Automated Classification of Inherited Retinal Diseases in Optical Coherence Tomography Images Using Few-shot Learning

Objective To develop a few-shot learning(FSL) approach for classifying optical coherence tomography(OCT) images in patients with inherited retinal disorders(IRDs).Methods In this study, an FSL model based on a student–teacher learning framework was designed to classify images. 2,317 images from 189 participants were included. Of these, 1,126 images revealed IRDs, 533 were normal samples, and 658 were control samples.Results The FSL model achieved a total accuracy of 0.974–0.983, total sensitivity of 0.934–0.957, total specificity of 0.984–0.990, and total F1 score of 0.935–0.957, which were superior to the total accuracy of the baseline model of 0.943–0.954, total sensitivity of 0.866–0.886, total specificity of 0.962–0.971,and total F1 score of 0.859–0.885. The performance of most subclassifications also exhibited advantages. Moreover, the FSL model had a higher area under curves(AUC) of the receiver operating characteristic(ROC) curves in most subclassifications.Conclusion This study demonstrates the effective use of the FSL model for the classification of OCT images from patients with IRDs, normal, and control participants with a smaller volume of data. The general principle and similar network architectures can also be applied to other retinal diseases with a low prevalence.

Convolutional Neural Network-Based Classificationof Multiple Retinal Diseases Using Fundus Images

Use of deep learning algorithms for the investigation and analysis of medical images has emerged as a powerful technique.The increase in retinal dis-eases is alarming as it may lead to permanent blindness if left untreated.Automa-tion of the diagnosis process of retinal diseases not only assists ophthalmologists in correct decision-making but saves time also.Several researchers have worked on automated retinal disease classification but restricted either to hand-crafted fea-ture selection or binary classification.This paper presents a deep learning-based approach for the automated classification of multiple retinal diseases using fundus images.For this research,the data has been collected and combined from three distinct sources.The images are preprocessed for enhancing the details.Six layers of the convolutional neural network(CNN)are used for the automated feature extraction and classification of 20 retinal diseases.It is observed that the results are reliant on the number of classes.For binary classification(healthy vs.unhealthy),up to 100%accuracy has been achieved.When 16 classes are used(treating stages of a disease as a single class),93.3%accuracy,92%sensitivity and 93%specificity have been obtained respectively.For 20 classes(treating stages of the disease as separate classes),the accuracy,sensitivity and specificity have dropped to 92.4%,92%and 92%respectively.

Stem cell therapy in retinal diseases

Alteration of the outer retina leads to various diseases such as age-related macular degeneration or retinitis pigmentosa characterized by decreased visual acuity and ultimately blindness.Despite intensive research in the field of retinal disorders,there is currently no curative treatment.Several therapeutic approaches such as cell-based replacement and gene therapies are currently in development.In the context of cell-based therapies,different cell sources such as embryonic stem cells,induced pluripotent stem cells,or multipotent stem cells can be used for transplantation.In the vast majority of human clinical trials,retinal pigment epithelial cells and photoreceptors are the cell types considered for replacement cell therapies.In this review,we summarize the progress made in stem cell therapies ranging from the pre-clinical studies to clinical trials for retinal disease.

Retinal imaging in inherited retinal diseases

Inherited retinal diseases(IRD)are a leading cause of blindness in the working age population.The advances in ocular genetics,retinal imaging and molecular biology,have conspired to create the ideal environment for establishing treatments for IRD,with the first approved gene therapy and the commencement of multiple therapy trials.The scope of this review is to familiarize clinicians and scientists with the current landscape of retinal imaging in IRD.Herein we present in a comprehensive and concise manner the imaging findings of:(I)macular dystrophies(MD)[Stargardt disease(ABCA4),X-linked retinoschisis(RS1),Best disease(BEST1),pattern dystrophy(PRPH2),Sorsby fundus dystrophy(TIMP3),and autosomal dominant drusen(EFEMP1)],(II)cone and cone-rod dystrophies(GUCA1A,PRPH2,ABCA4 and RPGR),(III)cone dysfunction syndromes[achromatopsia(CNGA3,CNGB3,PDE6C,PDE6H,GNAT2,ATF6],blue-cone monochromatism(OPN1LW/OPN1MW array),oligocone trichromacy,bradyopsia(RGS9/R9AP)and Bornholm eye disease(OPN1LW/OPN1MW),(IV)Leber congenital amaurosis(GUCY2D,CEP290,CRB1,RDH12,RPE65,TULP1,AIPL1 and NMNAT1),(V)rod-cone dystrophies[retinitis pigmentosa,enhanced S-Cone syndrome(NR2E3),Bietti crystalline corneoretinal dystrophy(CYP4V2)],(VI)rod dysfunction syndromes(congenital stationary night blindness,fundus albipunctatus(RDH5),Oguchi disease(SAG,GRK1),and(VII)chorioretinal dystrophies[choroideremia(CHM),gyrate atrophy(OAT)].

