- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
An enhanced diabetic retinopathy detection and classification approach using deep convolutional neural network
摘要： The objective of this study is to propose an alternative, hybrid solution method for diagnosing diabetic retinopathy from retinal fundus images. In detail, the hybrid method is based on using both image processing and deep learning for improved results. In medical image processing, reliable diabetic retinopathy detection from digital fundus images is known as an open problem and needs alternative solutions to be developed. In this context, manual interpretation of retinal fundus images requires the magnitude of work, expertise, and over-processing time. So, doctors need support from imaging and computer vision systems and the next step is widely associated with use of intelligent diagnosis systems. The solution method proposed in this study includes employment of image processing with histogram equalization, and the contrast limited adaptive histogram equalization techniques. Next, the diagnosis is performed by the classification of a convolutional neural network. The method was validated using 400 retinal fundus images within the MESSIDOR database, and average values for different performance evaluation parameters were obtained as accuracy 97%, sensitivity (recall) 94%, specificity 98%, precision 94%, FScore 94%, and GMean 95%. In addition to those results, a general comparison of with some previously carried out studies has also shown that the introduced method is efficient and successful enough at diagnosing diabetic retinopathy from retinal fundus images. By employing the related image processing techniques and deep learning for diagnosing diabetic retinopathy, the proposed method and the research results are valuable contributions to the associated literature.
关键词： Image processing,Deep learning,Convolutional neural network,Diabetic retinopathy
更新于2025-09-23 15:23:52
-
Computer-Assisted Diagnosis for Diabetic Retinopathy Based on Fundus Images Using Deep Convolutional Neural Network
摘要： Diabetic retinopathy (DR) is a complication of long-standing diabetes, which is hard to detect in its early stage because it only shows a few symptoms. Nowadays, the diagnosis of DR usually requires taking digital fundus images, as well as images using optical coherence tomography (OCT). Since OCT equipment is very expensive, it will benefit both the patients and the ophthalmologists if an accurate diagnosis can be made, based solely on reading digital fundus images. In the paper, we present a novel algorithm based on deep convolutional neural network (DCNN). Unlike the traditional DCNN approach, we replace the commonly used max-pooling layers with fractional max-pooling. Two of these DCNNs with a different number of layers are trained to derive more discriminative features for classification. After combining features from metadata of the image and DCNNs, we train a support vector machine (SVM) classifier to learn the underlying boundary of distributions of each class. For the experiments, we used the publicly available DR detection database provided by Kaggle. We used 34,124 training images and 1,000 validation images to build our model and tested with 53,572 testing images. The proposed DR classifier classifies the stages of DR into five categories, labeled with an integer ranging between zero and four. The experimental results show that the proposed method can achieve a recognition rate up to 86.17%, which is higher than previously reported in the literature. In addition to designing a machine learning algorithm, we also develop an app called 'Deep Retina.' Equipped with a handheld ophthalmoscope, the average person can take fundus images by themselves and obtain an immediate result, calculated by our algorithm. It is beneficial for home care, remote medical care, and self-examination.
关键词： deep convolutional neural network,mobile app,fractional max-pooling,support vector machine,diabetic retinopathy,fundus images,teaching-learning-based optimization
更新于2025-09-23 15:23:52
-
Retinal photoreceptor apoptosis is associated with impaired serum ionized calcium homeostasis in diabetic retinopathy: An in-vivo analysis
摘要： Purpose: The aim of this work was to study the association of serum ionized calcium with retinal photoreceptor apoptosis on spectral domain optical coherence tomography (SD-OCT) in diabetic retinopathy (DR). Methods: Sixty consecutive cases with Type 2 diabetes mellitus were categorized into three groups: no diabetic retinopathy; non-proliferative DR; proliferative DR. The eye with more severe form of the disease was considered. Twenty healthy controls were also included. Best corrected visual acuity (BCVA) was measured on logMAR scale. Retinal photoreceptor apoptosis was defined as disruption of retinal photoreceptor ellipsoid zone (EZ). Ellipsoid zone disruption was assessed using SD-OCT. Serum levels of total and ionized calcium were measured using standard protocol. Results: EZ disruption was found to be positively associated with serum total calcium and ionized calcium. Also, EZ disruption was found to be positively associated with logMAR BCVA. Conclusion: Increased serum ionized calcium induces retinal photoreceptor apoptosis resulting in increased EZ disruption in DR.
