Considering the small differences between different types in the diabetic retinopathy (DR) grading task, a retinopathy grading algorithm based on cross-layer bilinear pooling is proposed. Firstly, the input image is cropped according to the Hough circle transform (HCT), and then the image contrast is improved by the preprocessing method; then the squeeze excitation group residual network (SEResNeXt) is used as the backbone of the model, and a cross-layer bilinear pooling module is introduced for classification. Finally, a random puzzle generator is introduced in the training process for progressive training, and the center loss (CL) and focal loss (FL) methods are used to further improve the effect of the final classification. The quadratic weighted Kappa (QWK) is 90.84% in the Indian Diabetic Retinopathy Image Dataset (IDRiD), and the area under the receiver operating characteristic curve (AUC) in the Messidor-2 dataset (Messidor-2) is 88.54%. Experiments show that the algorithm proposed in this paper has a certain application value in the field of diabetic retina grading.
Humans ; Diabetic Retinopathy/diagnostic imaging* ; Algorithms ; ROC Curve ; Diabetes Mellitus

Microaneurysm is the initial symptom of diabetic retinopathy. Eliminating this lesion can effectively prevent diabetic retinopathy in the early stage. However, due to the complex retinal structure and the different brightness and contrast of fundus image because of different factors such as patients, environment and acquisition equipment, the existing detection algorithms are difficult to achieve the accurate detection and location of the lesion. Therefore, an improved detection algorithm of you only look once (YOLO) v4 with Squeeze-and-Excitation networks (SENet) embedded was proposed. Firstly, an improved and fast fuzzy c-means clustering algorithm was used to optimize the anchor parameters of the target samples to improve the matching degree between the anchors and the feature graphs; Then, the SENet attention module was embedded in the backbone network to enhance the key information of the image and suppress the background information of the image, so as to improve the confidence of microaneurysms; In addition, an spatial pyramid pooling was added to the network neck to enhance the acceptance domain of the output characteristics of the backbone network, so as to help separate important context information; Finally, the model was verified on the Kaggle diabetic retinopathy dataset and compared with other methods. The experimental results showed that compared with other YOLOv4 network models with various structures, the improved YOLOv4 network model could significantly improve the automatic detection results such as F-score which increased by 12.68%; Compared with other network models and methods, the automatic detection accuracy of the improved YOLOv4 network model with SENet embedded was obviously better, and accurate positioning could be realized. Therefore, the proposed YOLOv4 algorithm with SENet embedded has better performance, and can accurately and effectively detect and locate microaneurysms in fundus images.
Algorithms ; Diabetic Retinopathy/diagnostic imaging* ; Fundus Oculi ; Humans ; Microaneurysm/diagnostic imaging*

Metformin is a kind of biguanide hypoglycemic agent that has been widely used in patients with diabetes mellitus. In clinical practice, whether metformin should be stopped before Fundus fluorescein angiography (FFA) remains largely unclear. Some endocrinologists suggest stop metformin before FFA. However, ophthalmologists do not always adopt this opinion in their practice. This situation may lead to disputes between physicians and patients. This article analyzed contrast-induced nephropathy(CIN) and the related contrast agent, as well as the adverse reactions of fluorescein angiography. It pointed out that the discrepancy may be caused by misunderstanding of contrast agents used in FFA. For angiography using iodine contrast agent, metformin must be stopped because of the increased possibility of CIN, while for FFA using fluorescein sodium, no CIN has been reported yet. Therefore, the authors believe FFA is safe for diabetic patients with oral metformin and it is unnecessary to stop metformin before the examination.
Administration, Oral ; Contrast Media ; adverse effects ; Diabetes Mellitus, Type 2 ; diagnostic imaging ; drug therapy ; Diabetic Retinopathy ; diagnostic imaging ; Fluorescein Angiography ; adverse effects ; methods ; Humans ; Hypoglycemic Agents ; therapeutic use ; Kidney Diseases ; chemically induced ; Metformin ; therapeutic use

