In Vivo Imaging of Rodent Retina in Retinal Disease.

High-quality imaging of the retina is crucial to the diagnosis and monitoring of disease, as well as for evaluating the success of therapeutics in human patients and in preclinical animal models. Here, we describe the basic principles and methods for in vivo retinal imaging in rodents, including fundus imaging, fluorescein angiography, optical coherence tomography, fundus autofluorescence, and infrared imaging. After providing a concise overview of each method and detailing the retinal diseases and conditions that can be visualized through them, we will proceed to discuss the advantages and disadvantages of each approach. These protocols will facilitate the acquisition of optimal images for subsequent quantification and analysis. Additionally, a brief explanation will be given regarding the potential results and the clinical significance of the detected abnormalities.

Role of fundus autofluorescence imaging in the management of submacular hemorrhage.

PURPOSE: To evaluate the baseline characteristics of fundus autofluorescence (FAF) in patients with submacular hemorrhage (SMH). METHODS: This retrospective study included patients diagnosed with treatment-naive, foveal-involving subretinal hemorrhage (size > 2-disc diameters) of any etiology, presenting between June 2017 and June 2023. Only cases with good-quality color fundus photographs, optical coherence tomography (OCT) scans, and blue-light FAF images at baseline were included. SMH imaging characteristics were documented and correlated with treatment outcomes. A successful treatment outcome was defined as the reduction, displacement or clearance of the SMH from beneath the fovea. RESULTS: Nineteen cases of SMH (13 males, 6 females), ranging from 14 to 85 years, were analyzed. Neovascular age-related macular degeneration (nAMD) was the most common etiology (n = 11, 58%). Baseline visual acuity ranged from 6/9 to counting fingers at ½ meter, with a median presentation time of 7 days from symptom onset (range: 1-57 days). Treatment success was observed in 13 eyes (68%). Hypoautofluoroscence on FAF was significantly associated with SMH resolution (p = 0.021). However, no association was found between treatment success and clinical hemorrhage characteristics (p = 0.222), OCT findings (p = 0.222), or specific treatments (p > 0.05). Hypoautofluoroscence on FAF was the sole predictor of treatment success, as demonstrated by Spearman's correlation (r = 0.637; p = 0.003) and linear regression analysis (p = 0.003). CONCLUSION: FAF, in conjunction with color fundus photography and OCT, may provide valuable insights for clinicians in formulating treatment strategies for patients with SMH. Hypoautofluoroscence on FAF was a significant predictor of successful SMH resolution in this study.

Oxyntic Gland Neoplasms - From Adenoma to Advanced Gastric Cancer: A Review of 29 Cases.

PURPOSE: Oxyntic gland neoplasm (OGN) is a rare condition that can be classified as oxyntic gland adenoma (OGA) or gastric adenocarcinoma of fundic-gland type (GA-FG). GA-FG primarily presents as early gastric cancer, with only a few reported cases of advanced gastric cancer (AGC). We aimed to investigate the clinicopathological features of OGN and describe an aggressive variant. MATERIALS AND METHODS: We investigated a total of 29 cases, including a patient with double primary cases, diagnosed with OGN or differentiated-type adenocarcinoma with GA-FG morphology, between November 2016 and April 2022. We analyzed 54 pathological specimens and reviewed their clinicopathological, endoscopic, and histological features. The lesions were reclassified as OGA or GA-FG, and immunohistochemical (IHC) staining for MUC-5AC and MUC-6 was performed on available resected GA-FG cases. RESULTS: The median patient age was 65 years and males accounted for 58.6% of patients. Most cases occurred in the body horizontally (69.0%) and on the greater curvature side cross-sectionally (48.3%). Endoscopically, type 0-IIa (41.4%) and a subepithelial tumor-like appearance (24.1%) were the most common findings. Histologically, there were 8 cases of OGA (27.6%) and 21 cases of GA-FG (72.4%). In GA-FG, MUC-6 was positive in 13 cases (81.3%), whereas MUC-5AC was positive in 8 cases (50.0%). Three cases presented as AGCs. CONCLUSIONS: Although OGNs are generally considered low-grade, they can present as AGCs and may exhibit features of lymphovascular or perineural invasion. Recognizing the clinicopathological features and accurately diagnosing OGN are important for providing adequate treatment.

