Early Diabetic Retinopathy Research Articles

A Deep Learning System for Early Diabetic Retinopathy Diagnosis

A Deep Learning System for Early Diabetic Retinopathy Diagnosis

The application of artificial intelligence in diabetic retinopathy: progress and prospects.

In recent years, artificial intelligence (AI), especially deep learning models, has increasingly been integrated into diagnosing and treating diabetic retinopathy (DR). From delving into the singular realm of ocular fundus photography to the gradual development of proteomics and other molecular approaches, from machine learning (ML) to deep learning (DL), the journey has seen a transition from a binary diagnosis of "presence or absence" to the capability of discerning the progression and severity of DR based on images from various stages of the disease course. Since the FDA approval of IDx-DR in 2018, a plethora of AI models has mushroomed, gradually gaining recognition through a myriad of clinical trials and validations. AI has greatly improved early DR detection, and we're nearing the use of AI in telemedicine to tackle medical resource shortages and health inequities in various areas. This comprehensive review meticulously analyzes the literature and clinical trials of recent years, highlighting key AI models for DR diagnosis and treatment, including their theoretical bases, features, applicability, and addressing current challenges like bias, transparency, and ethics. It also presents a prospective outlook on the future development in this domain.

Multifocal electroretinogram changes after panretinal photocoagulation in early proliferative diabetic retinopathy.

Panretinal photocoagulation (PRP) impacts macular function in eyes with early proliferative diabetic retinopathy (PDR). Herein, we used the multifocal electroretinogram (mfERG) to objectively investigate this concept. In this prospective interventional case series, we enrolled patients with treatment-naive early PDR, absence of clinically significant macular edema, and requirement for PRP. All participants underwent detailed ocular examinations. We measured the best-corrected distance visual acuity (BCDVA), conducted optical coherence tomography imaging to measure central macular thickness (CMT), and performed mfERG at baseline and 3 months post-PRP. Amplitude and latency of the mfERG response were evaluated within the innermost four of the five concentric rings. We enrolled 29 eyes of 23 patients with a mean (standard deviation) age of 54.3 (8.8) years and male-to-female ratio of 1:1.3. The mean BCDVA was unchanged post-treatment (P > 0.05), and the BCDVA in 26 eyes (89.7%) was either improved or unchanged, whereas in three eyes (10.3%) it decreased. The mean CMT was unchanged post-PRP (P > 0.05). Concerning the mfERG, the mean P1 amplitudes decreased significantly in all four concentric rings from the foveola at 3 months post-PRP compared with baseline values (all P < 0.05); however, the latencies were unchanged (all P > 0.05). At baseline, BCDVA correlated significantly with both the amplitude (r = + 0.55; P < 0.05) and latency (r = - 0.38; P < 0.05) of the mfERG in the central ring, whereas a significant correlation was detected with only the amplitude at 3 months post-PRP (r = + 0.52; P < 0.05). Macular function was decreased 3 months post-PRP in patients with early PDR, as indicated by decreased amplitude of the mfERG, whereas the functional and anatomical parameters were stable. The mfERG served as an objective tool for measuring retinal function and predicting visual outcomes post-PRP in eyes with early PDR. A higher amplitude in the mfERG correlated substantially with a better visual outcome post-PRP. Further multi-center longitudinal studies with robust designs including different PDR severity levels may reveal additional objective after-effects of PRP.