PURPOSE: To evaluate the effects of posterior subtenon triamcinolone acetonide injection on refractory diabetic macular edema (DME) after intravitreal bevacizumab (IVB) injection failure. METHODS: Patients with DME and central subfield thickness (CST) >300 microm who did not respond to IVB injections were retrospectively included. Specifically, we enrolled patients who were diagnosed with refractory DME and who experienced an increase in CST after 1 to 2 IVB injections or no decrease after > or =3 consecutive IVB injections. One clinician injected 20 mg of triamcinolone acetonide into the posterior subtenon space. All patients received ophthalmic examinations at baseline and at 2, 4, and 6 months post-baseline. Examinations included Snellen visual acuity, intraocular pressure, and spectral-domain optical coherence tomography. RESULTS: Forty eyes of 34 patients were included. The average baseline CST was 476 microm. The average CST decreased to 368 microm at 2 months, 374 microm at 4 months, and 427 microm at 6 months (p < 0.001 for all results, Wilcoxon signed-rank test). The average intraocular pressure increased from 15.50 to 16.92 mmHg at 2 months but decreased to 16.30 mmHg at 4 months and 15.65 mmHg at 6 months. Logarithm of the minimum angle of resolution visual acuity improved from 0.56 to 0.50 at 2 months (p = 0.023), 0.50 at 4 months (p = 0.083), and 0.48 at 6 months (p = 0.133, Wilcoxon signed-rank test). No complications were detected. CONCLUSIONS: Posterior subtenon triamcinolone acetonide is an effective and safe treatment for reducing CST in DME refractory to IVB.
Aged ; Angiogenesis Inhibitors/*therapeutic use ; Bevacizumab/*therapeutic use ; Diabetic Retinopathy/diagnostic imaging/*drug therapy/physiopathology ; Female ; Glucocorticoids/*administration & dosage ; Humans ; Injections, Intraocular ; Intraocular Pressure/physiology ; Intravitreal Injections ; Macular Edema/diagnostic imaging/*drug therapy/physiopathology ; Male ; Middle Aged ; Retrospective Studies ; Tenon Capsule/*drug effects ; Tomography, Optical Coherence ; Treatment Failure ; Triamcinolone Acetonide/*administration & dosage ; Vascular Endothelial Growth Factor A/antagonists & inhibitors ; Visual Acuity/physiology

Both diabetic retinopathy (DR) and coronary heart disease (CHD) are clinically significant in diabetic patients. We investigated the correlation between the severity of DR and the presence and severity of CHD among type 2 diabetic patients. A total of 175 patients who were examined at the DR clinic and underwent dual-source computed tomography (DSCT) angiography within 6 months were included. The degree of DR was graded as no DR, nonproliferative DR (NPDR), and proliferative DR (PDR). The severity of CHD and the numbers of significant stenotic coronary artery on DSCT angiography according to DR grade were assessed. The mean Agatston Calcium Score (ACS) in patients with PDR was significantly higher than other groups (P < 0.001). The overall odds of an ACS increase were about 4.7-fold higher in PDR group than in no DR group (P < 0.001). In PDR group, in comparison with in no DR, the odds of having 1 or 2 arterial involvement were 3-fold higher (P = 0.044), and those of having 3 were 17-fold higher (P = 0.011). The c-index, one of the predictability values in regression analysis model, was not significantly increased when PDR was added to classical CHD risk factors (0.671 to 0.706, P = 0.111). Conclusively, patients with PDR develop a greater likelihood of not only having CHD, but being more severe nature. PDR has no additional effect to classical CHD risk factors for predicting CHD.
Aged ; Angiography ; Coronary Artery Disease/complications/*pathology ; Coronary Vessels/diagnostic imaging ; Diabetes Mellitus, Type 2/*complications ; Diabetic Retinopathy/complications/*diagnosis/diagnostic imaging ; Female ; Glomerular Filtration Rate ; Humans ; Linear Models ; Male ; Middle Aged ; Odds Ratio ; Risk Factors ; Severity of Illness Index ; Tomography, X-Ray Computed

Both diabetic retinopathy (DR) and coronary heart disease (CHD) are clinically significant in diabetic patients. We investigated the correlation between the severity of DR and the presence and severity of CHD among type 2 diabetic patients. A total of 175 patients who were examined at the DR clinic and underwent dual-source computed tomography (DSCT) angiography within 6 months were included. The degree of DR was graded as no DR, nonproliferative DR (NPDR), and proliferative DR (PDR). The severity of CHD and the numbers of significant stenotic coronary artery on DSCT angiography according to DR grade were assessed. The mean Agatston Calcium Score (ACS) in patients with PDR was significantly higher than other groups (P < 0.001). The overall odds of an ACS increase were about 4.7-fold higher in PDR group than in no DR group (P < 0.001). In PDR group, in comparison with in no DR, the odds of having 1 or 2 arterial involvement were 3-fold higher (P = 0.044), and those of having 3 were 17-fold higher (P = 0.011). The c-index, one of the predictability values in regression analysis model, was not significantly increased when PDR was added to classical CHD risk factors (0.671 to 0.706, P = 0.111). Conclusively, patients with PDR develop a greater likelihood of not only having CHD, but being more severe nature. PDR has no additional effect to classical CHD risk factors for predicting CHD.
Aged ; Angiography ; Coronary Artery Disease/complications/*pathology ; Coronary Vessels/diagnostic imaging ; Diabetes Mellitus, Type 2/*complications ; Diabetic Retinopathy/complications/*diagnosis/diagnostic imaging ; Female ; Glomerular Filtration Rate ; Humans ; Linear Models ; Male ; Middle Aged ; Odds Ratio ; Risk Factors ; Severity of Illness Index ; Tomography, X-Ray Computed