ContextCortisol has been suggested as a risk factor for choroidal thickening, which may lead to retinal changes.ObjectiveTo compare choroidal thickness measurements using optical coherence tomography (OCT) in patients with endogenous active Cushing’s syndrome (CS) and to evaluate the occurrence of retinal abnormalities in the same group of patients.DesignCross-sectional study.SettingOutpatient clinic.PatientsEleven female patients with CS in hypercortisolism state as determined by the presence of at least two abnormal measurements from urinary cortisol 24 h, no suppression of cortisol with low dose dexamethasone suppression test, and nocturnal salivary cortisol levels and 12 healthy controls.MethodsChoroidal and retinal morphology was assessed using OCT.Main outcome measuresChoroidal thickness measurements and the presence of retinal changes.ResultsThe mean subfoveal choroidal thickness was 372.96 ± 73.14 µm in the patients with CS and 255.63 ± 50.70 µm in the control group (p < 0.001). One patient (9.09%) presented with central serous chorioretinopathy and one patient (9.09%) with pachychoroid pigment epitheliopathy.ConclusionChoroidal thickness is increased in the eyes of patients with active CS compared to healthy and matched control. Also, 18.18% of patients presented with macular changes, possibly secondary to choroidal thickening. While further studies are necessary to confirm our findings, excess corticosteroid levels seem to have a significant effect on the choroid and might be associated with secondary retinal diseases.

No abstract

No abstract

Background Artificial intelligence and automated technology were first reported more than 70 years ago and nowadays provide unprecedented diagnostic accuracy, screening capacity, risk stratification, and workflow optimization. Diabetic retinopathy is an important cause of preventable blindness worldwide, and artificial intelligence technology provides precocious diagnosis, monitoring, and guide treatment. High-quality exams are fundamental in supervised artificial intelligence algorithms, but the lack of ground truth standards in retinal exams datasets is a problem. Main body In this article, ETDRS, NHS, ICDR, SDGS diabetic retinopathy grading, and manual annotation are described and compared in publicly available datasets. The various DR labeling systems generate a fundamental problem for AI datasets. Possible solutions are standardization of DR classification and direct retinal-finding identifications. Conclusion Reliable labeling methods also need to be considered in datasets with more trustworthy labeling.

No abstract