No abstract

Glaucoma, the leading cause of irreversible blindness worldwide, is a disease that damages the optic nerve. Current machine learning (ML) approaches for glaucoma detection rely on features such as retinal thickness maps; however, the high rate of segmentation errors when creating these maps increase the likelihood of faulty diagnoses. This paper proposes a new, comprehensive, and more accurate ML-based approach for population-level glaucoma screening. Our contributions include: (1) a multi-modal model built upon a large data set that includes demographic, systemic and ocular data as well as raw image data taken from color fundus photos (CFPs) and macular Optical Coherence Tomography (OCT) scans, (2) model interpretation to identify and explain data features that lead to accurate model performance, and (3) model validation via comparison of model output with clinician interpretation of CFPs. We also validated the model on a cohort that was not diagnosed with glaucoma at the time of imaging but eventually received a glaucoma diagnosis. Results show that our model is highly accurate (AUC 0.97) and interpretable. It validated biological features known to be related to the disease, such as age, intraocular pressure and optic disc morphology. Our model also points to previously unknown or disputed features, such as pulmonary capacity and retinal outer layers.

A 60-year-old woman with decreased visual acuity in her right eye and right-sided jaw claudication was found to have ocular ischemic syndrome secondary to complete occlusion of the brachiocephalic artery. Although jaw claudication is often considered to be pathognomonic for giant cell arteritis, it has a broad differential diagnosis including both vascular and nonvascular conditions.

A statistical approach to the evaluation of covariate effects on the receiver operating characteristic curves of diagnostic tests in glaucoma.

Background To determine accuracy of partial coherence interferometry (PCI) in patients with large inter-eye axial eye length (AEL) difference. Methods Patients undergoing cataract surgery at two academic medical centers with an inter-eye axial eye length (AEL) difference of > 0.30 mm were identified and were matched to control patients without inter-eye AEL difference > 0.30 mm on the basis of age, sex, and AEL. The expected post-operative refraction for the implanted IOL was calculated using SRK/T, Holladay II, and Hoffer Q formulae. The main outcome measures were the refractive prediction error and the equivalence of the refractive outcomes between the subjects and controls. Results Review of 2212 eyes from 1617 patients found 131 eyes of 93 patients which met inclusion criteria. These were matched to 131 control eyes of 115 patients. The mean AEL was 24.92 ± 1.50 mm. The mean absolute error (MAE) ranged from 0.47 D to 0.69 D, and was not statistically different between subjects and controls. The refractive prediction error was equivalent between the cases and controls, with no significant difference between the MAE for any formula, nor in the number of cases vs. controls with a refractive prediction error of at least 0.50 D or 1.00 D. Conclusions Among eyes in our study population, good-quality PCI data was equally accurate in patients with or without an inter-eye AEL difference > 0.30 mm. Confirmatory AEL measurements using different AEL measuring modalities in patients with a large inter-eye AEL difference may not be necessary.