Retinal vasculature provides an opportunity for direct observation of vessel morphology, which is linked to multiple clinical conditions. However, objective and quantitative interpretation of the retinal vasculature relies on precise vessel segmentation, which is time consuming and labor intensive. Artificial intelligence (AI) has demonstrated great promise in retinal vessel segmentation. The development and evaluation of AI-based models require large numbers of annotated retinal images. However, the public datasets that are usable for this task are scarce. In this paper, we collected a color fundus image vessel segmentation (FIVES) dataset. The FIVES dataset consists of 800 high-resolution multi-disease color fundus photographs with pixelwise manual annotation. The annotation process was standardized through crowdsourcing among medical experts. The quality of each image was also evaluated. To the best of our knowledge, this is the largest retinal vessel segmentation dataset for which we believe this work will be beneficial to the further development of retinal vessel segmentation.

Introduction Diabetic macular edema (DME) is one of the major sight-threatening complications of diabetic retinopathy, which is associated with retinal inflammation. However, it is still unknown whether DME is associated with systemic inflammation. The study aimed to investigate the association between systemic inflammatory and optical coherence tomography (OCT) biomarkers in patients with treatment-naïve center-involving diabetic macular edema (DME) and to further explore the role of systemic inflammation in DME. Methods Medical records including clinical characteristics and ophthalmic examinations were collected from patients with treatment-naïve center-involving DME. Systemic inflammation markers including systemic immune-inflammatory index (SII), neutrophil–lymphocyte ratio (NLR), and platelet–lymphocyte ratio (PLR) were calculated. OCT biomarkers, including intraretinal cyst (IRC) size, disorganization of retinal inner layers (DRIL), external limiting membrane (ELM)/ellipsoid zone (EZ) integrity, retinal hyperreflective foci (HRF), subretinal fluid (SRF) and vitreomacular (VM) status were evaluated manually. Correlation analysis and multivariable linear regression models were used to investigate the relationship between systemic inflammatory markers and OCT biomarkers. Results A total of 82 patients with treatment-naïve center-involving DME were included. The number of HRF on OCT was correlated with SII, NLR, and PLR and positively associated with SII (p < 0.001) in both univariate and multivariate linear regression analyses. The differences remained largely the same during subgroup analysis controlling DM duration, SRF, and ELM/EZ integrity. No significant association was observed between other OCT biomarkers and blood inflammatory markers. Conclusion Retinal HRF in diabetic macular edema is associated with blood inflammatory markers, which supports the theory of HRF’s inflammatory nature and emphasizes the important role of inflammation in DME. SII may be a potential marker for DME treatment decisions.

Background: Healthy Eating Index-2015 (HEI-2015), a healthy eating pattern proposed by Dietary Guidelines for Americans, is proven to be protective against various diseases. However, whether it is associated with age-related cataracts is unknown. Methods: This population-based cross-sectional study included 6395 participants from the National Health and Nutrition Examination Survey (NHANES) in the cycles from 2005 to 2008. HEI-2015 was calculated from 24-h dietary recall interviews, ranging from 0 to 100, and higher HEI-2015 represents better diet quality. Age-related cataract was generated from the questionnaire. The association between HEI-2015 and cataract was assessed with logistic regression models. Propensity score weighting, restricted cubic spline, and subgroup analysis were conducted to further explore the relationship. Results: 6395 participants were included in the study, with a mean [standard deviation (SD)] age of 48.7 (15.3) years and 3115 (48.7%) being male. HEI-2015 was negatively associated with cataracts after adjusting all covariates included, both as a continuous variable [odds ratio (OR): 0.991, 95% confidence interval (CI): 0.984–0.997, p = 0.006] and quartiles with the highest quartile compared to the lowest (OR: 0.739, 95% CI: 0.559–0.980, p = 0.035). After propensity score weighting, the association remained significant. Restricted cubic spline revealed no non-linear relationship (p for non-linearity 0.085). Subgroup analysis showed that there were no interaction effects. Conclusions: Adherence to the healthy eating pattern, HEI-2015, was associated with a lower risk of age-related cataracts.

