Two-stage deep neural network for diagnosing fungal keratitis via in vivo confocal microscopy images.

Timely and effective diagnosis of fungal keratitis (FK) is necessary for suitable treatment and avoiding irreversible vision loss for patients. In vivo confocal microscopy (IVCM) has been widely adopted to guide the FK diagnosis. We present a deep learning framework for diagnosing fungal keratitis using IVCM images to assist ophthalmologists. Inspired by the real diagnostic process, our method employs a two-stage deep architecture for diagnostic predictions based on both image-level and sequence-level information. To the best of our knowledge, we collected the largest dataset with 96,632 IVCM images in total with expert labeling to train and evaluate our method. The specificity and sensitivity of our method in diagnosing FK on the unseen test set achieved 96.65% and 97.57%, comparable or better than experienced ophthalmologists. The network can provide image-level, sequence-level and patient-level diagnostic suggestions to physicians. The results show great promise for assisting ophthalmologists in FK diagnosis.

Design and simulation of a compact polarization beam splitter based on dual-core photonic crystal fiber with elliptical gold layer.

For the polarization multiplexing requirements in all-optical networks, this work presents a compact all-fiber polarization beam splitter (PBS) based on dual-core photonic crystal fiber (PCF) and an elliptical gold layer. Numerical analysis using the finite element method (FEM) demonstrates that the mode modulation effect of the central gold layer effectively reduces the dimensions of the proposed PBS. By determining reasonable structural parameters of the proposed PCF, the coupling length ratio (CLR) between X- and Y-polarized super-modes can approach 2, achieving a minimal device length of 0.122 mm. The PBS exhibits a maximum extinction ratio (ER) of - 65 dB at 1.55 µm, with an operating bandwidth spanning 100 nm (1.5-1.6 µm) and a stable insertion loss (IL) of ~ 1.5 dB at 1.55 µm. Furthermore, the manufacture feasibility and performance verification scheme are also investigated. It is widely anticipated that the designed PBS will play a crucial role in the ongoing development process of miniaturization and integration of photonic devices.

Cross-Domain Social Rumor-Propagation Model Based on Transfer Learning.

Rumors in different topic domains have different text characteristics but similar emotional tendencies. To resolve the scarce-data problem in some rumor-topic domains, this study proposes a cross-domain rumor-propagation model, which is based on transfer learning. First, given the diversity and complexity of the rumor-propagation landscape, this study introduces a novel method, User-Retweet-Rumor2vec (URR2vec), which leverages the power of representation learning to uncover latent features within rumor topics. It also displays the forwarding relationship between users and rumors, user node information, and rumor-topic information in low-dimensional space. To capture the impact of human emotional cognition during rumor spreading, we also introduce a deep-learning model based on the natural language texts of rumor topics, which analyzes the sentiment in the text and uncovers the emotional correlations among users. Furthermore, a rumor-propagation prediction model based on the text-sentiment analysis-graph convolutional network (TSA-GCN) is proposed and pre-trained on existing rumor-topic data to ensure its prediction accuracy. Finally, considering the data sparsity at a rumor-topic outbreak, the trained propagation model is transferred to the rumor topic for prediction. Meanwhile, the rumor topic in different domains has different edges and conditional distribution, similar emotional characteristics, and network structure among the rumor topics. After fine-tuning the parameter and adding a domain adaptation layer in TSA-GCN, a domain adaptation model based on parameter and graph-structure migration is obtained.

Interest HD: An Interest Frame Model for Recommendation Based on HD Image Generation.

This work is inspired by high-definition (HD) image generation techniques. When the user's interests are viewed as different frames of varying clarity, the unclear parts of one interest frame can be clarified by other interest frames. The user's overall HD interest portrait can be viewed as a fusion of multiple interest frames through detail compensation. Based on this inspiration, we propose a model for generating HD interest portrait called interest frame for recommendation (IF4Rec). First, we present a fine-grained pixel-level user interest mining method, Pixel embedding (PE) uses positional coding techniques to mine atomic-level interest pixel matrices in multiple dimensions, such as time, space, and frequency. Then, using an atomic-level interest pixel matrix, we propose Item2Frame to generate several interest frames for a user. The similarity score of each item is calculated to fill the multi-interest pixel clusters, through an improved self-attention mechanism. Finally, stimulated by HD image generation techniques, we initially present an interest frame noise compensation method. By utilizing the multihead attention mechanism, pixel-level optimization and noise complementation are performed between multi-interest frames, and an HD interest portrait is achieved. Experiments show that our model mines users' interests well. On five publicly available datasets, our model outperforms the baselines.

