•Deep learning detects diabetic retinopathy without extracting vein structures and identifying to lesions in the retina.•Deep learning is capable of learning diabetic retinopathy even if training ...does not include the images from same datasets.•The overall network performance decreases as the field of view of the smartphone-based images get smaller.•Using retina images from different datasets improves the detection performance of deep learning network.
Diabetic Retinopathy (DR) may result in various degrees of vision loss and even blindness if not diagnosed in a timely manner. Therefore, having an annual eye exam helps early detection to prevent vision loss in earlier stages, especially for diabetic patients. Recent technological advances made smartphone-based retinal imaging systems available on the market to perform small-sized, low-powered, and affordable DR screening in diverse environments. However, the accuracy of DR detection depends on the field of view and image quality. Since smartphone-based retinal imaging systems have much more compact designs than a traditional fundus camera, captured images are likely to be the low quality with a smaller field of view. Our motivation in this paper is to develop an automatic DR detection model for smartphone-based retinal images using the deep learning approach with the ResNet50 network. This study first utilized the well-known AlexNet, GoogLeNet, and ResNet50 architectures, using the transfer learning approach. Second, these frameworks were retrained with retina images from several datasets including EyePACS, Messidor, IDRiD, and Messidor-2 to investigate the effect of using images from the single, cross, and multiple datasets. Third, the proposed ResNet50 model is applied to smartphone-based synthetic images to explore the DR detection accuracy of smartphone-based retinal imaging systems. Based on the vision-threatening diabetic retinopathy detection results, the proposed approach achieved a high classification accuracy of 98.6%, with a 98.2% sensitivity and a 99.1% specificity while its AUC was 0.9978 on the independent test dataset. As the main contributions, DR detection accuracy was improved using the transfer learning approach for the ResNet50 network with publicly available datasets and the effect of the field of view in smartphone-based retinal imaging was studied. Although a smaller number of images were used in the training set compared with the existing studies, considerably acceptable high accuracies for validation and testing data were obtained.
To define the incidence of pseudophakic macular edema (PME) after cataract surgery and to identify contributory risk factors.
Retrospective database study of electronic medical records (EMRs).
A ...total of 81984 eyes undergoing cataract surgery between December 2010 and December 2014 from 8 independent United Kingdom clinical sites.
Structured clinical data mandated by the EMR were anonymized and extracted for each eye undergoing cataract surgery including: perioperative visual acuity, copathologic features, simultaneous surgical procedures, and the presence or absence of a specified list of intraoperative complications. Diabetic status with matched Early Treatment Diabetic Retinopathy Study (ETDRS) grading also was mandated by the EMR. Eyes receiving prophylactic nonsteroidal anti-inflammatory drugs were excluded.
Diagnosis of cystoid macular edema or new-onset macular edema in patients with diabetes, recorded by a healthcare professional within 90 days of surgery.
Baseline incidence of PME in eyes without operative complications, diabetes, or risk factors was 1.17%. Eyes in which PME developed were more likely to be male, older, and to demonstrate risk factors. The relative risk (RR) was increased in eyes with capsule rupture with or without vitreous loss (RR, 2.61; 95% confidence interval CI, 1.57-4.34), a previous diagnosis of epiretinal membrane (RR, 5.60; 95% CI, 3.45-9.07), uveitis (RR, 2.88; 95% CI, 1.50-5.51), retinal vein occlusion (RR, 4.47; 95% CI, 2.56-5.92), or retinal detachment repair (RR, 3.93; 95% CI, 2.60-5.92). High myopia, age-related macular degeneration, or prostaglandin analog use were not shown to increase risk. Eyes with PME on average had poorer postoperative visual acuity, which persisted to the latest time point assessed, up to 24 weeks. Eyes from patients with diabetes, even in the absence of retinopathy, had an increased RR (RR, 1.80; 95% CI, 1.36-2.36) of new macular edema after surgery. The risk was higher in the presence of any diabetic retinopathy (DR; RR, 6.23; 95% CI, 5.12-7.58) and rose proportionately with increasing severity of DR.
