Efficacy of growth factor gene-modified stem cells for motor function after spinal cord injury in rodents: a systematic review and meta­analysis.

The efficacy of growth factor gene-modified stem cells in treating spinal cord injury (SCI) remains unclear. This study aims to evaluate the effectiveness of growth factor gene-modified stem cells in restoring motor function after SCI. Two reviewers searched four databases, including PubMed, Embase, Web of Science, and Scopus, to identify relevant records. Studies on rodents assessing the efficacy of transplanting growth factor gene-modified stem cells in restoring motor function after SCI were included. The results were reported using the standardized mean difference (SMD) with a 95% confidence interval (95% CI). Analyses showed that growth factor gene-modified stem cell transplantation improved motor function recovery in rodents with SCI compared to the untreated (SMD = 3.98, 95% CI 3.26-4.70, I2 = 86.8%, P < 0.0001) and stem cell (SMD = 2.53, 95% CI 1.93-3.13, I2 = 86.9%, P < 0.0001) groups. Using growth factor gene-modified neural stem/histone cells enhanced treatment efficacy. In addition, the effectiveness increased when viral vectors were employed for gene modification and high transplantation doses were administered during the subacute phase. Stem cells derived from the human umbilical cord exhibited an advantage in motor function recovery. However, the transplantation of growth factor gene-modified stem cells did not significantly improve motor function in male rodents (P = 0.136). Transplantation of growth factor gene-modified stem cells improved motor function in rodents after SCI, but claims of enhanced efficacy should be approached with caution. The safety of gene modification remains a significant concern, requiring additional efforts to enhance its clinical translatability.

Enriching contextualized language model from knowledge graph for biomedical information extraction.

Biomedical information extraction (BioIE) is an important task. The aim is to analyze biomedical texts and extract structured information such as named entities and semantic relations between them. In recent years, pre-trained language models have largely improved the performance of BioIE. However, they neglect to incorporate external structural knowledge, which can provide rich factual information to support the underlying understanding and reasoning for biomedical information extraction. In this paper, we first evaluate current extraction methods, including vanilla neural networks, general language models and pre-trained contextualized language models on biomedical information extraction tasks, including named entity recognition, relation extraction and event extraction. We then propose to enrich a contextualized language model by integrating a large scale of biomedical knowledge graphs (namely, BioKGLM). In order to effectively encode knowledge, we explore a three-stage training procedure and introduce different fusion strategies to facilitate knowledge injection. Experimental results on multiple tasks show that BioKGLM consistently outperforms state-of-the-art extraction models. A further analysis proves that BioKGLM can capture the underlying relations between biomedical knowledge concepts, which are crucial for BioIE.

Dopamine receptor D2 antagonism normalizes profibrotic macrophage-endothelial crosstalk in non-alcoholic steatohepatitis.

BACKGROUND & AIMS: Currently there is no effective treatment for liver fibrosis, which is one of the main histological determinants of non-alcoholic steatohepatitis (NASH). While Hippo/YAP (Yes-associated protein) signaling is essential for liver regeneration, its aberrant activation frequently leads to fibrosis and tumorigenesis. Unravelling "context-specific" contributions of YAP in liver repair might help selectively bypass fibrosis and preserve the pro-regenerative YAP function in hepatic diseases. METHODS: We used murine liver fibrosis and minipig NASH models, and liver biopsies from patients with cirrhosis. Single-cell RNA-sequencing (scRNA-Seq) was performed, and a G-protein-coupled receptor (GPCR) ligand screening system was used to identify cell-selective YAP inhibitors. RESULTS: YAP levels in macrophages are increased in the livers of humans and mice with liver fibrosis. The increase in type I interferon and attenuation of hepatic fibrosis observed in mice specifically lacking Yap1 in myeloid cells provided further evidence for the fibrogenic role of macrophage YAP. ScRNA-Seq further showed that defective YAP pathway signaling in macrophages diminished a fibrogenic vascular endothelial cell subset that exhibited profibrotic molecular signatures such as angiocrine CTGF and VCAM1 expression. To specifically target fibrogenic YAP in macrophages, we utilized a GPCR ligand screening system and identified a dopamine receptor D2 (DRD2) antagonist that selectively blocked YAP in macrophages but not hepatocytes. Genetic and pharmacological targeting of macrophage DRD2 attenuated liver fibrosis. In a large animal (minipig) NASH model recapitulating human pathology, the DRD2 antagonist blocked fibrosis and restored hepatic architecture. CONCLUSIONS: DRD2 antagonism selectively targets YAP-dependent fibrogenic crosstalk between macrophages and CTGF+VCAM1+ vascular niche, promoting liver regeneration over fibrosis in both rodent and large animal models. LAY SUMMARY: Fibrosis in the liver is one of the main histological determinants of non-alcoholic steatohepatitis (NASH), a disease paralleling a worldwide surge in metabolic syndromes. Our study demonstrates that a macrophage-specific deficiency in Yes-associated protein (YAP) attenuates liver fibrosis. Dopamine receptor D2 (DRD2) antagonism selectively blocks YAP in macrophages and thwarts liver fibrosis in both rodent and large animal models, and thus holds potential for the treatment of NASH.

