Validation of Automated Screening for Referable Diabetic Retinopathy With an Autonomous Diagnostic Artificial Intelligence System in a Spanish Population.

PURPOSE: The purpose of this study is to compare the diagnostic performance of an autonomous artificial intelligence (AI) system for the diagnosis of referable diabetic retinopathy (RDR) to manual grading by Spanish ophthalmologists. METHODS: Subjects with type 1 and 2 diabetes participated in a diabetic retinopathy (DR) screening program in 2011 to 2012 in Valencia (Spain), and two images per eye were collected according to their standard protocol. Mydriatic drops were used in all patients. Retinal images-one disc and one fovea centered-were obtained under the Medical Research Ethics Committee approval and de-identified. Exams were graded by the autonomous AI system (IDx-DR, Coralville, Iowa, United States), and manually by masked ophthalmologists using adjudication. The outputs of the AI system and manual adjudicated grading were compared using sensitivity and specificity for diagnosis of both RDR and vision-threatening diabetic retinopathy (VTDR). RESULTS: A total of 2680 subjects were included in the study. According to manual grading, prevalence of RDR was 111/2680 (4.14%) and of VTDR was 69/2680 (2.57%). Against manual grading, the AI system had a 100% (95% confidence interval [CI]: 97%-100%) sensitivity and 81.82% (95% CI: 80%-83%) specificity for RDR, and a 100% (95% CI: 95%-100%) sensitivity and 94.64% (95% CI: 94%-95%) specificity for VTDR. CONCLUSION: Compared to manual grading by ophthalmologists, the autonomous diagnostic AI system had high sensitivity (100%) and specificity (82%) for diagnosing RDR and macular edema in people with diabetes in a screening program. Because of its immediate, point of care diagnosis, autonomous diagnostic AI has the potential to increase the accessibility of RDR screening in primary care settings.

Use in clinical practice of an automated screening method of diabetic retinopathy that can be derived using a diagnostic artificial intelligence system. / Uso en la práctica clínica, de un método de cribado automatizado de retinopatía diabética derivable mediante un sistema de inteligencia artificial de diagnóstico.

BACKGROUND AND OBJECTIVE: To compare the diagnostic performance of an autonomous diagnostic artificial intelligence (AI) system for the diagnosis of derivable diabetic retinopathy (RDR) with manual classification. MATERIALS AND METHODS: Patients with type 1 and type 2 diabetes participated in a diabetic retinopathy (DR) screening program between 2011-2012. 2 images of each eye were collected. Unidentifiable retinal images were obtained, one centered on the disc and one on the fovea. The exams were classified with the autonomous AI system and manually by anonymous ophthalmologists. The results of the AI system and manual classification were compared in terms of sensitivity and specificity for the diagnosis of both (RDR) and diabetic retinopathy with decreased vision (VTDR). RESULTS: 10,257 retinal inages of 5,630 eyes of 2,680 subjects were included. According to the manual classification, the prevalence of RDR was 4.14% and that of VTDR 2.57%. The AI system recorded 100% (95% CI: 97-100%) sensitivity and 81.82% (95% CI: 80 -83%) specificity for RDR, and 100% (95% CI: 95-100%) of sensitivity and 94.64% (95% CI: 94-95%) of specificity for VTDR. CONCLUSIONS: Compared to the manual classification, the autonomous diagnostic AI system registered a high sensitivity (100%) and specificity (82%) in the diagnosis of RDR and macular edema in people with diabetes. Due to its immediate diagnosis, the autonomous diagnostic AI system can increase the accessibility of RDR screening in primary care settings.

External validation of a risk assessment model to adjust the frequency of eye-screening visits in patients with diabetes mellitus.

AIMS: To validate a sight-threatening diabetic retinopathy (STDR) risk assessment model to adjust the frequency of eye-screening visits in patients with diabetes mellitus. METHODS: Retrospective follow-up study of patients with diabetes mellitus attending a diabetes center. Anonimyzed data on gender, type and duration of diabetes, HbA1c, blood pressure and the presence and grade of diabetic retinopathy were gathered to estimate risk for STDR for each individual's worse eye over time by means of a prediction model. Receiver operating characteristics (ROC) analysis was performed to determine the diagnostic ability of the model, and a calibration graph was done to see the model fit. RESULTS: 508 screening intervals were analyzed, median diabetes duration was 10years, 87% were type 2 diabetes mellitus, and 3.1% developed STDR before the next screening visit. The area under the ROC curve was 0.74, and the calibration graph showed that model had a good fit. The reduction in screening frequency was 40% compared with fixed annual screening. CONCLUSIONS: Current prediction model used to estimate the risk of developing STDR in patients with diabetes performed well. A personalized screening frequency for diabetic retinopathy could be implemented in practice.

Evaluation of automated image analysis software for the detection of diabetic retinopathy to reduce the ophthalmologists' workload.

AIMS: To assess the safety and workload reduction of an automated 'disease/no disease' grading system for diabetic retinopathy (DR) within a systematic screening programme. METHODS: Single 45° macular field image per eye was obtained from consecutive patients attending a regional primary care based DR screening programme in Valencia (Spain). The sensitivity and specificity of automated system operating as 'one or more than one microaneurysm detection for disease presence' grader were determined relative to a manual grading as gold standard. Data on age, gender and diabetes mellitus were also recorded. RESULTS: A total of 5278 patients with diabetes were screened. The median age and duration of diabetes was 69 years and 6.9 years, respectively. Estimated prevalence of DR was 15.6%. The software classified 43.9% of the patients as having no DR and 26.1% as having ungradable images. Detection of DR was achieved with 94.5% sensitivity (95% CI 92.6- 96.5) and 68.8% specificity (95%CI 67.2-70.4). The overall accuracy of the automated system was 72.5% (95%CI 71.1-73.9). CONCLUSIONS: The present retinal image processing algorithm that can act as prefilter to flag out images with pathological lesions can be implemented in practice. Our results suggest that it could be considered when implementing DR screening programmes.

