ABSTRACT
BACKGROUND: Optic disc drusen (ODD) represent an important differential diagnosis of papilledema caused by intracranial hypertension, but their distinction may be difficult in clinical practice. The aim of this study was to train, validate, and test a dedicated deep learning system (DLS) for binary classification of ODD vs papilledema (including various subgroups within each category), on conventional mydriatic digital ocular fundus photographs collected in a large international multiethnic population. METHODS: This retrospective study included 4,508 color fundus images in 2,180 patients from 30 neuro-ophthalmology centers (19 countries) participating in the Brain and Optic Nerve Study with Artificial Intelligence (BONSAI) Group. For training and internal validation, we used 857 ODD images and 3,230 papilledema images, in 1,959 patients. External testing was performed on an independent data set (221 patients), including 207 images with ODD (96 visible and 111 buried), provided by 3 centers of the Optic Disc Drusen Studies Consortium, and 214 images of papilledema (92 mild-to-moderate and 122 severe) from a previously validated study. RESULTS: The DLS could accurately distinguish between all ODD and papilledema (all severities included): area under the receiver operating characteristic curve (AUC) 0.97 (95% confidence interval [CI], 0.96-0.98), accuracy 90.5% (95% CI, 88.0%-92.9%), sensitivity 86.0% (95% CI, 82.1%-90.1%), and specificity 94.9% (95% CI, 92.3%-97.6%). The performance of the DLS remained high for discrimination of buried ODD from mild-to-moderate papilledema: AUC 0.93 (95% CI, 0.90-0.96), accuracy 84.2% (95% CI, 80.2%-88.6%), sensitivity 78.4% (95% CI, 72.2%-84.7%), and specificity 91.3% (95% CI, 87.0%-96.4%). CONCLUSIONS: A dedicated DLS can accurately distinguish between ODD and papilledema caused by intracranial hypertension, even when considering buried ODD vs mild-to-moderate papilledema.
ABSTRACT
BACKGROUND: In the mammalian retina, intrinsically-photosensitive retinal ganglion cells (ipRGC) detect light and integrate signals from rods and cones to drive multiple non-visual functions including circadian entrainment and the pupillary light response (PLR). Non-visual photoreception and consequently non-visual sensitivity to light may change across child development. The PLR represents a quick and reliable method for examining non-visual responses to light in children. The purpose of this study was to assess differences in the PLRs to blue and red stimuli, measured one hour prior to bedtime, between children and adolescents. METHODS: Forty healthy participants (8-9 years, n = 21; 15-16 years, n = 19) completed a PLR assessment 1 h before their habitual bedtime. After a 1 h dim-light adaptation period (< 1 lx), baseline pupil diameter was measured in darkness for 30 s, followed by a 10 s exposure to 3.0 × 1013 photons/cm2/s of either red (627 nm) or blue (459 nm) light, and a 40 s recovery in darkness to assess pupillary re-dilation. Subsequently, participants underwent 7 min of dim-light re-adaptation followed by an exposure to the other light condition. Lights were counterbalanced across participants. RESULTS: Across both age groups, maximum pupil constriction was significantly greater (p < 0.001, ηp2 = 0.48) and more sustained (p < 0.001, ηp2 = 0.41) during exposure to blue compared to red light. For adolescents, the post-illumination pupillary response (PIPR), a hallmark of melanopsin function, was larger after blue compared with red light (p = 0.02, d = 0.60). This difference was not observed in children. Across light exposures, children had larger phasic (p < 0.01, ηp2 = 0.20) and maximal (p < 0.01, ηp2 = 0.22) pupil constrictions compared to adolescents. CONCLUSIONS: Blue light elicited a greater and more sustained pupillary response than red light in children and adolescents. However, the overall amplitude of the rod/cone-driven phasic response was greater in children than in adolescents. Our findings using the PLR highlight a higher sensitivity to evening light in children compared to adolescents, and continued maturation of the human non-visual photoreception/system throughout development.