A span-graph neural model for overlapping entity relation extraction in biomedical texts.

MOTIVATION: Entity relation extraction is one of the fundamental tasks in biomedical text mining, which is usually solved by the models from natural language processing. Compared with traditional pipeline methods, joint methods can avoid the error propagation from entity to relation, giving better performances. However, the existing joint models are built upon sequential scheme, and fail to detect overlapping entity and relation, which are ubiquitous in biomedical texts. The main reason is that sequential models have relatively weaker power in capturing long-range dependencies, which results in lower performance in encoding longer sentences. In this article, we propose a novel span-graph neural model for jointly extracting overlapping entity relation in biomedical texts. Our model treats the task as relation triplets prediction, and builds the entity-graph by enumerating possible candidate entity spans. The proposed model captures the relationship between the correlated entities via a span scorer and a relation scorer, respectively, and finally outputs all valid relational triplets. RESULTS: Experimental results on two biomedical entity relation extraction tasks, including drug-drug interaction detection and protein-protein interaction detection, show that the proposed method outperforms previous models by a substantial margin, demonstrating the effectiveness of span-graph-based method for overlapping relation extraction in biomedical texts. Further in-depth analysis proves that our model is more effective in capturing the long-range dependencies for relation extraction compared with the sequential models. AVAILABILITY AND IMPLEMENTATION: Related codes are made publicly available at http://github.com/Baxelyne/SpanBioER.

Integration of Transcriptomic and Proteomic Data Reveals the Possible Action Mechanism of the Antimicrobial Zhongshengmycin Against Didymella segeticola, the Causal Agent of Tea Leaf Spot.

Tea leaf spot, caused by the fungal phytopathogen Didymella segeticola, is an important foliar disease that can cause huge losses in the production and quality of tea, and there are no effective management measures to control the disease. This study screened a natural antimicrobial chemical for its activity against D. segeticola and studied its mode of action. Antifungal activity of the Streptomyces-derived antimicrobial zhongshengmycin (ZSM) against D. segeticola strain GZSQ-4 was assayed in vitro via the mycelial growth rate method. Optical microscopy and scanning and transmission electron microscopy were used to observe the morphological effects on hyphae treated with ZSM, with these studies complemented by transcriptomic, proteomic, and bioinformatic studies to identify the differentially expressed genes or differentially expressed proteins in hyphae treated with ZSM. Correlation analysis of transcriptomic and proteomic data were used to reveal the mode of action. The results indicated that ZSM could inhibit the growth of hyphae in vitro with a half-maximal effective concentration of 5.9 µg/ml, inducing some morphological changes in organelles, septa, and extracellular polysaccharides, targeting ribosomes to disturb translation, affecting the biosynthesis of some hyphal proteins at the messenger RNA and protein levels, and revealing correlations between findings from transcriptomes and proteomes.

Prediction of blood culture outcome using hybrid neural network model based on electronic health records.

