Performance of a Support Vector Machine Learning Tool for Diagnosing Diabetic Retinopathy in Clinical Practice.

PURPOSE: To examine the real-world performance of a support vector machine learning software (RetinaLyze) in order to identify the possible presence of diabetic retinopathy (DR) in patients with diabetes via software implementation in clinical practice. METHODS: 1001 eyes from 1001 patients-one eye per patient-participating in the Danish National Screening Programme were included. Three independent ophthalmologists graded all eyes according to the International Clinical Diabetic Retinopathy Disease Severity Scale with the exact level of disease being determined by majority decision. The software detected DR and no DR and was compared to the ophthalmologists' gradings. RESULTS: At a clinical chosen threshold, the software showed a sensitivity, specificity, positive predictive value and negative predictive value of 84.9% (95% CI: 81.8-87.9), 89.9% (95% CI: 86.8-92.7), 92.1% (95% CI: 89.7-94.4), and 81.0% (95% CI: 77.2-84.7), respectively, when compared to human grading. The results from the routine screening were 87.0% (95% CI: 84.2-89.7), 85.3% (95% CI: 81.8-88.6), 89.2% (95% CI: 86.3-91.7), and 82.5% (95% CI: 78.5-86.0), respectively. AUC was 93.4%. The reference graders Conger's Exact Kappa was 0.827. CONCLUSION: The software performed similarly to routine grading with overlapping confidence intervals, indicating comparable performance between the two groups. The intergrader agreement was satisfactory. However, evaluating the updated software alongside updated clinical procedures is crucial. It is therefore recommended that further clinical testing before implementation of the software as a decision support tool is conducted.

A cohort follow-up study for diabetic retinopathy screening incidence in the North Denmark Region.

AIMS: To evaluate diabetic retinopathy (DR) screening incidence in a universal healthcare system. METHODS: Registry-based cohort study based on a Danish regional population from 2009 to 2018. Individuals with diabetes were identified by medication. Screening attendance was estimated by surrogate measures using local and nationwide databases reported by cumulative incidence. RESULTS: 18,832 patients were included. By the end of the first year, the cumulative incidence of screening for DR was 60.2% and by the end of the second year 74.2%. The cumulative incidence was 93.9% overall, 97.7% for patients with type 1 diabetes (T1D) and 93.4% for patients with type 2 diabetes. Screening proportions per 1, 2 and 5 years were calculated. Females, patients with T1D, and patients attending screening at hospitals had a higher Hazard Ratio of 1.084, 1.157, and 1.573, respectively. The Cochran-Armitage trend test indicated increased screening frequency from 2009 to 2018. Validation of DR screening was done at hospitals with a mean positive predictive value of 86.78%. Cumulative incidence curves showed a small right shift when censoring the first, second and third screening visits. CONCLUSIONS: Nearly all patients were screened for DR over a 5-year timespan. Female patients with T1D who attended screening at hospitals were significantly more likely to be screened. Validation of screening visits at hospitals was reported with a high mean positive predictive value. Most other studies, to the best of our knowledge, only report screening attendance for patients already enrolled in a DR screening programme. This study describes the overall screening attendance for the total eligible diabetes population.

Biologic Therapy and Treatment Options in Diabetic Retinopathy with Diabetic Macular Edema.

Proliferative diabetic retinopathy and diabetic macular edema can be a potentially sightthreatening disease if not treated correctly. It is directly correlated to the duration of diabetes and how well managed the patients' diabetes is. In the last 15 years, the treatment of diabetic eye disease has taken a quantum leap in methodology due to the group of biological agents named antivascular endothelial growth factor (anti-VEGF). The introduction of the first biological agent has revolutionized the treatment, not only in diabetic eye disease but also across most inflammatory eye diseases, causing leakage of fluid from the blood vessels i.e., in age-related macular degeneration. The availability of these biological agents, despite their considerable costs, have significantly improved the outcomes measured in visual acuity compared to more traditional treatments of diabetic retinopathy in the form of sole laser treatment and glycemic control. The agents demonstrate a favorable safety profile, but if the rarest and most severe side effects occur, there is a potential total loss of vision. This review aims to make an overview of the current pharmaceutical therapeutic options in the treatment of diabetic macular edema. This includes laser therapy, intravitreal steroids, and a primary focus on intravitreal antivascular endothelial growth factors.