Development of Photobiomodulation and its Application in Retinal Diseases

Photobiomodulation(PBM)is a therap-eutic approach that utilizes low-energy laser or light to regulate biological tissues.The mechanism is to promote cytochrome C oxidase(CCO)through low energy light,regulate the REDOX of mitochondria,and then regulate the biological functions of tissues and cells.Compared with traditional laser,it has a higher safety.There are a large number of mitochondria in retinal tissue,and studies have shown that PBM has a good protective and regulatory effect on the mitochondrial functions of retina and optic nerve.Therefore,PBM is clinically applied to treat age-related macular degeneration,diabetic retinopathy and retinitis pigmentosa and other retinal diseases.In order to provide a new direction for the treatment of retinal diseases,this paper reviewed the main parameters,mechanisms and the research progress of PBM in the fundus indications.

Pyrroloquinoline quinone:a potential neuroprotective compound for neurodegenerative diseases targeting metabolism

Pyrroloquinoline quinone is a quinone described as a cofactor for many bacterial dehydrogenases and is reported to exert an effect on metabolism in mammalian cells/tissues.Pyrroloquinoline quinone is present in the diet being available in foodstuffs,conferring the potential of this compound to be supplemented by dietary administration.Pyrroloquinoline quinone’s nutritional role in mammalian health is supported by the extensive deficits in reproduction,growth,and immunity resulting from the dietary absence of pyrroloquinoline quinone,and as such,pyrroloquinoline quinone has been considered as a“new vitamin.”Although the classification of pyrroloquinoline quinone as a vitamin needs to be properly established,the wide range of benefits for health provided has been reported in many studies.In this respect,pyrroloquinoline quinone seems to be particularly involved in regulating cell signaling pathways that promote metabolic and mitochondrial processes in many experimental contexts,thus dictating the rationale to consider pyrroloquinoline quinone as a vital compound for mammalian life.Through the regulation of different metabolic mechanisms,pyrroloquinoline quinone may improve clinical deficits where dysfunctional metabolism and mitochondrial activity contribute to induce cell damage and death.Pyrroloquinoline quinone has been demonstrated to have neuroprotective properties in different experimental models of neurodegeneration,although the link between pyrroloquinoline quinone-promoted metabolism and improved neuronal viability in some of such contexts is still to be fully elucidated.Here,we review the general properties of pyrroloquinoline quinone and its capacity to modulate metabolic and mitochondrial mechanisms in physiological contexts.In addition,we analyze the neuroprotective properties of pyrroloquinoline quinone in different neurodegenerative conditions and consider future perspectives for pyrroloquinoline quinone’s potential in health and disease.

N-acetylserotonin alleviates retinal ischemia-reperfusion injury via HMGB1/RAGE/NF-κB pathway in rats