Utilizing deep neural networks to extract non-linearity as entities in PAM visible light communication with noise.

Herein, we propose a novel entity extraction neural network (EXNN) with a newly designed sampling convolution kernel approach and a deep learning-based structure to differentiate noise in visible light communication (VLC) systems. In particular, EXNN is used to extract linear and nonlinear distortion in a received signal as an entity and compensate for the signal by removing it. First, we designed a deep learning structure tailored for VLC systems, used experimentation to validate our algorithm's usability, and determined an appropriate range for the hyper-parameters that govern the PAM-8 system. Second, we compared our approach with existing finite impulse response (FIR) linear and Volterra nonlinear compensation algorithms via experiments. Below the hard-decision forward error correction (HD-FEC) threshold limit of 3.8×10-3, experimental results show that the use of the EXNN increased the operating range of the direct current (DC) bias and the voltage by ∼33.3% and by ∼50% under optimal conditions, respectively. Furthermore, under corresponding optimal power conditions, the proposed approach improved the Q factor of the VLC system by 0.36 and 1.57 dB compared with the aforementioned linear and nonlinear equalizers, respectively. To the best of our knowledge, this is the first time that a deep learning operator has been custom-designed for the VLC system and we have named the completely redesigned network with this sampling convolution kernel operator as EXNN.

Efficacy of Baduanjin exercise for rehabilitation after COVID-19: A protocol for systematic review and meta-analysis.

BACKGROUND: The study aims to evaluate the effectiveness and safety of Baduanjin exercise for rehabilitation after COVID-19. METHODS: The following electronic databases will be searched from establishment to Jan 2021: Cochrane Library, MEDLINE, EMBASE, Web of Science, Springer, World Health Organization International Clinical Trials Registry Platform, China National Knowledge Infrastructure, Wan-fang database, Chinese Scientific Journal Database, Chinese Biomedical Literature Databases, and other databases, All published randomized controlled trials about this topic will be included. Two independent researchers will operate article retrieval, duplication removing, screening, quality evaluation, and data analyses by Review Manager (V.5.3.5). Meta-analyses, subgroup analysis, and/or descriptive analysis will be performed based on the included data conditions. RESULTS: The results of this study will provide a combination of high-quality evidence for researchers in the current field of COVID-19 treatment and rehabilitation. CONCLUSION: The conclusion of this study will provide the evidence of whether Baduanjin is an effective and safe intervention for rehabilitation after COVID-19. PROSPERO REGISTRATION NUMBER: CRD42020181078.

Sulfur-Doped Alkylated Graphene Oxide as High-Performance Lubricant Additive.

Sulfur-doped graphene oxide (SA-GO) prepared by sulfuration and alkylation of graphene oxide is applied as an efficient green anti-wear additive for harsh operation conditions of engines. X-ray photoelectron spectroscopy analysis reveals the sulfur content of octadecylamine-modified SA-GO (sulfuration follows alkylation) is increased by 79 times compared with the reverse process that alkylation follows sulfuration, suggesting the preparation route is a key factor of the sulfuration process. The higher sulfur content and -C-S-C- sulfur bonding constitution result better lubrication effect, while the investigation of chain length of alkylation modification and concentration of the alkylated sulfur-doped graphene oxide indicates the octylamine-modified SA-GO shows smaller diameter of wear scar within the concentration range between 1 × 10-4 and 2.5 × 10-4 wt%. The decrement percent of wear scar diameter is 43.2% in 928 lubrication oil and 17.2% in PAO4 oil while the SA-GO modified by octylamine is applied with the concentrations 2.5 × 10-4 wt% in PAO4 and 1 × 10-4 wt% in 928 oil, respectively. The sulfur content in oil samples is only 0.006~0.001 wt%, which is much lower than the sulfur content standard recommended by ILSAC that is 0.5 wt%. The research work indicates the SA-GO additive is more feasible for the pollution treatment which focuses the substantial reduction of sulfur content in lubrication oil on the premise of improving lubrication capability.

