Search

An increasing number of patients suffering from diabetes require regular ophthalmological check-ups to diagnose and/or treat potential diabetic retinal disease. Some countries have already implemented systematic fundus assessments including artificial intelligence-based programs in order to detect sight-threatening retinopathy. The aim of this study was to improve the detection of diabetic fundus changes in Germany without examination by a doctor and to create an easy access to ophthalmological examinations.In this prospective monocentric study 93 patients in need for a routine check-up for diabetic retinopathy were included. The study participants took up an offer of an examination (visual examination, non-mydriatic camera-based fundus examination) without doctor-patient contact. Patient satisfaction with the organization and examinations was assessed using a questionnaire.The mean age was 53.5 years (SD 13.6 years, 49.5% female) and 17 eyes (18.3%) showed a diabetic retinopathy which was detected using a camera-based examination. Within the small sample, no patient had to repeat the examination due to poor image quality. All categories of the questionnaire showed a good to very good satisfaction, indicating a high acceptance of the other examination form that took place at the ophthalmologist's premises.In our study in an ophthalmological practice a high level of acceptance among the patients interested in the screening for diabetic retinopathy without any direct patient-doctor contact was achieved. Our study shows a very good acceptance and feasibility. Future use of artificial intelligence in clinical practice may help to be able to screen many more patients as in this study imaging quality was very good.HINTERGRUND: Eine regelmäßige ophthalmologische Untersuchung ist für immer mehr Menschen mit Diabetes gefordert, sei es zur Früherkennung und Kontrolle einer diabetischen Retinopathie. In manchen Ländern kommen bereits automatische Verfahren der Bildauswertung zur Detektion einer behandlungsbedürftigen diabetischen Retinopathie zur Anwendung. Ziel dieser Studie ist es, die Früherkennung diabetischer Augenerkrankungen ohne direkten Arzt-Patienten-Kontakt in Deutschland zu evaluieren und mögliche Barrieren für Patienten zu erkennen.In dieser prospektiven monozentrischen Studie wurden bisher 93 Patienten eingeschlossen, die sich für ein Screening auf eine diabetische Retinopathie vorstellten. Die Studienteilnehmer nahmen ein Angebot zu einer Untersuchung (Visuserhebung, non-mydriatische kamerabasierte Fundusuntersuchung) ohne Arzt-Patienten-Kontakt wahr. Mittels Fragebogen wurde die Zufriedenheit der Patienten erhoben.Das durchschnittliche Alter der eingeschlossenen Patienten (n = 93; 49,5 % weiblich) betrug 53,5 Jahre (29 bis 76 Jahre, SD: 13,6 Jahre). Bei 17 Augen (18,3 %) wurde mittels kamerabasierter Fundusuntersuchung eine diabetische Retinopathie festgestellt. In der Stichprobe musste sich kein Patient aufgrund schlechter Bildqualität erneut vorstellen. Insgesamt berichteten die Patienten eine hohe bis sehr hohe Zufriedenheit. Die Akzeptanz des Untersuchungsangebots war hoch.In dieser Studie konnte mit einer Screeninguntersuchung ohne direkten Arztkontakt eine hohe Akzeptanz erreicht werden. Qualifiziertes Fachpersonal ermöglichte eine hohe Bildqualität der erhobenen Fundusbilder, sodass der Einsatz von automatischen Algorithmen (künstliche Intelligenz) das Potenzial hat, die bisherigen Angebote zu ergänzen.