6635 A Call for High-Value Diabetic Retinopathy Screening and the Introduction of a Novel Predictive Model

Abstract Disclosure: P.J. Williams: None. C. Kim: None. K. Ziadeh: None. E. White: None. C.S. Chen-Milhone: None. L.D. Lugo: None. K.B. Ellis: None. M.S. Juzych: None. Introduction: Diabetic retinopathy (DR) remains one of the leading causes of preventable blindness in middle-aged and older populations. Late-stage DR is often visually symptomatic; however, early DR is generally asymptomatic and requires screening. For type 2 diabetes mellitus (T2DM), the American Diabetes Association and the National Institute of Diabetes and Digestive and Kidney Diseases recommend an initial dilated eye exam with follow-up every 1-3 years, or annually if diabetic retinopathy is detected. Despite these guidelines, providers and insurers in the United States routinely recommend annual screening regardless of other factors. Methods: A cross-sectional analysis with prospective data verification was done for new diabetic eye exam referrals from August 2021 through July 2022 to address screening discrepancies. Established patients and referrals unrelated to hyperglycemia were excluded. Data collection included duration of diabetes, last hemoglobin A1c (HbA1c), maximum HbA1c, age, race, gender, BMI, relevant medication use, and imaging, as well as the presence of hypertension, hypercholesterolemia, diabetic peripheral neuropathy (DPN), diabetic nephropathy, diabetic amputation, and DR and data was verified against objective records up to 2 years later. Logistical multivariate analysis was conducted by a trained statistician. Results: 474 consecutive patients were referred over 1 year. Among all referrals, 2.11% (10/474) of patients were prediabetic. Among the patients with diabetes, 5.39% (25/464) were type 1 and 94.6% (439/464) were type 2. Among patients with T2DM, the average age was 57.64±11.52 (range = 25-98) and females made up 52.16% (229/439). The average diabetes duration among T2DM patients with DR was 19.39±8.45 years (range = 11-51) compared to 6.12±5.77 years (range = 0-31 years) without DR. Maximum lifetime HbA1c (p=0.0016), duration of diabetes (p&amp;lt;0.0001), and presence of DPN (p = 0.0047) were the strongest predictors of DR in patients with T2DM. A predictive model for DR was created utilizing these variables and hypercholesterolemia (p=0.048) with a predictive accuracy of 89.5% (chi-squared value (4)=121.451, p&amp;lt;0.001, Nagelkerke R2=57%). Sensitivity and specificity measured 92.31% and 87.99% respectively. The area under the ROC curve measured 0.952 (CI 0.925 to 0.972, p&amp;lt;0.0001). Conclusion: Duration of diabetes, maximum HbA1c, and DPN are strongly associated with DR among patients with T2DM and should be tracked to guide screening. Among T2DM patients with a well-documented recent onset of diabetes, maximum HbA1c less than 8, and benign initial eye exam, annual diabetic eye exams are unlikely to reveal DR in the first 10 years after onset. Based on our data, annual diabetic screening exams are unnecessary for many patients, and upper limits of screening intervals should be considered in low-risk patients to increase high-value care. Presentation: 6/2/2024

ROS-Responsive Nanoparticles with Antioxidative Effect for the treatment of Diabetic Retinopathy

Diabetic retinopathy (DR) is a common microvascular complication of diabetes necessitating early intervention to impede progression, despite current clinical treatments focusing on advanced stages. Essential oils from Fructus Alpiniae zerumbet (EOFAZ) have demonstrated efficacy in protecting against high glucose (HG)-induced Müller cell activation and DR development. This study introduced a reactive oxidative species (ROS)-responsive drug delivery system (NPSPHE@EOFAZ) targeting early DR stages and oxidative stress. Our engineered nanoparticles effectively deliver EOFAZ into HG-exposed Müller cells by detecting and responding to elevated oxidative stress levels. The NPSPHE@EOFAZ significantly inhibited abnormal cell growth, reduced oxidative stress, and alleviated inflammation in vitro. In vivo experiments on diabetic mice with DR revealed that NPSPHE@EOFAZ mitigated early pathological changes by reducing oxidative stress and inflammation while also alleviating organ damage in the heart, liver, spleen, lung, and kidney. These findings underscore the potential of NPSPHE@EOFAZ as a promising antioxidant for early intervention in DR pathogenesis.

Evaluation of Foveal Vasculature by Optical Coherence Tomography Angiography after Pan-Retinal Photocoagulation versus Intravitreal Anti-VEGF Injections.

