OBJECTIVE

To assess the diagnostic accuracy of point-of-care screening using Medios™ Artificial Intelligence (AI) in the diagnosis of diabetic retinopathy (DR).

This is a multi-center, cross-sectional, instrument validation study among adult Filipinos with Type 2 diabetes seen at Endocrine specialty clinics from May to November 2021. Retinal images were captured by a minimally trained nurse using the Remedio Fundus on Phone (FOP). Images were interpreted separately by the Medios™ AI and three retina specialists. The primary outcome measure is the accuracy of Medios™™ AI in diagnosing DR compared to retina specialists’ findings using sensitivity and specificity, predictive values, and likelihood ratios.

A total of 182 subjects with Type 2 diabetes were included in the study. The sensitivity and specificity of the Medios™ AI in diagnosing any DR were 73.68% (95%CI, 57.99-85.03) and 83.74% (95%CI, 79.35-87.35), respectively, compared with the retinal specialists’ findings using the same images. The positive and negative predictive values were 34.57% (95%CI, 25.12-45.41) and 96.47% (95%CI, 93.62-98.07). The positive and negative likelihood ratios were 4.53 (95%CI, 4.26 4.82) and 0.31 (95%CI, 0.26-0.38). The overall diagnostic accuracy of Medios™ AI in detecting DR was 82.69% (95%CI, 78.47-86.23).

The Medios™ AI system showed acceptable diagnostic accuracy when used as a point-of-care screening tool in detecting DR among patients with Type 2 diabetes seen at Endocrine specialty clinics. This technology can be a useful screening tool for endocrinologists as it is relatively inexpensive, safe, and easily performed. It can also shorten the lead time from screening to referral and intervention.

OBJECTIVE

To determine if dilated smartphone photography with a 3D-printed adaptor for a fundus lens can be an alternative screening tool for diabetic retinopathy.

This was a single-center, prospective, comparative study of 102 eyes of patients with diabetes mellitus. DR screening was performed using binocular indirect ophthalmoscopy (BIO) as the reference standard, alongside two imaging modalities: a traditional fundus camera and a smartphone equipped with a 3D-printed adaptor. Coded images were evaluated by three masked retina specialists. Sensitivity, specificity, predictive values, and likelihood ratios were calculated for both imaging modalities.

DR gradings of both imaging modalities were compared to BIO and showed very good agreement (κ 0.94-0.96 95% CI 0.93-0.99). Both have high levels of interobserver reliability (kappa value 0.88-0.92 95% CI 0.84-0.96) and intraobserver reliability (kappa value 0.85-1.00 95% CI 0.64-1.00). Smartphone and Visucam 500 images can detect presence of DR with a sensitivity of 93.9% (95% CI 83.1-98.7) and 91.8% (95% CI 80.4- 97.7); and a specificity of 90.6% (95% CI 79.3-96.9) and 92.5% (95% CI 81.8-97.9), respectively, as well as the presence of vision threatening DR with a sensitivity of 100% (95% CI 91.2-100) and specificity of 85.5% (95% CI 74.2-93.1). Image quality was similar between the two imaging modalities (P=1.00).

Smartphone-based imaging with an attached 3D-printed adaptor offers high sensitivity and reliability comparable to a traditional fundus camera, and can be used an alternative for DR screening.

BACKGROUND: It remains unclear whether circulating malondialdehyde (MDA) levels change in people with diabetic retinopathy (DR). This systematic review compared circulating MDA levels in diabetic people with and without DR. METHODS: PubMed, Medline (Ovid), Embase (Ovid), and Web of Science were searched for case-control studies conducted before May 2022 in English that compared circulating MDA levels in people with and without DR. The following MeSH search terms were used: ("malondialdehyde" or "thiobarbituric acid reactive substances [TBARS]" or "lipid peroxidation" or "oxidative stress") and "diabetic retinopathy." Newcastle-Ottawa Quality Assessment Scale was used to evaluate the quality of the included studies. Random-effects pairwise meta-analysis pooled the effect size with standardized mean difference (SMD) and 95% confidence intervals (CIs). RESULTS: This meta-analysis included 29 case-control studies with 1680 people with DR and 1799 people with diabetes but not DR. Compared to people without DR, the circulating MDA levels were higher in those with DR (SMD, 0.897; 95% CI, 0.631 to 1.162; P  < 0.001). The study did not identify credible subgroup effects or publication bias and the sensitivity analysis confirmed the robustness of the study. CONCLUSIONS: Circulating MDA levels are higher in people with DR compared to those without. Future comparative studies that use more specific methods are required to draw firm conclusions. REGISTRATION PROSPERO; https://www.crd.york.ac.uk/PROSPERO/ ; No. CRD42022352640.
Humans ; Diabetic Retinopathy ; Malondialdehyde ; Oxidative Stress ; Case-Control Studies ; Diabetes Mellitus