PurposeTo determine real life quantitative changes in OCT biomarkers in a large set of treatment naive patients undergoing anti-VEGF therapy. For this purpose, we devised a novel deep learning based semantic segmentation algorithm providing, to the best of our knowledge, the first benchmark results for automatic segmentation of 11 OCT features including biomarkers that are in line with the latest consensus nomenclature of the AAO for age-related macular degeneration (AMD).DesignRetrospective study.ParticipantsSegmentation algorithm training set of 458 volume scans as well as single scans from 363 treatment naive patients for the analysis.MethodsTraining of a Deep U-net based semantic segmentation ensemble algorithm leveraging multiple deep convolutional neural networks for state of the art semantic segmentation performance as well as analyzing OCT features prior to, after 3 and 12 months of anti-VEGF therapy.Main outcome measuresF1 score for the segmentation efficiency and the quantified volumes of 11 OCT features.ResultsThe segmentation algorithm achieved high F1 scores of almost 1.0 for neurosensory retina and subretinal fluid on a separate hold out test set with unseen patients. The algorithm performed worse for subretinal hyperreflective material and fibrovascular PED, on par with drusenoid PED and better in segmenting fibrosis. In the evaluation of treatment naive OCT scans, significant changes occurred for intraretinal fluid (mean: 0.03µm3 to 0.01µm3, p3 to 0.01µm3, p3 to 0.01µm3, p3 to 0.09µm3, p=0.02) and central retinal thickness C0 (225.78µm3 to 169.40µm3).The amounts of intraretinal fluid, fibrovascular PED and ERM were predictive of poor outcome.ConclusionsThe segmentation algorithm allows efficient volumetric analysis of OCT scans. Anti-VEGF therapy provokes most potent changes in the first 3 months and afterwards only acts as a stabilizing agent. Furthermore, a gradual loss of RPE hints at a progressing decline of visual acuity even beyond month 12. Additional research is required to understand how these accurate OCT predictions can be leveraged for a personalized therapy regimen.PrécisNovel high performance segmentation algorithm shows most volumetric changes under anti-VEGF therapy in oct biomarkers occur in the first 3 months. Afterwards the injections seem only to serve as a stabilizing agent.

Background To study the enlargement rate of primary geographic atrophy (GA) before and after diagnosis of a secondary choroidal neovascularization (CNV) treated with anti-vascular endothelial growth factor (VEGF) therapy. Methods Five hundred twenty-two consecutive eyes with primary GA were screened for the development of a complicating secondary CNV. Geographic atrophy was measured on blue autofluorescence (BAF) by two readers and calculated into mean growth rate before and after CNV diagnosis. Results Ten eyes of six patients were included in the study (six study eyes with GA complicated by CNV, four GA only partner eyes). Follow-up was 1.42 ± 0.48 years before and 3.64 ± 2.73 years after CNV. There was no significant difference between mean growth rate before and after CNV (1.58 ± 0.99 vs. 1.39 ± 0.65 mm2/year; p = 0.44) or between study and partner eyes (p = 0.86). Over a mean time of 3.64 ± 2.73 years, a mean of 8.3 ± 2.8 anti-VEGF injections were given. No correlation between the amount of anti-VEGF injections and change in growth rate could be observed (r = 0.58; p = 0.23). Conclusion In this pilot study, primary GA enlargement did not seem to be influenced by a secondary CNV. No association between the intensity of anti-VEGF treatment and changes in atrophy enlargement rates were found. Further studies with larger sample sizes are warranted.

To report on long-term real-life outcomes of anti-vascular endothelial growth factor (anti-VEGF) therapy in neovascular age-related macular degeneration (nAMD) with optimal patient adherence.For this retrospective monocenter study, we identified 3217 eyes of 2793 patients that received a minimum of three intravitreal anti-VEGF injections for nAMD therapy between 2006 and 2014 at the University Eye Hospital Munich. From those, we included eyes with treatment-naïve nAMD, follow-up (FU) of ≥60 months and continuous adherence during FU. Primary measures were corrected visual acuity (VA), number of injections and visits as well as treatment regimen.We included 161 eyes of 125 patients with a mean FU of 8.0 ± 2.3 years. Mean VA at baseline was 60.1 letters (Snellen equivalent, 20/63). After the third year, mean VA declined constantly by 2-3 letters per year. After 5 and 8 years, 26.1% and 42.1% had lost at least 3 lines from baseline. Mean cumulative number of injections was 5.3 after the first year, and 23.9, 38.1, 48.5 after 5, 8, and 10 years. "Treat and extent" regimen with higher injection frequency correlated with better function. At time of last FU, 69.8% of eyes were under active treatment. Eyes with ≥70 letters at baseline correlated with better VA at the end of FU.Despite optimal patient adherence, visual function declined progressively in real-life nAMD therapy over long-term. The highest impact on treatment success is given by an early treatment start with individual but intensive anti-VEGF therapy.