This study aimed to compare macular vascular changes one and three months after treatment with either panretinal photocoagulation (PRP) or intravitreal bevacizumab (IVB). A total of 62 eyes with very severe non-proliferative diabetic retinopathy or early proliferative diabetic retinopathy without center-involved diabetic macular edema, were included in this retrospective study. Thirty-nine eyes were allocated to the PRP group, while 23 eyes were treated with IVB. Optical coherence tomography angiography (OCTA) was performed to measure foveal avascular zone (FAZ) characteristics as well as the densities of superficial and deep capillary plexuses (SCP and DCP). In the IVB group, the FAZ area and perimeter expanded at month one but returned to baseline level after three months. In the PRP group, however, the FAZ area and perimeter were rather steady. Changes in the FAZ area were significantly different between the treatment groups at month one (P = 0.02), but not at month three (P = 0.31). There was no significant difference in the change in FAZ circularity index between the two groups at each time point (P = 0.55 and P = 0.31). Similarly, changes in SCP density were not statistically significant between the two groups at both time points (all Ps 0.05). A comparison of the two treatment arms based on the mean change in DCP density revealed a significant difference at month one, but not at month three (P = 0.01 and P = 0.49, respectively). Although bevacizumab and PRP have different short-term macular vascular responses, both therapies have the ability to normalize or stabilize vascular measures over time.

Evaluating the performance of a non-uniform squash function in Capsule networks for early diabetic retinopathy detection using fundus image analysis

Diabetic retinopathy, known as one of the most crucial neurovascular barriers to diabetes, has affected the retina, leading to a potential risk of blindness. The harm is not isolated to personal health, with negative effects felt at the economic and social level in entire communities. The consequences of vision loss secondary to diabetic retinopathy on quality of life, independent living, and employment can be significant. It is necessary to adopt a comprehensive approach for effective management of underlying diabetes, which includes early detection, timely intervention, supportive care, and management of comorbid conditions. Critical to this is a precise classification of its stages for disease analysis. A manual diagnosis, on the other hand, requires qualified individuals to recognize the subtle nuances of the condition, making it a lengthy process. This study aims to develop a computer-assisted system for analyzing retinal fundus images to identify and classify different stages of diabetic retinopathy. A dataset of eye fundus images was used by the proposed network to classify the presence and severity of blindness and its stages. The proposed model uses a non-uniform squash function in capsule neural networks. The motive behind employing a non-uniform squash function was to encourage the model to allocate greater attention and stability to intricate images throughout the training phase. The proposed model's efficiency is assessed meticulously with pre-trained CNN methods like InceptionV3, VGG16, VGG19, and Xception. The proposed capsule network model has been revamped to address the shortcomings of prior research with care. The method stands out with an impressive 99.84 % training accuracy and 98.68 % validation accuracy. Diabetic retinopathy, though rare in occurrence, poses significant challenges due to the high cost of diagnosis, rendering it inaccessible to rural and remote communities with limited access to tests and check-ups.

VEGF-B prevents chronic hyperglycemia-induced retinal vascular leakage by regulating the CDC42-ZO1/VE-cadherin pathway.

