Artificial intelligence in glaucoma detection using color fundus photographs.

PURPOSE: To explore the potential of artificial intelligence (AI) for glaucoma detection using deep learning algorithm and evaluate its accuracy for image classification of glaucomatous optic neuropathy (GON) from color fundus photographs. METHODS: A total of 1375 color fundus photographs, 735 normal optic nerve head and 640 GON, were uploaded on the AI software for training, validation, and testing using deep learning model, which is based on Residual Network (Res Net) 50V2. For initial training and validation, 400 fundus images (200 normal and 200 GON) were uploaded and for the final training and testing 975 (535 normal and 440 GON) were uploaded later. Accuracy, sensitivity, and specificity were used to evaluate the image classification performance of the algorithm. Also, positive predictive value, negative predictive value, positive likelihood ratio, and negative likelihood ratio were calculated. RESULTS: The model used in the study showed an image classification accuracy of 81.3%, sensitivity of 83%, and specificity of 80% for the detection of GON. The false-negative grading was 17% and false-positive grading was 20% for the image classification of GON. Coexistence of glaucoma in patients with high myopia, early glaucoma in a small disc, and software misclassification of GON were the reasons for false-negative results. Physiological large cupping in a large disc, myopic or titled disc, and software misclassification of normal optic disc were the reasons for false-positive results. CONCLUSION: The model employed in this study achieved a good accuracy, and hence has a good potential in detection of GON using color fundus photographs.

Coexistence of open-angle glaucoma and sarcoidosis-associated optic neuropathy.

BACKGROUND: In cases with advanced glaucomatous disc changes, further changes associated with other optic neuropathies cannot be easily identified. We present a case of preexisting open-angle glaucoma and concurrent involvement of sarcoidosis-associated optic neuropathy. CASE PRESENTATION: A 53-year-old man presented with gradual visual loss in his left eye, which began 1 year ago and accelerated 3 months ago. The best-corrected visual acuity in the right eye was 20/20 and counting fingers in the left. Intraocular pressures (IOP) were 12 mmHg in the right eye and 34 mmHg in the left. We diagnosed him with advanced open-angle glaucoma in the left eye based on the advanced glaucomatous cupping of the left optic disc. The IOP in the left eye dropped to 10 mmHg and was well controlled with antiglaucomatous medication; however, his left optic disc developed pallor 3 months after the treatment. The patient was revealed to be diagnosed with sarcoidosis a month ago and had been treated with systemic corticosteroids thereafter by a pulmonologist. Orbital magnetic resonance imaging revealed sarcoidosis-associated optic neuropathy in the left eye. Subsequently, optic neuropathy occurred in his right eye. CONCLUSIONS: In eyes with advanced glaucomatous disc change, detecting the coexistence of other optic neuropathies can be difficult. This report highlights the importance of careful ophthalmic examinations and investigation for etiologies of other optic neuropathies if non-glaucomatous changes are suspected even in eyes with advanced glaucoma.

Sphenoid Bone Dysplasia: A Rare Cause of Compressive Optic Neuropathy Mimicking Glaucoma.

Fibrous dysplasia is a benign, rare bone disease in which bone is replaced by fibro-osseous tissue to varying degrees. It can present differently depending on the amount of compression caused by the fibro-osseous tissue. Patients are usually asymptomatic, but symptoms related to cranial nerve compression may occur. In this case report, we describe a 45-year-old woman with sphenoid bone dysplasia which compressed the optic nerve and caused unilateral optic disc cupping that mimicked glaucoma. Our case highlights the importance of including compressive etiologies associated with optic disc cupping in the differential diagnosis of glaucoma.

Bilateral compressive optic neuropathy and outer retinopathy due to optic canal hyperostosis in a child with isolated vitamin a deficiency.

