Learning Rates of Deep Nets for Geometrically Strongly Mixing Sequence.

The great success of deep learning poses an urgent challenge to establish the theoretical basis for its working mechanism. Recently, research on the convergence of deep neural networks (DNNs) has made great progress. However, the existing studies are based on the assumption that the samples are independent, which is too strong to be applied to many real-world scenarios. In this brief, we establish a fast learning rate for the empirical risk minimization (ERM) on DNN regression with dependent samples, and the dependence is expressed in terms of geometrically strongly mixing sequence. To the best of our knowledge, this is the first convergence result of DNN methods based on mixing sequences. This result is a natural generalization of the independent sample case.

Nano-Nd2O3 reduced soil bacterial community function by altering the relative abundance of rare and sensitive taxa.

Nanoparticulate-Nd2O3 (nano-Nd2O3) has been excessively utilized in agriculture, industry, and medicine. Hence, nano-Nd2O3 can have environmental implications. However, the impact of nano-Nd2O3 on alpha diversity, composition, and function of soil bacterial communities has not been thoroughly evaluated. We amended soil to achieve different concentrations of nano-Nd2O3 (0, 10, 50, and 100 mg kg-1 soil) and incubated the mesocosms for 60 days. On days 7 and 60 of the experiment, we measured the effect of nano-Nd2O3 on alpha diversity and composition of soil bacterial community. Further, the effect of nano-Nd2O3 on the function of soil bacterial community was assessed based on changes in the activities of the six potential enzymes that mediate the cycling of nutrients in the soil. Nano-Nd2O3 did not alter the alpha diversity and composition of the soil bacterial community; however, it negatively affected community function in a dose-dependent manner. Specifically, the activities of ß-1,4-glucosidase and ß-1,4-n-acetylglucosaminidase that mediate soil carbon and nitrogen cycling, respectively, were significantly affected on days 7 and 60 of the exposure. The effect of nano-Nd2O3 on the soil enzymes correlated with changes in relative abundances of the rare and sensitive taxa, viz., Isosphaerales, Isosphaeraceae, Ktedonobacteraceae, and Streptomyces. Overall, we provide information for the safe implementation of technological applications that use nano-Nd2O3.

Clinical, histological and molecular characteristics of Alport syndrome in Chinese children.

BACKGROUND: Alport syndrome is caused by COL4A3, COL4A4, or COL4A5 gene mutations. The present study aims to compare the clinicopathological features, gene mutations, and outcome of Chinese children with different forms of Alport syndrome. METHODS: One hundred twenty-eight children from 126 families diagnosed with Alport syndrome through pathological and genetic examination between 2003 and 2021 were included in this single-center retrospective study. The laboratory and clinicopathological features of the patients with different inheritance patterns were analyzed. The patients were followed-up for disease progression and phenotype-genotype correlation. RESULTS: Of the 126 Alport syndrome families, X-linked forms accounted for 77.0%, autosomal recessive for 11.9%, autosomal dominant for 7.1%, and digenic for 4.0%. Among the patients, 59.4% were males and 40.6% were females. Altogether, 114 different mutations were identified in 101 patients from 99 families by whole-exome sequencing, of which 68 have not been previously reported. The most prevalent type of mutation was glycine substitution, which was identified in 52.1%, 36.7%, and 60% of the patients with X-linked Alport syndrome, autosomal recessive and autosomal dominant Alport syndrome, respectively. At the end of a median follow up of 3.3 (1.8-6.3) years, Kaplan-Meier curves showed kidney survival was significantly lower in autosomal recessive compared to X-linked Alport syndrome (P = 0.004). Pediatric patients with Alport syndrome seldom presented extrarenal involvement. CONCLUSIONS: X-linked Alport syndrome is the most frequent form found in this cohort. Progression was more rapid in autosmal recessive than in X-linked Alport syndrome.

Outcomes of Microcatheter-Assisted Trabeculotomy for Glaucoma Associated With Sturge-Weber Syndrome and Phakomatosis Pigmentovascularis.