关键词： Ionized calcium,Retinal photoreceptor apoptosis,Diabetic retinopathy,Autophagy,Ellipsoid zone,Spectral-domain optical coherence tomography
更新于2025-09-23 15:23:52
-
Baicalin protects human retinal pigment epithelial cell lines against high glucose-induced cell injury by up-regulation of microRNA-145
摘要： Background: Diabetic retinopathy (DR) is a common complication of diabetes mellitus, which is a major reason of blindness. Baicalin (BAI) is a flavonoid extracted from Scutellaria baicalensis, whose pharmacological characterizes have been widely reported in various diseases. However, it remains unclear the effect of BAI on DR. The study aimed to confirm the protective effect of BAI on DR. Methods: ARPE-19 cells and HRMECs were exposed to the high glucose (HG) environment to construct a cell injury model. After treatment with HG and BAI, cell viability, apoptosis, inflammatory cytokines and ROS generations were determined in ARPE-19 cells and HRMECs. Subsequently, miR-145 inhibitor and its negative control were transfected into ARPE-19 cells, and the regulatory effects on HG-and BAI-co-treated cells were detected. NF-κB and p38MAPK signaling pathways were finally examined to state the underling mechanism. Results: HG treatment significantly induced ARPE-19 cells and HRMECs injury in vitro. BAI significantly promoted cell proliferation, reduced apoptosis, as well as inhibited the release of IL-1β, IL-6, IL-8 and ROS in HG- injured ARPE-19 cells and HRMECs. Additionally, the expression level of miR-145 was up-regulated in HG-and BAI-co-treated cells. More importantly, miR-145 inhibition reversed the protective effect of BAI on HG- injured ARPE-19 cells. Besides, we observed that BAI inhibited the activation of NF-κB and p38MAPK pathways by up-regulating miR-145. Conclusions: Results demonstrated that BAI exhibited the protective effect against HG-induced cell injury by up-regulation of miR-145.
关键词： microRNA-145,diabetic retinopathy,p38MAPK,NF-κB,baicalin
更新于2025-09-23 15:23:52
-
Assessment of diabetic teleretinal imaging program at the Portland Department of Veterans Affairs Medical Center
摘要： We conducted a retrospective chart review of 200 diabetic patients who had teleretinal imaging performed between January 1, 2010, and January 1, 2011, at Portland Department of Veterans Affairs (VA) Medical Center outpatient clinics to assess the effectiveness of the diabetic teleretinal imaging program. Twenty patients (10%) had diabetic retinopathy. Ninety percent of the available teleretinal imaging studies were of adequate quality for interpretation. In accordance with local VA policy at that time, all teleretinal imaging patients should have been referred for a dilated retinal examination the following year. Image readers referred 97.5% of the patients to eye clinics for subsequent eye examinations, but the imagers scheduled appointments for only 80% of these patients. The redundancy rate, i.e., patients who had an eye examination within the past 6 mo, was 11%; the duplicate recall rate, i.e., patients who had a second teleretinal imaging performed within 1 yr of the eye examination, was 37%. Rates of timely diabetic eye examinations at clinics with teleretinal imaging programs, particularly when teleretinal imaging and eye clinics were colocated at the same community-based outpatient clinic, were higher than at those without a teleretinal imaging program. We concluded that the Portland VA Medical Center's teleretinal imaging program was successful in increasing the screening rate for diabetic retinopathy.
关键词： teleretinal imaging program,diabetes mellitus,dilated eye examination,diabetic retinal examination,primary care,diabetic retinopathy,outpatient clinic,screening,teleretinal imaging,telehealth
更新于2025-09-23 15:22:29
-
Multimodal imaging in perifoveal exudative vascular anomalous complex with co-existent diabetic retinopathy
摘要： Perifoveal exudative vascular anomalous complex (PEVAC) is a recently described macular entity, which can be overlooked or confused with other retinal pathologies. On the basis of multimodal imaging features of 15 eyes with PEVAC, Sacconi et al.1 defined a PEVAC lesion as a unilateral, isolated, perifoveal, aneurysmal abnormality, occurring in otherwise healthy individuals, having characteristic angiographic and tomographic features and typically resistant to anti-vascular endothelial growth factor (anti-VEGF) therapy. Clinically, a PEVAC lesion is associated with intra-retinal fluid, haemorrhage and/or hard exudates. Querques et al.2 reported two cases of PEVAC and described the related angiographic and optical coherence tomography (OCT) features. In particular, there were no features of arterial hypertension, diabetes, or any other systemic or local vasculopathy, or signs of age-related macular degeneration. Multimodal imaging, in particular OCT-angiography along with OCT, fluorescein angiography and indocyanine angiography, have been used to provide a better understanding of the pathophysiological mechanisms of PEVAC. The purpose of this case report is to describe the imaging findings of a PEVAC lesion in a patient who had features of diabetic retinopathy.
