Polystyrene nanoplastics-induced intestinal barrier disruption via inflammation and apoptosis in zebrafish larvae (Danio Rerio).

Plastics are one of the most pervasive materials on Earth, to which humans are exposed daily. Polystyrene (PS) is a common plastic packaging material. However, the impact of PS on human health remains poorly understood. Therefore, this study aimed to identify intestinal damage induced by PS nanoplastics (PS-NPs) in zebrafish larvae which have a high homology with humans. Four days post fertilization (dpf), zebrafish larvae were exposed to 0-, 10-, and 50-ppm PS-NPs for 48 h Initially, to ascertain if 100 nm PS-NPs could accumulate in the gastrointestinal (GI) tract of zebrafish larvae, the larvae were exposed to red fluorescence-labeled PS-NPs, and at 6 dpf, the larvae were examined using a fluorescence microscope. Analysis of the fluorescence intensity revealed that the GI tract of larvae exposed to 50-ppm exhibited a significantly stronger fluorescence intensity than the other groups. Nonfluorescent PS-NPs were then used in further studies. Scanning electron microscopy (SEM) confirmed the spherical shape of the PS-NPs. Fourier-transform infrared spectroscopy (FT-IR) analysis revealed chemical alterations in the PS-NPs before and after exposure to larvae. The polydispersity index (PDI) value derived using a Zetasizer indicated a stable dispersion of PS-NPs in egg water. Whole-mount apoptotic signal analysis via TUNEL assay showed increased apoptosis in zebrafish larval intestines exposed to 50-ppm PS-NPs. Damage to the intestinal tissue was assessed by Alcian blue (AB) and hematoxylin and eosin (H&E) staining. AB staining revealed increased mucin levels in the zebrafish larval intestines. Thin larval intestinal walls with a decrease in the density of intestinal epithelial cells were revealed by H&E staining. The differentially expressed genes (DEGs) induced by PS-NPs were identified and analyzed. In conclusion, exposure to PS-NPs may damage the intestinal barrier of zebrafish larvae due to increased intestinal permeability, and the in vivo gene network may change in larvae exposed to PS-NPs.

Obesity is associated with incident chronic kidney disease in individuals with normal renal function.

BACKGROUND/AIMS: Obesity has known to be a modifiable risk factor associated with worse outcomes in chronic kidney disease (CKD), but few studies have examined the impact of obesity on CKD incidence in the general population. The purpose of this study was to investigate the role of body mass index (BMI) and waist-to-hip ratio (WHR) as predictors of incident CKD and to evaluate the impact of weight reduction on CKD prevention. METHODS: A total of 2,711 participants from a community-based cohort with normal renal function were prospectively analyzed. Among participants with obesity, we analyzed the change in WHR to evaluate the association of obesity reduction with CKD development. RESULTS: During a mean follow-up of 11.03 ± 4.22 years, incident CKD occurred in 190 (7.0%) participants. In the fully adjusted multivariable Cox proportional hazard models, the risk of incident CKD increased with higher BMI (hazard ratio, 1.06; 95% confidence interval, 1.00-1.11; p = 0.033) and higher WHR (hazard ratio, 1.33; 95% confidence interval, 1.07-1.66; p = 0.009). In the Kaplan-Meier analysis, cumulative adverse renal events were significantly more common in the maintained obesity group than in the reduced obesity group (p = 0.001). CONCLUSION: Both higher BMI and WHR were associated with development of CKD, but the magnitude of the effect of WHR was higher than that of BMI. Moreover, reducing obesity would be beneficial for renal prognosis.

Advanced materials for micro/nanorobotics.

Autonomous micro/nanorobots capable of performing programmed missions are at the forefront of next-generation micromachinery. These small robotic systems are predominantly constructed using functional components sourced from micro- and nanoscale materials; therefore, combining them with various advanced materials represents a pivotal direction toward achieving a higher level of intelligence and multifunctionality. This review provides a comprehensive overview of advanced materials for innovative micro/nanorobotics, focusing on the five families of materials that have witnessed the most rapid advancements over the last decade: two-dimensional materials, metal-organic frameworks, semiconductors, polymers, and biological cells. Their unique physicochemical, mechanical, optical, and biological properties have been integrated into micro/nanorobots to achieve greater maneuverability, programmability, intelligence, and multifunctionality in collective behaviors. The design and fabrication methods for hybrid robotic systems are discussed based on the material categories. In addition, their promising potential for powering motion and/or (multi-)functionality is described and the fundamental principles underlying them are explained. Finally, their extensive use in a variety of applications, including environmental remediation, (bio)sensing, therapeutics, etc., and remaining challenges and perspectives for future research are discussed.