Breaking the barriers: Methodology of implementation of a non-mydriatic ocular fundus camera in an emergency department.

Despite evidence that non-mydriatic fundus cameras are beneficial in non-ophthalmic settings, they are only available in a minority of hospitals in the US. The lag from research-based evidence to change in clinical practice highlights the complexities of implementation of new technology and practice. We describe the steps used to implement successfully a non-mydriatic ocular fundus camera combined with optical coherence tomography (OCT) in a general emergency department (ED) using Kotter's 8-Step Change Model. We prospectively collected the number of trained personnel in the ED, the number of imaging studies obtained each week during the first year following implementation, and we documented major achievements each month, as well as outcome measures, barriers to implementation and possible solutions. Between 12 and 42 patients were imaged per week, resulting in a total of 1274 patients imaged demonstrating sustained usage of non-mydriatic fundus camera/OCT in the ED one year after implementation. The implementation process was contingent upon multidisciplinary collaboration, extensive communication, coordinated training of staff, and continuous motivation. The future will likely include the use of artificial intelligence deep learning systems for automated interpretation of ocular imaging as an immediate diagnostic aid for ED or other non-eye care providers.

Human iPSC-based disease modeling studies identify a common mechanistic defect and potential therapies for AMD and related macular dystrophies.

Age-related macular degeneration (AMD) and related macular dystrophies (MDs) primarily affect the retinal pigment epithelium (RPE) in the eye. A hallmark of AMD/MDs that drives later-stage pathologies is drusen. Drusen are sub-RPE lipid-protein-rich extracellular deposits, but how drusen forms and accumulates is not known. We utilized human induced pluripotent stem cell (iPSC)-derived RPE from patients with AMD and three distinct MDs to demonstrate that reduced activity of RPE-secreted matrix metalloproteinase 2 (MMP2) contributes to drusen in multiple maculopathies in a genotype-agnostic manner by instigating sterile inflammation and impaired lipid homeostasis via damage-associated molecular pattern molecule (DAMP)-mediated activation of receptor for advanced glycation end-products (RAGE) and increased secretory phospholipase 2-IIA (sPLA2-IIA) levels. Therapeutically, RPE-specific MMP2 supplementation, RAGE-antagonistic peptide, and a small molecule inhibitor of sPLA2-IIA ameliorated drusen accumulation in AMD/MD iPSC-RPE. Ultimately, this study defines a causal role of the MMP2-DAMP-RAGE-sPLA2-IIA axis in AMD/MDs.

Retinal nerve fiber layer defects and chronic kidney disease: the Kailuan Eye Study.

AIM: To investigate whether retinal nerve fiber layer defects (RNFLDs) is a potential risk factor for chronic kidney disease (CKD) in Chinese adults. METHODS: The Kailuan Eye Study was a population-based study that included 14 440 participants. All participants underwent detailed assessments, RNFLDs were diagnosed using color fundus photographs. RESULTS: Overall, 12 507 participants [8533 males (68.23%)] had complete systemic examination data and at least one evaluable fundus photograph. RNFLDs were found in 621 participants [5.0%; 95% confidence interval (CI): 4.6%-5.34%], and 70 cases of multiple RNFLDs were found (11.27%). After adjusting multiple factors, RNFLDs was significantly associated with CKD severity, the ORs of CKD stage 3, stage 4 and stage 5 were 1.698, 4.167, and 9.512, respectively. Multiple RNFLDs were also associated with CKD severity after adjusting multiple factors, the ORs of CKD stage 3 and stage 5 were 4.465 and 11.833 respectively. Furthermore, 2294 participants had CKD (18.34%, 95%CI: 17.68%-18.99%). After adjusting for other factors, CKD presence was significantly correlated with the presence of RNFLDs. CONCLUSION: The strongest risk factors for RNFLDs are CKD and hypertension. Conversely, RNFLDs can be an ocular feature in patients with CKD. Fundoscopy can help detect systemic diseases, and assessment for RNFLDs should be considered in CKD patients.

Artificial Intelligence (AI)-Enhanced Detection of Diabetic Retinopathy From Fundus Images: The Current Landscape and Future Directions.