PURPOSE: Vitamin A plays a crucial role in rod phototransduction, with deficient levels manifesting as night blindness. Animal models have demonstrated bone dysplasia in the setting of hypovitaminosis A. We present a rare case of bony overgrowth leading to bilateral compressive optic neuropathy, combined with outer retinopathy, in a paediatric patient secondary to isolated vitamin A deficiency. METHODS: A single case report was conducted from Toronto, Canada. RESULTS: A 12-year-old boy with known autism spectrum disorder presented with a 9-month history of progressive painless vision loss. Vision was 20/300 and hand motion in the right and left eye, respectively. Fundus photography demonstrated bilateral optic atrophy and yellow lesions notably in the right eye far periphery. Optical coherence tomography (OCT) imaging demonstrated thinning of the retinal nerve fibre layer, alterations in the ellipsoid zone, as well as retinal pigment epithelium deposits. Computed tomography imaging demonstrated sphenoid bone thickening with narrow optic canals and moderate optic atrophy bilaterally. Full-field electroretinogram (ERG) demonstrated mildly reduced dark adapted (DA) 0.01 b-wave amplitudes and electronegative configuration of DA 3.0 and DA 10.0 ERG; the light adapted ERGs were normal. The patient was treated with pulse vitamin A therapy. Subsequently, the DA ERG normalized, outer retinal changes reversed and vision stabilised; no surgical intervention was conducted. CONCLUSION: This case represents a rare presentation of compressive optic neuropathy with concomitant outer retinopathy secondary to isolated vitamin A deficiency. Despite improvement in outer retinal integrity on OCT imaging and ERG testing results following vitamin A supplementation, no functional improvement was obtained due to severe optic atrophy.

Yield of investigations in patients with questionable nonglaucomatous optic neuropathy.

OBJECTIVE: Nonglaucomatous optic neuropathy (NGON) may often be mistaken for normal-tension glaucoma. The distinction between these two entities is important in determining treatment, prognosis, and need for further investigations. We report characteristics of a cohort of patients referred for neuro-ophthalmologic consultation to distinguish between glaucomatous (GON) and NGON. METHODS: Retrospective chart review of patients presenting to a tertiary neuro-ophthalmology practice investigated for GON versus NGON between 2018 and 2020. Patients were classified as GON or NGON based on presence of optic disc pallor, degree of cupping, central visual acuity, and/or when investigations yielded a cause of NGON. RESULTS: Eighty-three patients were enrolled. Seventy-one patients (86%) were deemed to have possible NGON after initial evaluation and having undergone neuroimaging. Of these, 14 patients (19.7%) were determined to have NGON and 7 patients (9.9%) both GON and NGON. The most common causes of NGON were undetermined (8), previous optic neuritis (4), and neurovascular conflict (2). The yield of neuroimaging was low, with abnormalities seen in only 4 of 71 patients (5.6%). No patients with bilateral cupping and no relative afferent pupillary defect (RAPD) had abnormalities on imaging, and no patients with positive imaging required intervention. Patients with NGON had lower intraocular pressure and cup-to-disc ratio and usually had optic disc pallor (86%). Nerve fibre bundle defects were most common in both GON and NGON, with more nasal defects seen in GON. CONCLUSIONS: When NGON resembles GON, the underlying cause is often undetermined, and the yield of neuroimaging is low. The yield of investigating patients with bilateral cupping and no RAPD for NGON is especially low.

Validation of an autonomous artificial intelligence-based diagnostic system for holistic maculopathy screening in a routine occupational health checkup context.

PURPOSE: This study aims to evaluate the ability of an autonomous artificial intelligence (AI) system for detection of the most common central retinal pathologies in fundus photography. METHODS: Retrospective diagnostic test evaluation on a raw dataset of 5918 images (2839 individuals) evaluated with non-mydriatic cameras during routine occupational health checkups. Three camera models were employed: Optomed Aurora (field of view - FOV 50º, 88% of the dataset), ZEISS VISUSCOUT 100 (FOV 40º, 9%), and Optomed SmartScope M5 (FOV 40º, 3%). Image acquisition took 2 min per patient. Ground truth for each image of the dataset was determined by 2 masked retina specialists, and disagreements were resolved by a 3rd retina specialist. The specific pathologies considered for evaluation were "diabetic retinopathy" (DR), "Age-related macular degeneration" (AMD), "glaucomatous optic neuropathy" (GON), and "Nevus." Images with maculopathy signs that did not match the described taxonomy were classified as "Other." RESULTS: The combination of algorithms to detect any abnormalities had an area under the curve (AUC) of 0.963 with a sensitivity of 92.9% and a specificity of 86.8%. The algorithms individually obtained are as follows: AMD AUC 0.980 (sensitivity 93.8%; specificity 95.7%), DR AUC 0.950 (sensitivity 81.1%; specificity 94.8%), GON AUC 0.889 (sensitivity 53.6% specificity 95.7%), Nevus AUC 0.931 (sensitivity 86.7%; specificity 90.7%). CONCLUSION: Our holistic AI approach reaches high diagnostic accuracy at simultaneous detection of DR, AMD, and Nevus. The integration of pathology-specific algorithms permits higher sensitivities with minimal impact on its specificity. It also reduces the risk of missing incidental findings. Deep learning may facilitate wider screenings of eye diseases.