Humans ; Diabetic Retinopathy/pathology* ; Retina ; Diabetes Mellitus

Objective To investigate the neuroprotective effect of methylene blue on diabetic retinopathy in rats. Methods Thirty SD rats were randomly divided into blank, control and experimental groups. The control and experimental groups were induced with diabetes by streptozotocin (STZ) intraperitoneal injection. After 6 weeks of successful modeling, the experimental group received intravitreal injection of methylene blue at a dose of [0.2 mg/(kg.d)], while the control group received an equal amount of dimethyl sulfoxide (DMSO) intravitreal injection, both continuously injected for 7 days. ELISA was used to detect the levels of retinal superoxide dismutase (SOD), 8-iso-prostaglandin F2alpha (iPF2α) and interleukin-1β (IL-1β) in rats. Western blot analysis was used to detect the expression of retinal extracellular signal-regulated kinase 1/2 phosphorylation (p-ERK1/2) and phosphorylated protein kinase B (p-AKT), and PAS staining was used to detect retinal morphological changes. Results Compared with the blank group rats, the retinal SOD activity in the control and experimental group rats was significantly reduced. iPF2α, IL-1β and p-ERK1/2 level increased, while p-AKT level decreased. Compared with the control group, the SOD activity of the experimental group rats increased. iPF2α and IL-1β level went down, while p-ERK1/2 and p-AKT level went up significantly. The overall thickness of the retinal layer and the number of retinal ganglion cells were significantly reduced. Conclusion Methylene blue improves diabetic retinopathy in rats by reducing retinal oxidative stress and enhancing ERK1/2 and AKT phosphorylation.
Rats ; Animals ; Diabetic Retinopathy/metabolism* ; Proto-Oncogene Proteins c-akt/metabolism* ; Mitogen-Activated Protein Kinase 3/metabolism* ; Interleukin-1beta/metabolism* ; Methylene Blue/pharmacology* ; Phosphorylation ; Rats, Sprague-Dawley ; MAP Kinase Signaling System ; Diabetes Mellitus, Experimental/drug therapy* ; Superoxide Dismutase/metabolism*

Ginsenoside Rg_1, one of the main active components of precious traditional Chinese medicine Ginseng Radix et Rhizoma, has the anti-oxidative stress, anti-inflammation, anti-aging, neuroprotection, and other pharmacological effects. Diabetic retinopathy(DR), the most common complication of diabetes, is also the main cause of impaired vision and blindness in the middle-aged and the elderly. The latest research shows that ginsenoside Rg_1 can protect patients against DR, but the protection and the mechanism are rarely studied. This study mainly explored the protective effect of ginsenoside Rg_1 against DR in type 2 diabetic mice and the mechanism. High fat diet(HFD) and streptozotocin(STZ) were used to induce type 2 diabetes in mice, and hematoxylin-eosin(HE) staining was employed to observe pathological changes in the retina of mice. The immunohistochemistry was applied to study the localization and expression of nucleotide-binding oligomerization domain-like receptors 3(NLRP3) and vascular endothelial growth factor(VEGF) in retina, and Western blot was used to detect the expression of nuclear factor-kappa B(NF-κB), p-NF-κB, NLRP3, caspase-1, interleukin-1β(IL-1β), transient receptor potential channel protein 6(TRPC6), nuclear factor of activated T-cell 2(NFAT2), and VEGF in retina. The results showed that ginsenoside Rg_1 significantly alleviated the pathological injury of retina in type 2 diabetic mice. Immunohistochemistry results demonstrated that ginsenoside Rg_1 significantly decreased the expression of NLRP3 and VEGF in retinal ganglion cells, middle plexiform layer, and outer plexiform layer in type 2 diabetic mice. According to the Western blot results, ginsenoside Rg_1 significantly lowered the expression of p-NF-κB, NLRP3, caspase-1, IL-1β, TRPC6, NFAT2, and VEGF in retina of type 2 diabetic mice. These findings suggest that ginsenoside Rg_1 can significantly alleviate DR in type 2 diabetic mice, which may be related to inhibition of NLRP3 inflammasome and VEGF. This study provides experimental evidence for the clinical application of ginsenoside Rg_1 in the treatment of DR.
Aged ; Animals ; Diabetes Mellitus, Experimental/metabolism* ; Diabetes Mellitus, Type 2/genetics* ; Diabetic Retinopathy/genetics* ; Ginsenosides/pharmacology* ; Humans ; Inflammasomes/metabolism* ; Mice ; Middle Aged ; NF-kappa B/metabolism* ; NLR Family, Pyrin Domain-Containing 3 Protein/metabolism* ; Signal Transduction ; Vascular Endothelial Growth Factor A/genetics*

Objective The main objective of this study is to determine the agreement between the clinical staging of diabetic retinopathy (DR) with fluorescein angiography (FA) staging in an actual clinic.
Diabetic Retinopathy ; Fluorescein Angiography ; Ophthalmoscopy ; 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*

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

Diabetic retinopathy (DR), one of the chronic complications of diabetes, is a serious and irreversible blinding disease. It is difficult to detect in the early stage, to control in the progressive stage, to operate in the advanced stage of DR. Recently, the "14th Five-year plan" for National Eye Health proposed to "improve the management mode of chronic eye disease, and build a chronic disease management system". The project team used artificial intelligence technology based on cloud platform, joint outpatient service, virtual ward to explore the comprehensive management of DR from the aspects of early screening, multidisciplinary collaborative diagnosis and treatment, and refined blood glucose management during perioperative period. In the future, it is urgent to integrate DR chronic disease management with other systemic chronic diseases to reduce the blindness caused by DR.
Humans ; Diabetic Retinopathy/diagnosis* ; Artificial Intelligence ; Mass Screening ; Blindness/prevention & control* ; Diabetes Mellitus