Diabetic retinopathy (DR) remains a leading cause of vision loss worldwide, with early detection critical for preventing irreversible damage. This review explores the current landscape and future directions of artificial intelligence (AI)-enhanced detection of DR from fundus images. Recent advances in deep learning and computer vision have enabled AI systems to analyze retinal images with expert-level accuracy, potentially transforming DR screening. Key developments include convolutional neural networks achieving high sensitivity and specificity in detecting referable DR, multi-task learning approaches that can simultaneously detect and grade DR severity, and lightweight models enabling deployment on mobile devices. While these AI systems show promise in improving the efficiency and accessibility of DR screening, several challenges remain. These include ensuring generalizability across diverse populations, standardizing image acquisition and quality, addressing the "black box" nature of complex models, and integrating AI seamlessly into clinical workflows. Future directions in the field encompass explainable AI to enhance transparency, federated learning to leverage decentralized datasets, and the integration of AI with electronic health records and other diagnostic modalities. There is also growing potential for AI to contribute to personalized treatment planning and predictive analytics for disease progression. As the technology continues to evolve, maintaining a focus on rigorous clinical validation, ethical considerations, and real-world implementation will be crucial for realizing the full potential of AI-enhanced DR detection in improving global eye health outcomes.

A deep-learning pipeline for the diagnosis and grading of common blinding ophthalmic diseases based on lesion-focused classification model.

Background: Glaucoma (GLAU), Age-related Macular Degeneration (AMD), Retinal Vein Occlusion (RVO), and Diabetic Retinopathy (DR) are common blinding ophthalmic diseases worldwide. Purpose: This approach is expected to enhance the early detection and treatment of common blinding ophthalmic diseases, contributing to the reduction of individual and economic burdens associated with these conditions. Methods: We propose an effective deep-learning pipeline that combine both segmentation model and classification model for diagnosis and grading of four common blinding ophthalmic diseases and normal retinal fundus. Results: In total, 102,786 fundus images of 75,682 individuals were used for training validation and external validation purposes. We test our model on internal validation data set, the micro Area Under the Receiver Operating Characteristic curve (AUROC) of which reached 0.995. Then, we fine-tuned the diagnosis model to classify each of the four disease into early and late stage, respectively, which achieved AUROCs of 0.597 (GL), 0.877 (AMD), 0.972 (RVO), and 0.961 (DR) respectively. To test the generalization of our model, we conducted two external validation experiments on Neimeng and Guangxi cohort, all of which maintained high accuracy. Conclusion: Our algorithm demonstrates accurate artificial intelligence diagnosis pipeline for common blinding ophthalmic diseases based on Lesion-Focused fundus that overcomes the low-accuracy of the traditional classification method that based on raw retinal images, which has good generalization ability on diverse cases in different regions.

Minocycline-induced retinal pigment epithelium hyperpigmentation masquerading as age-related macular degeneration: Case presentation and proposed mechanism.

Purpose: We describe the case of an 80-year-old man with bilateral minocycline-induced retinal pigment epithelium (RPE) hyperpigmentation, which initially masqueraded as AMD. Secondarily, using multimodal imaging features, we propose a mechanism for the development of minocycline-induced RPE hyperpigmentation. Observations: The patient was referred with concern for AMD given the presence of macular drusenoid deposits on optical coherence tomography. However, funduscopic evaluation showed dense granular parafoveal hyperpigmentation, with a diffuse slate-colored hyperpigmentation throughout the peripheral fundus. Short-wavelength fundus autofluorescence of the macula disclosed no irregularities (as would be expected with drusen) while on near-infrared reflectance (NIR) imaging, numerous hyperreflective foci were noted corresponding to the hyperpigmented granules observed clinically (as would instead be seen with melanin deposits). Clinical examination was notable for blue-gray hyperpigmentation of the lower and upper extremities, as well as of the face, periorbital skin, and sclera. Upon further questioning, the patient disclosed daily oral minocycline use for 15 years for acne rosacea, confirming a diagnosis of minocycline-induced hyperpigmentation of the RPE. Conclusions: Multimodal imaging can be useful for differentiating minocycline-induced RPE hyperpigmentation from similar masquerade entities. Timely diagnosis can prevent progressive vision loss.

Multimodal optical imaging of iris flocculi in three consecutive generations: a case report.