Artificial Intelligence Applications in Diabetic Retinopathy: What We Have Now and What to Expect in the Future.

Diabetic retinopathy (DR) is a major complication of diabetes mellitus and is a leading cause of vision loss globally. A prompt and accurate diagnosis is crucial for ensuring favorable visual outcomes, highlighting the need for increased access to medical care. The recent remarkable advancements in artificial intelligence (AI) have raised high expectations for its role in disease diagnosis and prognosis prediction across various medical fields. In addition to achieving high precision comparable to that of ophthalmologists, AI-based diagnosis of DR has the potential to improve medical accessibility, especially through telemedicine. In this review paper, we aim to examine the current role of AI in the diagnosis of DR and explore future directions.

Prognostic Impact of Mucin Expression in Curatively Resected Ampulla of Vater Cancer.

Introduction: Mucins play a pivotal role in epithelial carcinogenesis; however, their role remains elusive in ampulla of Vater (AoV) cancer, regardless of histological subtype. Therefore, we investigated the clinical significance of MUC1, MUC2, MUC5AC, and MUC6 expression in AoV cancer. Methods: Using samples from 68 patients with AoV cancer, we performed immunohistochemical staining for MUC1, MUC2, MUC5AC, and MUC6 using a tissue microarray. Subsequently, we analyzed their expression patterns in relation to clinicopathological parameters and patient outcomes. Results: Of the patients, 98.5% exhibited positive expression for MUC1, while MUC2, MUC5AC, and MUC6 were expressed in 44.1%, 47.1%, and 41.2% of the patients, respectively. Correlation analyses between mucin expression and clinicopathological factors revealed no significant associations, except between MUC5AC expression and N stage. Univariate analysis demonstrated significant associations between MUC5AC expression and overall survival (OS). Multivariate analysis further confirmed that MUC5AC expression was a significant predictor of OS, along with the N stage. However, MUC5AC expression was not meaningfully associated with recurrence-free survival (RFS). The patients positive for MUC5AC expression had a considerably shorter OS than those with negative expression. Conclusions: Our study provides insights into the clinical impact of mucins on AoV cancer, regardless of the histological subtype. Although MUC1 expression is universal, MUC5AC expression is a significant prognostic indicator that correlates with lymph node metastasis and poor OS. These results emphasize the possible utility of MUC5AC as a biomarker for extensive lymph node dissection and the prognostic evaluation of patients with AoV cancer.

FQ-UWF: Unpaired Generative Image Enhancement for Fundus Quality Ultra-Widefield Retinal Images.

Ultra-widefield (UWF) retinal imaging stands as a pivotal modality for detecting major eye diseases such as diabetic retinopathy and retinal detachment. However, UWF exhibits a well-documented limitation in terms of low resolution and artifacts in the macular area, thereby constraining its clinical diagnostic accuracy, particularly for macular diseases like age-related macular degeneration. Conventional supervised super-resolution techniques aim to address this limitation by enhancing the resolution of the macular region through the utilization of meticulously paired and aligned fundus image ground truths. However, obtaining such refined paired ground truths is a formidable challenge. To tackle this issue, we propose an unpaired, degradation-aware, super-resolution technique for enhancing UWF retinal images. Our approach leverages recent advancements in deep learning: specifically, by employing generative adversarial networks and attention mechanisms. Notably, our method excels at enhancing and super-resolving UWF images without relying on paired, clean ground truths. Through extensive experimentation and evaluation, we demonstrate that our approach not only produces visually pleasing results but also establishes state-of-the-art performance in enhancing and super-resolving UWF retinal images. We anticipate that our method will contribute to improving the accuracy of clinical assessments and treatments, ultimately leading to better patient outcomes.

Comparing generative and retrieval-based chatbots in answering patient questions regarding age-related macular degeneration and diabetic retinopathy.