BACKGROUND: Blood cultures are often performed to detect patients who has a serious illness without infections and patients with bloodstream infections. Early positive blood culture prediction is important, as bloodstream infections may cause inflammation of the body, even organ failure or death. However, existing work mainly adopts statistical models with laboratory indicators, and fails to make full use of textual description information from EHRs. METHODS: We study the problem of positive blood culture prediction by using neural network model. Specifically, we first construct dataset from raw EHRs. Then we propose a hybrid neural network which incorporates attention based Bi-directional Long Short-Term Memory and Autoencoder networks to fully capture the information in EHRs. RESULTS: In order to evaluate the proposed method, we constructe a dataset which consists of totally 5963 patients who had one or more blood cultures tests during hospitalization. Experimental results show that the proposed neural model gets 91.23% F-measure for this task. CONCLUSIONS: The comparison results of different models demonstrated the effectiveness of our model. The proposed model outperformed traditional statistical models.

Whole Genome Sequences of the Tea Leaf Spot Pathogen Didymella segeticola.

The fungal pathogen Didymella segeticola (basionym Phoma segeticola) causes leaf spot on tea (Camellia sinensis), which leads to a loss in tea leaf production in Guizhou Province, China. D. segeticola isolate GZSQ-4 was sequenced using Illumina HiSeq and Pacific Biosciences (PacBio) RS technologies, and then assembled to approximately 33.4 Mbp with a scaffold N50 value of approximately 2.3 Mbp. In total, 10,893 genes were predicted using the Nonredundant, Gene Ontology, Clusters of Orthologous Groups, Kyoto Encyclopedia of Genes and Genomes, and SWISS-PROT databases. The whole-genome sequence of D. segeticola will provide a resource for future research on host-pathogen interactions, determination of trait-specific genes, pathogen evolution, and plant-host adaptation mechanisms.

A hybrid neural network model for predicting kidney disease in hypertension patients based on electronic health records.

BACKGROUND: Disease prediction based on Electronic Health Records (EHR) has become one hot research topic in biomedical community. Existing work mainly focuses on the prediction of one target disease, and little work is proposed for multiple associated diseases prediction. Meanwhile, a piece of EHR usually contains two main information: the textual description and physical indicators. However, existing work largely adopts statistical models with discrete features from numerical physical indicators in EHR, and fails to make full use of textual description information. METHODS: In this paper, we study the problem of kidney disease prediction in hypertension patients by using neural network model. Specifically, we first model the prediction problem as a binary classification task. Then we propose a hybrid neural network which incorporates Bidirectional Long Short-Term Memory (BiLSTM) and Autoencoder networks to fully capture the information in EHR. RESULTS: We construct a dataset based on a large number of raw EHR data. The dataset consists of totally 35,332 records from hypertension patients. Experimental results show that the proposed neural model achieves 89.7% accuracy for the task. CONCLUSIONS: A hybrid neural network model was presented. Based on the constructed dataset, the comparison results of different models demonstrated the effectiveness of the proposed neural model. The proposed model outperformed traditional statistical models with discrete features and neural baseline systems.

Long short-term memory RNN for biomedical named entity recognition.

BACKGROUND: Biomedical named entity recognition(BNER) is a crucial initial step of information extraction in biomedical domain. The task is typically modeled as a sequence labeling problem. Various machine learning algorithms, such as Conditional Random Fields (CRFs), have been successfully used for this task. However, these state-of-the-art BNER systems largely depend on hand-crafted features. RESULTS: We present a recurrent neural network (RNN) framework based on word embeddings and character representation. On top of the neural network architecture, we use a CRF layer to jointly decode labels for the whole sentence. In our approach, contextual information from both directions and long-range dependencies in the sequence, which is useful for this task, can be well modeled by bidirectional variation and long short-term memory (LSTM) unit, respectively. Although our models use word embeddings and character embeddings as the only features, the bidirectional LSTM-RNN (BLSTM-RNN) model achieves state-of-the-art performance - 86.55% F1 on BioCreative II gene mention (GM) corpus and 73.79% F1 on JNLPBA 2004 corpus. CONCLUSIONS: Our neural network architecture can be successfully used for BNER without any manual feature engineering. Experimental results show that domain-specific pre-trained word embeddings and character-level representation can improve the performance of the LSTM-RNN models. On the GM corpus, we achieve comparable performance compared with other systems using complex hand-crafted features. Considering the JNLPBA corpus, our model achieves the best results, outperforming the previously top performing systems. The source code of our method is freely available under GPL at https://github.com/lvchen1989/BNER .