Subject(s)
Light , Pupil , Humans , Adolescent , Child , Male , Female , Pupil/physiology , Pupil/radiation effects , Reflex, Pupillary/physiology , Reflex, Pupillary/radiation effectsABSTRACT
BACKGROUND: Pediatric papilledema often reflects an underlying severe neurologic disorder and may be difficult to appreciate, especially in young children. Ocular fundus photographs are easy to obtain even in young children and in nonophthalmology settings. The aim of our study was to ascertain whether an improved deep-learning system (DLS), previously validated in adults, can accurately identify papilledema and other optic disk abnormalities in children. METHODS: The DLS was tested on mydriatic fundus photographs obtained in a multiethnic pediatric population (<17 years) from three centers (Atlanta-USA; Bucharest-Romania; Singapore). The DLS's multiclass classification accuracy (ie, normal optic disk, papilledema, disks with other abnormality) was calculated, and the DLS's performance to specifically detect papilledema and normal disks was evaluated in a one-vs-rest strategy using the AUC, sensitivity and specificity, with reference to expert neuro-ophthalmologists. RESULTS: External testing was performed on 898 fundus photographs: 447 patients; mean age, 10.33 (231 patients ≤10 years of age; 216, 11-16 years); 558 normal disks, 254 papilledema, 86 other disk abnormalities. Overall multiclass accuracy of the DLS was 89.6% (range, 87.8%-91.6%). The DLS successfully distinguished "normal" from "abnormal" optic disks (AUC 0.99 [0.98-0.99]; sensitivity, 87.3% [84.9%-89.8%]; specificity, 98.5% [97.6%-99.6%]), and "papilledema" from "normal and other" (AUC 0.99 [0.98-1.0]; sensitivity, 98.0% [96.8%-99.4%]; specificity, 94.1% (92.4%-95.9%)]. CONCLUSIONS: Our DLS reliably distinguished papilledema from normal optic disks and other disk abnormalities in children, suggesting it could be utilized as a diagnostic aid for the assessment of optic nerve head appearance in the pediatric age group.
Subject(s)
Deep Learning , Papilledema , Adult , Humans , Child , Child, Preschool , Papilledema/diagnosis , Fundus Oculi , Artificial Intelligence , Optic Nerve , BrainABSTRACT
BACKGROUND: Myopia, commonly known as near-sightedness, has emerged as a global epidemic, impacting almost one in three individuals across the world. The increasing prevalence of myopia during early childhood has heightened the risk of developing high myopia and related sight-threatening eye conditions in adulthood. This surge in myopia rates, occurring within a relatively stable genetic framework, underscores the profound influence of environmental and lifestyle factors on this condition. In this comprehensive narrative review, we shed light on both established and potential environmental and lifestyle contributors that affect the development and progression of myopia. MAIN BODY: Epidemiological and interventional research has consistently revealed a compelling connection between increased outdoor time and a decreased risk of myopia in children. This protective effect may primarily be attributed to exposure to the characteristics of natural light (i.e., sunlight) and the release of retinal dopamine. Conversely, irrespective of outdoor time, excessive engagement in near work can further worsen the onset of myopia. While the exact mechanisms behind this exacerbation are not fully comprehended, it appears to involve shifts in relative peripheral refraction, the overstimulation of accommodation, or a complex interplay of these factors, leading to issues like retinal image defocus, blur, and chromatic aberration. Other potential factors like the spatial frequency of the visual environment, circadian rhythm, sleep, nutrition, smoking, socio-economic status, and education have debatable independent influences on myopia development. CONCLUSION: The environment exerts a significant influence on the development and progression of myopia. Improving the modifiable key environmental predictors like time spent outdoors and engagement in near work can prevent or slow the progression of myopia. The intricate connections between lifestyle and environmental factors often obscure research findings, making it challenging to disentangle their individual effects. This complexity underscores the necessity for prospective studies that employ objective assessments, such as quantifying light exposure and near work, among others. These studies are crucial for gaining a more comprehensive understanding of how various environmental factors can be modified to prevent or slow the progression of myopia.
Subject(s)
Myopia , Child, Preschool , Child , Humans , Prospective Studies , Myopia/epidemiology , Myopia/genetics , Myopia/prevention & control , Refraction, Ocular , Accommodation, Ocular , Circadian RhythmABSTRACT
Age-related sleep and circadian rhythm disturbances may be due to altered nonvisual photoreception. Here, we investigated the temporal dynamics of light-induced melatonin suppression in young and older individuals. In a within-subject design study, young and older participants were exposed for 60 min (0030-0130 at night) to nine narrow-band lights (range: 420-620 nm). Plasma melatonin suppression was calculated at 15, 30, 45, and 60 min time intervals. Individual spectral sensitivity of melatonin suppression and photoreceptor contribution were predicted for each interval and age group. In young participants, melanopsin solely drove melatonin suppression at all time intervals, with a peak sensitivity at 485.3 nm established only after 15 min of light exposure. Conversely, in older participants, spectral light-driven melatonin suppression was best explained by a more complex model combining melanopsin, S-cone, and M-cone functions, with a stable peak (~500 nm) at 30, 45, and 60 min of light exposure. Aging is associated with a distinct photoreceptor contribution to melatonin suppression by light. While in young adults melanopsin-only photoreception is a reliable predictor of melatonin suppression, in older individuals this process is jointly driven by melanopsin, S-cone, and M-cone functions. These findings offer new prospects for customizing light therapy for older individuals.
