Early Detection of Optic Nerve Changes on Optical Coherence Tomography Using Deep Learning for Risk-Stratification of Papilledema and Glaucoma.

BACKGROUND: The use of artificial intelligence is becoming more prevalence in medicine with numerous successful examples in ophthalmology. However, much of the work has been focused on replicating the works of ophthalmologists. Given the analytical potentials of artificial intelligence, it is plausible that artificial intelligence can detect microfeatures not readily distinguished by humans. In this study, we tested the potential for artificial intelligence to detect early optic coherence tomography changes to predict progression toward papilledema or glaucoma when no significant changes are detected on optical coherence tomography by clinicians. METHODS: Prediagnostic optical coherence tomography of patients who developed papilledema (n = 93, eyes = 166) and glaucoma (n = 187, eyes = 327) were collected. Given discrepancy in average cup-to-disc ratios of the experimental groups, control groups for papilledema (n = 254, eyes = 379) and glaucoma (n = 441, eyes = 739) are matched by cup-to-disc ratio. Publicly available Visual Geometry Group-19 model is retrained using each experimental group and its respective control group to predict progression to papilledema or glaucoma. Images used for training include retinal nerve fiber layer thickness map, extracted vertical tomogram, ganglion cell thickness map, and ILM-RPE thickness map. RESULTS: Trained model was able to predict progression to papilledema with a precision of 0.714 and a recall of 0.769 when trained with retinal nerve fiber layer thickness map, but not other image types. However, trained model was able to predict progression to glaucoma with a precision of 0.682 and recall of 0.857 when trained with extracted vertical tomogram, but not other image types. Area under precision-recall curve of 0.826 and 0.785 were achieved for papilledema and glaucoma models, respectively. CONCLUSIONS: Computational and analytical power of computers have become an invaluable part of our lives and research endeavors. Our proof-of-concept study showed that artificial intelligence (AI) algorithms have the potential to detect early changes on optical coherence tomography for prediction of progression that is not readily observed by clinicians. Further research may help establish possible AI models that can assist with early diagnosis or risk stratification in ophthalmology.

Pathologic Study of Supernumerary Orbital Band in Type I Duane Syndrome.

BACKGROUND/AIMS: Accessory orbital bands are relatively rare and very few reports detail histopathology. Cases in the literature describe the composition of the bands as muscular and/or fibrous. The composition of the supernumerary band lying deep in the medial rectus muscle in a patient with type I Duane syndrome was investigated. METHODS: Histochemical stains were used in conjunction with polarized light for differentiating compressed collagen from muscle. Immunohistochemistry was used for verification of the presence of muscle. RESULTS: Compressed collagen appeared red using Masson trichrome staining. Collagen was positively identified by illumination with polarized light on several stains including the underutilized Sirius red dye. CONCLUSIONS: The findings of dense collagen fibers in the fibrotic band with focal striated muscle correlated with the restrictive strabismus. In concert with other cases in the literature, it is proposed that the fibrous bands are generally associated with restrictive strabismus. Bands that are muscular may or may not be associated with strabismus. Special techniques are needed to positively identify compressed collagen.

Bilateral Congenital Posterior Capsular Defects and Ectopic Cataracts.

Preexisting posterior capsular defects are rare and can be associated with infantile cataracts. The authors review possible etiologies of this condition and emphasize the importance of cautious preoperative planning and surgery.

EN FACE OPTICAL COHERENCE TOMOGRAPHY OF MACULAR MICROCYSTS DUE TO OPTIC NEUROPATHY FROM NEUROMYELITIS OPTICA.

PURPOSE: To describe the multimodal imaging, including en face optical coherence tomography (OCT) and OCT angiography, findings of a case of macular microcysts associated with neuromyelitis optica. METHODS: Findings on clinical examination, color fundus photography, fluorescein angiography, fundus autofluorescence, visual fields, and OCT including en face OCT and OCT angiography are presented. RESULTS: A 12-year-old African American boy presented with bilateral optic atrophy from neuromyelitis optica. Clinical examination was notable for bilateral optic nerve head pallor. Visual fields of both eyes showed generalized depression. Fluorescein angiography and fundus autofluorescence were unremarkable. Spectral domain OCT B-scan images showed characteristic paracentral, hyporeflective, microcystic lesions in the inner nuclear layer of both eyes, and en face OCT images demonstrated a corresponding pattern of large paracentral cysts radiating peripherally into smaller diffuse cysts. Optical coherence tomographic angiography of the superficial and deep retinal capillary plexuses was unremarkable. CONCLUSION: Macular microcysts have been associated with various forms of optic atrophy, including neuromyelitis optica. Spectral domain and en face OCT imaging of the microcysts demonstrated a very characteristic pattern. Normal fluorescein and OCT angiography suggest that nonvascular causes, such as Müller cell degeneration, might contribute to the etiologic mechanism.

Binocular inhibition in strabismic patients is associated with diminished quality of life.

PURPOSE: To characterize the relationship between binocular summation (BiS) and binocular inhibition (BI) on the quality of life of adults and children with strabismus. METHODS: Strabismus patients at a single center from 2010 to 2012 were prospectively enrolled. A BiS score was measured using ETDRS and Sloan low-contrast visual acuity (LCA) protocols at 2.5% and 1.25% contrast. Patients were categorized as having BiS (binocular better than better-eye visual acuity by ≥5 letters), BI (binocular worse than better-eye visual acuity by ≥5 letters), or otherwise indeterminate visual acuity (a difference between binocular visual acuity and monocular visual acuity of the better eye of <5 letters). Quality of life was evaluated by the National Eye Institute Visual Functioning Questionnaire 25 (VFQ-25), 20-item Adult Strabismus Questionnaire (AS-20), and the Amblyopia and Strabismus Questionnaire. RESULTS: A total of 108 patients were included. There was no significant BiS or BI for high-contrast ETDRS or 2.5% LCA tests; however, a mean BiS score of -2.14 ± 7.0 letters for 1.25% LCA demonstrated significant binocular inhibition (P = 0.004) for this contrast level. The mean composite VFQ-25 score was significantly lower in subjects with BI on ETDRS (80 ± 19 vs 57 ± 7 for subjects with BiS and BI, resp. [P = 0.03]), 2.5% LCA (81 ± 14 vs 66 ± 16 for subjects with BiS and BI, resp. [P = 0.01]), and 1.25% LCA tests (91 ± 9 vs 72 ± 14 for subjects with BiS and BI, resp. [P = 0.005]). After accounting for potential covariates, significant association persisted for BI, demonstrated by 1.25% LCA (P = 0.01). With BI demonstrable at 2.5%, AS-20 scores were also significantly lower (P = 0.04). CONCLUSIONS: Strabismic patients with BI had significantly lower quality-of -life scores than those who did not, even after accounting for potential covariates and the absence of diplopia.