Preliminary study on pathogenesis of bronchial asthma in children.

BACKGROUND: The etiology and pathogenesis of bronchial asthma remain unclear. This study is to investigate the risk factors related to bronchial asthma onset in children from genetics and immunology and preliminarily reveal the pathogenesis of bronchial asthma in children. METHODS: Real-time quantitative PCR was adopted to detect the expression level of TRPV1 gene and mRNA and enzyme-linked immunosorbent assay method to the total immunoglobulin E level and levels of IL-4, IL-5, and IFN-γ in serum in peripheral venous blood for children in two groups. Logistic regression analysis was applied to analyze the most essential factors inducing bronchial asthma in children. RESULTS: The mRNA level of TRPV1 in peripheral blood in the case group was higher than that in the control group (P < 0.01). The levels of IL-4, IL-5, and eosinophils in serum in the case group were markedly higher than those in the control group (P < 0.01), while IFN-γ level in the case group was lower than that in the control group (P < 0.01). The results of logistic regression analysis indicated that TRPV1 expression level, IL-4 level, and rs4790522 site mutation were the main risk factors inducing bronchial asthma in children. CONCLUSION: The levels of TRPV1 gene expression and Th1/Th2 cytokines have a close relationship with asthma onset in children, which provides theoretical evidences for molecular targeted treatment in children with bronchial asthma.

Endothelial H2S-AMPK dysfunction upregulates the angiocrine factor PAI-1 and contributes to lung fibrosis.

Dysfunction of the vascular angiocrine system is critically involved in regenerative defects and fibrosis of injured organs. Previous studies have identified various angiocrine factors and found that risk factors such as aging and metabolic disorders can disturb the vascular angiocrine system in fibrotic organs. One existing key gap is what sense the fibrotic risk to modulate the vascular angiocrine system in organ fibrosis. Here, using human and mouse data, we discovered that the metabolic pathway hydrogen sulfide (H2S)-AMP-activated protein kinase (AMPK) is a sensor of fibrotic stress and serves as a key mechanism upregulating the angiocrine factor plasminogen activator inhibitor-1 (PAI-1) in endothelial cells to participate in lung fibrosis. Activation of the metabolic sensor AMPK was inhibited in endothelial cells of fibrotic lungs, and AMPK inactivation was correlated with enriched fibrotic signature and reduced lung functions in humans. The inactivation of endothelial AMPK accelerated lung fibrosis in mice, while the activation of endothelial AMPK with metformin alleviated lung fibrosis. In fibrotic lungs, endothelial AMPK inactivation led to YAP activation and overexpression of the angiocrine factor PAI-1, which was positively correlated with the fibrotic signature in human fibrotic lungs and inhibition of PAI-1 with Tiplaxtinin mitigated lung fibrosis. Further study identified that the deficiency of the antioxidative gas metabolite H2S accounted for the inactivation of AMPK and activation of YAP-PAI-1 signaling in endothelial cells of fibrotic lungs. H2S deficiency was involved in human lung fibrosis and H2S supplement reversed mouse lung fibrosis in an endothelial AMPK-dependent manner. These findings provide new insight into the mechanism underlying the deregulation of the vascular angiocrine system in fibrotic organs.

Nonautoregressive Encoder-Decoder Neural Framework for End-to-End Aspect-Based Sentiment Triplet Extraction.