BACKGROUND/AIMS: To compare the performance of generative versus retrieval-based chatbots in answering patient inquiries regarding age-related macular degeneration (AMD) and diabetic retinopathy (DR). METHODS: We evaluated four chatbots: generative models (ChatGPT-4, ChatGPT-3.5 and Google Bard) and a retrieval-based model (OcularBERT) in a cross-sectional study. Their response accuracy to 45 questions (15 AMD, 15 DR and 15 others) was evaluated and compared. Three masked retinal specialists graded the responses using a three-point Likert scale: either 2 (good, error-free), 1 (borderline) or 0 (poor with significant inaccuracies). The scores were aggregated, ranging from 0 to 6. Based on majority consensus among the graders, the responses were also classified as 'Good', 'Borderline' or 'Poor' quality. RESULTS: Overall, ChatGPT-4 and ChatGPT-3.5 outperformed the other chatbots, both achieving median scores (IQR) of 6 (1), compared with 4.5 (2) in Google Bard, and 2 (1) in OcularBERT (all p ≤8.4×10-3). Based on the consensus approach, 83.3% of ChatGPT-4's responses and 86.7% of ChatGPT-3.5's were rated as 'Good', surpassing Google Bard (50%) and OcularBERT (10%) (all p ≤1.4×10-2). ChatGPT-4 and ChatGPT-3.5 had no 'Poor' rated responses. Google Bard produced 6.7% Poor responses, and OcularBERT produced 20%. Across question types, ChatGPT-4 outperformed Google Bard only for AMD, and ChatGPT-3.5 outperformed Google Bard for DR and others. CONCLUSION: ChatGPT-4 and ChatGPT-3.5 demonstrated superior performance, followed by Google Bard and OcularBERT. Generative chatbots are potentially capable of answering domain-specific questions outside their original training. Further validation studies are still required prior to real-world implementation.

Retinal Disease Diagnosis Using Deep Learning on Ultra-Wide-Field Fundus Images.

Ultra-wide-field fundus imaging (UFI) provides comprehensive visualization of crucial eye components, including the optic disk, fovea, and macula. This in-depth view facilitates doctors in accurately diagnosing diseases and recommending suitable treatments. This study investigated the application of various deep learning models for detecting eye diseases using UFI. We developed an automated system that processes and enhances a dataset of 4697 images. Our approach involves brightness and contrast enhancement, followed by applying feature extraction, data augmentation and image classification, integrated with convolutional neural networks. These networks utilize layer-wise feature extraction and transfer learning from pre-trained models to accurately represent and analyze medical images. Among the five evaluated models, including ResNet152, Vision Transformer, InceptionResNetV2, RegNet and ConVNext, ResNet152 is the most effective, achieving a testing area under the curve (AUC) score of 96.47% (with a 95% confidence interval (CI) of 0.931-0.974). Additionally, the paper presents visualizations of the model's predictions, including confidence scores and heatmaps that highlight the model's focal points-particularly where lesions due to damage are evident. By streamlining the diagnosis process and providing intricate prediction details without human intervention, our system serves as a pivotal tool for ophthalmologists. This research underscores the compatibility and potential of utilizing ultra-wide-field images in conjunction with deep learning.

Isobavachin, a main bioavailable compound in Psoralea corylifolia, alleviates lipopolysaccharide-induced inflammatory responses in macrophages and zebrafish by suppressing the MAPK and NF-κB signaling pathways.

ETHNOPHARMACOLOGICAL RELEVANCE: Psoralea corylifolia L. (PC) is widely used in traditional medicines to treat inflammatory and infectious diseases. Isobavachin (IBC) is a bioavailable prenylated flavonoid derived from PC that has various biological properties. However, little information is available on its anti-inflammatory effects and mechanisms of action. AIM OF THE STUDY: In this study, we aimed to determine the anti-inflammatory effects of IBC in vitro and in vivo by conducting a mechanistic study using murine macrophages. MATERIALS AND METHODS: We evaluated the modulatory effects of IBC on the production of pro-inflammatory cytokines and mediators in murine macrophages. In addition, we examined whether IBC inhibits lipopolysaccharide (LPS)-induced inflammatory responses in a zebrafish model. Alterations in inflammatory response-associated genes and proteins were determined using quantitative reverse transcriptional polymerase chain reaction (RT-qPCR) and Western blotting analysis. RESULTS: IBC markedly reduced the overproduction of inflammatory mediators, pro-inflammatory cytokines, inducible nitric oxide synthase (iNOS), cyclooxygenase-2 (COX-2), phosphorylation of mitogen-activated protein kinase (MAPK) and nuclear translocation of nuclear factor-kappa B (NF-κB) in macrophages induced by lipopolysaccharides (LPS). In addition, excessive NO, ROS, and neutrophil level induced by LPS, were suppressed by IBC treatment in a zebrafish inflammation model. CONCLUSIONS: Collectively, bioavailable IBC inhibited on the inflammatory responses by LPS via MAPK and NF-κB signaling pathways in vitro and in vivo, suggesting that it may be a potential modulatory agent against inflammatory disorders.

