scholarly journals Variability in grading diabetic retinopathy from stereo fundus photographs: comparison of physician and lay readers.

1977 ◽  
Vol 61 (3) ◽  
pp. 192-201 ◽  
Author(s):  
R. C. Milton ◽  
J. P. Ganley ◽  
R. H. Lynk
2017 ◽  
Vol 2017 ◽  
pp. 1-14 ◽  
Author(s):  
Mathilde M. Goudot ◽  
Anne Sikorav ◽  
Oudy Semoun ◽  
Alexandra Miere ◽  
Camille Jung ◽  
...  

Purpose. To evaluate the capacity of OCT angiography (OCTA) for detecting infraclinical lesions in parafoveal capillaries in diabetic patients without diabetic retinopathy (DR). Methods. This prospective observational cross-sectional case-control study analyzed the superficial and deep capillary plexuses (SCP and DCP) on macular OCTA scans (3 × 3 mm) centered on the fovea. We compared 22 diabetic patients (34 eyes included) without DR diagnosis on color fundus photographs, with 22 age- and gender-matched nondiabetic controls (40 eyes included). Qualitative analysis concerned morphological ischemic capillary alterations. Quantitative analysis measured foveal avascular zone (FAZ) size, parafoveal capillary density, and enlargement coefficient of FAZ between SCP and DCP. Results. Neither the qualitative nor quantitative parameters were significantly different between both groups. No microaneurysms or venous tortuosity was observed in any of the analyzed images. On the SCP, the mean FAZ area was 0.322 ± 0.125 mm2 in diabetic patients and 0.285 ± 0.150 mm2 in controls, P=0.31. On the DCP, the mean FAZ area was 0.444 ± 0.153 mm2 in cases and 0.398 ± 0.138 mm2 in controls, P=0.20. Conclusion. OCTA did not detect infraclinical qualitative or quantitative differences in parafoveal capillaries of diabetic patients without DR in comparison with nondiabetic controls.


Stroke ◽  
2020 ◽  
Vol 51 (Suppl_1) ◽  
Author(s):  
Ka-ho Wong ◽  
Cecilia Peterson ◽  
Rock Theodore ◽  
Kinga aitken ◽  
Michael Dela Cruz ◽  
...  

Background: Diabetic retinopathy is a common microvascular complication of diabetes. Previous research has shown that the macrovascular complications of diabetes, including stroke, are often comorbid with shared and, possibly, synergistic pathology. Methods: This is a secondary analysis of the subgroup of patients who enrolled in the ACCORD Eye study of ACCORD. The primary outcome is stroke during follow-up. The primary predictor was presence of diabetic retinopathy on the Early Treatment Diabetic Retinopathy Study Severity Scale as assessed from seven-field stereoscopic fundus photographs at study baseline. We fit adjusted Cox models to the primary outcome to provide hazard ratios for stroke and included interaction terms with the ACCORD randomization arms. Results: We included 2,828 patients with a mean (SD) age of 62.1 years and 61.8% were male. The primary outcome of stroke was met by 117 patients during a mean (SD) of 5.4 (1.8) years of follow-up. Diabetic retinopathy was present in 874/2,828 (30.9%) of patients at baseline, and was more common in patients with stroke versus without stroke (41.0 vs 30.5%, p=0.016). In the Cox model, adjusted for baseline patient age, gender, race, total cholesterol, Hgb A1c, smoking, and randomization arm, we found that diabetic retinopathy remained associated with incident stroke (HR 1.60, 95% CI 1.10-2.32, p=0.015) (Figure 1). This association was not affected by randomization to the ACCORD glucose intervention (p=0.305), lipid intervention (p=0.546), or blood pressure intervention (p=0.422). Conclusion: Diabetic retinopathy is associated with an increased risk of stroke, which suggests that the microvascular pathology inherent to diabetic retinopathy has larger cardiovascular implications.