PURPOSE: To evaluate mid-term efficacy and safety of ab externo Microcatheter-assisted trabeculotomy (MAT) for early-onset glaucoma associated with Sturge-Weber syndrome (SWS) and phakomatosis pigmentovascularis (PPV). DESIGN: Retrospective, non-comparative, interventional case series. METHODS: Medical records of consecutive SWS- or PPV-associated glaucoma patients who had undergone ab externo MAT between August 2017 and April 2020 at Beijing Children's Hospital were reviewed. Success was defined as an intraocular pressure (IOP) of <21 mmHg with (qualified success) or without (complete success) the use of antiglaucoma medication. RESULTS: Overall, 13 eyes (12 patients) with SWS and 9 eyes (8 patients) with PPV were included, with a mean age of 12.8 ± 15.8 months at the time of surgery and a mean follow-up time of 39.5 ±10.4 months. Both the SWS (26.5 ± 5.3 mmHg at baseline vs 16.5 ± 5.0 mmHg at the last visit; P < .001) and PPV (29.2 ± 7.5 mmHg vs 23.4 ± 4.7 mmHg; P = .014) subsets achieved a statistically significant fall in IOP following surgery. The Kaplan-Meier survival rate of complete (qualified) success after 42 months was 76.2% (87.5%) and 22.2% (40.0%) for eyes with SWS and PPV, respectively. Complications were minimal. Phakomatosis pigmentovascularis was associated with worse surgical outcomes. CONCLUSIONS: Ab externo MAT is an effective and safe treatment for early-onset glaucoma associated with SWS, but a gradual increase in IOP over time was noted in some patients. Ab externo MAT has limited efficacy for early-onset glaucoma associated with PPV in the mid-term.

Integrated bioinformatics and in silico approaches reveal the biological targets and molecular mechanisms of 1,25-dihydroxyvitamin D against COVID-19 and diabetes mellitus.

Coronavirus disease 2019 (COVID-19) and diabetes mellitus (DM) are two major diseases threatening human health. The susceptibility of DM patients to COVID-19 and their worse outcomes have forced us to explore efficient routes to combat COVID-19/DM. As the most active form of Vitamin D, 1,25-dihydroxyvitamin D (1,25(OH)2D) has been shown a beneficial effect in the treatment of COVID-19/DM. However, the anti-COVID-19/DM mechanisms of 1,25(OH)2D remain unclear. In this study, an approach combining network pharmacology and molecular docking was performed to reveal the potential hub target genes and underlying mechanisms of 1,25(OH)2D in the treatment of COVID-19/DM. The hub targets and interaction pathways related to 1,25(OH)2D were identified by integrating the key 1,25(OH)2D-target-signaling pathway-COVID-19/DM networks. Fifteen hub targets of 1,25(OH)2D against COVID-19DM were determined, including EGFR, PIK3R1, PIK3CA, STAT3, MAPK1, ESR1, HSP90AA1, LCK, MTOR, IGF1, AR, NFKB1, PIK3CB, PTPN1, and MAPK14. An enrichment analysis of the hub targets further revealed that the effect of 1,25(OH)2D against COVID-19/DM involved multiple biological processes, cellular components, molecular functions and biological signaling pathways. Molecular docking disclosed that 1,25(OH)2D docked nicely with the hub target proteins, including EGFR, PIK3R1, and PIK3CA. These findings suggested that the potential mechanisms of 1,25(OH)2D against COVID-19/DM may be related to multiple biological targets and biological signaling pathways.

The Hematopoietic Function of Medicinal Wine Maoji Jiu Revealed in Blood Deficiency Model Rats.

Maoji Jiu (MJ), a medicinal wine, has been used commonly by the Chinese to enrich and nourish the blood. In this study, the aim is to examine the hematopoietic function of MJ and investigate its hematopoietic regulation mechanism. Thirty-six female Sprague-Dawley rats (200 ± 20 g) were randomly divided into six groups with six rats in each group. The blood deficiency model was induced by injecting hypodermically with N-acetylphenylhydrazine (APH) and injecting intraperitoneally with cyclophosphamide (CTX), and treatment drugs were given by oral gavage twice a day for continuous 10 days from the start of the experiments. The administration of MJ improved the levels of white blood cells (WBCs), red blood cells (RBCs), hemoglobin (HGB), and hematocrit (HCT) in the blood deficiency model rats. Hematopoietic effect involves regulating the antioxidant activity in the liver and the levels of Bcl-2, Bax, erythropoietin (EPO), transforming growth factor-beta-1 (TGF-ß1), and macrophage colony-stimulating factor (M-CSF) mRNA in spleen tissues to enhance extramedullary hematopoiesis. This study suggests that MJ has a beneficial effect on blood deficiency model rats.