Ocular Involvement and Treatment Pattern in Korean Patients with Marfan Syndrome: A Population-Based Study.

PURPOSE: This study aimed to investigate the incidence and prevalence of, and treatment patterns for ocular complications in Korean patients with Marfan syndrome. METHODS: Incidence and prevalence of Marfan syndrome was calculated from 2010 to 2018, based on data from the Korean National Health Insurance Service (KNHIS). Diagnosis codes (for cataract, ectopia lentis, retinal detachment, etc.) and surgery reimbursement codes (lensectomy, phacoemulsification, buckling, vitrectomy, etc.) in the patients with Marfan syndrome were retrieved by complete enumeration of the data. RESULTS: The annual prevalence of Marfan syndrome adjusted by age and sex was gradually increased from 2.44 per 100,000 in 2010 to 4.36 per 100,000 in 2018. The age group of 10-19 years showed the highest prevalence. The prevalence of ectopia lentis was 21.7%, of whom 43.0% underwent surgeries. Surgery for RD was performed in 253 (14.1%) of 2044 patients during the study period. CONCLUSION: Although the most prevalent ophthalmologic manifestation was ectopia lentis, total prevalence rate of RD was more than 10% in the study period; thus, regular fundus examination is recommended for the patients with Marfan syndrome.

Efficacy and safety of three-dimensional visualization technology in assisting ablation therapy for hepatocellular carcinoma： A Meta-analysis / 临床肝胆病杂志

ObjectiveTo systematically evaluate the efficacy and safety of three-dimensional visualization technology in assisting ablation therapy for hepatocellular carcinoma. MethodsThis study was conducted according to PRISMA guidelines， with a PROSPERO registration number of CRD42023488398. PubMed， Embase， Web of Science， the Cochrane Library， CNKI， Wanfang Data， VIP， and CBM were searched for Chinese and English articles on three-dimensional visualization technology in assisting ablation therapy for hepatocellular carcinoma published up to March 2023. After quality assessment and data extraction of the studies included， RevMan 5.4 software was used to perform the meta-analysis. ResultsA total of 11 studies were included， with 972 patients in total， among whom 447 underwent ablation assisted by three-dimensional visualization technology （3D group） and 525 underwent ablation assisted by traditional two-dimensional imaging technology （2D group）. The meta-analysis showed that compared with the 2D group， the 3D group had significantly higher success rate of first-time ablation treatment （odds ratio ［OR］=5.43， 95% confidence interval ［CI］： 2.64‍ ‍—‍ ‍11.18， P<0.001）， technical efficiency （OR=6.15， 95%CI： 3.23‍ ‍—‍ ‍11.70， P<0.001）， and complete ablation rate （OR=2.50， 95%CI： 1.08‍ ‍—‍ ‍5.78， P=0.03）， as well as significantly lower incidence rate of major complications （OR=0.45， 95%CI： 0.24‍ ‍—‍ ‍0.87， P=0.02）， local recurrence rate （OR=0.35， 95%CI： 0.17‍ ‍—‍ ‍0.72， P=0.004）， and local tumor progression rate （OR=0.29， 95%CI： 0.16‍ ‍—‍ ‍0.50， P<0.001）， while there was no significant difference in the incidence rate of mild complications between the two groups （P>0.05）. ConclusionThree-dimensional visualization technology is safe and feasible in assisting ablation therapy for hepatocellular carcinoma and can improve ablation rate and reduce the incidence rate of serious complications， local recurrence rate， and local tumor progression rate， thereby showing an important application value in clinical practice.

Is Kidney Function Associated with Age-Related Macular Degeneration?: Findings from the Asian Eye Epidemiology Consortium.