Non-proliferative diabetic retinopathy (NPDR) is the early stage of diabetic retinopathy (DR) and is a chronic oxidative stress-related ocular disease. Few treatments are approved for early DR. This study aimed to investigate the pathogenic mechanisms underlying the retinal micro-vasculopathy induced by diabetes and to explore an early potential for treating early DR in a mouse model. The mouse model of type 1 diabetes was established by intraperitoneal injection of streptozotocin (STZ, 180 mg/kg), which was used as the early DR model. The body weight and blood glucose mice were measured regularly; The retinal vascular leakage in the early DR mice was determined by whole-mount staining; Label-free quantitative proteomic analysis and bioinformatics were used to explore the target proteins and signaling pathways associated with the retinal tissues of early DR mice; To detect the effects of target protein on endothelial cell proliferation, migration, and tube formation, knockdown and overexpression of VEGF-B were performed in human retinal vascular endothelial cells (HRECs); Western blotting was used to detect the expression of target proteins invitro and invivo; Meanwhile, the therapeutic effect of VEGF-B on vascular leakage has also been evaluated invitro and invivo. The protein expressions of vascular endothelial growth factor (VEGF)-B and the Rho GTPases family member CDC42 were reduced in the retinal tissues of early DR. VEGF-B upregulated the expression of CDC42/ZO1/VE-cadherin and prevented hyperglycemia-induced vascular leakage in HRECs. Standard intravitreal VEGF-B injections improved the retinal vascular leakage and neurovascular response in early DR mice. Our findings demonstrated, for the first time, that in diabetes, the retinal vessels are damaged due to decreased VEGF-B expression through downregulation of CDC42/ZO1/VE-cadherin expression. Therefore, VEGF-B could be used as a novel therapy for early DR.

A Novel Automated System for Early Diabetic Retinopathy Detection and Severity Classification

ABSTRACTDiabetes is a common and serious global disease that damages blood vessels in the eye, leading to vision loss. Early and accurate diagnosis of this issue is crucial to reduce the risk of visual impairment. The typical deep learning (DL) methods for diabetic retinopathy (DR) grading are often time‐consuming, resulting in unsatisfactory detection performance due to inadequate representation of lesion features. To overcome these challenges, this research proposes a new automated mechanism for detecting and classifying DR, aiming to identify DR severities and different stages. To figure out and capture feature characteristics from DR samples, a conjugated attention mechanism and vision transformer are utilized within a collective net model, which automatically generates feature maps for diagnosing DR. These extracted feature maps are then fused through the feature fusion function in a fused attention net model, calculating attention weights to produce the most powerful feature map. Finally, the DR cases are identified and discriminated using the kernel extreme learning machine (KELM) model. For evaluating DR severity, our work utilizes four different benchmark datasets: APTOS 2019, MESSIDOR‐2 dataset, DiaRetDB1 V2.1, and DIARETDB0 datasets. To illuminate data noise and unwanted variations, two preprocessing steps are carried out, which include contrast enhancement and illumination correction. The experimental results evaluated using well‐known indicators demonstrate that the suggested method achieves a higher accuracy of 99.63% compared to other baseline methods. This research contributes to the development of powerful DR screening techniques that are less time‐consuming and capable of automatically identifying DR severity levels at a premature level.

Retinal microvascular changes in patients with pancreatitis and their clinical significance

Acute pancreatitis, a common exocrine inflammatory disease affecting the pancreas, is characterized by intense abdominal pain and multiple organ dysfunction. However, the alterations in retinal blood vessels among individuals with acute pancreatitis remain poorly understood. This study employed optical coherence tomography angiography (OCTA) to examine the superficial and deep retinal blood vessels in patients with pancreatitis. Sixteen patients diagnosed with pancreatitis (32 eyes) and 16 healthy controls (32 eyes) were recruited from the First Affiliated Hospital of Nanchang University for participation in the study. Various ophthalmic parameters, such as visual acuity, intraocular pressure, and OCTA image for retina consisting of the superficial retinal layer (SRL) and the deep retinal layer (DRL), were recorded for each eye. The study observed the superficial and deep retinal microvascular ring (MIR), macrovascular ring (MAR), and total microvessels (TMI) were observed. Changes in retinal vascular density in the macula through annular partitioning (C1–C6), hemispheric quadrant partitioning (SR, SL, IL, and IR), and early diabetic retinopathy treatment studies (ETDRS) partitioning methods (R, S, L, and I). Correlation analysis was employed to investigate the relationship between retinal capillary density and clinical indicators. Our study revealed that in the superficial retinal layer, the vascular density of TMI, MIR, MAR, SR, IR, S, C2, C3 regions were significantly decreased in patients group compared with the normal group. For the deep retinal layer, the vascular density of MIR, SR, S, C1, C2 regions also reduced in patient group. The ROC analysis demonstrated that OCTA possesses significant diagnostic performance for pancreatitis. In conclusion, patients with pancreatitis may have retinal microvascular dysfunction, and OCTA can be a valuable tool for detecting alterations in ocular microcirculation in pancreatitis patients in clinical practice.