Practical Application of Artificial Intelligence Technology in Glaucoma Diagnosis.

Purpose: By comparing the performance of different models between artificial intelligence (AI) and doctors, we aim to evaluate and identify the optimal model for future usage of AI. Methods: A total of 500 fundus images of glaucoma and 500 fundus images of normal eyes were collected and randomly divided into five groups, with each group corresponding to one round. The AI system provided diagnostic suggestions for each image. Four doctors provided diagnoses without the assistance of the AI in the first round and with the assistance of the AI in the second and third rounds. In the fourth round, doctor B and doctor D made diagnoses with the help of the AI and the other two doctors without the help of the AI. In the last round, doctor A and doctor B made diagnoses with the help of AI and the other two doctors without the help of the AI. Results: Doctor A, doctor B, and doctor D had a higher accuracy in the diagnosis of glaucoma with the assistance of AI in the second (p=0.036, p=0.003, and p ≤ 0.000) and the third round (p=0.021, p ≤ 0.000, and p ≤ 0.000) than in the first round. The accuracy of at least one doctor was higher than that of AI in the second and third rounds, in spite of no detectable significance (p=0.283, p=0.727, p=0.344, and p=0.508). The four doctors' overall accuracy (p=0.004 and p ≤ 0.000) and sensitivity (p=0.006 and p ≤ 0.000) as a whole were significantly improved in the second and third rounds. Conclusions: This "Doctor + AI" model can clarify the role of doctors and AI in medical responsibility and ensure the safety of patients, and importantly, this model shows great potential and application prospects.

Compared With Girls, Boys' Psychological Symptoms Are More Likely to Be Influenced by Lifestyle in Chinese Middle School Students.

Purpose: To explore the relationship between Chinese middle school students' lifestyles and psychological symptoms and compare the differences between genders. Method: Using the method of stratified cluster sampling, 14,356 students aged 13-18 years from 8 cities in China were selected as participants. The Multidimensional Sub-health Questionnaire of Adolescents was used to estimate psychological symptoms. Chi-square test and Logistic regression analysis were used to analyze the relationship between lifestyle and psychological symptoms. Results: The psychological symptom detection rate of Chinese middle school students was 21.37%. The detection rate of psychological symptoms for boys was 22.1%, which was significantly higher than for girls (20.64%, χ2 = 4.608, P < 0.05). According to the Logistic regression analysis, factors that have a positive correlation with the demonstration of psychological symptoms (P < 0.01) include: how the students go to school (by foot or on the vehicle) (OR = 1.16, 95% CI: 1.07-1.25); inadequate time for sleep (OR = 1.48, 95% CI: 1.28-1.72); video watching time ≥2 h/d (OR = 1.25, 95% CI: 1.13-1.39); never exercise (OR = 2.39, 95% CI: 2.07-2.76); never participate in extracurricular exercises (OR = 1.45, 95% CI: 1.27-1.66); have breakfast occasionally (OR = 1.35, 95% CI:1.22-1.50); never have breakfast (OR = 1.90, 95% CI: 1.62-2.24); always have snacks (OR = 1.27, 95% CI: 1.13-1.44); always drink sugared beverages (OR = 1.37, 95% CI: 1.22-1.55); picky with food occasionally (OR = 1.22, 95% CI: 1.11-1.33). Conclusions: There was a positive correlation between unhealthy lifestyle and the occurrence of psychological symptoms, and boys are more easily influenced by lifestyles than girls.

Metabonomics Study of the Hematopoietic Effect of Medicinal Wine Maoji Jiu on a Blood Deficiency Rat Model by Ultra-High-Performance Liquid Chromatography Coupled to Quadrupole Time-of-Flight Mass Spectrometry and a Pattern Recognition Approach.