Visual loss and optic neuropathy in a patient with Klinefelter's syndrome, open-angle glaucoma, vitamin B12 (cobalamin) and folate deficiency.

A 54-year-old man with Klinefelter's syndrome presented to the neuro-ophthalmology clinic with progressive painless visual blurring in the right eye over 2 years. He was receiving intramuscular testosterone therapy for hypogonadism and hypromellose for dry eye. Acuity was reduced bilaterally, and the right optic nerve head appeared pale and asymmetrically cupped. Optical coherence tomography revealed loss of retinal nerve fibre layer thickness in the right eye and visual field testing showed a developing right-ring scotoma. Blood tests showed vitamin B12 and folate deficiencies and polycythaemia. The patient was managed with intramuscular hydroxocobalamin, oral folate administration and re-initiation of his glaucoma medication. In Klinefelter's syndrome, signs of comorbid deficiency can be masked by the polycythaemic effect of testosterone therapy. For patients on long-term testosterone therapy, such as those with Klinefelter's syndrome, we recommend baseline ophthalmic examination and assessment, including intraocular pressure measurement, pachymetry, gonioscopy and screening 24-2 visual field testing.

Glaucoma Mimickers: A major review of causes, diagnostic evaluation, and recommendations.

Background: Glaucoma is characterized by chronic progressive optic neuropathy with corresponding visual field changes, with or without raised intraocular pressure (IOP). When diagnosing glaucoma or monitoring its progression, the examiner has to rely on the appearance of the optic disc, characteristic retinal nerve fiber layer defects, and corresponding visual field defects. However, similar changes and symptoms may be observed in several other disorders of the optic nerve and retina that may mimic glaucoma, often leading to misdiagnosis. Methods and result: The consequences of misdiagnosis not only result in improper treatment that may impact vision but also would negatively affect the overall health, psychological well-being of the patient, and may have considerable economic implications.Conclusion: The current review describes various conditions that mimic glaucoma and the features that help differentiate these conditions from glaucoma.

The role of clinical examination of the optic nerve head in glaucoma today.

PURPOSE OF REVIEW: The study was undertaken to assess the current importance of clinical examination of the optic nerve head (ONH) in glaucoma management. RECENT FINDINGS: ONH changes in glaucoma are easily assessed with good specificity on examination, whereas imaging has not been able to identify pallor of the neuroretinal rim (NRR), disc hemorrhages, or vascular signs of acquired cupping. Glaucomatous neuropathy in primary open angle glaucoma (POAG) has extensive databases on imaging machines, however, other glaucomas such as primary angle closure glaucoma (PACG), juvenile open angle glaucoma (JOAG), congenital, and secondary glaucomas have been reported to have a significantly different ONH morphology. Clinical evaluation with knowledge of the underlying cause of glaucoma allows easy diagnosis of any abnormality, whereas the absence of an imaging database in such eyes makes diagnosis difficult on optical coherence tomography (OCT) and so forth. In eyes with congenital anomalies and dysplastic discs, clinical examination and history provide better identification of glaucomatous damage. Staging of glaucoma on ONH examination when perimetry is unreliable provides a means of determining the amount of damage and appropriate reduction of intraocular pressure (IOP). SUMMARY: Clinical examination of the ONH is very relevant for screening, diagnosis, staging, and management of all glaucomas, and is irreplaceable in detecting nonglaucomatous causes of ONH cupping and visual field defects similar to those in glaucoma.