AIM:To observe the effects of N-acetylserotonin(NAS)administration on retinal ischemia-reperfusion(RIR)injury in rats and explore the underlying mechanisms involving the high mobility group box 1(HMGB1)/receptor for advanced glycation end-products(RAGE)/nuclear factor-kappa B(NF-κB)signaling pathway.METHODS:A rat model of RIR was developed by increasing the pressure of the anterior chamber of the eye.Eighty male Sprague Dawley were randomly divided into five groups:sham group(n=8),RIR group(n=28),RIR+NAS group(n=28),RIR+FPS-ZM1 group(n=8)and RIR+NAS+FPS-ZM1 group(n=8).The therapeutic effects of NAS were examined by hematoxylin-eosin(H&E)staining,and retinal ganglion cells(RGCs)counting.The expression of interleukin 1 beta(IL-1β),HMGB1,RAGE,and nod-like receptor 3(NLRP3)proteins and the phosphorylation of nuclear factorkappa B(p-NF-κB)were analyzed by immunohistochemistry staining and Western blot analysis.The expression of HMGB1 protein was also detected by enzyme-linked immunosorbent assay(ELISA).RESULTS:H&E staining results showed that NAS significantly reduced retinal edema and increased the number of RGCs in RIR rats.With NAS therapy,the HMGB1 and RAGE expression decreased significantly,and the activation of the NF-κB/NLRP3 pathway was antagonized along with the inhibition of p-NF-κB and NLRP3 protein expression.Additionally,NAS exhibited an anti-inflammatory effect by reducing IL-1βexpression.The inhibitory of RAGE binding to HMGB1 by RAGE inhibitor FPS-ZM1 led to a significant decrease of p-NF-κB and NLRP3 expression,so as to the IL-1βexpression and retinal edema,accompanied by an increase of RGCs in RIR rats.CONCLUSION:NAS may exhibit a neuroprotective effect against RIR via the HMGB1/RAGE/NF-κB signaling pathway,which may be a useful therapeutic target for retinal disease.

Retinal layer segmentation using gradient feature calculation in OCT

Retinal diseases pose significant challenges to global healthcare systems,necessitating accurate and efficient diagnostic methods.Optical Coherence Tomography(OCT)has emerged as a valuable tool for diagnosing and monitoring retinal conditions due to its noncontact and noninvasive nature.This paper presents a novel retinal layering method based on OCT images,aimed at enhancing the accuracy of retinal lesion diagnosis.The method utilizes gradient analysis to effectively identify and segment retinal layers.By selecting a column of pixels as a segmentation line and utilizing gradient information from adjacent pixels,the method initiates and proceeds with the layering process.This approach addresses potential issues arising from partial layer overlapping,minimizing deviations in layer segmentation.Experimental results demonstrate the efficacy of the proposed method in accurately segmenting eight retinal boundaries,with an average absolute position deviation of 1.75 pixels.By providing accurate segmentation of retinal layers,this approach contributes to the early detection and management of ocular conditions,ultimately improving patient outcomes and reducing the global burden of vision-related ailments.

Accuracy improvement for classifying retinal OCT images by diseases using deep learning-based selective denoising approach

In ophthalmology,retinal optical coherence tomography(OCT)images with noticeable structural features help identify human eyes as healthy or diseased.The recently hot arti ficial intelligence(AI)realized this recognition process automatically.However,speckle noise in the original retinal OCT image reduces the accuracy of disease classi fication.This study presents a timesaving approach based on deep learning to improve classi fication accuracy by removing the noise from the original dataset.Firstly,four pre-trained convolutional neural networks(CNNs)from the ImageNet Large Scale Visual Recognition Challenge(ILSVRC)were trained to classify the original images into two categories:The noise reduction required(NRR)and the noise-free(NF)images.Among the CNNs,VGG19 BN performed best with 98%accuracy and 99%recall.Then,we used the block-matching and 3D filtering(BM3D)algorithm to denoise the NRR images.Those noise-removed NRR and the NF images form the processed dataset.The quality of images in the dataset is prominently ameliorated after denoising,which is valid to improve the models'performance.The original and processed datasets were tested on the four pre-trained CNNs to evaluate the effectiveness of our proposed approach.We have compared the CNNs,and the results show the performance of the CNNs trained with the processed dataset is improved by an average of 2.04%,5.19%,and 5.10%under overall accuracy(OA),Macro F1-score,and Micro F1-score,respectively.Especially for DenseNet161,the OA is improved to 98.14%.Our proposed method demonstrates its effectiveness in improving classi fication accuracy and opens a new solution to reduce denoising time-consuming for large datasets.