Maoji Jiu (MJ) is a kind of medicinal wine that has been widely used by Chinese people for many years to nourish and promote blood circulation. The purpose of this study was to investigate the hematopoietic effect of MJ on the metabolism of blood deficient rats and to explore the underlying hematopoietic regulation mechanisms. Blood deficiency model rats were induced by subcutaneous injection of N-acetylphenylhydrazine (APH) and intraperitoneal injection of cyclophosphamide (CTX). The plasma metabolic fingerprints of blood deficiency model rats with and without MJ treatment were obtained by using metabonomics based on ultra-high-performance liquid chromatography coupled to quadrupole time-of-flight mass spectrometry (UHPLC-QTOF/MS). Orthogonal partial least squares-discriminant analysis (OPLS-DA) was used to evaluate the hematopoietic effect of MJ and identify potential biomarkers in the plasma of blood deficiency model rats. The levels of white blood cells (WBC), red blood cells (RBC) and hemoglobin (HGB) and the activity of antioxidant capacity showed a recovery trend to the control group after MJ treatment, while the dose of 10 mL/kg showed the best effect. In this study, thirteen potential biomarkers were identified, which were mainly related to seven metabolic pathways, including linoleic acid metabolism, d-glutamine and d-glutamate metabolism, alanine, aspartate and glutamate metabolism, tryptophan metabolism, pyrimidine metabolism, porphyrin and chlorophyll metabolism and arginine biosynthesis. Metabolomics was applied frequently to reflect the physiological and metabolic state of organisms comprehensively, indicating that the rapid plasma metabonomics may be a potentially powerful tool to reveal the efficacy and enriching blood mechanism of MJ.

Preparation, Characterization, and Performance Evaluation of Petroleum Asphalt Modified with Bio-Asphalt Containing Furfural Residue and Waste Cooking Oil.

The study aims to analyze the feasibility of proposing waste cooking oil and industrial waste furfural residue as raw materials to prepare bio-asphalt as partial substitutes for petroleum asphalt, so as to reduce the cost of pavement construction and decrease the consumption of non-renewable resources. In this study, 90# petroleum asphalt was partially substituted with the bio-asphalt in different proportions to prepare biomass-modified petroleum asphalt, the performance of which was first evaluated based on three indices: penetration, softening point, and ductility. Comparison of the crystal structures of the bio-asphalt and furfural residue were enabled by X-ray diffraction, and the blending mechanism and microscopic morphologies of the biomass-substituted asphalt mixtures were characterized by Fourier transform infrared spectroscopy and scanning electron microscopy. The results showed that the bio-asphalt was hydrophobic and exhibited excellent compatibility with 90# petroleum asphalt. The partial substitution of petroleum asphalt with bio-asphalt improved the low-temperature crack resistance of the asphalt by adversely affecting the high-temperature stability of the asphalt; however, when the bio-asphalt content was 8 wt.%, the performance parameters of the biomass-modified asphalt met the requirements of the 90# petroleum asphalt standard.

Clinical and Renal Histology Findings and Different Responses to Induction Treatment Affecting the Long-Term Renal Outcomes of Children With ANCA-Associated Vasculitis: a Single-Center Cohort Analysis.

Introduction: Antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV) is relatively rare in children. This article aimed to analyze clinical and renal histology findings and different responses to induction treatment associated with the long-term renal outcomes in children with AAV in a single center. Methods: All pediatric patients with AAV admitted to Tongji Hospital from January 2002 to January 2021 were included in the study. The demographic, clinical, pathological, laboratory, and treatment data and outcomes were collected and analyzed to identify predictors associated with response to induction treatment and progression to end-stage renal disease (ESRD). Results: In total, 48 children with AAV were included in this cohort; 81.25% of them were women, and 91.7% were microscopic polyangiitis (MPA). Kidney involvement was found in 45 patients (93.75%). The most common histopathological subtype was crescentic form in this cohort according to Berden's classification. In total, 34 patients (70.8%) showed eGFR <60 ml/min/1.73 m2 at the time of diagnosis. Complete and partial remission was achieved in 8 patients (16.7%) and 19 patients (39.6%), respectively, following 6-month induction treatment. Half of the patients eventually progressed to ESRD at a mean time of (13.04 ± 15.83) months after diagnosis. The independent predictors of nonremission following induction treatment and progression to ESRD were baseline eGFR <60 ml/min/1.73 m2 and hypertension at diagnosis. Renal survival significantly decreased over time in patients with renal sclerotic subtypes or those with nonremission following induction treatment by Kaplan-Meier curve estimation. Conclusions: Our study demonstrates that women, MPA, and crescentic subtypes are predominant in pediatric AAV in China. Initial renal failure (eGFR <60 ml/min/1.73 m2), hypertension, sclerotic pathological subtype, and nonremission following induction treatment are predictive of long-term renal outcomes.

SMRT sequencing of the full-length transcriptome of Gekko gecko.