PURPOSE: Chronic kidney disease (CKD) may elevate susceptibility to age-related macular degeneration (AMD) because of shared risk factors, pathogenic mechanisms, and genetic polymorphisms. Given the inconclusive findings in prior studies, we investigated this association using extensive datasets in the Asian Eye Epidemiology Consortium. DESIGN: Cross-sectional study. PARTICIPANTS: Fifty-one thousand two hundred fifty-three participants from 10 distinct population-based Asian studies. METHODS: Age-related macular degeneration was defined using the Wisconsin Age-Related Maculopathy Grading System, the International Age-Related Maculopathy Epidemiological Study Group Classification, or the Beckman Clinical Classification. Chronic kidney disease was defined as estimated glomerular filtration rate (eGFR) of less than 60 ml/min per 1.73 m2. A pooled analysis using individual-level participant data was performed to examine the associations between CKD and eGFR with AMD (early and late), adjusting for age, sex, hypertension, diabetes, body mass index, smoking status, total cholesterol, and study groups. MAIN OUTCOME MEASURES: Odds ratio (OR) of early and late AMD. RESULTS: Among 51 253 participants (mean age, 54.1 ± 14.5 years), 5079 had CKD (9.9%). The prevalence of early AMD was 9.0%, and that of late AMD was 0.71%. After adjusting for confounders, individuals with CKD were associated with higher odds of late AMD (OR, 1.46; 95% confidence interval [CI], 1.11-1.93; P = 0.008). Similarly, poorer kidney function (per 10-unit eGFR decrease) was associated with late AMD (OR, 1.12; 95% CI, 1.05-1.19; P = 0.001). Nevertheless, CKD and eGFR were not associated significantly with early AMD (all P ≥ 0.149). CONCLUSIONS: Pooled analysis from 10 distinct Asian population-based studies revealed that CKD and compromised kidney function are associated significantly with late AMD. This finding further underscores the importance of ocular examinations in patients with CKD. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Insulin resistance is associated with incident chronic kidney disease in population with normal renal function.

Background: Insulin resistance is prevalent in chronic kidney disease and may accelerate the progression of chronic kidney disease. This study aimed to investigate whether insulin resistance is associated with the development of incident chronic kidney disease in a population with normal renal function. Methods: A total of 3,331 participants from a community-based cohort with normal renal function were prospectively analyzed. We determined the relationship of insulin resistance indices with the incident chronic kidney disease using the Cox proportional hazard model and Kaplan-Meier survival analysis. Results: During a mean follow-up of 11.03 ± 4.22 years, incident chronic kidney disease occurred in 414 participants (12.4%). The high homeostasis model assessment-insulin resistance level group had an increased risk of incident chronic kidney disease (hazard ratio [HR], 1.40; 95% confidence interval [CI], 1.13-1.74; p = 0.002) compared to the normal group after adjustment for age, sex, history of hypertension, body mass index, total cholesterol, alcohol drinking status, smoking status, and baseline estimated glomerular filtration rate. The risk of incident chronic kidney disease also increased with the lower quantitative insulin sensitivity check index level (HR, 0.62; 95% CI, 0.41-0.92; p = 0.02) and higher leptin-adiponectin ratio level (HR, 1.23; 95% CI, 1.06-1.42; p = 0.006). Conclusion: Higher insulin resistance indices are associated with the incidence of chronic kidney disease. Our data suggest that increased insulin resistance may be involved in the development of incident chronic kidney disease in a population with normal renal function.

Fatal Hypermagnesemia in Patients Taking Magnesium Hydroxide.

Hypermagnesemia is a rare but potentially fatal electrolyte disorder often overlooked because of its unfamiliarity. Magnesium is regulated through a balance of bone, intestinal absorption, and renal excretion. Hypermagnesemia typically arises from excessive magnesium intake or reduced renal excretion; however, it also occurs in patients with normal kidney function. Herein, we report two cases of hypermagnesemia in patients taking magnesium hydroxide for constipation. The first case involved an 82-year-old woman with end-stage renal disease who developed metabolic encephalopathy due to hypermagnesemia, after taking 3,000 mg of magnesium hydroxide daily for constipation. Her magnesium level was 9.9 mg/dL. Her treatment involved discontinuing magnesium hydroxide and continuing hemodialysis, which led to her recovery. In the second case, a 50-year-old woman with a history of cerebral hemorrhage and mental retardation developed hypermagnesemia despite having normal renal function. She was also taking magnesium hydroxide for constipation, and her magnesium level was 11.0 mg/dL. She experienced cardiac arrest while preparing for continuous renal replacement therapy (CRRT). After achieving return of spontaneous circulation, CRRT was initiated, and her magnesium level showed a decreasing trend. However, vital signs and lactate levels did not recover, leading to death. These cases highlight the importance of prompt diagnosis and intervention for hypermagnesemia and the need to regularly monitor magnesium levels in individuals receiving magnesium-containing preparations, especially those with impaired kidney function.

