D and ≤−3.0D, n = 24), low myopes (>−3.0D and 0.5D and R = −0.26, P R = 0.37, P R = 0.43, P P = 0.45), while constriction amplitudes to red light were greater in hyperopes compared to emmetropes (P = 0.04) at moderate to bright light intensities (12.25–14.0 Log photons/cm²/s). Our results demonstrate that low and moderate myopia do not alter pupillary responses to ramping-up blue- and red-light stimuli in healthy older individuals. Conversely, pupillary responses to red light should be interpreted cautiously in hyperopic eyes.

Background/aimsEarly detection and treatment of glaucoma can delay vision loss. In this study, we evaluate the performance of handheld chromatic pupillometry (HCP) for the objective and rapid detection of functional loss in glaucoma.MethodsIn this clinic-based, prospective study, we enrolled 149 patients (median (IQR) years: 68.5 (13.6) years) with confirmed glaucoma and 173 healthy controls (55.2 (26.7) years). Changes in pupil size in response to 9 s of exponentially increasing blue (469 nm) and red (640 nm) light-stimuli were assessed monocularly using a custom-built handheld pupillometer. Pupillometric features were extracted from individual traces and compared between groups. Features with the highest classification potential, selected using a gradient boosting machine technique, were incorporated into a generalised linear model for glaucoma classification. Receiver operating characteristic curve analyses (ROC) were used to compare the performance of HCP, optical coherence tomography (OCT) and Humphrey Visual Field (HVF).ResultsPupillary light responses were altered in glaucoma compared with controls. For glaucoma classification, HCP yielded an area under the ROC curve (AUC) of 0.94 (95% CI 0.91 to 0.96), a sensitivity of 87.9% and specificity of 88.4%. The classification performance of HCP in early-moderate glaucoma (visual field mean deviation (VFMD) > -12 dB; AUC=0.91 (95% CI 0.87 to 0.95)) was similar to HVF (AUC=0.91) and reduced compared with OCT (AUC=0.97; p=0.01). For severe glaucoma (VFMD ≤ -12 dB), HCP had an excellent classification performance (AUC=0.98, 95% CI 0.97 to 1) that was similar to HVF and OCT.ConclusionHCP allows for an accurate, objective and rapid detection of functional loss in glaucomatous eyes of different severities.

Background/aimsTo evaluate the association of reported time outdoors and light exposure patterns with myopia among children aged 9 years from the Growing Up in Singapore Towards Healthy Outcomes birth cohort.MethodsWe assessed reported time outdoors (min/day), light exposure patterns and outdoor activities of children aged 9 years (n=483) with a questionnaire, the FitSight watch and a 7-day activity diary. Light levels, the duration, timing and frequency of light exposure were assessed. Cycloplegic spherical equivalent (SE), myopia (SE≤−0.5 D) and axial length (AL) of paired eyes were analysed using generalised estimating equations.ResultsIn this study, 483 (966 eyes) multiethnic children (50.0% boys, 59.8% Chinese, 42.2% myopic) were included. Reported time outdoors (mean±SD) was 100±93 min/day, and average light levels were 458±228 lux. Of the total duration children spent at light levels of ≥1000 lux (37±19 min/day), 76% were spent below 5000 lux. Peak light exposure occurred at mid-day. Children had 1.7±1.0 light exposure episodes/day. Common outdoor activities were walks, neighbourhood play and swimming. Greater reported time outdoors was associated with lower odds of myopia (OR=0.82, 95% CI 0.70 to 0.95/hour increase daily; p=0.009). Light levels, timing and frequency of light exposures were not associated with myopia, SE or AL (p>0.05).ConclusionReported time outdoors, light levels and number of light exposure episodes were low among Singaporean children aged 9 years. Reported time outdoors was protective against myopia but not light levels or specific light measures. A multipronged approach to increase time outdoors is recommended in the combat against the myopia epidemic.

Contact lens is a major risk factor for microbial keratitis among contact lens wearers. Chemical strategies that can prevent microbial adhesion and biofilm formation are required to improve a wearer's hygiene and safety. Taking advantage of the material-independent properties of a polydopamine (pDA) coating, we investigated the role of covalent/noncovalent interactions of the antimicrobials and pDA in conferring long-term antimicrobial activities. The developed antimicrobial contact lenses not only retain their antibacterial efficiency against different bacterial strains for 2 weeks but also inhibit microbial adhesion and biofilm formation on the lens surfaces. The designed antimicrobial coatings were found to be safe for ocular cell lines. Moreover, the antimicrobial coatings did not affect the functional and surface properties of coated contact lenses. This methodology can be used to protect the contact lenses from microbial contamination for prolonged periods and has the potential to be extended for designing antimicrobial coatings for other medical devices as well.