Subject(s)
Melatonin , Young Adult , Humans , Aged , Photoreceptor Cells, Vertebrate , Retinal Cone Photoreceptor Cells , Rod Opsins , Circadian Rhythm/physiology , AgingABSTRACT
PURPOSE: The Fundus photography vs Ophthalmoscopy Trial Outcomes in the Emergency Department (FOTO-ED) studies showed that ED providers poorly recognized funduscopic findings in patients in the ED. We tested a modified version of the Brain and Optic Nerve Study Artificial Intelligence (BONSAI) deep learning system on nonmydriatic fundus photographs from the FOTO-ED studies to determine if the deep learning system could have improved the detection of papilledema had it been available to ED providers as a real-time diagnostic aid. DESIGN: Retrospective secondary analysis of a cohort of patients included in the FOTO-ED studies. METHODS: The testing data set included 1608 photographs obtained from 828 patients in the FOTO-ED studies. Photographs were reclassified according to the optic disc classification system used by the deep learning system ("normal optic discs," "papilledema," and "other optic disc abnormalities"). The system's performance was evaluated by calculating the area under the receiver operating characteristic curve (AUC), sensitivity, and specificity using a 1-vs-rest strategy, with reference to expert neuro-ophthalmologists. RESULTS: The BONSAI deep learning system successfully distinguished normal from abnormal optic discs (AUC 0.92 [95% confidence interval {CI} 0.90-0.93]; sensitivity 75.6% [73.7%-77.5%] and specificity 89.6% [86.3%-92.8%]), and papilledema from normal and others (AUC 0.97 [0.95-0.99]; sensitivity 84.0% [75.0%-92.6%] and specificity 98.9% [98.5%-99.4%]). Six patients with missed papilledema in 1 eye were correctly identified by the deep learning system as having papilledema in the other eye. CONCLUSIONS: The BONSAI deep learning system was able to reliably identify papilledema and normal optic discs on nonmydriatic photographs obtained in the FOTO-ED studies. Our deep learning system has excellent potential as a diagnostic aid in EDs and non-ophthalmology clinics equipped with nonmydriatic fundus cameras. NOTE: Publication of this article is sponsored by the American Ophthalmological Society.
ABSTRACT
BACKGROUND: There is no consensus on reporting light characteristics in studies investigating non-visual responses to light. This project aimed to develop a reporting checklist for laboratory-based investigations on the impact of light on non-visual physiology. METHODS: A four-step modified Delphi process (three questionnaire-based feedback rounds and one face-to-face group discussion) involving international experts was conducted to reach consensus on the items to be included in the checklist. Following the consensus process, the resulting checklist was tested in a pilot phase with independent experts. FINDINGS: An initial list of 61 items related to reporting light-based interventions was condensed to a final checklist containing 25 items, based upon consensus among experts (final n = 60). Nine items were deemed necessary to report regardless of research question or context. A description of each item is provided in the accompanying Explanation and Elaboration (E&E) document. The independent pilot testing phase led to minor textual clarifications in the checklist and E&E document. INTERPRETATION: The ENLIGHT Checklist is the first consensus-based checklist for documenting and reporting ocular light-based interventions for human studies. The implementation of the checklist will enhance the impact of light-based research by ensuring comprehensive documentation, enhancing reproducibility, and enabling data aggregation across studies. FUNDING: Network of European Institutes for Advanced Study (NETIAS) Constructive Advanced Thinking (CAT) programme; Sir Henry Wellcome Postdoctoral Fellowship (Wellcome Trust, 204686/Z/16/Z); Netherlands Organisation for Health Research and Development VENI fellowship (2020-09150161910128); U.S. Department of Defense Grant (W81XWH-16-1-0223); National University of Singapore (NUHSRO/2022/038/Startup/08); and National Research Foundation Singapore (NRF2022-THE004-0002).
