Metadata information and fundus image fusion neural network for hyperuricemia classification in diabetes.

OBJECTIVE: In diabetes mellitus patients, hyperuricemia may lead to the development of diabetic complications, including macrovascular and microvascular dysfunction. However, the level of blood uric acid in diabetic patients is obtained by sampling peripheral blood from the patient, which is an invasive procedure and not conducive to routine monitoring. Therefore, we developed deep learning algorithm to detect noninvasively hyperuricemia from retina photographs and metadata of patients with diabetes and evaluated performance in multiethnic populations and different subgroups. MATERIALS AND METHODS: To achieve the task of non-invasive detection of hyperuricemia in diabetic patients, given that blood uric acid metabolism is directly related to estimated glomerular filtration rate(eGFR), we first performed a regression task for eGFR value before the classification task for hyperuricemia and reintroduced the eGFR regression values into the baseline information. We trained 3 deep learning models: (1) metadata model adjusted for sex, age, body mass index, duration of diabetes, HbA1c, systolic blood pressure, diastolic blood pressure; (2) image model based on fundus photographs; (3)hybrid model combining image and metadata model. Data from the Shanghai General Hospital Diabetes Management Center (ShDMC) were used to develop (6091 participants with diabetes) and internally validated (using 5-fold cross-validation) the models. External testing was performed on an independent dataset (UK Biobank dataset) consisting of 9327 participants with diabetes. RESULTS: For the regression task of eGFR, in ShDMC dataset, the coefficient of determination (R2) was 0.684±0.07 (95 % CI) for image model, 0.501±0.04 for metadata model, and 0.727±0.002 for hybrid model. In external UK Biobank dataset, a coefficient of determination (R2) was 0.647±0.06 for image model, 0.627±0.03 for metadata model, and 0.697±0.07 for hybrid model. Our method was demonstrably superior to previous methods. For the classification of hyperuricemia, in ShDMC validation, the area, under the curve (AUC) was 0.86±0.013for image model, 0.86±0.013 for metadata model, and 0.92±0.026 for hybrid model. Estimates with UK biobank were 0.82±0.017 for image model, 0.79±0.024 for metadata model, and 0.89±0.032 for hybrid model. CONCLUSION: There is a potential deep learning algorithm using fundus photographs as a noninvasively screening adjunct for hyperuricemia among individuals with diabetes. Meanwhile, combining patient's metadata enables higher screening accuracy. After applying the visualization tool, it found that the deep learning network for the identification of hyperuricemia mainly focuses on the fundus optic disc region.

Integration of Vitreous Lipidomics and Metabolomics for Comprehensive Understanding of the Pathogenesis of Proliferative Diabetic Retinopathy.

As a vision-threatening complication of diabetes mellitus (DM), proliferative diabetic retinopathy (PDR) is associated with sustained metabolic disorders. Herein, we collected the vitreous cavity fluid of 49 patients with PDR and 23 control subjects without DM for metabolomics and lipidomics analyses. Multivariate statistical methods were performed to explore relationships between samples. For each group of metabolites, gene set variation analysis scores were generated, and we constructed a lipid network by using weighted gene co-expression network analysis. The association between lipid co-expression modules and metabolite set scores was investigated using the two-way orthogonal partial least squares (O2PLS) model. A total of 390 lipids and 314 metabolites were identified. Multivariate statistical analysis revealed significant vitreous metabolic and lipid differences between PDR and controls. Pathway analysis showed that 8 metabolic processes might be associated with the development of PDR, and 14 lipid species were found to be altered in PDR patients. Combining metabolomics and lipidomics, we identified fatty acid desaturase 2 (FADS2) as an important potential contributor to the pathogenesis of PDR. Collectively, this study integrates vitreous metabolomics and lipidomics to comprehensively unravel metabolic dysregulation and identifies genetic variants associated with altered lipid species in the mechanistic pathways for PDR.

Rapid Detection and Differentiation of Legionella pneumophila and Non-Legionella pneumophila Species by Using Recombinase Polymerase Amplification Combined With EuNPs-Based Lateral Flow Immunochromatography.

Legionella, a waterborne pathogen, is the main cause of Legionnaires' disease. Therefore, timely and accurate detection and differentiation of Legionella pneumophila and non-Legionella pneumophila species is crucial. In this study, we develop an easy and rapid recombinase polymerase amplification assay combined with EuNPs-based lateral flow immunochromatography (EuNPs-LFIC-RPA) to specifically distinguish Legionella pneumophila and non-Legionella pneumophila. We designed primers based on the mip gene of Legionella pneumophila and the 5S rRNA gene of non-Legionella pneumophila. The recombinase polymerase amplification reaction could go to completion in 10 min at 37°C, and the amplification products could be detected within 5 min with EuNPs-LFIC strips. Using a florescent test strip reader, the quantitative results were achieved by reading the colored signal intensities on the strips. The sensitivity was 1.6 × 101 CFU/ml, and a linear standard linear curve plotted from the test strip reader had a correlation coefficient for the determination of Legionella pneumophila (R 2 = 0.9516). Completed concordance for the presence or absence of Legionella pneumophila by EuNPs-LFIC-RPA and qPCR was 97.32% (κ = 0.79, 95% CI), according to an analysis of practical water samples (n = 112). In short, this work shows the feasibility of EuNPs-LFIC-RPA for efficient and rapid monitoring of Legionella pneumophila and non-Legionella pneumophila in water samples.