Despite the current hype in almost every area of the society and medicine, artificial intelligence (AI) is not very new. Machine learning (ML), one of the first AI methods, was described in the 50s, followed recently by an extraordinary technical development of computers and processing power leading to the current achievements of deep learning (DL). DL is the current state-of-art technique in ML, particularly well adapted for image analysis. DL techniques have allowed new algorithms to make predictions on various outcomes and diagnosis, based on « learning » (training) on large datasets, using either a supervised (labelled data) or un-supervised (unlabelled data) approach. As in other areas of DL, the input data is used for training purposes and needs to be selected according to the highest clinical diagnostic standards, because it represents the « ground truth » or the « reference standard ». All the results provided by a DL algorithm will be compared to this ground truth. After the training phase, the performance of the algorithm is tested, first via internal validation (cross-validation) and, more importantly, via external testing (in totally novel datasets). The datasets used for training, internal validation and external testing need to be distinct and should not intersect at any time. The performance of an algorithm is expressed as diagnostic accuracy, sensitivity, specificity and area under the receiving operating curve (AUC), compared to the reference standard. The current, extraordinary boom of AI requires a high need of rigorous, controlled, prospective evaluations to demonstrate the impact of AI systems in health outcomes. In response to this need, AI consensus groups have recently elaborated international guidelines regarding AI interventions, including instructions and skills required for use of AI systems, including considerations for the handling of input and output data, the human–AI interaction and analysis of error cases [1].

A method for rapid and objective assessment of ocular lens density and transmittance is needed for research and clinical practice. The aim of this study was to determine whether the Purkinje image-based technique can be used for objective and accurate quantification of spectral density and transmittance of ocular media (the mainly crystalline lens) in visible light. Twenty-six individuals (10 young, 9 middle-aged and 7 older individuals) participated in this study. Spectral lens density was evaluated by detecting the intensity of the IVth Purkinje image for different wavelengths. Subsequently, optical density index (ODI), the area under the curve in the lens density spectrum, was calculated and ODIs were compared with clinical lens opacification scales assessed subjectively using a slit lamp. Spectral lens transmittance was estimated from the lens density spectrum. Lens densities were higher in the short wavelength region of the visible spectrum across all age groups. ODI was highly correlated with the clinical opacification scale, while lens transmittance decreased with aging. Our results showed that spectral transmittance of the human crystalline lens can be easily estimated from optical density spectra evaluated objectively and rapidly using the Purkinje image-based technique. Our results provide clinicians and scientists with an accurate, rapid and objective technique for quantification of lens transmittance.

Objective To compare the diagnostic performance of an artificial intelligence deep learning system with that of expert neuro-ophthalmologists in classifying optic disc appearance. Methods The deep learning system was previously trained and validated on 14,341 ocular fundus photographs from 19 international centers. The performance of the system was evaluated on 800 new fundus photographs (400 normal optic discs, 201 papilledema [disc edema from elevated intracranial pressure], 199 other optic disc abnormalities) and compared with that of 2 expert neuro-ophthalmologists who independently reviewed the same randomly presented images without clinical information. Area under the receiver operating characteristic curve, accuracy, sensitivity, and specificity were calculated. Results The system correctly classified 678 of 800 (84.7%) photographs, compared with 675 of 800 (84.4%) for Expert 1 and 641 of 800 (80.1%) for Expert 2. The system yielded areas under the receiver operating characteristic curve of 0.97 (95% confidence interval [CI] = 0.96-0.98), 0.96 (95% CI = 0.94-0.97), and 0.89 (95% CI = 0.87-0.92) for the detection of normal discs, papilledema, and other disc abnormalities, respectively. The accuracy, sensitivity, and specificity of the system's classification of optic discs were similar to or better than the 2 experts. Intergrader agreement at the eye level was 0.71 (95% CI = 0.67-0.76) between Expert 1 and Expert 2, 0.72 (95% CI = 0.68-0.76) between the system and Expert 1, and 0.65 (95% CI = 0.61-0.70) between the system and Expert 2. Interpretation The performance of this deep learning system at classifying optic disc abnormalities was at least as good as 2 expert neuro-ophthalmologists. Future prospective studies are needed to validate this system as a diagnostic aid in relevant clinical settings. ANN NEUROL 2020;88:785-795.