Aspect-based sentiment triplet extraction (ASTE) aims at recognizing the joint triplets from texts, i.e., aspect terms, opinion expressions, and correlated sentiment polarities. As a newly proposed task, ASTE depicts the complete sentiment picture from different perspectives to better facilitate real-world applications. Unfortunately, several major challenges, such as the overlapping issue and long-distance dependency, have not been addressed effectively by the existing ASTE methods, which limits the performance of the task. In this article, we present an innovative encoder-decoder framework for end-to-end ASTE. Specifically, the ASTE task is first modeled as an unordered triplet set prediction problem, which is satisfied with a nonautoregressive decoding paradigm with a pointer network. Second, a novel high-order aggregation mechanism is proposed for fully integrating the underlying interactions between the overlapping structure of aspect and opinion terms. Third, a bipartite matching loss is introduced for facilitating the training of our nonautoregressive system. Experimental results on benchmark datasets show that our proposed framework significantly outperforms the state-of-the-art methods. Further analysis demonstrates the advantages of the proposed framework in handling the overlapping issue, relieving long-distance dependency and decoding efficiency.

Effects of Various Cell Surface Engineering Reactions on the Biological Behavior of Mammalian Cells.

Cell surface engineering technologies can regulate cell function and behavior by modifying the cell surface. Previous studies have mainly focused on investigating the effects of cell surface engineering reactions and materials on cell activity. However, they do not comprehensively analyze other cellular processes. This study exploits covalent bonding, hydrophobic interactions, and electrostatic interactions to modify the macromolecules succinimide ester-methoxy polyethylene glycol (NHS-mPEG), distearoyl phosphoethanolamine-methoxy polyethylene glycol (DSPE-mPEG), and poly-L-lysine (PLL), respectively, on the cell surface. This work systematically investigates the effects of the three surface engineering reactions on the behavior of human umbilical vein endothelial cells (HUVECs) and human skin fibroblasts, including viability, growth, proliferation, cell cycle, adhesion, and migration. The results reveals that the PLL modification method notably affects cell viability and G2/M arrest and has a short modification duration. However, the DSPE-mPEG and NHS-mPEG modification methods have little effect on cell viability and proliferation but have a prolonged modification duration. Moreover, the DSPE-mPEG modification method highly affects cell adherence. Further, the NHS-mPEG modification method can significantly improve the migration ability of HUVECs by reducing the area of focal adhesions. The findings of this study will contribute to the application of cell surface engineering technology in the biomedical field.

Double Modification of Poly(urethane-urea): Toward Healable, Tear-Resistant, and Mechanically Robust Elastomers for Strain Sensors.

Polyurethane elastomers with mechanical robustness, tear resistance, and healing efficiency hold great potential in wearable sensors and soft robots. However, achieving excellent mechanical properties and healable capability simultaneously remains highly desirable but exclusive. Herein, we propose a straightforward procedure for double modification of poly(urethane-urea) (PUU) via thiolactone chemistry, and two different dynamic cross-linking bonds (disulfide linkages and Zn2+/imidazole coordination) are successively incorporated into the side chain of PUU, producing double cross-linking elastomers (PUU-I/Zn-S). The synergy between disulfide linkages and Zn2+/imidazole coordination forms a robust and dynamic network, endowing PUU-I/Zn-S with excellent mechanical and healing properties. The tensile stress, elongation at break, and toughness of the resultant elastomer can reach 44.06 MPa, 1000%, and 181.93 MJ m-3, respectively. Meanwhile, PUU-I/Zn-S exhibits outstanding tearing resistance with a tearing energy of 42.1 kJ m-2. The PUU-I/Zn-S can restore its mechanical robustness after self-healing at room temperature (25 ± 2 °C) or 60 °C and even maintain 91% of its original tensile strength after reprocessing two times. Additionally, PUU-I/Zn-S-based strain sensors are fabricated by introducing conductive nanofillers and demonstrate remarkable sensing capability to diverse human body motions. This work demonstrates a simple and feasible method for the postfunctionalization and enhancement of polyurethane and provides some insights into reconciling the traditional contradictory properties of mechanical robustness and healing efficiency.