Optic Disc Cupping Due to Dolichoectatic Internal Carotid Artery Optic Nerve Compression.

BACKGROUND: To determine whether significant compression of the optic nerve by the internal carotid artery (ICA) can produce an optic neuropathy with optic disc cupping that resembles glaucoma in patients without elevated intraocular pressure (IOP). METHODS: This was a retrospective case series of patients referred to neuro-ophthalmology for a possible nonglaucomatous optic neuropathy. Patients were included in the study if they had preserved visual acuity, optic disc-related visual field defects, optic nerve cupping, IOP less than 21 mm Hg, open angles, and unequivocal radiological compression of the ipsilateral optic nerve by an intracranial blood vessel. RESULTS: Three patients were included with a mean age of 56.3 (range 29-82) years. Patient 1 was a 58-year-old man incidentally noted to have left optic nerve cupping on a routine examination. He had an inferior arcuate defect and the left prechiasmatic optic nerve was elevated and compressed by a tortuous left ICA. Patient 2 was a 29-year-old man with a normal-tension glaucoma (NTG) diagnosis for 7 years in the right eye treated with latanoprost. He had a superior greater than inferior arcuate defect and there was vascular compression of the optic nerve between the supraclinoid ICA and A1 segment of the anterior cerebral artery. Patient 3 was an 82-year-old woman with an NTG diagnosis for 10 years who had progression of her visual field defects despite low IOPs. MRI showed mass effect on the right optic nerve by a dolichoectatic right supraclinoid ICA. CONCLUSIONS: Significant compression of the optic nerve by a normal, tortuous, or dolichoectatic ICA may result in an optic neuropathy with optic disc cupping that resembles glaucoma.

Temporal Raphe Sign in Elderly Patients With Large Optic Disc Cupping: Its Evaluation as a Predictive Factor for Glaucoma Conversion.

PURPOSE: To determine baseline clinical features associated with conversion to glaucoma in elderly patients with large optic-disc cupping. DESIGN: Retrospective cohort study. METHODS: Seventy-two eyes of 72 untreated elderly (≥65-year-old) patients with large vertical cup-to-disc ratio (CDR ≥0.7) and without any other glaucomatous findings were included. They had undergone a full ophthalmologic examination twice per year for at least 5 years. The optic nerve head (ONH), peripapillary retinal nerve fiber layer (RNFL), and macular ganglion cell-inner plexiform layer (GCIPL) were imaged with Cirrus high-definition optical coherence tomography (OCT). Presence of temporal raphe sign on the OCT's GCIPL thickness map was assessed as one of the morphologic factors. Conversion to normal-tension glaucoma (NTG) was defined as structural or functional deterioration on either red-free RNFL photography or standard automated perimetry, respectively. The utility of the baseline factors associated with conversion to NTG were identified. RESULTS: During the 5.5-year follow-up, 19 eyes (26.4%) converted to NTG. There were no significant differences in demographics, systemic factors, intraocular pressure factors, or OCT parameters between the nonconverters and converters. Interestingly, the temporal raphe sign was observed in the converters (18/19, 94.7%) much more frequently than in the nonconverters (3/53, 5.7%, P < .001) at baseline. A Cox proportional hazards model indicated the significant influences of temporal raphe sign positivity (hazard ratio 6.823, 95% confidence interval 2.574, 18.088, P < .001) on conversion to NTG. CONCLUSIONS: In elderly subjects with large CDR, temporal raphe sign positivity on the baseline macular GCIPL thickness map was associated with faster conversion to NTG.

Bruch's Membrane Opening Minimum Rim Width Provides Objective Differentiation between Glaucoma and Nonglaucomatous Optic Neuropathies.