Background: Multiple pigmented epithelial cysts at the edge of pupils, that is, iris flocculi, in both eyes, are rare ocular diseases. It has been demonstrated that this disease can be attributed to mutations in the smooth muscle α-actin 2 (ACTA2) gene, which mainly affects the function of smooth muscle cells (SMCs). SMCs are components of the iris, aorta, and several other systemic organs. In addition, iris flocculi are strongly correlated with familial thoracic aortic aneurysm and dissection (TAAD), which is caused by the mutation of amino acid 149 in the ACTA2 gene. Case description: A 6-month-old Chinese boy was found to have iris flocculi during ocular fundus screening for premature infants. His mother, a 30-year-old Chinese woman with a history of aortic dissection, underwent an ophthalmic examination and was found to have iris flocculi. Whole exome sequencing revealed a heterozygous c.445C > T (p. Arg149Cys) mutation in ACTA2 in both the boy and his mother. After his family history was traced, the boy's grandfather was diagnosed with similar iris flocculi. Due to the absence of any ocular complications caused by iris flocculi in the cases, no special treatment was given, and regular follow-up was recommended. Conclusion: We reported one case of familial iris flocculi caused by a heterozygous missense mutation in ACTA2 (p. Arg149Cys) and presented multimodal optical images of both the iris and fundus in three consecutive generations. This case report enriched the clinical features of retinal vasculature and macula associated with the mutation in the amino acid 149 of the ACTA2 gene.

Multiple-ResNet GAN: An enhanced high-resolution image generation method for translation from fundus structure image to fluorescein angiography.

Fluorescein angiography (FA) is a diagnostic method for observing the vascular circulation in the eye. However, it poses a risk to patients. Therefore, generative adversarial networks have been used to convert retinal fundus structure images into FA images. Existing high-resolution image generation methods employ complex deep network models that are challenging to optimize, which leads to issues such as blurred lesion boundaries and poor capture of microleakage and microvessels. In this study, we propose a multiple-ResNet generative adversarial network (GAN) to improve model training, thereby enhancing the ability to generate high-resolution FA images. First, the structure of the multiple-ResNet generator is designed to enhance detail generation in high-resolution images. Second, the Gaussian error linear unit (GELU) activation function is used to help the model converge rapidly. The effectiveness of the multiple-ResNet is verified using the publicly available Isfahan MISP dataset. Experimental results show that our method outperforms other methods, achieving better quantitative results with a mean structural similarity of 0.641, peak signal-to-noise ratio of 18.25, and learned perceptual image patch similarity of 0.272. Compared with state-of-the-art methods, the results showed that using the multiple-ResNet framework and GELU activation function can improve the generation of detailed regions in high-resolution FA images.

Use of a Fundus Image-Based Titration Strategy for Selective Retina Therapy for Central Serous Chorioretinopathy.

Background/Objectives: This study evaluated the clinical outcomes of selective retina therapy (SRT) for treating central serous chorioretinopathy. A fundus image-based titration method was used for laser irradiation. Methods: This retrospective cohort study included 29 eyes (29 patients) that underwent SRT for CSC. Both the pulse energy and number of micropulses were adjusted according to the fundus image. Mean best-corrected visual acuity (BCVA), central foveal thickness (CFT), and subretinal fluid (SRF) height were measured 1, 2, 3, 4, and 6 months after SRT. Mean deviation (MD) was measured using microperimetry at 3 and 6 months post-treatment. Results: At 6 months after SRT treatment, SRF was completely resolved in 89.7% of cases (26/29 eyes). The mean Snellen BCVA significantly improved from 0.34 ± 0.31 logMAR (logarithm of the minimum angle of resolution) (20/40) at baseline to 0.24 ± 0.24 logMAR (20/32) at 6 months (p = 0.009). The 0.1 improvement in mean BCVA is equivalent to a 5-letter gain on the ETDRS chart. The mean CFT decreased significantly from 309.31 ± 81.6 µm at baseline to 211.07 ± 50.21 µm at 6 months (p < 0.001). The mean SRF height also decreased significantly from 138.36 ± 56.78 µm at baseline to 23.75 ± 61.19 µm at 6 months (p < 0.001). The mean MD was improved from -1.56 ± 1.47 dB at baseline to -1.03 ± 2.43 dB at 6 months (p = 0.07) after treatment. Conclusions: SRT using fundus image-based titration can yield favorable functional and anatomical outcomes in the treatment of CSC.