BACKGROUND: Nonophthalmologist physicians do not confidently perform direct ophthalmos-copy. The use of artificial intelligence to detect papilledema and other optic-disk abnormalities from fundus photographs has not been well studied METHODS: We trained, validated, and externally tested a deep-learning system to classify optic disks as being normal or having papilledema or other abnormalities from 15,846 retrospectively collected ocular fundus photographs that had been obtained with pharmacologic pupillary dilation and various digital cameras in persons from multiple ethnic populations. Of these photographs, 14,341 from 19 sites in 11 coun-tries were used for training and validation, and 1505 photographs from 5 other sites were used for external testing. Performance at classifying the optic-disk ap-pearance was evaluated by calculating the area under the receiver-operating-characteristic curve (AUC), sensitivity, and specificity, as compared with a reference standard of clinical diagnoses by neuro-ophthalmologists RESULTS: The training and validation data sets from 6779 patients included 14,341 photo-graphs: 9156 of normal disks, 2148 of disks with papilledema, and 3037 of disks with other abnormalities. The percentage classified as being normal ranged across sites from 9.8 to 100%; the percentage classified as having papilledema ranged across sites from zero to 59.5%. In the validation set, the system discriminated disks with papilledema from normal disks and disks with nonpapilledema abnor-malities with an AUC of 0.99 (95% confidence interval [CI], 0.98 to 0.99) and normal from abnormal disks with an AUC of 0.99 (95% CI, 0.99 to 0.99). In the external-testing data set of 1505 photographs, the system had an AUC for the detection of papilledema of 0.96 (95% CI, 0.95 to 0.97), a sensitivity of 96.4% (95% CI, 93.9 to 98.3), and a specificity of 84.7% (95% CI, 82.3 to 87.1).CONCLUSIONSA deep-learning system using fundus photographs with pharmacologically dilated pupils differentiated among optic disks with papilledema, normal disks, and disks with nonpapilledema abnormalities. (Funded by the Singapore National Medical Research Council and the SingHealth Duke–NUS Ophthalmology and Visual Sci-ences Academic Clinical Program.)

Purpose To evaluate the ability of chromatic pupillometry to reveal abnormal pupillary responses to light in patients with early-stage primary open-angle glaucoma (POAG) and to test whether the degree of pupillometric impairment correlates with structural hallmarks of optic nerve damage in the disease. Design Cross-sectional study. Participants Forty-six patients with early-stage POAG (63.4±8.3 years, 63% male, 87% ethnic-Chinese) and 90 age-matched healthy controls (61.4±8.6 years, 34% male, 89% ethnic-Chinese). Patients with POAG had a visual field mean deviation (VFMD) of −6 decibels or better on automated perimetry. Methods Each participant underwent a monocular 2-minute exposure to blue light (462 nm) followed by another 2-minute exposure to red light (638 nm) using a modified Ganzfeld dome equipped with a light-emitting diode lighting system. The light stimuli intensity was increased logarithmically to evaluate the combined extrinsic and intrinsic response of intrinsically photosensitive retinal ganglion cells (ipRGCs). Light-induced changes in horizontal pupil diameter were assessed monocularly using infrared pupillography. Main Outcome Measures Baseline-adjusted, light-induced pupillary constriction amplitudes were calculated, and individual irradiance-response curves were constructed for each stimulus. Pupillary constriction amplitudes were compared between groups and across light intensities using a linear mixed model analysis. The linear relationship between pupillometric parameters and different structural and functional features of glaucoma was assessed using Pearson's correlation analysis. Results Light-induced pupillary constriction was reduced in patients with early-stage POAG compared with controls at moderate to high irradiances (≥11 Log photons/cm 2 /s) of blue ( P = 0.003) and red ( P r = 0.51, P r = 0.45, P = 0.002) in patients with POAG but not in controls. Conversely, pupillometric parameters were not correlated with visual field scores in patients with early-stage POAG. Conclusions Patients with early-stage POAG exhibit reduced pupillary responses to moderate and high irradiances of blue and red lights. This wavelength-independent functional alteration correlates with structural thinning of the RNFL and could be the consequence of dysfunction or loss of melanopsin expressing ipRGCs in the early stages of the disease.