Tokay Gecko (Gekko gecko) is a rare and endangered medicinal animal in China. Its dry body has been used as an anti-asthmatic agent for two thousand years. To date, the genome and transcriptome of this species remain poorly understood. Here, we adopted single molecule real-time (SMRT) sequencing to obtain full-length transcriptome data and characterized the transcriptome structure. We identified 882,273 circular consensus (CCS) reads, including 746,317 full-length nonchimeric (FLNC) reads. The transcript cluster analysis revealed 212,964 consensus sequences, including 203,994 high-quality isoforms. In total, 111,372 of 117,888 transcripts were successfully annotated against eight databases (Nr, eggNOG, Swiss-Prot, GO, COG, KOG, Pfam and KEGG). Furthermore, 23,877 alternative splicing events, 169,128 simple sequence repeats (SSRs), 10,437 lncRNAs and 7,932 transcription factors were predicted across all transcripts. To our knowledge, this report is the first to document the G. gecko transcriptome using SMRT sequencing. The full-length transcript data might accelerate transcriptome research and lay the foundation for further research on G. gecko.

CUBN gene mutations may cause focal segmental glomerulosclerosis (FSGS) in children.

BACKGROUND: Imerslund-Gräsbeck Syndrome (IGS) is mainly caused by CUBN gene biallelic mutations. Proteinuria accompanies IGS specific symptoms in about half of the patients, isolated proteinuria is rarely reported. Here we present 3 patients with isolated proteinuria and focal segmental glomerulosclerosis (FSGS) caused by CUBN gene biallelic pathogenic variants. METHOD: Whole exome sequencing was performed on three children with isolated proteinuria. CUBN gene biallelic pathogenic variants were found and then verified by sanger sequencing. Their clinical, pathological and molecular genetic characteristics were analyzed and correlated accordingly. RESULTS: All three children presented with isolated proteinuria, no megaloblastic anemia. Their urine levels of ß2 microglobulin were normal or slightly higher. Renal biopsies showed focal segmental glomerulosclerosis with mild glomerular mesangial hypercellularity, partial effacement of foot processes and podocyte microvillation. Two of them were found to carry compound heterozygous mutations and one homozygous mutation of CUBN gene. Totally four CUBN gene biallelic pathogenic variants were identified, including c.9287 T > C (p.L3096P), c.122 + 1G > A, c.7906C > T (p.R2636*), c.10233G > A (p.W3411*). Except for intron splice-site mutation, all other variants are located in highly conserved sites of CUB domain for binding to albumin. CONCLUSION: The results demonstrate that CUBN gene mutations may cause isolated proteinuria pathologically presented as FSGS. Our cases extend the spectrum of renal manifestation and genotype of CUBN gene mutations.

A hierarchical deep learning approach with transparency and interpretability based on small samples for glaucoma diagnosis.

The application of deep learning algorithms for medical diagnosis in the real world faces challenges with transparency and interpretability. The labeling of large-scale samples leads to costly investment in developing deep learning algorithms. The application of human prior knowledge is an effective way to solve these problems. Previously, we developed a deep learning system for glaucoma diagnosis based on a large number of samples that had high sensitivity and specificity. However, it is a black box and the specific analytic methods cannot be elucidated. Here, we establish a hierarchical deep learning system based on a small number of samples that comprehensively simulates the diagnostic thinking of human experts. This system can extract the anatomical characteristics of the fundus images, including the optic disc, optic cup, and appearance of the retinal nerve fiber layer to realize automatic diagnosis of glaucoma. In addition, this system is transparent and interpretable, and the intermediate process of prediction can be visualized. Applying this system to three validation datasets of fundus images, we demonstrate performance comparable to that of human experts in diagnosing glaucoma. Moreover, it markedly improves the diagnostic accuracy of ophthalmologists. This system may expedite the screening and diagnosis of glaucoma, resulting in improved clinical outcomes.

REFUGE Challenge: A unified framework for evaluating automated methods for glaucoma assessment from fundus photographs.