Tumor-associated antigen-based personalized dendritic cell vaccine in solid tumor patients.

Tumor-associated antigens (TAAs) have been tested in various clinical trials in cancer treatment but the patterns of specific T cell response to personalized TAA immunization remains to be fully understood. We report antigen-specific T cell responses in patients immunized with dendritic cell vaccines pulsed with personalized TAA panels. Tumor samples from patients were first analyzed to identify overexpressed TAAs. Autologous DCs were then transfected with pre-manufactured mRNAs encoding the full-length TAAs, overexpressed in the patients' tumors. Patients with glioblastoma multiforme (GBM) or advanced lung cancer received DC vaccines transfected with personalized TAA panels, in combination with low-dose cyclophosphamide, poly I:C, imiquimod and anti-PD-1 antibody. Antigen-specific T cell responses were measured. Safety and efficacy were evaluated. A total of ten patients were treated with DC vaccines transfected with personalized TAA panels containing 3-13 different TAAs. Among the seven patients tested for anti-TAA T cell responses, most of the TAAs induced antigen-specific CD4+ and/or CD8+ T cell responses, regardless of their expression levels in the tumor tissues. No Grade III/IV adverse events were observed among these patients. Furthermore, the treated patients were associated with favorable overall survival when compared to patients who received standard treatment in the same institution. Personalized TAA immunization-induced-specific CD4+ and CD8+ T cell responses without obvious autoimmune adverse events and was associated with favorable overall survival. These results support further studies on DC immunization with personalized TAA panels for combined immunotherapeutic regimens in solid tumor patients.Trial registration ClinicalTrials.gov, NCT02709616 (March, 2016), NCT02808364 (June 2016), NCT02808416 (June, 2016).

A Multi-Element Expression Score Is A Prognostic Factor In Glioblastoma Multiforme.

PURPOSE: Glioblastoma multiforme (GBM) is a highly malignant tumor of the central nervous system. Although primary GBM patients receive extensive therapies, tumors may recur within months, and there is no objective and scientific method to predict prognosis. Adoptive immunotherapy holds great promise for GBM treatment. However, the expression profiles of the tumor-associated antigens (TAAs) and tumor immune microenvironment (TME) genes used in immunotherapy of GBM patients have not been fully described. The present study aimed to develop a predictive tool to evaluate patient survival based on full analysis of the expression levels of TAAs and TME genes. METHODS: Expression profiles of a panel of 87 TAAs and 8 TME genes significantly correlated with poor prognosis were evaluated in 44 GBM patients and 10 normal brain tissues using quantitative real-time polymerase chain reaction (qRT-PCR). A linear formula (the LASSO algorithm based in the R package) weighted by regression coefficients was used to develop a multi-element expression score to predict prognosis; this formula was cross-validated by the leave-one-out method in different GBM cohorts. RESULTS: After analysis of gene expression, clinical features, and overall survival (OS), a total of 8 TAAs (CHI3L1, EZH2, TRIOBP, PCNA, PIK3R1, PRKDC, SART3 and EPCAM), 1 TME gene (FOXP3) and 4 clinical features (neutrophil-to-lymphocyte (NLR), number of basophils (BAS), age and treatment with standard radiotherapy and chemotherapy) were included in the formula. There were significant differences between high and low scoring groups identified using the formula in different GBM cohorts (TCGA (n=732) and GEO databases (n=84)), implying poor and good prognosis, respectively. CONCLUSION: The multi-element expression score was significantly associated with OS of GBM patients. The improve understanding of TAAs and TMEs and well-defined formula could be implemented in immunotherapy for GBM to provide better care.

Multiplex confounding factor correction for genomic association mapping with squared sparse linear mixed model.