The Correlations between the Intensity of Histopathological Ubiquitin-Specific Protease 11 Staining and Progression of Prostate Cancer.

BACKGROUND: Ubiquitin-specific protease 11 (USP11), one of the principal phosphatase and tensin homolog (PTEN) deubiquitinases, can reserve PTEN polyubiquitination to maintain PTEN protein integrity and inhibit PI3K/AKT pathway activation. The aim of the current study was to investigate the associations between immunohistochemical USP11 staining intensities and prognostic indicators in individuals with prostate cancer. METHODS: Tissue microarrays (TMAs) were performed for human prostate cancer and normal tissue (control) samples. Data on patient's age, Gleason score, plasma prostate-specific antigen (PSA) titer, disease stage, and presence of seminal vesicles, lymph nodes, and surgical margin involvement were collected. A pathologist who was blinded to the clinical outcome data scored the TMA for USP11 staining intensity as either positive or negative. RESULTS: Cancerous tissues exhibited lower USP11 staining intensity, whereas the neighboring benign peri-tumoral tissues showed higher USP11 staining intensity. The degree of USP11 staining intensity was lower in patients with a higher PSA titer, higher Gleason score, or more advanced disease stage. Patients who showed positive USP11 staining were more likely to have more optimal clinical and biochemical recurrence-free survival statistics. CONCLUSIONS: USP11 staining intensity in patients with prostate cancer is negatively associated with several prognostic factors such as an elevated PSA titer and a high Gleason score. It also reflects both biochemical and clinical recurrence-free survival in such patients. Thus, USP11 staining is a valuable prognostic factor in patients with prostate cancer.

Multi-Scale Learning with Sparse Residual Network for Explainable Multi-Disease Diagnosis in OCT Images.

In recent decades, medical imaging techniques have revolutionized the field of disease diagnosis, enabling healthcare professionals to noninvasively observe the internal structures of the human body. Among these techniques, optical coherence tomography (OCT) has emerged as a powerful and versatile tool that allows high-resolution, non-invasive, and real-time imaging of biological tissues. Deep learning algorithms have been successfully employed to detect and classify various retinal diseases in OCT images, enabling early diagnosis and treatment planning. However, existing deep learning algorithms are primarily designed for single-disease diagnosis, which limits their practical application in clinical settings where OCT images often contain symptoms of multiple diseases. In this paper, we propose an effective approach for multi-disease diagnosis in OCT images using a multi-scale learning (MSL) method and a sparse residual network (SRN). Specifically, the MSL method extracts and fuses useful features from images of different sizes to enhance the discriminative capability of a classifier and make the disease predictions interpretable. The SRN is a minimal residual network, where convolutional layers with large kernel sizes are replaced with multiple convolutional layers that have smaller kernel sizes, thereby reducing model complexity while achieving a performance similar to that of existing convolutional neural networks. The proposed multi-scale sparse residual network significantly outperforms existing methods, exhibiting 97.40% accuracy, 95.38% sensitivity, and 98.25% specificity. Experimental results show the potential of our method to improve explainable diagnosis systems for various eye diseases via visual discrimination.

Large language models and their impact in ophthalmology.

The advent of generative artificial intelligence and large language models has ushered in transformative applications within medicine. Specifically in ophthalmology, large language models offer unique opportunities to revolutionise digital eye care, address clinical workflow inefficiencies, and enhance patient experiences across diverse global eye care landscapes. Yet alongside these prospects lie tangible and ethical challenges, encompassing data privacy, security, and the intricacies of embedding large language models into clinical routines. This Viewpoint highlights the promising applications of large language models in ophthalmology, while weighing up the practical and ethical barriers towards their real-world implementation. This Viewpoint seeks to stimulate broader discourse on the potential of large language models in ophthalmology and to galvanise both clinicians and researchers into tackling the prevailing challenges and optimising the benefits of large language models while curtailing the associated risks.