Sensors ◽  
2019 ◽  
Vol 19 (3) ◽  
pp. 695 ◽  
Author(s):  
Emil Saeed ◽  
Maciej Szymkowski ◽  
Khalid Saeed ◽  
Zofia Mariak

Hard exudates are one of the most characteristic and dangerous signs of diabetic retinopathy. They can be marked during the routine ophthalmological examination and seen in color fundus photographs (i.e., using a fundus camera). The purpose of this paper is to introduce an algorithm that can extract pathological changes (i.e., hard exudates) in diabetic retinopathy. This was a retrospective, nonrandomized study. A total of 100 photos were included in the analysis—50 sick and 50 normal eyes. Small lesions in diabetic retinopathy could be automatically diagnosed by the system with an accuracy of 98%. During the experiments, the authors used classical image processing methods such as binarization or median filtration, and data was read from the d-Eye sensor. Sixty-seven patients (39 females and 28 males with ages ranging between 50 and 64) were examined. The results have shown that the proposed solution accuracy level equals 98%. Moreover, the algorithm returns correct classification decisions for high quality images and low quality samples. Furthermore, we consider taking retina photos using mobile phones rather than fundus cameras, which is more practical. The paper presents an innovative approach. The results are introduced and the algorithm is described.


Author(s):  
Mehmet Akif Cifci

The complication of people with diabetes causes an illness known as Diabetic Retinopathy (DR). It is very widespread among middle-aged and older people. As diabetes progresses, patients' vision may deteriorate and cause DR. People to lose their vision because of this illness. To cope with DR, early detection is needed. Patients will have to be checked by doctors regularly, which is a waste of time and energy. DR can be divided into two groups: non-proliferative (NPDR) while the other is proliferative (PDR). In this study, machine learning (ML) techniques are used to diagnose DR early. These are PNN, SVM, Bayesian Classification, and K-Means Clustering. These techniques will be evaluated and compared with each other to choose the best methodology. A total of 300 fundus photographs are processed for training and testing. The features are extracted from these raw images using image processing techniques. After an experiment, it is concluded that PNN has an accuracy of about 89%, Bayes Classifications 94%, SVM 97%, and K-Means Clustering 87%. The preliminary results prove that SVM is the best technique for early detection of DR.


2017 ◽  
Vol 9 (1) ◽  
pp. 43-50 ◽  
Author(s):  
Raba Thapa ◽  
Sanyam Bajimaya ◽  
Eli Pradhan ◽  
Govinda Paudyal

Introduction: Diabetic retinopathy (DR) is the emerging cause of blindness in the developing world. Timely detection of DR could save vision from its avoidable blinding condition. Objective: To assess the accuracy of DR grading in fundus photographs by the allied ophthalmic personnel (AOP) as compared to ophthalmologist at a community setting in Nepal. Materials and methods: Fundus photographs of known diabetes subjects attending for DR screening were graded by two groups of AOP and ophthalmologist. Agreement for DR grading by the AOP versus ophthalmologist was assessed using kappa coefficient (k). Results: Fundus photographs of 864 eyes of 435 subjects with diabetes were evaluated in the study. The agreement was substantial for detection of normal versus abnormal retina by both the AOP 1 and AOP 2. For normal versus abnormal macula, the agreement was substantial for AOP 1 and moderate for AOP 2. The agreement for grading macular exudates, retinal hemorrhage, venous beading ranged from moderate to substantial for both the AOPs. There was overall substantial agreement for diagnosing cases with or without DR and CSME by both the AOP 1 and AOP 2. The agreement ranged from fair to moderate for diagnosing other stages of NPDR by both the AOPs. Conclusion: Allied ophthalmic personnel with training could be a first level DR screener and referral of vision threatening DR. Three out of five diabetics could be managed at community level and thus reduce work load of ophthalmologist. This DR screening modality can be useful in other resource limited countries. 


Sign in / Sign up

Export Citation Format

Share Document