Vitrectomy for cases of diabetic retinopathy.

Microvascular complications of diabetic retinopathy (DR) may require surgical intervention in the form of vitrectomy. Since its inception, diabetic vitrectomy has evolved with introduction of better instruments, newer techniques, and smaller port sizes. Common indications for diabetic vitrectomy include nonresolving vitreous hemorrhage, tractional retinal detachment, epiretinal membrane, progression of fibrovascular membranes despite laser therapy, recalcitrant diabetic macular edema, and neovascular glaucoma. Preoperative systemic stabilization is essential prior to planning surgery. Surgical techniques commonly used in diabetic vitrectomy are segmentation, delamination, and rarely en-bloc dissection. Modification in surgical techniques such as chandelier-assisted bimanual dissection and pharmacological adjuvants improve surgical outcomes in these patients. Prognosis in these patients could be improved with early intervention. Studies evaluating the outcome of vitrectomy in patients with early proliferative DR are required to understand the appropriate time of intervention in patients. Treatment aimed at arresting the progression of DR and gene therapy are avenues that need further evaluation. The following review will focus on covering the epidemiology of DR, indications of vitrectomy, preoperative considerations, surgical procedures of diabetic vitrectomy, methods of membrane dissection, pharmacological adjuvants to vitrectomy, outcomes of diabetic vitrectomy, and future directions of diabetic vitrectomy.

Research on the role of exosomes secreted by immortalized adipose-derived mesenchymal stem cells differentiated into pericytes in the repair of high glucose-induced retinal vascular endothelial cell damage

Diabetic retinopathy, a leading cause of vision impairment, is marked by microvascular complications in the retina, including pericyte loss, a key indicator of early-stage disease. This study explores the therapeutic potential of exosomes derived from immortalized adipose-mesenchymal stem cells differentiated into pericyte-like cells in restoring the function of mouse retinal microvascular endothelial cells damaged by high glucose conditions, thereby contributing to the understanding of early diabetic retinopathy intervention strategies. To induce immortalized adipose-mesenchymal stem cells differentiation into pericyte-like cells, the study employed pericyte growth supplement. And confirmed the success of cell differentiation through the detection of α-smooth muscle actin and neural/glial antigen 2 expression by Western blot and immunofluorescence. Exosomes were isolated from the culture supernatant of immortalized adipose-mesenchymal stem cells using ultracentrifugation and characterized through Western blot for exosomal markers (CD9, CD81, and TSG101), transmission electron microscopy, and nanoparticle tracking analysis. Their influence on mouse retinal microvascular endothelial cells under high glucose stress was assessed through various functional assays. Findings revealed that exosomes, especially those from pericyte-like immortalized adipose-mesenchymal stem cells, were efficiently internalized by retinal microvascular endothelial cells and effectively counteracted high glucose-induced apoptosis. These exosomes also mitigated the rise in reactive oxygen species levels and suppressed the migratory and angiogenic properties of retinal microvascular endothelial cells, as demonstrated by Transwell and tube formation assays, respectively. Furthermore, they preserved endothelial barrier function, reducing hyperglycemia-induced permeability. At the molecular level, qRT-PCR analysis showed that exosome treatment modulated the expression of critical genes involved in angiogenesis (VEGF-A, ANG2, MMP9), inflammation (IL-1β, TNF-α), gap junction communication (CX43), and cytoskeletal regulation (ROCK1), with the most prominent effects seen with exosomes from pericyte-like immortalized adipose-mesenchymal stem cells. High glucose increased the expression of pro-angiogenic and pro-inflammatory markers, which were effectively normalized post-exosome treatment. In conclusion, this research highlights the reparative capacity of exosomes secreted by pericyte-like differentiated immortalized adipose-mesenchymal stem cells in reversing the detrimental effects of high glucose on retinal microvascular endothelial cells. By reducing apoptosis, oxidative stress, inflammation, and abnormal angiogenic behavior, these exosomes present a promising avenue for therapeutic intervention in early diabetic retinopathy. Future studies can focus on elucidating the precise molecular mechanisms and exploring their translational potential in vivo.