Complications of Ziv-Aflibercept in Choroidal and Retinal Vascular Diseases

Introduction: The modern ophthalmology trends are changing rapidly every day with the introduction of much newer studies and research. Numerous anti-vascular endothelial growth factors (VEGF) are utilized as the mainstay in the treatment of intraocular vascular pathologies. The rationale of this study is to add to the literature regarding the safety and efficacy profile of the ziv-aflibercept as there is insubstantial data in patients with intraocular vascular pathologies being treated with this injection with prime focus on the complications of the injection. Materials and Methods: A prospective observational study was conducted at Opthalmology Department, Lahore General Hospital, Lahore between 14 August 2018 and 23 December 2019. Patients with choroidal and retinal vascular diseases like diabetic macular edema (DME), age-related macular degeneration (AMD) and retinal vein occlusion (RVO) who had no active infection of eye and had no history of myocardial infarction or cerebrovascular accident were added in this study. Results: Best-corrected visual acuity was significantly improved at 4, 8, and 12 weeks as compared to the baseline (p Conclusion: The use of ziv-aflibercept injection via intravitreal route under aseptic conditions for choroidal and retinal vascular diseases is effective as well as safe with mild and treatable ocular side effects.

Artificial intelligence in individualized retinal disease management

Owing to the rapid development of modern computer technologies,artificial intelligence(AI)has emerged as an essential instrument for intelligent analysis across a range of fields.AI has been proven to be highly effective in ophthalmology,where it is frequently used for identifying,diagnosing,and typing retinal diseases.An increasing number of researchers have begun to comprehensively map patients’retinal diseases using AI,which has made individualized clinical prediction and treatment possible.These include prognostic improvement,risk prediction,progression assessment,and interventional therapies for retinal diseases.Researchers have used a range of input data methods to increase the accuracy and dependability of the results,including the use of tabular,textual,or image-based input data.They also combined the analyses of multiple types of input data.To give ophthalmologists access to precise,individualized,and high-quality treatment strategies that will further optimize treatment outcomes,this review summarizes the latest findings in AI research related to the prediction and guidance of clinical diagnosis and treatment of retinal diseases.

Indirubin alleviates retinal neurodegeneration through the regulation of PI3K/AKT signaling

Retinal neurodegenerative disease is a leading cause of blindness among the elderly in developed countries,including glaucoma,diabetic retinopathy,traumatic optic neuropathy and optic neuritis,etc.The current clinical treatment is not very effective.We investigated indirubin,one of the main bioactive components of the traditional Chinese medicine Danggui Longhui Pill,in the present study for its role in retinal neurodegeneration.Indirubin exhibited no detectable tissue toxicity in vivo or cytotoxicity in vitro.Moreover,indirubin improved visual function and ameliorated retinal neurodegeneration in mice after optic nerve crush injury in vivo.Furthermore,indirubin reduced the apoptosis of retinal ganglion cells induced by oxidative stress in vitro.In addition,indirubin significantly suppressed the increased production of intracellular reactive oxygen species and the decreased activity of superoxide dismutase induced by oxidative stress.Mechanically,indirubin played a neuroprotective role by regulating the PI3K/AKT/BAD/BCL-2 signaling.In conclusion,indirubin protected retinal ganglion cells from oxidative damage and alleviated retinal neurodegeneration induced by optic nerve crush injury.The present study provides a potential therapeutic medicine for retinal neurodegenerative diseases.

Intravitreal stem cell paracrine properties as a potential neuroprotective therapy for retinal photoreceptor neurodegenerative diseases

Retinal degenerations are the leading causes of irreversible visual loss worldwide. Many pathologies included under this umbrella involve progressive degeneration and ultimate loss of the photoreceptor cells, with age-related macular degeneration and inherited and ischemic retinal diseases the most relevant. These diseases greatly impact patients' daily lives, with accompanying marked social and economic consequences. However, the currently available treatments only delay the onset or slow progression of visual impairment, and there are no cures for these photoreceptor diseases. Therefore, new therapeutic strategies are being investigated, such as gene therapy, optogenetics, cell replacement, or cell-based neuroprotection. Specifically, stem cells can secrete neurotrophic, immunomodulatory, and anti-angiogenic factors that potentially protect and preserve retinal cells from neurodegeneration. Further, neuroprotection can be used in different types of retinal degenerative diseases and at different disease stages, unlike other potential therapies. This review summarizes stem cell-based paracrine neuroprotective strategies for photoreceptor degeneration, which are under study in clinical trials, and the latest preclinical studies. Effective retinal neuroprotection could be the next frontier in photoreceptor diseases, and the development of novel neuroprotective strategies will address the unmet therapeutic needs.