Screening coverage for diabetic retinopathy using a three-field digital non-mydriatic fundus camera.

AIMS: Guidelines for regular screening of diabetic retinopathy (DR) have been published in the Spanish and European literature since 1992, but screening for DR is still in its early stages in Spain. The aim of this paper is to estimate the prevalence of screening coverage for DR and prevalence of DR itself using three-field digital non-mydriatic fundus photography to determine whether these guidelines had been implemented. METHODS: Data on age, gender, diabetes and previous eye examinations were recorded on a specially designed questionnaire. Three 45 degrees digital images per eye were taken using a three-field digital non-mydriatic fundus camera with two photographic procedures (both eyes versus the eye with the poorer visual acuity). RESULTS: A total of 183 patients with diabetes participated. The median age and duration of diabetes was 63 years and 10 years, respectively. Only six patients (3.3%) could not be completely graded. Screening coverage for DR was 38.5% in patients with type 2 diabetes and a duration less than 5 years versus those with longer diabetes duration (P=0.007); 20.5% of these patients had DR. CONCLUSIONS: This study highlights the need for heightened awareness of the importance of screening for retinopathy in people with type 2 diabetes and duration of diabetes under 5 years.

Lanreotide Autogel for persistent diabetic macular edema.

Diffuse diabetic macular edema is usually refractory to conventional treatment. We investigated the effectiveness of a new prolonged lanreotide formulation in patients with bad-controlled diabetes and persistent cystoid macular edema. Our findings suggest that monthly subcutaneous injections of lanreotide Autogel offered an effective treatment alternative in these patients.

Duration of diabetes and screening coverage for retinopathy among patients with type 2 diabetes.

AIMS: Despite reporting of the Wisconsin Epidemiologic Study of Diabetic Retinopathy(1) and the Diabetic Retinopathy Awareness Program(2) that diabetes duration was a significant predictor for adherence to vision care guidelines, reports of estimates of screening coverage for diabetic retinopathy taking into account diabetes duration have been lagging. This article estimates considering diabetes duration, the prevalence of diabetic retinopathy and screening coverage for diabetic retinopathy among type 2 diabetic patients. METHODS: As part of a treatment program at a High-Resolution Diabetes Center in Spain, type 2 diabetic patients attending the center from January 2003 to January 2005 were invited to participate in the study. Data on age, sex, and diabetes were recorded into a questionnaire, as was information about previous eye examinations. Polaroid(R) photographs were taken of the eye fundus with the poorest visual acuity using a nonmydriatic retinal camera. RESULTS: A total of 217 type 2 diabetic patients entered the program. The average age and duration of diabetes was 60.9 years and 7 years, respectively. Screening coverage for diabetic retinopathy was higher in those with a longer duration of diabetes (chi(2) = 36.5; p = 0.001). Fifty percent of patients had developed some retinopathy within the first 5 years after the diagnosis of the disease, but only 26.1% had received a previous fundus examination. CONCLUSIONS: These results argue for screening programs for people with type 2 diabetes mellitus focused on the subgroup of patients with diabetes duration of 5 years or less.

Use of the Honan balloon as a compression device for intravitreal triamcinolone acetonide injection.

Obtaining a lower intraocular pressure (IOP) before an intravitreal injection is important to avoid the danger of an immediate increase in IOP. This article describes the use of the Honan balloon to reduce the IOP before an intravitreal triamcinolone acetonide injection instead of the other procedures described in the literature, which may increase the risk of complications. The results of the technique in 19 patients with retinal vein occlusion are reported. Honan balloon was applied at 30 mm Hg for 10 minutes and the injection was then performed no more than 5 minutes after the removal of the Honan balloon. In no more than 10 minutes, this procedure resulted in a significant reduction in IOP. No complications such as central retinal artery occlusion were observed after the injection.

Removal of benzyl alcohol from a commercially available triamcinolone acetonide suspension for intravitreal use.

Purification of commercially available formulations of triamcinolone acetonide is important to avoid potential toxic effects of the vehicle for intravitreal use. In 2004, a simple and rapid method to remove most of the vehicle with a centrifuge was reported. The aim of this article is to study the degree to which benzyl alcohol can be eliminated with this method. By means of a gas chromatographic procedure, it has been proven that centrifugation is suitable for removing most benzyl alcohol (ie, up to 95.5% at 10,000 rpm, 5 minutes), and it is better at modifying the concentration of the drug than other proposed methods (ie, decantation, filtering methods, or both).

A simple and rapid method for purification of triamcinolone acetonide suspension for intravitreal injection.

Purification of triamcinolone acetonide suspension is important to avoid the potential toxic effects of the vehicle. The aim of this article is to describe a simple and rapid technique to remove the vehicle, instead of the long procedure described in the literature. Triamcinolone acetonide suspension was sedimented by density gradient centrifugation. In a few minutes, this new technique yielded a clear supernatant that was easily replaced by a nontoxic sterile solution. More prolonged procedures may not be convenient for unplanned treatments and may also contribute to a greater chance for contamination of the triamcinolone acetonide suspension.