OctoVision: A Smart System for Diabetic Retinopathy Disease Detection

Diabetic retinopathy (DR) is the major cause of vision impairment and blindness in diabetics. Early detection and treatments are critical in preventing irreparable retinal damage. Manual detection of diabetic retinopathy by an ophthalmologist takes a long time, and patients must suffer greatly during this time. This paper presents an automated approach for rapid DR detection using the DenseNet-121 architecture. Our model achieves an accuracy exceeding 80%, with a precision score of 81% and a recall score of 86%, indicating its high effectiveness in detecting DR. Additionally, we developed a server-based implementation where the trained model is deployed. Images captured by a camera are uploaded to a cloud server, which processes them and sends back a diagnostic response. This study contributes to continuing efforts to create efficient and reliable techniques for early DR identification, resulting in earliermanagement and better patient outcomes.

Improved diabetic retinopathy severity classification using squeeze-and-excitation and sparse light weight multi-level attention u-net with transfer learning from xception.

Diabetic Retinopathy (DR) is a significant cause of vision loss in diabetic patients, making early detection and accurate severity classification essential for effective management and prevention. This study aims to develop an enhanced DR severity classification approach using advanced model architectures and transfer learning to improve diagnostic accuracy and support better patient care. We propose a novel model, Xception Squeeze-and-Excitation Sparse Lightweight Multi-Level Attention U-Net (XceSE_SparseLwMLA-UNet), designed to classify DR severity using fundus images from the Messidor 1 and Messidor 2 datasets. The XceSE_SparseLwMLA-UNet integrates several advanced mechanisms: the Squeeze-and-Excitation (SE) mechanism for adaptive feature recalibration, the Sparse Lightweight Multi-Level Attention (SparseLwMLA) mechanism for effective contextual information integration, and transfer learning from the Xception architecture to enhance feature extraction capabilities. The SE mechanism refines channel-wise feature responses, while SparseLwMLA enhances the model's ability to identify complex DR patterns. Transfer learning utilizes pre-trained weights from Xception to improve generalization across DR severity levels. The proposed XceSE_SparseLwMLA-UNet model demonstrates superior performance in DR severity classification, achieving higher accuracy and improved multi-class F1 scores compared to existing models. The model's color-coded segmentation outputs offer interpretable visual representations, aiding medical professionals in assessing DR severity levels. The XceSE_SparseLwMLA-UNet model shows promise for advancing early DR diagnosis and management by enhancing classification accuracy and providing valuable visual insights. Its integration of advanced architectural features and transfer learning contributes to better patient care and improved visual health outcomes.

Mesenchymal stem cells-derived exosomes alleviate senescence of retinal pigment epithelial cells by activating PI3K/AKT-Nrf2 signaling pathway in early diabetic retinopathy