Subject(s)
Checklist , Research Design , United States , Humans , Reproducibility of Results , ConsensusABSTRACT
Purpose: The purpose of this study was to isolate the structural components of the ex vivo porcine iris tissue and to determine their biomechanical properties. Methods: The porcine stroma and dilator tissues were separated, and their dimensions were assessed using optical coherence tomography (OCT). The stroma underwent flow test (n = 32) to evaluate for permeability using Darcy's Law (ΔP = 2000 Pa, A = 0.0391 mm2), and both tissues underwent stress relaxation experiments (ε = 0.5 with initial ramp of δε = 0.1) to evaluate for their viscoelastic behaviours (n = 28). Viscoelasticity was characterized by the parameters ß (half width of the Gaussian distribution), τm (mean relaxation time constant), E0 (instantaneous modulus), and E∞ (equilibrium modulus). Results: For the stroma, the hydraulic permeability was 9.49 ± 3.05 × 10-6 mm2/Pa · s, and the viscoelastic parameters were ß = 2.50 ± 1.40, and τm = 7.43 ± 4.96 s, with the 2 moduli calculated to be E0 = 14.14 ± 6.44 kPa and E∞ = 6.08 ± 2.74 kPa. For the dilator tissue, the viscoelastic parameters were ß = 2.06 ± 1.33 and τm = 1.28 ± 1.27 seconds, with the 2 moduli calculated to be E0 = 9.16 ± 3.03 kPa and E∞ = 5.54 ± 1.98 kPa. Conclusions: We have established a new protocol to evaluate the biomechanical properties of the structural layers of the iris. Overall, the stroma was permeable and exhibited smaller moduli than those of the dilator muscle. An improved characterization of iris biomechanics may form the basis to further our understanding of angle closure glaucoma.
Subject(s)
Glaucoma, Angle-Closure , Iris , Swine , Animals , Iris/physiology , Biomechanical Phenomena/physiology , Tomography, Optical CoherenceABSTRACT
Purpose: To assess differences in the pupillary light responses (PLRs) to blue and red evening lights between children and adolescents. Methods: Forty healthy participants (8-9 years, n=21; 15-16 years, n=19) completed a PLR assessment 1 h before their habitual bedtime. After a 1 h dim-light adaptation period (<1 lux), baseline pupil diameter was measured in darkness for 30 s, followed by a 10 s exposure to 3.0×1013 photons/cm2/s of either red (627 nm) or blue (459 nm) light, and a 40 s recovery in darkness to assess pupillary re-dilation. Subsequently, participants underwent 7 min of dim-light re-adaptation followed by an exposure to the other light condition. Lights were counterbalanced across participants. Results: Across both age groups, maximum pupil constriction was significantly greater (p< 0.001, ηp2=0.48) and more sustained (p< 0.001, ηp2=0.41) during exposure to blue compared to red light. For adolescents, the post-illumination pupillary response (PIPR), a hallmark of melanopsin function, was larger after blue compared with red light (p= 0.02, d=0.60). This difference was not observed in children. Across light exposures, children had larger phasic (p< 0.01, ηp2=0.20) and maximal (p< 0.01, ηp2=0.22) pupil constrictions compared to adolescents. Conclusions: Blue light elicited a greater and more sustained pupillary response than red light across participants. However, the overall amplitude of the rod/cone-driven phasic response was greater in children than in adolescents. Our findings using the PLR highlight a higher sensitivity to evening light in children compared to adolescents, and continued maturation of the human non-visual photoreception/system throughout development.