Glaucoma is one of the leading causes of irreversible but preventable blindness in working age populations. Color fundus photography (CFP) is the most cost-effective imaging modality to screen for retinal disorders. However, its application to glaucoma has been limited to the computation of a few related biomarkers such as the vertical cup-to-disc ratio. Deep learning approaches, although widely applied for medical image analysis, have not been extensively used for glaucoma assessment due to the limited size of the available data sets. Furthermore, the lack of a standardize benchmark strategy makes difficult to compare existing methods in a uniform way. In order to overcome these issues we set up the Retinal Fundus Glaucoma Challenge, REFUGE (https://refuge.grand-challenge.org), held in conjunction with MICCAI 2018. The challenge consisted of two primary tasks, namely optic disc/cup segmentation and glaucoma classification. As part of REFUGE, we have publicly released a data set of 1200 fundus images with ground truth segmentations and clinical glaucoma labels, currently the largest existing one. We have also built an evaluation framework to ease and ensure fairness in the comparison of different models, encouraging the development of novel techniques in the field. 12 teams qualified and participated in the online challenge. This paper summarizes their methods and analyzes their corresponding results. In particular, we observed that two of the top-ranked teams outperformed two human experts in the glaucoma classification task. Furthermore, the segmentation results were in general consistent with the ground truth annotations, with complementary outcomes that can be further exploited by ensembling the results.

Mixed Maximum Loss Design for Optic Disc and Optic Cup Segmentation with Deep Learning from Imbalanced Samples.

Glaucoma is a serious eye disease that can cause permanent blindness and is difficult to diagnose early. Optic disc (OD) and optic cup (OC) play a pivotal role in the screening of glaucoma. Therefore, accurate segmentation of OD and OC from fundus images is a key task in the automatic screening of glaucoma. In this paper, we designed a U-shaped convolutional neural network with multi-scale input and multi-kernel modules (MSMKU) for OD and OC segmentation. Such a design gives MSMKU a rich receptive field and is able to effectively represent multi-scale features. In addition, we designed a mixed maximum loss minimization learning strategy (MMLM) for training the proposed MSMKU. This training strategy can adaptively sort the samples by the loss function and re-weight the samples through data enhancement, thereby synchronously improving the prediction performance of all samples. Experiments show that the proposed method has obtained a state-of-the-art breakthrough result for OD and OC segmentation on the RIM-ONE-V3 and DRISHTI-GS datasets. At the same time, the proposed method achieved satisfactory glaucoma screening performance on the RIM-ONE-V3 and DRISHTI-GS datasets. On datasets with an imbalanced distribution between typical and rare sample images, the proposed method obtained a higher accuracy than existing deep learning methods.

Development and Validation of a Deep Learning System to Detect Glaucomatous Optic Neuropathy Using Fundus Photographs.

Importance: A deep learning system (DLS) that could automatically detect glaucomatous optic neuropathy (GON) with high sensitivity and specificity could expedite screening for GON. Objective: To establish a DLS for detection of GON using retinal fundus images and glaucoma diagnosis with convoluted neural networks (GD-CNN) that has the ability to be generalized across populations. Design, Setting, and Participants: In this cross-sectional study, a DLS for the classification of GON was developed for automated classification of GON using retinal fundus images obtained from the Chinese Glaucoma Study Alliance, the Handan Eye Study, and online databases. The researchers selected 241 032 images were selected as the training data set. The images were entered into the databases on June 9, 2009, obtained on July 11, 2018, and analyses were performed on December 15, 2018. The generalization of the DLS was tested in several validation data sets, which allowed assessment of the DLS in a clinical setting without exclusions, testing against variable image quality based on fundus photographs obtained from websites, evaluation in a population-based study that reflects a natural distribution of patients with glaucoma within the cohort and an additive data set that has a diverse ethnic distribution. An online learning system was established to transfer the trained and validated DLS to generalize the results with fundus images from new sources. To better understand the DLS decision-making process, a prediction visualization test was performed that identified regions of the fundus images utilized by the DLS for diagnosis. Exposures: Use of a deep learning system. Main Outcomes and Measures: Area under the receiver operating characteristics curve (AUC), sensitivity and specificity for DLS with reference to professional graders. Results: From a total of 274 413 fundus images initially obtained from CGSA, 269 601 images passed initial image quality review and were graded for GON. A total of 241 032 images (definite GON 29 865 [12.4%], probable GON 11 046 [4.6%], unlikely GON 200 121 [83%]) from 68 013 patients were selected using random sampling to train the GD-CNN model. Validation and evaluation of the GD-CNN model was assessed using the remaining 28 569 images from CGSA. The AUC of the GD-CNN model in primary local validation data sets was 0.996 (95% CI, 0.995-0.998), with sensitivity of 96.2% and specificity of 97.7%. The most common reason for both false-negative and false-positive grading by GD-CNN (51 of 119 [46.3%] and 191 of 588 [32.3%]) and manual grading (50 of 113 [44.2%] and 183 of 538 [34.0%]) was pathologic or high myopia. Conclusions and Relevance: Application of GD-CNN to fundus images from different settings and varying image quality demonstrated a high sensitivity, specificity, and generalizability for detecting GON. These findings suggest that automated DLS could enhance current screening programs in a cost-effective and time-efficient manner.