Genome-wide Association Study has presented a promising way to understand the association between human genomes and complex traits. Many simple polymorphic loci have been shown to explain a significant fraction of phenotypic variability. However, challenges remain in the non-triviality of explaining complex traits associated with multifactorial genetic loci, especially considering the confounding factors caused by population structure, family structure, and cryptic relatedness. In this paper, we propose a Squared-LMM (LMM2) model, aiming to jointly correct population and genetic confounding factors. We offer two strategies of utilizing LMM2 for association mapping: 1) It serves as an extension of univariate LMM, which could effectively correct population structure, but consider each SNP in isolation. 2) It is integrated with the multivariate regression model to discover association relationship between complex traits and multifactorial genetic loci. We refer to this second model as sparse Squared-LMM (sLMM2). Further, we extend LMM2/sLMM2 by raising the power of our squared model to the LMMn/sLMMn model. We demonstrate the practical use of our model with synthetic phenotypic variants generated from genetic loci of Arabidopsis Thaliana. The experiment shows that our method achieves a more accurate and significant prediction on the association relationship between traits and loci. We also evaluate our models on collected phenotypes and genotypes with the number of candidate genes that the models could discover. The results suggest the potential and promising usage of our method in genome-wide association studies.

Different binding sites of bovine organic anion-transporting polypeptide1a2 are involved in the transport of different fluoroquinolones.

Because of their wide distribution and capability of transporting a large variety of compounds, organic anion-transporting polypeptides (OATPs) have been extensively recognized as crucial players in absorption, distribution, and excretion of various drugs. OATP1A2 was the first cloned human OATP and has been found to transport wide range of endogenous and exogenous compounds. Bovine Oatp1a2 (bOatp1a2) shares high homology with human OATP1A2 and is considered the functional ortholog of the latter. Previous study in our laboratory demonstrated that bOatp1a2 transport of estrone-3-sulfate (ES) exhibited biphasic saturation kinetics. In the present study, we investigated the transport function of bOatp1a2 for four different quinolone antibacterial agents (enrofloxacin, levofloxacin, norfloxacin, and ciprofloxacin) and found that all the tested fluoroquinolones can be transported by bOatp1a2. Further studies showed that different binding sites are responsible for the transport of different fluoroquinolones. Both ciprofloxacin and norfloxacin exhibited biphasic saturation kinetics. The Kms of the high- and low-affinity components for ciprofloxacin were 3.80 ± 0.85 µM and 182 ± 31 µM, respectively, while those for norfloxacin were 24.7 ± 0.1 µM and 393 ± 79 µM, respectively. Enrofloxacin and levofloxacin showed an inhibitory effect on the uptake of only the high concentration of ES and thus may be transported by the low-affinity site for ES. Interestingly, enrofloxacin and levofloxacin demonstrated an activation effect on ES uptake at the high-affinity binding site. These results suggested that multiple binding sites within the structure of bOatp1a2 may be responsible for the uptake of different quinolone antimicrobial agents.

Cloning and functional characterization of the pig (Sus scrofa) organic anion transporting polypeptide 1a2.

1. Organic anion transporting polypeptides (OATPs) are a family of transporter proteins that have been extensively recognized as key determinants of absorption, distribution, metabolism and excretion of various drugs. Human OATP1A2 has been demonstrated to transport wide spectrum of endogenous and exogenous compounds. Study on OATP1A2 orthologues of other species, however, is still limited. 2. Here, we described the cloning and functional characterization of a member of the OATP/Oatp family member obtained from pig (Sus scrofa) liver. Sequence analysis suggested that it has a high homology with human OATP1A2 and bovine Oatp1a2. Prototypic substrates estrone-3-sulfate (E-3-S) and taurocholic acid were transported by the protein. The transport of these two substrates is pH-dependent, with lower pH showing higher uptake function. Kinetic study showed the transport of these two substrates have a Km of 42.5 ± 12.1 and 33.1 ± 8.7 µM, respectively. Pig Slco1a2 has the highest expression level in the liver, and to a less extend in the brain and small intestine. 3. In conclusion, an OATP member was cloned from pig liver. Sequence analysis and phylogenic study revealed it as an orthologue of human OATP1A2. Its kinetic characteristic for prototypic substrates and organ distribution are similar with that of OATP1A2.