Multi-label deep learning for comprehensive optic nerve head segmentation through data of fundus images.

Early diagnosis and continuous monitoring of patients with eye diseases are critical in computer-aided detection (CAD) techniques. Semantic segmentation, a key component in computer vision, enables pixel-level classification and provides detailed information about objects within images. In this study, we present three U-Net models designed for multi-class semantic segmentation, leveraging the U-Net architecture with transfer learning. To generate ground truth for the HRF dataset, we combine two U-Net models, namely MSU-Net and BU-Net, to predict probability maps for the optic disc and cup regions. Binary masks are then derived from these probability maps to extract the optic disc and cup regions from retinal images. The dataset used in this study includes pre-existing blood vessels and manually annotated peripapillary atrophy zones (alpha and beta) provided by expert ophthalmologists. This comprehensive dataset, integrating existing blood vessels and expert-marked peripapillary atrophy zones, fulfills the study's objectives. The effectiveness of the proposed approach is validated by training nine pre-trained models on the HRF dataset comprising 45 retinal images, successfully segmenting the optic disc, cup, blood vessels, and peripapillary atrophy zones (alpha and beta). The results demonstrate 87.7 % pixel accuracy, 87 % Intersection over Union (IoU), 86.9 % F1 Score, 85 % mean IoU (mIoU), and 15 % model loss, significantly contributing to the early diagnosis and monitoring of glaucoma and optic nerve disorders.

Detection and diagnosis of diabetic eye diseases using two phase transfer learning approach.

Background: Early diagnosis and treatment of diabetic eye disease (DED) improve prognosis and lessen the possibility of permanent vision loss. Screening of retinal fundus images is a significant process widely employed for diagnosing patients with DED or other eye problems. However, considerable time and effort are required to detect these images manually. Methods: Deep learning approaches in machine learning have attained superior performance for the binary classification of healthy and pathological retinal fundus images. In contrast, multi-class retinal eye disease classification is still a difficult task. Therefore, a two-phase transfer learning approach is developed in this research for automated classification and segmentation of multi-class DED pathologies. Results: In the first step, a Modified ResNet-50 model pre-trained on the ImageNet dataset was transferred and learned to classify normal diabetic macular edema (DME), diabetic retinopathy, glaucoma, and cataracts. In the second step, the defective region of multiple eye diseases is segmented using the transfer learning-based DenseUNet model. From the publicly accessible dataset, the suggested model is assessed using several retinal fundus images. Our proposed model for multi-class classification achieves a maximum specificity of 99.73%, a sensitivity of 99.54%, and an accuracy of 99.67%.

Diabetic Retinopathy Features Segmentation without Coding Experience with Computer Vision Models YOLOv8 and YOLOv9.

Computer vision is a powerful tool in medical image analysis, supporting the early detection and classification of eye diseases. Diabetic retinopathy (DR), a severe eye disease secondary to diabetes, accompanies several early signs of eye-threatening conditions, such as microaneurysms (MAs), hemorrhages (HEMOs), and exudates (EXs), which have been widely studied and targeted as objects to be detected by computer vision models. In this work, we tested the performances of the state-of-the-art YOLOv8 and YOLOv9 architectures on DR fundus features segmentation without coding experience or a programming background. We took one hundred DR images from the public MESSIDOR database, manually labelled and prepared them for pixel segmentation, and tested the detection abilities of different model variants. We increased the diversity of the training sample by data augmentation, including tiling, flipping, and rotating the fundus images. The proposed approaches reached an acceptable mean average precision (mAP) in detecting DR lesions such as MA, HEMO, and EX, as well as a hallmark of the posterior pole of the eye, such as the optic disc. We compared our results with related works in the literature involving different neural networks. Our results are promising, but far from being ready for implementation into clinical practice. Accurate lesion detection is mandatory to ensure early and correct diagnoses. Future works will investigate lesion detection further, especially MA segmentation, with improved extraction techniques, image pre-processing, and standardized datasets.

[Prevalence of diabetic retinopathy assessed using two-field mydriatic fundus photography]. / Otsenka rasprostranennosti diabeticheskoi retinopatii s pomoshch'yu dvukhpol'nogo midriaticheskogo fundus-fotografirovaniya.