Effect of Yiyuan moxibustion on urodynamics and TRPV4/ATP/P2X5 signal pathway of bladder tissue in rats with neurogenic bladder after sacral cord injury. / ç›Šå ƒç¸å¯¹éª¶é«“æŸä¼¤åŽç¥žç»æºæ€§è†€èƒ±å¤§é¼ å°¿æµåŠ¨åŠ›å­¦åŠè†€èƒ±ç»„ç»‡TRPV4/ATP/P2X5信号通路的影响.

OBJECTIVES: To observe the effect of Yiyuan moxibustion on urodynamics and the expressions of transient receptor potential vanilloid 4 (TRPV4), adenosine triphosphate (ATP), tyrosine protein kinase KIT (C-Kit) and adenosine triphosphate receptor P2X5 in bladder tissue of rats with detrusor reflex-free neurogenic bladder (NB) after sacral cord injury (SCI), so as to explore its mechanism in promoting the recovery of urination function of NB rats. METHODS: Female SD rats were randomly divided into sham operation, model, Yiyuan moxibustion, Yiyuan moxibustion+inhibitor (combination) and inhibitor groups, with 12 rats in each group. The model of detruser reflex-free NB after sacral SCI was established by modified Hassan Shaker spinal cord transection method. The behavioral score of Basso Beasttie Bresnahan (BBB) and urodynamic indexes were used to evaluate the model of rats after operation. Fifteen days after modeling, Yiyuan moxibustion was applied to "Shenque" (CV8) and "Guanyuan" (CV4) for 20 min, once daily for 14 days. Rats of the inhibitor and combination groups were given intravesical instillation of HC067047 (1 mL, 1 µmol/L, 30 min). After the interventions, urodynamics was used to evaluate the bladder function of rats. HE staining was used to observe the morphology of bladder tissue. ATP content in bladder tissue was detected by colorimetric method. The positive expression rates of C-Kit and their receptor P2X5 in bladder tissue were observed by immunofluorescence double labeling method, and TRPV4, C-Kit, and P2X5 protein expression levels in bladder tissue were detected by Western blot. RESULTS: Compared with the sham operation group, the maximum bladder capacity and bladder compliance of rats in the model group were increased (P<0.01), the leak point pressure, ATP content, the possitive expression rates of C-Kit and P2X5, and the protein expression levels of TRPV4, C-Kit, P2X5 in bladder tissue were decreased (P<0.01). In comparison with the model and combination groups, the Yiyuan moxibustion group showed a decrease in maximum bladder capacity and bladder compliance (P<0.01), an increase in leakage point pressure, ATP content, the possitive expression rates of C-Kit and P2X5, and TRPV4, C-Kit, and P2X5 protein expression levels (P<0.01, P<0.05)；However, these indicators showed opposite trends in the inhibitor group (P<0.01, P<0.05). CONCLUSIONS: Yiyuan moxibustion can improve the urodynamics and bladder function in rats with bladder detrusor nonreflective after SCI, which may be related to its effect in activating the TRPV4 channel in bladder tissue, promoting the release of ATP from bladder epithelium, thus increasing the expression of bladder Cajal interstitial cells and their purinergic P2X5 receptors.

[Efficacy differences between electroacupuncture and moxibustion for neurogenic bladder after spinal cord injury: a randomized controlled trial].

