Search

Center-involved diabetic macular edema (ci-DME) is a major cause of vision loss. Although the gold standard for diagnosis involves 3D imaging, 2D imaging by fundus photography is usually used in screening settings, resulting in high false-positive and false-negative calls. To address this, we train a deep learning model to predict ci-DME from fundus photographs, with an ROC-AUC of 0.89 (95% CI: 0.87-0.91), corresponding to 85% sensitivity at 80% specificity. In comparison, retinal specialists have similar sensitivities (82-85%), but only half the specificity (45-50%, p

Purpose: To evaluate the economic impact of treating severe infectious keratitis (IK) at one tertiary referral center in Thailand by analyzing the direct costs of treatment and estimating the indirect costs, and to determine whether cultured organisms had any effect on treatment expenditure. Methods: A retrospective study was conducted of patients with severe IK who had been hospitalized between January 2014 and December 2021 in Rajavithi Hospital. Data from medical records were collected from the time of the patients' admission until the point at which they were discharged and treated in the outpatient department and their IK was completely healed, or until evisceration/enucleation was performed. The direct costs of treatment included fees for services, medical professionals and investigation, as well as for operative and non-operative treatment. The indirect costs consisted of patients' loss of wages, and costs of travel and food. Results: A total of 335 patients were studied. The median direct, indirect and total costs were US$65.2, range US$ 6.5–1,119.1, US$314.5, range US$50.8–1,067.5, and US$426.1, range 57.5–1,971.5 respectively. There was no statistically significant difference between direct, indirect, or total treatment costs for culture-negative and culture-positive patients. Among those who were positive, fungal infections entailed the highest total cost of treatment, and this difference was statistically significant (p<0.001). In terms of direct and indirect costs, patients with fungal infections had the greatest direct costs, and this figure was statistically significant (p = 0.001); however, those with parasitic infections had the highest indirect treatment costs, and this was also statistically significant (p<0.001). Conclusion: Severe IK can cause serious vision impairment or blindness. Indirect costs represented the majority of the expense at 73.8%. There was no difference between direct, indirect, and total treatment costs for patients who were culture-negative or positive. Among the latter, fungal infections resulted in the highest total cost of treatment. [ABSTRACT FROM AUTHOR]

Purpose: To demonstrate changes in three-dimensional choroidal volume with enhanced depth imaging optical coherence tomography (EDI-OCT) in patients with recurrent stage of Vogt-Koyanagi-Harada disease (VKH). Materials and Methods: This prospective comparative case series included 9 patients with recurrent VKH, 10 patients with quiet VKH, and 15 healthy controls after sample size was calculated. All VKH cases with recurrences underwent raster scanning with EDI-OCT at active and inactive stages of the disease. Results: All choroidal parameters in the active stage significantly reduced when the inflammation subsided: total choroidal volume (P =.02), central choroidal volume (P =.01), central choroidal thickness (P =.03). The changes in central choroidal volume over the resolution phase were more pronounced than the changes in central choroidal thickness in 56% of cases. Two cases presenting with only subclinical posterior segment recurrence had their choroidal parameters recovered after prompt treatment. Conclusions: In the recurrent stage of VKH, alteration in choroidal volume was evident by EDI-OCT even in an absence of anterior segment inflammation. Central choroidal volume may serve as a biomarker for detecting choroidal morphological change. [ABSTRACT FROM AUTHOR]

Deep learning algorithms have been used to detect diabetic retinopathy (DR) with specialist-level accuracy. This study aims to validate one such algorithm on a large-scale clinical population, and compare the algorithm performance with that of human graders. A total of 25,326 gradable retinal images of patients with diabetes from the community-based, nationwide screening program of DR in Thailand were analyzed for DR severity and referable diabetic macular edema (DME). Grades adjudicated by a panel of international retinal specialists served as the reference standard. Relative to human graders, for detecting referable DR (moderate NPDR or worse), the deep learning algorithm had significantly higher sensitivity (0.97 vs. 0.74, p < 0.001), and a slightly lower specificity (0.96 vs. 0.98, p < 0.001). Higher sensitivity of the algorithm was also observed for each of the categories of severe or worse NPDR, PDR, and DME (p < 0.001 for all comparisons). The quadratic-weighted kappa for determination of DR severity levels by the algorithm and human graders was 0.85 and 0.78 respectively (p < 0.001 for the difference). Across different severity levels of DR for determining referable disease, deep learning significantly reduced the false negative rate (by 23%) at the cost of slightly higher false positive rates (2%). Deep learning algorithms may serve as a valuable tool for DR screening. [ABSTRACT FROM AUTHOR]