Long-term fundus fluorescence angiography and real-time diagnosis of retinal diseases in non-human primate-animal models

Fluorescein angiography(FA)is a standard imaging modality for evaluating vascular abnormalities in retina-related diseases,which is recognized as the major cause of vision loss.Long-term and real-time fundus angiography is of great importance in preclinical research,nevertheless remaining big challenges up to present.In this study,we demonstrate that long-term fluorescence imaging of retinal vessels is enabled through a kind of fluorescent nanoagents,which is made of small-sized(hydrodynamic diameter:∼3 nm)silicon nanoparticles(SiNPs)featuring strong fluorescence,robust photostability,lengthened blood residency and negligible toxicity.In particular,the presented SiNPs-based nanoagents are capable of imaging retinal capillaries in∼10 min,which is around 10-fold longer than that(∼1 min)of fluorescein sodium(FS,known as the most widely used contrast agents for FA in clinic).Taking cynomolgus macaques as non-human primate-animal model,we further demonstrate the feasibility of real-time diagnosis of retinal diseases(e.g.,age-related macular degeneration(AMD))through dynamic monitoring of vascular dysfunction.

Human retinal secretome:A cross-link between mesenchymal and retinal cells

In recent years,mesenchymal stem cells(MSC)have been considered the most effective source for regenerative medicine,especially due to released soluble paracrine bioactive components and extracellular vesicles.These factors,collectively called the secretome,play crucial roles in immunomodulation and in improving survival and regeneration capabilities of injured tissue.Recently,there has been a growing interest in the secretome released by retinal cytotypes,especially retinal pigment epithelium and Müller glia cells.The latter trophic factors represent the key to preserving morphofunctional integrity of the retina,regulating biological pathways involved in survival,function and responding to injury.Furthermore,these factors can play a pivotal role in onset and progression of retinal diseases after damage of cell secretory function.In this review,we delineated the importance of cross-talk between MSCs and retinal cells,focusing on common/induced secreted factors,during experimental therapy for retinal diseases.The cross-link between the MSC and retinal cell secretomes suggests that the MSC secretome can modulate the retinal cell secretome and vice versa.For example,the MSC secretome can protect retinal cells from degeneration by reducing oxidative stress,autophagy and programmed cell death.Conversely,the retinal cell secretome can influence the MSC secretome by inducing changes in MSC gene expression and phenotype.

Blood-retinal barrier as a converging pivot in understanding the initiation and development of retinal diseases

Clinical ophthalmologists consider each retinal disease as a completely unique entity.However,various retinal diseases,such as uveitis,age-related macular degeneration,diabetic retinopathy,and primary open-angle glaucoma,share a number of common pathogenetic pathways.Whether a retinal disease initiates from direct injury to the blood-retinal barrier(BRB)or a defect/injury to retinal neurons or glia that impairs the BRB secondarily,the BRB is a pivotal point in determining the prognosis as self-limiting and recovering,or developing and progressing to a clinical phenotype.The present review summarizes our current knowledge on the physiology and cellular and molecular pathology of the BRB,which underlies its pivotal role in the initiation and development of common retinal diseases.

The role of Toll-like receptors in retinal ischemic diseases

Toll-like receptors（TLRs） are commonly referred to a series of evolutionary conserved receptors which recognize and respond to various microbes and endogenous ligands.Growing evidence has demonstrated that the expression of TLRs in the retina is regulated during retinal ischemic diseases,including ischemia-reperfusion injury,glaucoma,diabetic retinopathy（DR） and retinopathy of prematurity（ROP）.TLRs can be expressed in multiple cells in the retina,such as glial cells,retinal pigment epithelium（RPE）,as well as photoreceptor cells and endothelium cells.Activation of TLRs in retina could initiate a complex signal transduction cascade,induce the production of inflammatory cytokines and regulate the level of costimulatory molecules,which play prominent roles in the pathogenesis of retinal ischemic diseases.In this review,we summarized current studies about the relationship between TLRs and ischemic retinopathy.A greater understanding of the effect of TLRs on ischemic injuries may contribute to the development of specific TLR targeted therapeutic strategies in these conditions.