PURPOSE: A challenging clinical scenario is distinguishing between normal tension glaucoma (NTG) and non-glaucomatous optic neuropathies (NGON). The key to the assessment remains identifying the presence of optic nerve head cupping. Recent optical coherence tomography (OCT) measurements now allow objective assessment of cupping by minimum rim width at Bruch's membrane opening (MRW-BMO). This study assessed the hypothesis that the MRW-BMO measurement quantifies cupping and therefore can differentiate between NTG and NGON. DESIGN: Diagnostic evaluation with area under the curve. METHODS: Setting: multicenter tertiary hospitals and outpatient clinics. PATIENT POPULATION: 81 eyes of 81 patients were enrolled, 27 with NTG and 54 with NGON, including ischemic optic neuropathy, previous optic neuritis, and compressive and inherited optic neuropathies. All NGON patients with intraocular pressure >21 mm Hg, narrow drainage angles, or a family history of glaucoma were excluded. Observational procedure: optic disc OCT images were obtained of both the retinal nerve fiber layer thickness and the MRW-BMO. MAIN OUTCOME MEASUREMENTS: the utility of the MRW-BMO in differentiating GON from NGON was assessed using the area under the curve (AUC) estimated from a logistic regression model. RESULTS: The 5-fold cross-validated AUC for glaucoma versus nonglaucoma from logistic regression models using MRW-BMO values from all sectors was 0.95 (95% confidence interval: 0.86-1.00). CONCLUSIONS: The measurement of MRW-BMO effectively differentiates between NTG and NGON with a high level of sensitivity and specificity. Incorporating this measurement into routine glaucoma assessment may provide a robust method of assisting clinicians to improve diagnosis and therefore treatment of optic nerve diseases.

The Relationship Between Optic Nerve Cup-to-Disc Ratio and Retinal Nerve Fiber Layer Thickness in Suspected Pediatric Glaucoma.

PURPOSE: To evaluate the relationship between optic nerve cup-to-disc ratio and peripapillary retinal nerve fiber layer (RNFL) thickness in suspected pediatric glaucoma with large cup-to-disc ratios. METHODS: This was a retrospective study undertaken at a single academic institution. Eighty-six eyes of 43 patients who presented with large (≥ 0.5) cup-to-disc ratios in both eyes and without elevated intraocular pressure were evaluated using spectral-domain optical coherence tomography. Global and sectoral peripapillary RNFL thickness measurements, Bruch's membrane opening size, refractive error in spherical equivalents, and intraocular pressure levels were recorded for all patients. Cup-to-disc ratios were manually derived using digital fundus images (D-cup-to-disc ratio). Parameters were compared between gender or race by t tests or analysis of variance. The differences in the relationship among the clinical parameters between two eyes were assessed using generalized estimation equation modeling followed by Pearson's correlation analysis. RESULTS: Forty-three patients (25 boys and 18 girls) with a mean age of 9.3 ± 2.7 years (range: 5 to 15 years) were included. The mean global peripapillary RNFL thickness and the D-cup-to-disc ratio of study eyes were 99.0 ± 9.2 µm and 0.66 ± 0.03, respectively. The peripapillary RNFL thickness was found to be correlated with refractive error (r = 0.404; P = .008) and Bruch's membrane opening size (r = 0.410; P = .008) but not with cup-to-disc ratios (r = 0.029; P = .858) or patient age (r = -0.044; P = .797). CONCLUSIONS: In patients with suspected pediatric glaucoma who present with large cup-to-disc ratios, RNFL thickness does not correlate with the degree of optic nerve cupping. Myopic refractive errors and Bruch's membrane opening size need to be taken into consideration to prevent misinterpretation of peripapillary RNFL measurements. [J Pediatr Ophthalmol Strabismus. 2020;57(2):90-96.].

Current concepts of cerebrospinal fluid dynamics and the translaminar cribrosa pressure gradient: a paradigm of optic disk disease.