Early detection of diabetic retinopathy (DR) is an urgent ophthalmological problem in Russia and globally. PURPOSE: This study assesses the prevalence of asymptomatic retinopathy and attempts to identify risk groups for its development in patients with type 1 and 2 diabetes mellitus (T1DM and T2DM). MATERIAL AND METHODS: The study involved clinics from 5 cities in the Russian Federation and it included 367 patients with DM, 34.88% men and 65.12% women, aged 50.88±20.55 years. 34.88% of patients suffered from T1DM, 65.12% suffered from T2DM, the average duration of the disease was 9.02±7.22 years. 58.31% of patients had a history of arterial hypertension, 13.08% had a history of smoking. The primary endpoint was the frequency of detection of diabetic changes in the eye fundus of patients with T1DM and T2DM in general; the secondary endpoint - same but separately, and for T2DM patients depending on the duration of the disease. The exploratory endpoint was the assessment of the influence of various factors on the development of DR. The patients underwent visometry (modified ETDRS table), biomicroscopy, mydriatic fundus photography according to the «2 fields¼ protocol. RESULTS: The average detection rate of DR was 12.26%, primarily observed in patients with T2DM (13.81%), women (9.26%), in both eyes (8.17%). Among patients with DR, 26 (19.55%) had glycated hemoglobin (HbA1c) level exceeding 7.5% (p=0.002), indicating a direct relationship between this indicator and the incidence of DR. Logistic regression analysis showed that the duration of diabetes of more than 10 years has a statistically significant effect on the development of DR. In the modified model for odds estimation, the likelihood of developing DR is increased by the duration of DM for more than 10 years; increased blood pressure; HbA1c level >7.5%. CONCLUSION: The obtained results, some of which will be presented in subsequent publications, highlight the effectiveness of using two-field mydriatic fundus photography as a screening for DR.

The Rationale of sub-hepatic drainage on a specialist biliary unit: a review of 6140 elective and urgent laparoscopic cholecystectomies and bile duct explorations.

BACKGROUND: Drains are used to reduce abdominal collections after procedures where such risk exists. Using abdominal drains after cholecystectomy has been controversial since the open surgery era. Universally accepted indications and agreement exist that routine drainage is unnecessary but the role of selective drainage remains undetermined. This study evaluates the indications and benefits of sub-hepatic drainage in patients undergoing laparoscopic cholecystectomy (LC) and bile duct exploration (BDE) in a specialist unit with a large biliary emergency workload. METHODS: Prospectively collected data from 6,140 LCs with a 46.6% emergency workload over 30 years was reviewed. Demographic factors, pre-operative presentations, imaging and operative details in patients with and without drains were compared. Sub-hepatic drains were inserted after all transductal explorations, subtotal cholecystectomies, almost all open conversions and 94% of LC for empyemas. Adverse or beneficial postoperative drain-related outcomes were analysed. RESULTS: Abdominal drains were utilised in 3225/6140 (52.5%). Patients were significantly older with more males. 59.4% were emergency admissions. Preoperative imaging showed thick-walled gallbladders in 25.2% and bile duct stones or dilatation in 36.2%. At operation they had cystic duct stones in 19.8%, acute cholecystitis, empyema or mucocele in 28.4% and operative difficulty grades III or higher in 59%. 38% underwent BDE, 5.4% had fundus-first dissection and the operating times were longer ( 80 vs.45 min). Drain related complications were rare; 3 abdominal pains after anaesthetic recovery settling when drains were removed, 2 drain site infections and one re-laparoscopy to retrieve a retracted drain. 55.8% of 43 bile leaks and 35% of 20 other collections in patients with drains resolved spontaneously. CONCLUSIONS: The utilisation of drains in this study was relatively high due to the high emergency workload and interest in BDE. While drains allowed early detection of bile leakage, avoiding some complications and monitoring conservative management to allow early reinterventions, the study has identified operative criteria that could potentially limit drain insertion through a selective policy.

Bio-spectroscopic investigation linking changes of retinal structure with short-term administration of Amiodarone and revealing the ameliorative effect of vitamin E supplementation.