ABSTRACT
PURPOSE: To evaluate factors influencing stabilisation of myopia in the Singapore Cohort of Risk factors for Myopia. METHODS: We evaluated the longitudinal natural history of 424 myopic participants from 1999 to 2022. The outcome was the change in myopia from the adolescence follow-up visit (aged 12-19 years) to the adulthood follow-up visit (aged 26-33 years). Association of predictive factors, including baseline spherical error, gender, ethnicity, parental myopia, time outdoor, near work and age at adolescence, was examined with the dichotomous outcome of adult myopia progression (≤ -1.00 dioptres (D) over 10 years) using multiple logistic regression and progression in linear regression models. RESULTS: For the primary outcome, the mean rate of progression of the outcome was found to be -0.04±0.09 D per year from the adolescent to the adulthood follow-up visits. 82.3% (95% CI 78.3% to 85.8%) had myopia stabilisation, with progression of less than 1.00 D over 10 years while 61.3% (95% CI 56.5% to 66.0%) of the subjects had progression of less than 0.50 D. In logistic regression models, both male gender (p=0.035) and non-Chinese ethnicity (p=0.032) were more likely to achieve myopia stabilisation while in linear multivariate regression models, males had a significantly slower degree of myopia progression (p=0.021). CONCLUSION: 5 in 6 Singaporean young adults had myopia stabilisation. Male gender is 2 times and non-Chinese ethnicities are 2.5 times more likely to achieve myopia stabilisation. However, a proportion of myopes continue to exhibit a clinically significant degree of progression in adulthood.
ABSTRACT
BACKGROUND/AIMS: To assess pupillary light responses (PLRs) in eyes with high myopia (HM) and evaluate the ability of handheld chromatic pupillometry (HCP) to identify glaucomatous functional loss in eyes with HM. METHODS: This prospective, cross-sectional study included 28 emmetropes (EM), 24 high myopes without glaucoma (HM) and 17 high myopes with confirmed glaucoma (HMG), recruited at the Singapore National Eye Center. Monocular PLRs were evaluated using a custom-built handheld pupillometer that recorded changes in horizontal pupil radius in response to 9 s of exponentially increasing blue (469.1 nm) and red (640.1 nm) lights. Fifteen pupillometric features were compared between groups. A logistic regression model (LRM) was used to distinguish HMG eyes from non-glaucomatous eyes (EM and HM). RESULTS: All pupillometric features were similar between EM and HM groups. Phasic constriction to blue (p<0.001) and red (p=0.006) lights, and maximum constriction to blue light (p<0.001) were reduced in HMG compared with EM and HM. Pupillometric features of melanopsin function (postillumination pupillary response, PIPR area under the curve (AUC) 0-12 s (p<0.001) and PIPR 6 s (p=0.01) to blue light) were reduced in HMG. Using only three pupillometric features, the LRM could classify glaucomatous from non-glaucomatous eyes with an AUC of 0.89 (95% CI 0.77 to 1.00), sensitivity 94.1% (95% CI 82.4% to 100.0%) and specificity 78.8% (95% CI 67.3% to 90.4%). CONCLUSION: PLRs to ramping-up light stimuli are unaltered in highly myopic eyes without other diagnosed ocular conditions. Conversely, HCP can distinguish glaucomatous functional loss in eyes with HM and can be a useful tool to detect/confirm the presence of glaucoma in patients with HM.
ABSTRACT
In myopic eyes, pathological remodelling of collagen in the posterior sclera has mostly been observed ex vivo. Here we report the development of triple-input polarization-sensitive optical coherence tomography (OCT) for measuring posterior scleral birefringence. In guinea pigs and humans, the technique offers superior imaging sensitivities and accuracies than dual-input polarization-sensitive OCT. In 8-week-long studies with young guinea pigs, scleral birefringence was positively correlated with spherical equivalent refractive errors and predicted the onset of myopia. In a cross-sectional study involving adult individuals, scleral birefringence was associated with myopia status and negatively correlated with refractive errors. Triple-input polarization-sensitive OCT may help establish posterior scleral birefringence as a non-invasive biomarker for assessing the progression of myopia.
Subject(s)
Myopia , Sclera , Adult , Humans , Animals , Guinea Pigs , Sclera/diagnostic imaging , Sclera/pathology , Birefringence , Cross-Sectional Studies , Myopia/diagnostic imaging , Myopia/pathology , BiomarkersSubject(s)
Eye Diseases , Sleep , Humans , Eye Diseases/epidemiology , Public Health , Outcome Assessment, Health Care , Health PromotionABSTRACT
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.