关键词： aneurysm,perifoveal exudative vascular anomalous complex,diabetic retinopathy
更新于2025-09-23 15:22:29
-
HMGB1 siRNA can reduce damage to retinal cells induced by high glucose in vitro and in vivo
摘要： Background: Diabetic retinopathy (DR), one of the most common complications of late-phase diabetes, is associated with many risk factors, among which continuous low-grade inflammation is one of the principal ones. As such, lowering inflammation levels and maintain the viability of human retinal endothelial cells (HRECs) are critical for DR therapy. HMGB1 is a well-known proinflammatory cytokine. However, whether HMGB1 small interfering RNA (siRNA) can protect retina cells under a high-glucose environment from morphological changes and functional abnormalities remain undetermined. We aimed to investigate the effect of HMGB1 siRNA on retinal cells in DR. Materials and methods: A total of 80 adult Wistar rats were randomly divided into four groups (n=20 each): normal control, diabetes mellitus (DM), scrambled (Scr) siRNA, and HMGB1 siRNA. Rats in the DM, Scr siRNA, and siRNA groups were established by intraperitoneal injection of streptozotocin. At 16 weeks after injection, rats in the siRNA and Scr-siRNA groups were intravitreally injected with 2 μL HMGB1 siRNA and 2 μL Scr-siRNA, while rats in the control and DM groups were intravitreally injected with the same dose of sterile saline. At 1 week after injections, we performed the following experiments. Immunohistochemical staining and real-time quantitative polymerase chain reaction were performed to test HMGB1 protein and messenger RNA expression in retinas. We performed TUNEL assays to detect retinal cell apoptosis and electroretinography to detect retinal function. In HRECs treated with high glucose, proliferation, morphology, apoptosis, superoxide dismutase (SOD), and reactive oxygen species production were detected. Western blot was applied to determine the expressions of HMGB1 and its related protein and apoptosis protein. Results: Intravitreal injection of HMGB1 siRNA reduced protein and messenger RNA expression of HMGB1 (both P,0.05). Intravitreal injection of HMGB1 siRNA reduced apoptosis of retinal cells (P,0.05), protected morphological changes in the retina, and improved the function of the retina (P,0.05). In HRECs treated with high glucose, HMGB1 siRNA pretreatment increased cell viability, reduced cell apoptosis, and reduced oxidative damage to cells (all P,0.05). Western blot detection found that HMGB1 siRNA pretreatment can inhibit the expression of cleaved caspase 3 and improve the expression of BCL2 (P,0.05). HMGB1 and NFκB expression increased in a time-dependent manner in the high-glucose environment and IKKβ and NFκB protein expression decreased significantly after HMGB1 silencing. Conclusion: As a therapeutic target, HMGB1 siRNA can reduce retinal cell damage induced by high glucose in vitro and in vivo and delay DR progress through the HMGB1–IKKβ–NFκB signaling pathway.
关键词： small interfering RNA,diabetic retinopathy,human retinal endothelial cells,inhibitor of nuclear factor κB,nuclear factor κB,high-mobility group box 1
更新于2025-09-23 15:22:29
-
The assessment of autofluorescence of the crystalline lens in diabetic patients and healthy controls: can it be used as a screening test?