Long term use of Amiodarone (AMIO) is associated with the development of ocular adverse effects. This study investigates the short term effects, and the ameliorative consequence of vitamin E on retinal changes that were associated with administration of AMIO. This is accomplished by investigating both retinal structural and conformational characteristics using Fourier transform infrared spectroscopy (FTIR) and Fundus examination. Three groups of healthy rabbits of both sexes were used; the first group served as control. The second group was orally treated with AMIO (160 mg /kg body weight) in a daily basis for two weeks. The last group orally received AMIO as the second group for two weeks then, oral administration of vitamin E (100 mg/kg body weight) for another two weeks as well. FTIR results revealed significant structural and conformational changes in retinal tissue constituents that include lipids and proteins due to AMIO administration. AMIO treatment was associated with fluctuated changes (increased/decreased) in the band position and bandwidth of NH, OH, and CH bonds. This was concomitant with changes in the percentage of retinal protein constituents in particularly α-helix and Turns. AMIO facilitates the formation of intra-molecular hydrogen bonding and turned retinal lipids to be more disordered structure. In conclusion, the obtained FTIR data together with principal component analysis provide evidence that administration of vitamin E following the treatment with AMIO can ameliorate these retinal changes and, these biophysical changes are too early to be detected by Fundus examination.

CA-ViT: Contour-Guided and Augmented Vision Transformers to Enhance Glaucoma Classification Using Fundus Images.

Glaucoma, a predominant cause of visual impairment on a global scale, poses notable challenges in diagnosis owing to its initially asymptomatic presentation. Early identification is vital to prevent irreversible vision impairment. Cutting-edge deep learning techniques, such as vision transformers (ViTs), have been employed to tackle the challenge of early glaucoma detection. Nevertheless, limited approaches have been suggested to improve glaucoma classification due to issues like inadequate training data, variations in feature distribution, and the overall quality of samples. Furthermore, fundus images display significant similarities and slight discrepancies in lesion sizes, complicating glaucoma classification when utilizing ViTs. To address these obstacles, we introduce the contour-guided and augmented vision transformer (CA-ViT) for enhanced glaucoma classification using fundus images. We employ a Conditional Variational Generative Adversarial Network (CVGAN) to enhance and diversify the training dataset by incorporating conditional sample generation and reconstruction. Subsequently, a contour-guided approach is integrated to offer crucial insights into the disease, particularly concerning the optic disc and optic cup regions. Both the original images and extracted contours are given to the ViT backbone; then, feature alignment is performed with a weighted cross-entropy loss. Finally, in the inference phase, the ViT backbone, trained on the original fundus images and augmented data, is used for multi-class glaucoma categorization. By utilizing the Standardized Multi-Channel Dataset for Glaucoma (SMDG), which encompasses various datasets (e.g., EYEPACS, DRISHTI-GS, RIM-ONE, REFUGE), we conducted thorough testing. The results indicate that the proposed CA-ViT model significantly outperforms current methods, achieving a precision of 93.0%, a recall of 93.08%, an F1 score of 92.9%, and an accuracy of 93.0%. Therefore, the integration of augmentation with the CVGAN and contour guidance can effectively enhance glaucoma classification tasks.

FTSNet: Fundus Tumor Segmentation Network on Multiple Scales Guided by Classification Results and Prompts.

The segmentation of fundus tumors is critical for ophthalmic diagnosis and treatment, yet it presents unique challenges due to the variability in lesion size and shape. Our study introduces Fundus Tumor Segmentation Network (FTSNet), a novel segmentation network designed to address these challenges by leveraging classification results and prompt learning. Our key innovation is the multiscale feature extractor and the dynamic prompt head. Multiscale feature extractors are proficient in eliciting a spectrum of feature information from the original image across disparate scales. This proficiency is fundamental for deciphering the subtle details and patterns embedded in the image at multiple levels of granularity. Meanwhile, a dynamic prompt head is engineered to engender bespoke segmentation heads for each image, customizing the segmentation process to align with the distinctive attributes of the image under consideration. We also present the Fundus Tumor Segmentation (FTS) dataset, comprising 254 pairs of fundus images with tumor lesions and reference segmentations. Experiments demonstrate FTSNet's superior performance over existing methods, achieving a mean Intersection over Union (mIoU) of 0.8254 and mean Dice (mDice) of 0.9042. The results highlight the potential of our approach in advancing the accuracy and efficiency of fundus tumor segmentation.