Pseudophakic macular edema occurs commonly after phacoemulsification cataract surgery, even in the absence of complications and risk factors. This large retrospective study using structured EMR data quantified the RRs of PME and the risk with increasing ETDRS severity of DR. It highlights the need for prophylactic therapy, especially in those groups of eyes with the highest RRs.
Age-related macular degeneration AMD is one of the leading causes of blindness in the elderly population. An advanced form of AMD known as neovascular AMD (nAMD) is implicated as the main attributor ...of visual loss among these patients. The hallmark feature of nAMD is the presence of neovascular structures known as choroidal neovascular membranes (CNVs), along with fluid exudation, hemorrhages, and subretinal fibrosis. These pathological changes eventually result in anatomical and visual loss. A type of proangiogenic factor known as vascular endothelial growth factor (VEGF) has been known to mediate the pathological process behind nAMD. Therefore, therapy has transitioned over the years from laser therapy that ablates the lesions to using Anti-VEGF to target the pathology directly. In this work, we provide an overview of current and emerging therapies for the treatment of nAMD. Currently approved Anti-VEGF agents include ranibizumab, aflibercept, and brolucizumab. Bevacizumab, also an Anti-VEGF agent, is used to manage nAMD even though this is an off-label use. While Anti-VEGF agents have provided a favorable prognosis for nAMD, they are associated with a substantial financial burden for patients and the healthcare system, due to their high cost as well as the need for frequent repeat treatments and visits. Emerging therapies and studies aim to extend the intervals between required treatments and introduce new treatment modalities that would improve patients’ compliance and provide superior results.
Gut microbiome and diabetic retinopathy Jabbehdari, Sayena; Sallam, Ahmed B.
European journal of ophthalmology,
09/2022, Letnik:
32, Številka:
5
Journal Article
Recenzirano
Odprti dostop
During the last decades, the incidence of diabetes and a variety of complications such as diabetic retinopathy (DR) and cardiovascular diseases have been increased exponentially. Gut bacterial ...composition -microbiota – has been associated with the pathobiology of many inflammatory and metabolic disorders such as diabetes. Gut microbiota plays a crucial role in preserving the metabolic and immune homeostasis, protecting against pathogens and regulating host immunity; however, gut microbiome ecosystem can be altered by lifestyle, cigarette smoking, dietary patterns, and oxidative stress.
Herein, we present a hypothesis on the potential complex association between gut microbiota and DR as one of the microvascular complications of diabetes.
•Cataract surgery can improve vision in cataractous eyes with retinitis pigmentosa (RP).•Intraoperative complications were similar between eyes with RP and control eyes.•RP eyes experienced ...postoperative cystoid macular edema more frequently.
To report visual acuity (VA) outcomes, intraoperative and postoperative complications of isolated cataract surgery in eyes with retinitis pigmentosa (RP), compared with non-RP-affected eyes.
Retrospective clinical cohort study.
A total of 113,389 eyes underwent cataract surgery between July 2003 and March 2015 at 8 clinical sites in the United Kingdom. Eyes with RP as the only comorbid pathology and eyes without any ocular comorbidities (controls) undergoing cataract surgery were compared. VA at 4 to 12 weeks postoperatively and rates of intraoperative and postoperative complications are reported.
Seventy-two eyes had RP. The mean age in the RP group was 57 ± 15 compared to 75 ± 10 in controls (P < .001). Females represented 46% of RP cases and 60% of controls (P = .06). Preoperative VA (mean LogMAR = 1.03 vs 0.59, P < .001) and postoperative VA (0.71 vs 0.14, P < .001) were worse in RP group. The mean VA gain was 0.25 ± 0.60 LogMAR in RP vs 0.43 ± 0.48 LogMAR in controls (P < .001). There were no significant differences in the rate of intraoperative pupil expansion use, posterior capsular tears, or zonular dialysis. Postoperative cystoid macular edema developed in 6.9% of RP eyes and 1% of controls (P < .001). The need for IOL repositioning or exchange was not statistically different between the two groups.