Game Theory-Based Dynamic Weighted Ensemble for Retinal Disease Classification

An automated retinal disease detection system has long been in exis-tence and it provides a safe,no-contact and cost-effective solution for detecting this disease.This paper presents a game theory-based dynamic weighted ensem-ble of a feature extraction-based machine learning model and a deep transfer learning model for automatic retinal disease detection.The feature extraction-based machine learning model uses Gaussian kernel-based fuzzy rough sets for reduction of features,and XGBoost classifier for the classification.The transfer learning model uses VGG16 or ResNet50 or Inception-ResNet-v2.A novel ensemble classifier based on the game theory approach is proposed for the fusion of the outputs of the transfer learning model and the XGBoost classifier model.The ensemble approach significantly improves the accuracy of retinal disease pre-diction and results in an excellent performance when compared to the individual deep learning and feature-based models.

Lycium barbarum glycopeptide(wolfberry extract)slows N-methyl-N-nitrosourea-induced degradation of photoreceptors

Photoreceptor cell degeneration leads to blindness, for which there is currently no effective treatment. Our previous studies have shown that Lycium barbarum(L. barbarum) polysaccharide(LBP) protects degenerated photoreceptors in rd1, a transgenic mouse model of retinitis pigmentosa. L. barbarum glycopeptide(Lb GP) is an immunoreactive glycoprotein extracted from LBP. In this study, we investigated the potential protective effect of Lb GP on a chemically induced photoreceptor-degenerative mouse model. Wild-type mice received the following: oral administration of Lb GP as a protective pre-treatment on days 1–7;intraperitoneal administration of 40 mg/kg N-methylN-nitrosourea to induce photoreceptor injury on day 7;and continuation of orally administered Lb GP on days 8–14. Treatment with Lb GP increased photoreceptor survival and improved the structure of photoreceptors, retinal photoresponse, and visual behaviors of mice with photoreceptor degeneration. Lb GP was also found to partially inhibit the activation of microglia in N-methyl-N-nitrosourea-injured retinas and significantly decreased the expression of two pro-inflammatory cytokines. In conclusion, Lb GP effectively slowed the rate of photoreceptor degeneration in N-methyl-N-nitrosourea-injured mice, possibly through an anti-inflammatory mechanism, and has potential as a candidate drug for the clinical treatment of photoreceptor degeneration.

Association between central serous chorioretinopathy and Helicobacter pylori infection: a systematic review and Meta-analysis

AIM:To investigate the association between central serous chorioretinopathy(CSC)and Helicobacter pylori(Hp)by summarizing all available evidence.METHODS:The Scopus,Embase,EBSCO,PubMed,Web of Science,and Cochrane Library databases for all relevant studies published from inception to October 2022 were searched,and manually searched for relevant reference lists as a supplement.Studies investigating the association between CSC and Hp infection were included.Finally,8 case-control studies were included in the Meta-analysis after study selection.RESULTS:The results showed no significant correlation between Hp infection and CSC[odds ratio(OR)1.89,95%confidential interval(CI)0.58–6.15,I2=96%,P=0.29].After subgroup analysis based on the degree of development of the study(developing/developed countries),it was found that the results of the two subgroups were the same as the whole,and no significant difference between the two subgroups existed.Meta-regression showed that the effect of sample size on heterogeneity among studies was more prominent(P<0.01,adjusted R^(2)=89.72%),which can explain 89.72%of the sources of heterogeneity.CONCLUSION:This Meta-analysis reveals no significant correlation between Hp infection and CSC,which still warrants further well-designed extensive sample studies to reach a more reliable conclusion and promote a better understanding of the treatment of CSC.