OBJECTIVE: To compare the clinical efficacy between electroacupuncture(EA) and moxibustion for neurogenic bladder (NB) after spinal cord injury (SCI). METHODS: One hundred and twenty patients with NB after SCI were randomly divided into an EA group, a moxibustion group, and an intermittent catheterization group, with 40 patients in each group. The patients in the intermittent catheterization group were treated with routine treatment and intermittent catheterization, while the patients in the EA group and the moxibustion group were treated with additional treatments of EA (discontinuous wave, with a frequency of 1.3-1.6 Hz, and intensity based on patient tolerance) and moxibustion, respectively. The acupoints used in both groups were Zhongji (CV 3) and Guanyuan (CV 4), bilateral Zusanli (ST 36), Yinlingquan (SP 9), and Baliao points. Each session lasted for 30 min, once daily, six times a week, for a total of six weeks.The maximum bladder capacity (MBC), residual urine vdume (RUV), detrusor pressure (Pdet) during the filling phase, bladder compliance (BC), maximum renal pelvis separation width of both kidneys, urine white blood cell count, TCM syndrome score, and World Health Organization quality of life assessment-BREF (WHOQOL-BREF) score were compared before and after treatment in the 3 groups. The number of patients in each group who achieved bladder functional balance was recorded, and the clinical efficacy was assessed after treatment. RESULTS: After treatment, the MBC, Pdet, BC, and WHOQOL-BREF scores in the EA group and the moxibustion group were increased (P<0.05), while the RUV, maximum renal pelvis separation width of both kidneys, urine white blood cell count, and TCM syndrome scores were decreased (P<0.05, P<0.01). In the intermittent catheterization group, MBC, RUV, maximum renal pelvis separation width of both kidneys, and urine white blood cell count were decreased (P<0.05), while BC and WHOQOL-BREF score were increased (P<0.05) after treatment. After treatment, the MBC, Pdet, BC, and WHOQOL-BREF scores in the EA group and the moxibustion group were higher than those in the intermittent catheterization group (P<0.05), while the RUV and TCM syndrome scores were lower than those in the intermittent catheterization group (P<0.05). Moreover, after treatment, the MBC and Pdet in the moxibustion group were higher than those in the EA group (P<0.05), while the RUV, maximum renal pelvis separation width of both kidneys, and TCM syndrome score in the EA group were lower than those in the moxibustion group (P<0.05). The number of patients who achieved bladder functional balance after treatment in the EA group and the moxibustion group was higher than that in the intermittent catheterization group (P<0.05). The cured and effective rate was 85.0% (34/40) in the EA group and 82.5% (33/40) in the moxibustion group, which were both higher than 65.0% (26/40) in the intermittent catheterization group (P<0.05), there was no significant difference between the EA group and the moxibustion group (P>0.05). CONCLUSION: EA and moxibustion could effectively improve the functional state of bladder in patients with NB after SCI. EA is more effective in reducing residual urine volume and excessive activity of the urethral sphincter, and relieving TCM syndromes, while moxibustion is more effective in increasing the pressure of the detrusor during the filling period and establishing the detrusor reflex.

Multi-supervised bidirectional fusion network for road-surface condition recognition.

Rapid developments in automatic driving technology have given rise to new experiences for passengers. Safety is a main priority in automatic driving. A strong familiarity with road-surface conditions during the day and night is essential to ensuring driving safety. Existing models used for recognizing road-surface conditions lack the required robustness and generalization abilities. Most studies only validated the performance of these models on daylight images. To address this problem, we propose a novel multi-supervised bidirectional fusion network (MBFN) model to detect weather-induced road-surface conditions on the path of automatic vehicles at both daytime and nighttime. We employed ConvNeXt to extract the basic features, which were further processed using a new bidirectional fusion module to create a fused feature. Then, the basic and fused features were concatenated to generate a refined feature with greater discriminative and generalization abilities. Finally, we designed a multi-supervised loss function to train the MBFN model based on the extracted features. Experiments were conducted using two public datasets. The results clearly demonstrated that the MBFN model could classify diverse road-surface conditions, such as dry, wet, and snowy conditions, with a satisfactory accuracy and outperform state-of-the-art baseline models. Notably, the proposed model has multiple variants that could also achieve competitive performances under different road conditions. The code for the MBFN model is shared at https://zenodo.org/badge/latestdoi/607014079.

Optimizing Attention for Sequence Modeling via Reinforcement Learning.

Attention has been shown highly effective for modeling sequences, capturing the more informative parts in learning a deep representation. However, recent studies show that the attention values do not always coincide with intuition in tasks, such as machine translation and sentiment classification. In this study, we consider using deep reinforcement learning to automatically optimize attention distribution during the minimization of end task training losses. With more sufficient environment states, iterative actions are taken to adjust attention weights so that more informative words receive more attention automatically. Results on different tasks and different attention networks demonstrate that our model is of great effectiveness in improving the end task performances, yielding more reasonable attention distribution. The more in-depth analysis further reveals that our retrofitting method can help to bring explainability for baseline attention.