摘要： Background: Our purpose was to demonstrate if measuring lens autofluorescence (AF) with a scanning confocal biomicroscope may be used to identify subjects with undiagnosed type II diabetes mellitus (DM), and hence, for it to be used as a marker for the severity of diabetic retinopathy in diabetic patients. Patients and methods: In this cross-sectional, comparative study, lens AF was measured with scanning confocal lens fluorescence biomicroscope in diabetic and healthy groups. Full ophthalmological examination was performed. Blood tests of fasting plasma glucose, and glycosylated hemoglobin were also analyzed. The correlation between lens AF results and blood tests was evaluated in both groups. The cutoff value for the diagnosis of DM using lens AF was investigated. Results: The study included 191 subjects with a mean age of 52.09±6.75 years. One hundred and seven (56.0%) subjects were female, and 84 (44.0%) were male. Eighty-two (42.9%) patients had type II DM, and 109 (57.1%) subjects self-reported as normal. The fluorescence ratio (FR) values ranged from 0.09 to 0.46 (0.23±0.06) in the total group. Mean FR measurements of diabetic subjects were significantly higher (0.27±0.06) than those without DM (0.20±0.05), ( p=0.001). A statistically significant correlation was found between glycosylated hemoglobin, fasting plasma glucose, and FR. The cutoff point for the FR according to the presence of DM was found to be 0.24 and above ( p=0.001), with a sensitivity of 71.95% and a specificity of 80.73%. Conclusion: Measuring AF of human lens as an indirect evidence of increased advanced glycation end products may helpful in detecting impaired glucose metabolism. Our results show highly significant correlation between possibility of DM and FR.
关键词： fasting plasma glucose,diabetic retinopathy,screening,HgA1c,lens autofluorescence,diabetes
更新于2025-09-23 15:22:29
-
Employing Image Processing Techniques and Artificial Intelligence for Automated Eye Diagnosis Using Digital Eye Fundus Images
摘要： Blindness usually comes from two main causes, glaucoma and diabetes. Robust mass screening is performed for diagnosing, such as screening that requires a cost-effective method for glaucoma and diabetic retinopathy and integrates well with digital medical imaging, image processing, and administrative processes. For addressing all these issues, we propose a novel low-cost automated glaucoma and diabetic retinopathy diagnosis system, based on features extraction from digital eye fundus images. This paper proposes a diagnosis system for automated identification of healthy, glaucoma, and diabetic retinopathy. Using a combination of local binary pattern features, Gabor filter features, statistical features, and color features which are then fed to an artificial neural network and support vector machine classifiers. In this work, the classifier identifies healthy, glaucoma, and diabetic retinopathy images with an accuracy of 91.1%,92.9%, 92.9%, and 92.3% and sensitivity of 91.06%, 92.6%, 92.66%, and 91.73% and specificity of 89.83%, 91.26%, 91.96%, and 89.16% for ANN, and an accuracy of 90.0%,92.94%, 95.43%, and 97.92% and sensitivity of 89.34%, 93.26%, 95.72%, and 97.93% and specificity of 95.13%, 96.68%, 97.88%, and 99.05% for SVM, based on 5, 10, 15, and 31 number of selected features. The proposed system can detect glaucoma, diabetic retinopathy and normal cases with high accuracy and sensitivity using selected features, the performance of the system is high due to using of a huge fundus database.
关键词： and Classification,Diagnosis,Diabetic Retinopathy,Glaucoma,Eye Fundus
更新于2025-09-23 15:22:29
-
The Prevalence and Risk Factors for Diabetic Retinopathy in Shiraz, Southern Iran
摘要： Globally, diabetic retinopathy (DR) is one of the leading causes of blindness, that diminishes quality of life. This study aimed to describe the prevalence of DR, and its associated risk factors. This cross-sectional study was carried out among 478 diabetic patients in a referral center in Fars province, Iran. The mean±standard deviation age of the participants was 56.64±12.45 years old and DR prevalence was 32.8%. In multivariable analysis, lower education levels (adjusted odds ratio [aOR], 0.43; 95% confidence interval [CI], 0.24 to 0.76), being overweight (aOR, 1.70; 95% CI, 1.02 to 2.83) or obese (aOR, 1.88; 95% CI, 1.09 to 3.26), diabetes duration of 10 to 20 years (aOR, 2.35; 95% CI, 1.48 to 3.73) and over 20 years (aOR, 5.63; 95% CI, 2.97 to 10.68), receiving insulin (aOR, 1.99; 95% CI, 1.27 to 3.10), and having chronic diseases (aOR, 1.71; 95% CI, 1.02 to 2.85) were significantly associated with DR. In conclusion, longer diabetes duration and obesity or having chronic diseases are strongly associated with DR suggesting that control of these risk factors may reduce both the prevalence and impact of retinopathy in Iran.
关键词： Diabetic retinopathy,Prevalence,Diabetes,Diabetes duration,Risk factors,Insulin
更新于2025-09-23 15:22:29