Modern advances in measuring and studying cerebrospinal fluid dynamics have furthered our understanding of intracranial pressure and its pathophysiological effects on the eye. In particular, the cerebrospinal fluid pressure and composition within the optic nerve subarachnoid space are key factors in diseases of the optic disk. Intracranial pressure and intraocular pressure establish a pressure gradient across the lamina cribrosa. Alterations in this translaminar cribrosa pressure difference induce structural deformations in the lamina cribrosa manifested clinically by the appearance of optic disk edema or optic disk cupping. Much has been learned about papilledema (i.e., optic disk edema due to elevated intracranial pressure) from clinical observations and studies on patients with idiopathic intracranial hypertension. Furthermore, optic nerve subarachnoid space hydrodynamics and translaminar cribrosa pressure difference are postulated to contribute to the pathogenesis of optic disk edema observed in spaceflight-associated neuroocular syndrome. Recently, a substantial body of literature has accumulated suggesting low intracranial pressure may be a risk factor for the development of glaucomatous optic disk cupping within the context of the translaminar cribrosa pressure difference and posterior scleral biomechanics.

Periventricular Leukomalacia in Patients With Pseudo-glaucomatous Cupping.

PURPOSE: Periventricular leukomalacia (PVL) is a structural loss of white matter pathways that carry visual information from the lateral geniculate bodies to the visual cortex. It is observed radiologically in patients with a history of prematurity and is associated with visual field (VF) defects and optic disc cupping. Advances in perinatal care have improved survival for premature babies, so many now present as adolescents and adults to comprehensive eye doctors who are unaware of the relationship of cupping, field defects, and prematurity and who may diagnose manifest or suspected normal tension glaucoma. We describe 2 such patients to raise awareness of this entity. DESIGN: Case series. METHODS: Review of clinical information of 2 patients identified during clinical practice. Charts were reviewed for gestational age, optic nerve appearance, intraocular pressure (IOP), and sequelae of prematurity. Magnetic resonance imaging (MRI), optical coherence tomography (OCT), VF, and optic disc photographs were reviewed. RESULTS: Two young patients with a history of prematurity presented with enlarged cup-to-disc ratio and normal IOP. OCT thinning was most prominent superiorly, with VF defects more notable inferior and homonymous. No progression on VF or OCT was noted in the index case over almost 4 years. CONCLUSIONS: Periventricular leukomalacia should be added to the differential diagnosis of normal tension glaucoma (NTG) when there is a history of prematurity. Careful examination of the optic nerve will assist in differentiating from NTG. Specifically, horizontal cupping with minimal or no nasal displacement of vessels, and superior optic nerve thinning with inferior VF defects, suggest PVL.

An Association Between Large Optic Nerve Cupping and Cognitive Function.

PURPOSE: To determine if a larger cup-to-disc ratio is associated with poor cognitive function in postmenopausal women without glaucoma or ocular hypertension. METHODS: We used data from the Women's Health Initiative (WHI) hormone trial, originally designed to test effects of hormone therapy (HT) on various health outcomes. Large cup-to-disc ratio was defined as greater than 0.6 in either eye based on stereoscopic optic nerve photographs. Global cognitive function was assessed annually by Modified Mini-Mental State Examination (3MSE) in the WHI Memory Study. Exclusions were no information on optic nerve grading; no 3MSE scores at the time of the eye examination, ocular hypertension (intraocular pressure >23 mm Hg, Goldmann applanation tonometry), or glaucoma medication use. A generalized linear model for log-transformed 3MSE scores was used for determining the association between large cup-to-disc ratio and 3MSE scores, adjusting for age, race, diabetes, body mass index, cardiovascular disease, smoking, HT randomization, education, and diabetic retinopathy. RESULTS: Analyses included 1636 women (mean age ± standard deviation, 69.57 ± 3.64 years; 90.39% white). Of those, 122 women had large cup-to-disc ratio. The mean 3MSE scores in women with vs without large cup-to-disc ratio were 95.4 ± 6 vs 96.6 ± 5. In the adjusted model, women with large cup-to-disc ratio had statistically significantly lower 3MSE scores, compared with those without large cup-to-disc ratio, yielding the predicted mean difference in 3MSE scores of 0.75 with a standard error of 0.05 units (P = .04). CONCLUSIONS: Postmenopausal women who had large cup-to-disc ratio without glaucoma or ocular hypertension exhibited lower global cognitive function. Further investigation is warranted. NOTE: Publication of this article is sponsored by the American Ophthalmological Society.