Subject(s)
Artificial Intelligence , Optic Disk , Humans , Brain/diagnostic imaging , Retina , AgingABSTRACT
Purpose: To evaluate the duration-dependent and synergetic impact of high-intensity light (HL) and unrestricted vision (UnV) on lens-induced myopia (LIM) development in chickens. Methods: Myopia was induced in one eye in chicks (10 groups, n = 126) from day 1 posthatching (D1) until day 8 (D8) using -10 diopter (D) lenses. Fellow eyes remained uncovered as controls. Nine groups were exposed daily to 2, 4, or 6 hours of HL (15,000 lux), UnV (removal of -10 D lens), or both (HL + UnV). 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 interocular difference (IOD = experimental eye - control eye) ± SEM. Results: By D8, LIM increased AL (0.36 ± 0.04 mm), myopic refraction (-9.02 ± 0.37 D), and choroidal thinning (-90.27 ± 16.44 µm) in the LIM group (all, P < 0.001). Compared to the LIM group, exposure to 2, 4, or 6 hours of HL, UnV, or HL + UnV reduced myopic refraction in a duration-dependent manner, with UnV being more effective than HL (P < 0.05). Only 6 hours of HL + UnV (not 2 or 4 hours) prevented LIM and was more effective than UnV (P = 0.004) or HL (P < 0.001) in reducing myopic refraction and more effective than HL (P < 0.001) in reducing axial elongation. Conclusions: Daily exposure to 2, 4, or 6 hours of HL, UnV, or HL + UnV reduced lens-induced myopic refraction in a duration-dependent manner in chickens. Only 6 hours of HL + UnV completely stopped LIM development. The synergetic effect of HL and UnV is dependent on the duration of the interventions.
Subject(s)
Chickens , Myopia , Animals , Animals, Newborn , Myopia/prevention & control , Eye , Vision, Ocular , Refraction, Ocular , Choroid , Disease Models, AnimalABSTRACT
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.
ABSTRACT
BACKGROUND: The examination of the optic nerve head (optic disc) is mandatory in patients with headache, hypertension, or any neurological symptoms, yet it is rarely or poorly performed in general clinics. We recently developed a brain and optic nerve study with artificial intelligence-deep learning system (BONSAI-DLS) capable of accurately detecting optic disc abnormalities including papilledema (swelling due to elevated intracranial pressure) on digital fundus photographs with a comparable classification performance to expert neuro-ophthalmologists, but its performance compared to first-line clinicians remains unknown. METHODS: In this international, cross-sectional multicenter study, the DLS, trained on 14,341 fundus photographs, was tested on a retrospectively collected convenience sample of 800 photographs (400 normal optic discs, 201 papilledema and 199 other abnormalities) from 454 patients with a robust ground truth diagnosis provided by the referring expert neuro-ophthalmologists. The areas under the receiver-operating-characteristic curves were calculated for the BONSAI-DLS. Error rates, accuracy, sensitivity, and specificity of the algorithm were compared with those of 30 clinicians with or without ophthalmic training (6 general ophthalmologists, 6 optometrists, 6 neurologists, 6 internists, 6 emergency department [ED] physicians) who graded the same testing set of images. RESULTS: With an error rate of 15.3%, the DLS outperformed all clinicians (average error rates 24.4%, 24.8%, 38.2%, 44.8%, 47.9% for general ophthalmologists, optometrists, neurologists, internists and ED physicians, respectively) in the overall classification of optic disc appearance. The DLS displayed significantly higher accuracies than 100%, 86.7% and 93.3% of clinicians (n = 30) for the classification of papilledema, normal, and other disc abnormalities, respectively. CONCLUSIONS: The performance of the BONSAI-DLS to classify optic discs on fundus photographs was superior to that of clinicians with or without ophthalmic training. A trained DLS may offer valuable diagnostic aid to clinicians from various clinical settings for the screening of optic disc abnormalities harboring potentially sight- or life-threatening neurological conditions.
Subject(s)
Deep Learning , Optic Disk , Papilledema , Humans , Optic Disk/diagnostic imaging , Artificial Intelligence , Retrospective Studies , Cross-Sectional StudiesABSTRACT
BACKGROUND AND OBJECTIVES: The 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. METHODS: This 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. RESULTS: A 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. DISCUSSION: Our 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.
Subject(s)
Optic Disk Drusen , Optic Disk , Papilledema , Humans , Optic Disk/diagnostic imaging , Optic Disk/pathology , Papilledema/diagnostic imaging , Optic Disk Drusen/diagnosis , Optic Disk Drusen/diagnostic imaging , Artificial Intelligence , Retrospective Studies , Cross-Sectional Studies , Tomography, Optical Coherence/methodsABSTRACT
BACKGROUND/AIMS: Early 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. METHODS: In 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). RESULTS: Pupillary 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. CONCLUSION: HCP allows for an accurate, objective and rapid detection of functional loss in glaucomatous eyes of different severities.