Anti-Diabetic Effect of a Shihunine-Rich Extract of Dendrobium loddigesii on 3T3-L1 Cells and db/db Mice by Up-Regulating AMPK-GLUT4-PPARα.

The stems of Dendrobium loddigesii, a Chinese herb, are often used to treat diabetes and its polar extract is rich in shihunine, a water-soluble Orchidaceae alkaloid, but little is known about the anti-diabetes effects and mechanism of shihunine. This study investigated the anti-diabetic effect of a shihunine-rich extract of D. loddigesii (DLS) based on 3T3-L1 cells and db/db mice. The underlying mechanisms were primarily explored using Western blot analysis and immunohistochemical staining. The 3T3-L1 cell experiments showed that DLS can reduce the intracellular accumulation of oil droplets as well as triglycerides (p < 0.001) and promote the 2-[N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino]-2deoxyglucose (2-NBDG) uptake of 3T3-L1 cells (p < 0.001). The animal experiments confirmed that after 8 weeks of DLS treatment, the body weight, fasting blood sugar, and serum lipid levels of mice were significantly lowered, and the oral glucose tolerance test and serum insulin level were significantly improved compared to the no-treatment diabetes mellitus group. Further histomorphology observation led to the conclusion that the quantities of islet cells were significantly increased and the increase in adipose cell size was significantly suppressed. The immunohistochemical test of pancreatic tissue revealed that DLS inhibited the expression of cleaved cysteine aspartic acid-specific protease 3 (cleaved caspase-3). Western blot experiments showed that DLS had agonistic effects on adenosine monophosphate (AMP)-activated protein kinase phosphorylation (p-AMPK) and increased the expression levels of peroxisome proliferator-activated receptor α (PPARα) and glucose transporter 4 (GLUT4) in liver or adipose tissues. These data suggest that the shihunine-rich extract of D. loddigesii is an anti-diabetic fraction of D. loddigesii. Under our experimental condition, DLS at a dose of 50 mg/kg has good anti-diabetic efficacy.

Microcatheter-assisted Trabeculotomy for Primary Congenital Glaucoma After Failed Glaucoma Surgeries.

PURPOSE: To evaluate the effectiveness of microcatheter-assisted trabeculotomy (MAT) to treat primary congenital glaucoma after failed previous glaucoma surgeries. MATERIALS AND METHODS: Retrospective, noncomparative, interventional case series conducted at Beijing Tongren Eye Center, China. Outcome measures were compared between 3 groups: successful and complete (≥330 degrees) MAT; successful and partial (<330 degrees) MAT; or cases converted to traditional trabeculotomy when the Schlemm's canal could not be catheterized >180 degrees. Success was defined as final intraocular pressure ≤21 mm Hg, with (qualified success) or without (complete success) glaucoma medications. RESULTS: In total, 74 eyes of 63 consecutive patients were included. MAT was performed in 50 eyes (67.6%). Postoperative intraocular pressure and number of glaucoma drops (17.7±8.6 mm Hg, 0.6±1.2 medications) was significantly less than the preoperative values (35.3±7.2 mm Hg, 2.7±0.8 medications; P<0.001). Cumulative probabilities of qualified and complete success were 84.0% and 80.0% at 3-year follow-up with no difference between complete and partial trabeculotomies. MAT was not successfully performed in 24 eyes (32.4%), requiring conversion to traditional trabeculotomy and associated with greater incidence of previous surgeries (P<0.001), earlier age of disease onset (P=0.024) and worse corneal transparency (P=0.010). Cumulative probabilities of qualified and complete success were 37.0% and 29.2% at 3-year follow-up. CONCLUSIONS: Both complete and partial MAT achieved significant pressure reduction in cases of primary congenital glaucoma with previous failed glaucoma surgeries in intermediate term.