Cataract surgery can improve vision in eyes with RP and cataract. Intraoperative complications were similar to control eyes; however, RP eyes experienced more frequent postoperative cystoid macular edema.
To quantify the risk of pseudophakic cystoid macular edema (pCME) in fellow-eye cataract surgery and to determine risk factors, including prior first-eye pCME.
Retrospective, clinical database study.
...Patients undergoing bilateral nonsimultaneous cataract surgeries in 8 UK National Health Service clinical centers between July 2003 and March 2015.
We excluded patients with a history of diabetic macular edema (DME) or CME and perioperative topical nonsteroidal anti-inflammatory drug use in either eye. We calculated the overall risk of pCME and used Poisson model with robust estimation of standard error to identify potential risk factors for pCME in the fellow eye.
The risk of postoperative clinical pCME in the fellow eye.
A total of 54 209 patients were included. The mean age was 74.6 ± 10.4 years, and 38.8% were male. The fellow eye developed pCME in 544 patients (1%). The risk of fellow-eye pCME among patients without first-eye pCME was 0.9%. However, the risk of fellow-eye pCME among those with first-eye pCME was 10.7%. In the fully adjusted model, we found that the risk factors for the development of fellow-eye pCME were first-eye pCME (RR, 8.55, 95% confidence interval CI, 6.19-11.8), epiretinal membrane (ERM) (RR, 4.1, CI, 2.63-6.19), history of retinal vein occlusion (RR, 2.94, CI, 1.75-4.93), diabetes without history of DME (RR, 2.08, CI, 1.73-2.5), advanced cataract (RR, 1.75, CI, 1.16-2.65), prostaglandin analogue use preoperatively (RR, 1.49, CI, 1.13-1.97), and male sex (RR, 1.19, CI, 1.0-1.41).
History of pCME in the first-operated eye is the strongest independent risk factor for the development of pCME in the fellow eye. Our findings may guide clinicians in counseling patients on the risk of pCME before performing cataract surgery in the fellow eye and help in identifying high-risk patients who may benefit from prophylactic therapy.
Proprietary or commercial disclosure may be found after the references.
Retinitis pigmentosa (RP) is often undetected in its early stages. Artificial intelligence (AI) has emerged as a promising tool in medical diagnostics. Therefore, we conducted a systematic review and ...meta-analysis to evaluate the diagnostic accuracy of AI in detecting RP using various ophthalmic images. We conducted a systematic search on PubMed, Scopus, and Web of Science databases on December 31, 2022. We included studies in the English language that used any ophthalmic imaging modality, such as OCT or fundus photography, used any AI technologies, had at least an expert in ophthalmology as a reference standard, and proposed an AI algorithm able to distinguish between images with and without retinitis pigmentosa features. We considered the sensitivity, specificity, and area under the curve (AUC) as the main measures of accuracy. We had a total of 14 studies in the qualitative analysis and 10 studies in the quantitative analysis. In total, the studies included in the meta-analysis dealt with 920,162 images. Overall, AI showed an excellent performance in detecting RP with pooled sensitivity and specificity of 0.985 95%CI: 0.948–0.996, 0.993 95%CI: 0.982–0.997 respectively. The area under the receiver operating characteristic (AUROC), using a random-effect model, was calculated to be 0.999 95%CI: 0.998–1.000; P < 0.001. The Zhou and Dendukuri I² test revealed a low level of heterogeneity between the studies, with I2 = 19.94% for sensitivity and I2 = 21.07% for specificity. The bivariate I² 20.33% also suggested a low degree of heterogeneity. We found evidence supporting the accuracy of AI in the detection of RP; however, the level of heterogeneity between the studies was low.