Diabetic retinopathy (DR) is a major cause of vision loss and blindness in adults. Cellular senescence was involved in the pathogenesis of early-stage DR and is positively correlated with progression. Thus, our study aimed at exploring the effect and potential mechanism of Mesenchymal stem cells-derived exosomes (MSCs-EXOs) on Retinal Pigment Epithelial (RPE) cells senescence at an early stage of DR in vivo and in vitro. ARPE-19 cells were incubated in high glucose (HG) medium mixed with MSCs-EXOs to observe the changes in cell viability. Senescence-associated β-galactosidase (SA-β-gal) staining, Western blot and qRT-PCR were used to assess the expression of senescence-related genes and antioxidant mediators. Quantitative Real-Time polymerase chain reaction (qRT-PCR), Optical coherence tomography (OCT) Hematoxylin and eosin (HE) staining and Electroretinogram (ERG) were respectively used to verify cellular senescence, the structure and function of the retina. Our findings demonstrated that MSCs-EXOs inhibited HG-induced senescence in ARPE-19 cells. Furthermore, MSCs-EXOs reduced HG-induced cell apoptosis and oxidative stress levels while promoting cell proliferation. Mechanistically, HG suppressed PI3K/AKT phosphorylation as well as nuclear factor erythroid 2-related factor 2 (Nrf2) expression along with its downstream target gene expression in ARPE-19 cells. However, MSCs-EXOs reversed these changes by alleviating cellular senescence while enhancing antioxidant activity. In line with our results in vitro, MSCs-EXOs significantly ameliorated hyperglycemia-induced senescence in DR mice by downregulating mRNA expression of P53, P21, P16, and SASP. Additionally, MSCs-EXOs improved the functional and structural integrity of the retina in DR mice. Our study revealed the protective effect of MSCs-EXOs on cellular senescence, offering new insights for the treatment of DR.

Isolation of Mouse Retinal Capillaries and Subendothelial Matrix for Stiffness Measurement Using Atomic Force Microscopy.

Retinal capillary degeneration is a clinical hallmark of the early stages of diabetic retinopathy (DR). Our recent studies have revealed that diabetes-induced retinal capillary stiffening plays a crucial and previously unrecognized causal role in inflammation-mediated degeneration of retinal capillaries. The increase in retinal capillary stiffness results from the overexpression of lysyl oxidase, an enzyme that crosslinks and stiffens the subendothelial matrix. Since tackling DR at the early stage is expected to prevent or slow down DR progression and associated vision loss, subendothelial matrix, and capillary stiffness represent relevant and novel therapeutic targets for early DR management. Further, direct measurement of retinal capillary stiffness can serve as a crucial preclinical validation step for the development of new imaging techniques for non-invasive assessment of retinal capillary stiffness in animal and human subjects. With this view in mind, we here provide a detailed protocol for the isolation and stiffness measurement of mouse retinal capillaries and subendothelial matrix using atomic force microscopy.

Deciphering the connection between microvascular damage and neurodegeneration in early diabetic retinopathy.

Diabetic retinopathy (DR), a common diabetes complication leading to vision loss, presents early clinical signs linked to retinal vasculature damage, affecting the neural retina at advanced stages. However, vascular changes and potential effects on neural cells before clinical diagnosis of DR are less well understood. To study the earliest stages of DR we performed histological phenotyping and quantitative analysis on postmortem retinas from 10 donors with diabetes and without signs of DR (such as microaneurysms and haemorrhages), plus 3 controls and 1 DR case, focusing on capillary loss in the deeper (DVP) and superficial vascular plexuses (SVP) and neural retina effects. The advanced DR case exhibited profound vascular and neural damage, whereas the ten randomly selected donors with diabetes appeared superficially normal. The SVP was indistinguishable from the controls. In contrast, over half of the retinas from donors with diabetes showed capillary dropout in the DVP and increased capillary diameter. However, we could not detect any localised neural cell loss in the vicinity of dropout capillaries. Instead, we observed a subtle pan-retinal loss of inner nuclear layer (INL) cells in all diabetes cases (p<0.05), independent of microvascular damage. In conclusion, our findings demonstrate a novel histological biomarker for early-stage diabetes-related damage in human postmortem retina, common in people with diabetes before clinical DR diagnosis. Furthermore, the mismatch between capillary dropout and neural loss questions the notion of microvascular loss directly causing neurodegeneration at the earliest stages of DR, so diabetes may affect the two readouts independently.