Ring -opening of polythiolactones to construct protein-imprinted nanospheres with high recognition and regulation capabilities.

Developing and preparing novel protein-imprinted nanomaterials with high recognition ability remains challenging because it is difficult to controllably and orderly design and arrange functional groups on the imprinted polymer layers of protein-imprinted nanomaterials to improve their protein identification. Herein, we present a new technology using rationally designed polythiolactone-decorated magnetic nanospheres as the precursor of multifunctionalized imprinted materials. Moreover, the strategy of ring-opening the polythiolactione layers using primary amines with terminal alcohols, acids and pyrrolidines introduces abundant recognition sites, which enhance the recognition for template proteins through multiple hydrogen-bonding and hydrophobic interactions. Thiols generated in situ by the ring-opening reaction provide sufficient crosslinking sites proximate to each recognition site for the formation of imprinting cavities, endowing the imprinted nanospheres with promising regulation capabilities. Based on the rational design, the imprinted nanospheres can be prepared conveniently and present tunable rebinding capacity and specificity for bovine serum albumin (BSA). The maximum saturated rebinding capacity of imprinted materials for BSA is up to 285 ± 15 mg g-1 and the highest imprinting factor reaches 5.79. The simple and versatile strategy demonstrated in this study shows promise for the design of other protein-imprinted materials with high recognition ability.

[Effects of eye acupuncture on motor evoked potential and somatosensory evoked potential in patients with incomplete spinal cord injury based on neuroelectrophysiological technology].

OBJECTIVE: To observe the effects of eye acupuncture on motor evoked potential (MEP) and somatosensory evoked potential (SEP) in the patients with incomplete spinal cord injury so as to evaluate its clinical efficacy. METHODS: According to the random number table, 90 patients were divided into exercise therapy group, eye acupuncture group and eye acupuncture combined exercise therapy group (combined treatment group), 30 cases in each. In the exercise therapy group, patients were treated with the routine exercise and occupational therapy. Patients of the eye acupuncture group were treated with eye acupuncture at upper jiao region, lower jiao region, liver region and kidney region bilaterally. Patients of the combined treatment group were given the routine exercise and occupational therapy combined with eye acupuncture. All the treatments were conducted once daily, 7 days as one treatment course for 4 treatment courses. Before treatment and 4 weeks after treatment, the motor function, light touch sensation and pinprick sensation, injury grade and clinical efficacy were assessed separately, using the criteria developed by the American Spinal Injury Association. The modified Barthel index(MBI) was adopted to evaluate the activities of daily livings. By monitoring SEP and MEP, the neurophysiological conditions were assessed for spinal cord injury. RESULTS: The total effective rate was 56.7% (17/30), 66.7% (20/30) and 90.0% (27/30) in the exercise therapy group, the eye acupuncture group and the combined treatment group, respectively. The total effective rate in the combined treatment group was higher than those in the other two groups (P<0.05). Compared with those before treatment, the scores of motor function, light tough sensation and pinprick sensation were all increased after treatment in three groups (P<0.05), MBI score was increased in both the exercise therapy group and the combined treatment group (P<0.05), and the latency of SEP (N11, N20, N23, P38) and the Cortical (hand region), Csp, Cortical (leg region) and Lsp of MEP were all shortened in the three groups separately (P<0.05). After treatment, compared with the exercise therapy group, the score of motor function was increased (P<0.05), MBI score decreased (P<0.05) and MEP latency shortened (P<0.05) in the eye acupuncture group. After treatment, compared with the exercise therapy group and the eye acupuncture group, the scores of motor function, light touch sensation and pinprick sensation, as well as MBI score were all increased (P<0.05), and the latency of SEP (N11,N20,N23,P38) and MEP shortened (P<0.05) in the combined treatment group. CONCLUSION: In treatment of incomplete spinal cord injury, eye acupuncture combined with exercise therapy can significantly increase the excitability of sensory and motor nerve conduction in the spinal cord and cerebral cortex of patients, effectively promote the recovery of patients' motor and sensory function and improve the activities of daily living.