Background Myasthenia gravis (MG) is a chronic autoimmune neuromuscular disease characterized by weakness of bodily skeletal muscles. Office-based diagnostic tests such as repetitive nerve stimulation (RNS), single fiber electromyography (SFEMG), and the ice test, are used to refine the differential clinical diagnosis of this disease. Evaluating the clinical sensitivity and specificity of these tests, however, may be confounded by lack of a gold standard, non-blinding, incorporation bias, use of non-representative populations and retrospective data. Objective In this study comprising a large Asian cohort of 127 patients recruited from a Neuro-ophthalmology clinic, we minimized aforementioned confounders and tested the diagnostic value of 3 office-based tests against 2 reference standards of MG by virtue of clinical features, antibody assay and response to treatment. Results Regardless of the reference standard used, the ice and SFEMG tests displayed a higher sensitivity (86.0 to 97.3%) compared to the RNS test (21.3 to 30.6%). Conversely, the specificity of the ice (31.3%) and SFEMG (21.7% and 17.2%) tests were reduced compared to the RNS test (82.6% and 84.4%). The combined use of the ice test and SFEMG, improved the specificity of MG diagnosis to 63.6% and 64.3%, without affecting the sensitivity of those tests. Conclusion Our findings indicate, in an Asian population, high sensitivity of the SFEMG test and suggest the ice test as a valid, affordable and less technically demanding approach to diagnose MG with ocular involvement. Both ice test and SFEMG alone, however, yielded poor specificity. We suggest that the combination of SFEMG and ice test provides a more reliable diagnosis of MG.

Photic stimulation of rods, cones and intrinsically photosensitive melanopsin-containing retinal ganglion cells (ipRGCs) mediates non-visual light responses, including entrainment of circadian rhythms and pupillary light reflex. Unlike visual responses to photic stimulation, the cerebral correlates of non-visual light responses in humans remains elusive. In this study, we used functional magnetic resonance imaging (fMRI) in 14 healthy young participants, to localize cerebral regions which are differentially activated by metameric light that gave rise to different levels of melanopic excitation. Mean blood oxygen-level dependent (BOLD) responses disclosed bilateral activation of the frontal eye fields during exposure to light geared towards melanopsin. Furthermore, multivariate pattern analyses showed distinct bilateral pattern activity in the inferior temporal gyri and the caudate nuclei. Taken together, our findings suggest that melanopsin-based photoreception activates a cerebral network including frontal regions, classically involved in attention and ocular motor responses.

Objectif Les troubles du sommeil et du rythme circadien sont des caracteristiques communes du vieillissement. Ces modifications peuvent resulter d’un entrainement inapproprie de l’horloge circadienne par la lumiere. Ici, nous etudions l’impact du vieillissement sur la dynamique de la suppression de la melatonine plasmatique a la lumiere. Methodes Huit sujets sains âges (55 a 63 ans) et cinq jeunes (24 a 27 ans) ont ete exposes a 9 lumieres monochromatiques de longueurs d’onde differentes (420 a 620 nm, within subject design) a une intensite lumineuse fixe de 3,16 × 1013 photons/cm2/s de 00 30–01 30 h. Nous avons construit les spectres de sensibilite a quatre intervalles de 15 minutes de cette suppression de la melatonine, chez les jeunes vs les plus âges. Resultats La suppression de la melatonine suit un pattern differe chez les plus âges. Chez les jeunes, le pic de sensibilite a ete observe entre ∼488 nm apres 15 min d’exposition et ∼480 nm apres 30 a 60 min. Chez les plus âges, un pic stable a ∼494 nm est detecte qu’apres 30 min. La modelisation non lineaire suggere que la contribution photoreceptrice melanopique augmente avec le temps dans les deux groupes, mais demeure plus elevee chez les jeunes. Conclusion Le vieillissement est associe a une reponse temporelle differee des effets non-visuels a la lumiere. Les modifications de la dynamique de suppression de la melatonine pourraient etre dues a une contribution alteree des photorecepteurs et/ou un input different au niveau de l’horloge centrale avec le vieillissement.