Clinical serum lipidomic profiling revealed potential lipid biomarkers for early diabetic retinopathy

Diabetic retinopathy (DR) is a serious complication of diabetes featuring abnormal lipid metabolism. However, the specific lipid molecules associated with onset and progression remain unclear. We used a broad-targeted lipidomics approach to assess the lipid changes that occur before the proliferative retinopathy stage and to identify novel lipid biomarkers to distinguish between patients without DR (NDR) and with non-proliferative DR (NPDR). Targeted lipomics analysis was carried out on serum samples from patients with type I diabetes, including 20 NDRs and 20 NPDRs. The results showed that compared with the NDR group, 102 lipids in the NPDR group showed specific expressions. Four lipid metabolites including TAG58:2-FA18:1 were obtained using the Least Absolute Shrink And Selection Operator (LASSO) and Support Vector Machine Recursive Feature Elimination (SVM-RFE) methods. The four-lipid combination diagnostic models showed good predictive ability in both the discovery and validation sets, and were able to distinguish between NDR patients and NPDR patients. The identified lipid markers significantly improved diagnostic accuracy within the NPDR group. Our findings help to better understand the complexity and individual differences of DR lipid metabolism.

Enhanced diabetic retinopathy detection and classification using fundus images with ResNet50 and CLAHE-GAN

Diabetic retinopathy (DR), a progressive eye disorder, can lead to irreversible vision impairment ranging from no DR to severe DR, necessitating precise identification for early treatment. This study introduces an innovative deep learning (DL) approach, surpassing traditional methods in detecting DR stages. It evaluated two scenarios for training DL models on balanced datasets. The first employed image enhancement via contrast limited adaptive histogram equalization (CLAHE) and a generative adversarial network (GAN), while the second did not involve any image enhancement. Tested on the Asia pacific tele-ophthalmology society 2019 blindness detection (APTOS-2019 BD) dataset, the enhanced model (scenario 1) reached 98% accuracy and a 99% Cohen kappa score (CKS), with the non-enhanced model (scenario 2) achieving 95.4% accuracy and a 90.5% CKS. The combination of CLAHE and GAN, termed CLANG, significantly boosted the model's performance and generalizability. This advancement is pivotal for early DR detection and intervention, offering a new pathway to prevent irreversible vision loss in diabetic patients.

Advances in Structural and Functional Retinal Imaging and Biomarkers for Early Detection of Diabetic Retinopathy.

Diabetic retinopathy (DR), a vision-threatening microvascular complication of diabetes mellitus (DM), is a leading cause of blindness worldwide that requires early detection and intervention. However, diagnosing DR early remains challenging due to the subtle nature of initial pathological changes. This review explores developments in multimodal imaging and functional tests for early DR detection. Where conventional color fundus photography is limited in the field of view and resolution, advanced quantitative analysis of retinal vessel traits such as retinal microvascular caliber, tortuosity, and fractal dimension (FD) can provide additional prognostic value. Optical coherence tomography (OCT) has also emerged as a reliable structural imaging tool for assessing retinal and choroidal neurodegenerative changes, which show potential as early DR biomarkers. Optical coherence tomography angiography (OCTA) enables the evaluation of vascular perfusion and the contours of the foveal avascular zone (FAZ), providing valuable insights into early retinal and choroidal vascular changes. Functional tests, including multifocal electroretinography (mfERG), visual evoked potential (VEP), multifocal pupillographic objective perimetry (mfPOP), microperimetry, and contrast sensitivity (CS), offer complementary data on early functional deficits in DR. More importantly, combining structural and functional imaging data may facilitate earlier detection of DR and targeted management strategies based on disease progression. Artificial intelligence (AI) techniques show promise for automated lesion detection, risk stratification, and biomarker discovery from various imaging data. Additionally, hematological parameters, such as neutrophil-lymphocyte ratio (NLR) and neutrophil extracellular traps (NETs), may be useful in predicting DR risk and progression. Although current methods can detect early DR, there is still a need for further research and development of reliable, cost-effective methods for large-scale screening and monitoring of individuals with DM.