Clinical application of artificial intelligence is limited due to the lack of interpretability and expandability in complex clinical settings. We aimed to develop an eye diseases screening system with improved interpretability and expandability based on a lesion-level dissection and tested the clinical expandability and auxiliary ability of the system. The four-hierarchical interpretable eye diseases screening system (IEDSS) based on a novel structural pattern named lesion atlas was developed to identify 30 eye diseases and conditions using a total of 32,026 ultra-wide field images collected from the Second Affiliated Hospital of Zhejiang University, School of Medicine (SAHZU), the First Affiliated Hospital of University of Science and Technology of China (FAHUSTC), and the Affiliated People's Hospital of Ningbo University (APHNU) in China between November 1, 2016 to February 28, 2022. The performance of IEDSS was compared with ophthalmologists and classic models trained with image-level labels. We further evaluated IEDSS in two external datasets, and tested it in a real-world scenario and an extended dataset with new phenotypes beyond the training categories. The accuracy (ACC), F1 score and confusion matrix were calculated to assess the performance of IEDSS. IEDSS reached average ACCs (aACC) of 0·9781 (95%CI 0·9739-0·9824), 0·9660 (95%CI 0·9591-0·9730) and 0·9709 (95%CI 0·9655-0·9763), frequency-weighted average F1 scores of 0·9042 (95%CI 0·8957-0·9127), 0·8837 (95%CI 0·8714-0·8960) and 0·8874 (95%CI 0·8772-0·8972) in datasets of SAHZU, APHNU and FAHUSTC, respectively. IEDSS reached a higher aACC (0·9781, 95%CI 0·9739-0·9824) compared with a multi-class image-level model (0·9398, 95%CI 0·9329-0·9467), a classic multi-label image-level model (0·9278, 95%CI 0·9189-0·9366), a novel multi-label image-level model (0·9241, 95%CI 0·9151-0·9331) and a lesion-level model without Adaboost (0·9381, 95%CI 0·9299-0·9463). In the real-world scenario, the aACC of IEDSS (0·9872, 95%CI 0·9828-0·9915) was higher than that of the senior ophthalmologist (SO) (0·9413, 95%CI 0·9321-0·9504, p = 0·000) and the junior ophthalmologist (JO) (0·8846, 95%CI 0·8722-0·8971, p = 0·000). IEDSS remained strong performance (ACC = 0·8560, 95%CI 0·8252-0·8868) compared with JO (ACC = 0·784, 95%CI 0·7479-0·8201, p= 0·003) and SO (ACC = 0·8500, 95%CI 0·8187-0·8813, p = 0·789) in the extended dataset. IEDSS showed excellent and stable performance in identifying common eye conditions and conditions beyond the training categories. The transparency and expandability of IEDSS could tremendously increase the clinical application range and the practical clinical value of it. It would enhance the efficiency and reliability of clinical practice, especially in remote areas with a lack of experienced specialists. National Natural Science Foundation Regional Innovation and Development Joint Fund (U20A20386), Key research and development program of Zhejiang Province (2019C03020), Clinical Medical Research Centre for Eye Diseases of Zhejiang Province (2021E50007).

Longitudinal disease progression evaluation between follow-up examinations relies on precise registration of medical images. Compared to other medical imaging methods, color fundus photograph, a common retinal examination, is easily affected by eye movements while shooting, for which we think it is necessary to develop a reliable longitudinal registration method for this modality. Thus, the purpose of this study was to propose a robust registration method for longitudinal color fundus photographs and establish a longitudinal retinal registration dataset. In the proposed algorithm, radiation-variation insensitive feature transform (RIFT) feature points were calculated and aligned, followed by further refinement using a normalized total gradient (NTG). Experiments and ablation analyses were conducted on both public and private datasets, using the mean registration error and registration success plot as the main evaluation metrics. The results showed that our proposed method was comparable to other state-of-the-art registration algorithms and was particularly accurate for longitudinal images with disease progression. We believe the proposed method will be beneficial for the longitudinal evaluation of fundus images.

Background: Malnutrition, characterized by degeneration of body composition due to reduced intake or inflammation, shares some common mechanisms with age-related macular degeneration (AMD), while their associations remain unexplored. Methods: This cohort study utilized data from the UK Biobank. Participants without pre-existing AMD and with complete malnutrition data were included. Cox regression was employed to evaluate the longitudinal association. An ElasticNet model was used to derive a metabolomic signature of malnutrition, which was subsequently assessed for association with AMD. Malnutrition and the metabolomic signature were further tested for associations with photoreceptor thinning. Structural equation modeling was applied to delineate underlying mechanisms. Results: A total of 444 681 participants (mean age: 56.4 ± 8.1 years; 45.8% male) were included, with 32 086 (7.2%) diagnosed with malnutrition at baseline. Over a median follow-up of 13.6 years, 10 009 AMD cases were identified. Malnutrition was associated with an increased risk of AMD (hazard ratio [HR]: 1.221, 95% CI, 1.144-1.304, p < .001). The metabolomic signature of malnutrition, derived from 127 metabolites, was significantly associated with AMD risk (per SD increase: HR: 1.073, 95% CI, 1.037-1.110, p < .001) and thinner photoreceptor layer (β = −.214, 95% CI, −0.314 to −0.114, p < .001). Structural equation modeling revealed that malnutrition increased AMD risk partially through metabolomic changes that induced photoreceptor thinning. Conclusions: Malnutrition in middle-aged adults was significantly associated with increased risk of AMD, which was mediated by metabolomic alterations that impaired photoreceptor health.