Purpose of review The aim of this review is to highlight novel artificial intelligence-based methods for the detection of optic disc abnormalities, with particular focus on neurology and neuro-ophthalmology. Recent findings Methods for detection of optic disc abnormalities on retinal fundus images have evolved considerably over the last few years, from classical ophthalmoscopy to artificial intelligence-based identification methods being applied to retinal imaging with the aim of predicting sight and life-threatening complications of underlying brain or optic nerve conditions. Summary Artificial intelligence and in particular newly developed deep-learning systems are playing an increasingly important role for the detection and classification of acquired neuro-ophthalmic optic disc abnormalities on ocular fundus images. The implementation of automatic deep-learning methods for detection of abnormal optic discs, coupled with innovative hardware solutions for fundus imaging, could revolutionize the practice of neurologists and other non-ophthalmic healthcare providers.

International audience; BACKGROUND: Myoclonic epilepsy with ragged-red fibers (MERRF, OMIM, #545000) is a rare neurological condition mostly caused by the m.8344A>G mitochondrial DNA pathogenic variant, which can variably affect multiple tissues, including the retina and optic nerve. We report detection of visually asymptomatic neuroretinal loss in 3 patients with genetically confirmed MERRF, using spectral domain optical coherence tomography (SD-OCT).METHODS: All patients underwent a complete ophthalmic examination including assessments of visual acuity, color vision, pupillary reactions, extraocular movements, applanation tonometry, slit-lamp, and dilated fundus examinations. Standard automated perimetry or Goldmann kinetic perimetry was performed, as well as fundus photographs and SD-OCT of the optic nerve head and macula.RESULTS: Despite the absence of visual symptoms in all patients, and normal visual acuity and visual fields in 1 patient, the 3 genetically confirmed patients (point mutations m.8344A>G; age range: 18-62 years) with MERRF-related neurological manifestations, displayed thinning of the retinal nerve fiber layer and variable alterations of the macular ganglion cell complex.CONCLUSIONS: Visually asymptomatic patients with genetically confirmed MERRF can display features of structural neuroretinal loss, quantifiable with SD-OCT. Further investigations are needed to establish whether OCT can assess early neurodegeneration in MERRF.

Beyond image formation, the light that is detected by retinal photoreceptors influences subcortical functions, including circadian timing, sleep, and arousal. The physiology of nonimage-forming (NIF) photoresponses in humans is not well understood; therefore, the development of therapeutic interventions based on this physiology, such as bright light therapy to treat chronobiological disorders, remains challenging.Thirty-nine participants were exposed to 60 minutes of either continuous light (n = 8) or sequences of 2-millisecond light flashes (n = 31) with different interstimulus intervals (ISIs; ranging from 2.5 to 240 seconds). Melatonin phase shift and suppression, along with changes in alertness and sleepiness, were assessed.We determined that the human circadian system integrates flash sequences in a nonlinear fashion with a linear rise to a peak response (ISI = 7.6 ± 0.53 seconds) and a power function decrease following the peak of responsivity. At peak ISI, flashes were at least 2-fold more effective in phase delaying the circadian system as compared with exposure to equiluminous continuous light 3,800 times the duration. Flashes did not change melatonin concentrations or alertness in an ISI-dependent manner.We have demonstrated that intermittent light is more effective than continuous light at eliciting circadian changes. These findings cast light on the phenomenology of photic integration and suggest a dichotomous retinohypothalamic network leading to circadian phase shifting and other NIF photoresponses. Further clinical trials are required to judge the practicality of light flash protocols.Clinicaltrials.gov NCT01119365.National Heart, Lung, and Blood Institute (1R01HL108441-01A1) and Department of Veterans Affairs Sierra Pacific Mental Illness Research, Education, and Clinical Center.

Deep learning has recently gained high interest in ophthalmology, due to its ability to detect clinically significant features for diagnosis and prognosis. Despite these significant advances, little is known about the ability of various deep learning systems to be embedded within ophthalmic imaging devices, allowing automated image acquisition. In this work, we will review the existing and future directions for "active acquisition" embedded deep learning, leading to as high quality images with little intervention by the human operator. In clinical practice, the improved image quality should translate into more robust deep learning-based clinical diagnostics. Embedded deep learning will be enabled by the constantly improving hardware performance with low cost. We will briefly review possible computation methods in larger clinical systems. Briefly, they can be included in a three-layer framework composed of edge, fog and cloud layers, the former being performed at a device-level. Improved edge layer performance via "active acquisition" serves as an automatic data curation operator translating to better quality data in electronic health records (EHRs), as well as on the cloud layer, for improved deep learning-based clinical data mining., This work has been submitted to "Therapeutic Advances in Ophthalmology" for possible publication 17 pages, 5 figures