Association Between Three Atopic Triad and Retinal Vein Occlusion Risk: A Nationwide Population-Based Study.

PURPOSE: To evaluate the association between three allergic diseases (allergic dermatitis, allergic rhinitis, and asthma) and the development of retinal vein occlusion (RVO), a major retinal disease that causes visual impairment. METHOD: This study used data obtained from the Korean National Health Insurance Claims database between 2009 and 2018. The association between the three atopic triads (allergic dermatitis, allergic rhinitis, and asthma) and the occurrence of sight-threatening RVO, as determined by diagnostic and treatment codes, were analyzed. Multivariate adjusted Cox regression analysis was used to determine the hazard ratios (HRs) and 95% confidence intervals for RVO development in the presence of allergic disease. RESULTS: In this population-based study, 2,160,195 (54.6%) individuals were male, 1,794,968 (45.4%) were female, and 620,938 (15.7%) were diagnosed with allergic diseases. Patients with either asthma or allergic rhinitis had a greater risk of RVO (adjusted hazard ratio (aHR) = 1.101, 95% confidence interval [CI] = 1.029-1.178 for asthma; aHR = 1.181, 95% CI = 1.147-1.215 for allergic rhinitis) compared to those without asthma or allergic rhinitis; however, patients with atopic dermatitis did not show a significant association with RVO (aHR = 1.071, 95% CI = 0.889-1.290), after adjusting for other risk factors. CONCLUSION: Our study revealed that allergic rhinitis, asthma, and coexisting multiple allergic conditions were associated with an increased risk of RVO. Thus, it may be advisable to suggest an ophthalmological examination for patients with allergies due to the increased possibility of the occurrence of retinal vascular disease.

Discriminative-Region Multi-Label Classification of Ultra-Widefield Fundus Images.

Ultra-widefield fundus image (UFI) has become a crucial tool for ophthalmologists in diagnosing ocular diseases because of its ability to capture a wide field of the retina. Nevertheless, detecting and classifying multiple diseases within this imaging modality continues to pose a significant challenge for ophthalmologists. An automated disease classification system for UFI can support ophthalmologists in making faster and more precise diagnoses. However, existing works for UFI classification often focus on a single disease or assume each image only contains one disease when tackling multi-disease issues. Furthermore, the distinctive characteristics of each disease are typically not utilized to improve the performance of the classification systems. To address these limitations, we propose a novel approach that leverages disease-specific regions of interest for the multi-label classification of UFI. Our method uses three regions, including the optic disc area, the macula area, and the entire UFI, which serve as the most informative regions for diagnosing one or multiple ocular diseases. Experimental results on a dataset comprising 5930 UFIs with six common ocular diseases showcase that our proposed approach attains exceptional performance, with the area under the receiver operating characteristic curve scores for each class spanning from 95.07% to 99.14%. These results not only surpass existing state-of-the-art methods but also exhibit significant enhancements, with improvements of up to 5.29%. These results demonstrate the potential of our method to provide ophthalmologists with valuable information for early and accurate diagnosis of ocular diseases, ultimately leading to improved patient outcomes.

Self-FI: Self-Supervised Learning for Disease Diagnosis in Fundus Images.

Self-supervised learning has been successful in computer vision, and its application to medical imaging has shown great promise. This study proposes a novel self-supervised learning method for medical image classification, specifically targeting ultra-wide-field fundus images (UFI). The proposed method utilizes contrastive learning to pre-train a deep learning model and then fine-tune it with a small set of labeled images. This approach reduces the reliance on labeled data, which is often limited and costly to obtain, and has the potential to improve disease detection in UFI. This method employs two contrastive learning techniques, namely bi-lateral contrastive learning and multi-modality pre-training, to form positive pairs using the data correlation. Bi-lateral learning fuses multiple views of the same patient's images, and multi-modality pre-training leverages the complementary information between UFI and conventional fundus images (CFI) to form positive pairs. The results show that the proposed contrastive learning method achieves state-of-the-art performance with an area under the receiver operating characteristic curve (AUC) score of 86.96, outperforming other approaches. The findings suggest that self-supervised learning is a promising direction for medical image analysis, with potential applications in various clinical settings.