Search

Rui Zheng,1,* Yaxian Jiang,1,* Cheng Yan,2,* Yikun Li,1 Xiaozhou Song,3 Pengcheng Zheng4 1Department of Clinical Laboratory, The First People’s Hospital of Yunnan Province, The Affiliated Hospital of Kunming University of Science and Technology, Kunming, Yunnan, People’s Republic of China; 2Medical School, Kunming University of Science and Technology, The First People’s Hospital of Yunnan, Kunming, People’s Republic of China; 3Department of Infection Control, The First People’s Hospital of Yunnan Province, The Affiliated Hospital of Kunming University of Science and Technology, Kunming, Yunnan, People’s Republic of China; 4Department of Pharmacy, The First People’s Hospital of Yunnan Province, The Affiliated Hospital of Kunming University of Science and Technology, Kunming, Yunnan, People’s Republic of China*These authors contributed equally to this workCorrespondence: Pengcheng Zheng, Department of Pharmacy, The First People’s Hospital of Yunnan Province, No. 157 of Jingbi Road, Kunming, People’s Republic of China, Tel +86 013888116045, Email zpc1997@163.com Xiaozhou Song, Department of Infection Control, The First People’s Hospital of Yunnan Province, No. 157 of Jingbi Road, Kunming, People’s Republic of China, Tel +86 013888144965, Email 13888144965@163.comIntroduction: Ventilator-associated pneumonia (VAP) is an ICU (intensive care unit)-acquired pulmonary parenchymal infection that is complicated by mechanical ventilation and is associated with high morbidity and mortality. Klebsiella pneumoniae (KPN) is known to asymptomatically colonize the gastrointestinal tract and may increase the incidence of corresponding VAP. Our study aims were to investigate the exact origin of the carbapenem-resistant Klebsiella pneumoniae (CRKP) causing VAP in our patient.Methods: Various environmental samples, including the patient’s anal swab, were collected in order to find the source of the bacteria. Minimum inhibitory concentrations (MICs) for antimicrobial agents were determined according to the guidelines of the Clinical and Laboratory Standards Institute (CLSI); resistant genes were detected by using PCR and sequencing; clone relationships were analyzed by using multilocus-sequence typing (MLST) and pulsed field gel electrophoresis (PFGE). The IAP values were obtained via urinary catheter.Results: One CRKP strain was detected in the patient’s anal swab; this strain was confirmed with the same gene type as the strain isolated from the sputum. We found that the patient’s intra-abdominal pressure (IAP) was 29.41, 27.06, 24.12, and 22.66 mmHg; the IAP was either equal to or above 12 mmHg, on the operation day and the following three days. Intra-abdominal hypertension (IAH) occurred during the patient’s hospitalization and was considered to be caused by the surgical procedure. Meanwhile, we found that there was a correlation between IAH and the detection of CRKP in the sputum. The findings suggested that his VAP was caused by intestinal colonial KPN, and not from the environment.Discussion: Our research illustrated that the ST11 KPC-2-producing strain colonized the intestinal tract and caused the development of VAP when the IAP was elevated. Routine screening for the intestinal carriage of CRKP, among patients in ICUs, can limit and prevent current and future outbreaks.Keywords: ventilator-associated pneumonia, intra-abdominal pressure, intra-abdominal hypertension, Klebsiella pneumoniae, healthcare-associated infection

Large language models (LLMs) have formulated a blueprint for the advancement of artificial general intelligence. Its primary objective is to function as a human-centric (helpful, honest, and harmless) assistant. Alignment with humans assumes paramount significance, and reinforcement learning with human feedback (RLHF) emerges as the pivotal technological paradigm underpinning this pursuit. Current technical routes usually include \textbf{reward models} to measure human preferences, \textbf{Proximal Policy Optimization} (PPO) to optimize policy model outputs, and \textbf{process supervision} to improve step-by-step reasoning capabilities. However, due to the challenges of reward design, environment interaction, and agent training, coupled with huge trial and error cost of large language models, there is a significant barrier for AI researchers to motivate the development of technical alignment and safe landing of LLMs. The stable training of RLHF has still been a puzzle. In the first report, we dissect the framework of RLHF, re-evaluate the inner workings of PPO, and explore how the parts comprising PPO algorithms impact policy agent training. We identify policy constraints being the key factor for the effective implementation of the PPO algorithm. Therefore, we explore the PPO-max, an advanced version of PPO algorithm, to efficiently improve the training stability of the policy model. Based on our main results, we perform a comprehensive analysis of RLHF abilities compared with SFT models and ChatGPT. The absence of open-source implementations has posed significant challenges to the investigation of LLMs alignment. Therefore, we are eager to release technical reports, reward models and PPO codes, aiming to make modest contributions to the advancement of LLMs.

Prompting methods such as Chain-of-Thought (CoT) have shed new light on enhancing the reasoning capabilities of large language models, and researchers have extensively explored the generation process of rationales and answers. However, they have overlooked the potential challenges posed by the poor quality of reasoning problems, which may influence the reasoning performance significantly. In this work, we propose Self-Polish (SP), a novel method that facilitates the model's problem-solving process by prompting them to progressively refine the given problems to be more comprehensible and solvable. Specifically, the method teaches models to eliminate irrelevant information, rearrange the logic structure and organize local conditions into new ones parallelly. SP is orthogonal to all other prompting methods, making it convenient to integrate with state-of-the-art techniques for further improvement. We conduct thorough experiments on five benchmarks to illustrate the effectiveness of the proposed method. For example, with Text-davinci-003, our method boosts the performance of standard few-shot prompting by $8.0\%$ on GSM8K and $17.8\%$ on MultiArith; it also improves the performance of CoT by $6.0\%$ on GSM8K and $6.0\%$ on MathQA, respectively. Furthermore, our method also showcases impressive performance on robustness evaluation., Preprint

Models trained with empirical risk minimization (ERM) are revealed to easily rely on spurious correlations, resulting in poor generalization. Group distributionally robust optimization (group DRO) can alleviate this problem by minimizing the worst-case loss over pre-defined groups. While promising, in practice factors like expensive annotations and privacy preclude the availability of group labels. More crucially, when taking a closer look at the failure modes of out-of-distribution generalization, the typical procedure of reweighting in group DRO loses efficiency. Hinged on the limitations, in this work, we reformulate the group DRO framework by proposing Q-Diversity. Characterized by an interactive training mode, Q-Diversity relaxes the group identification from annotation into direct parameterization. Furthermore, a novel mixing strategy across groups is presented to diversify the under-represented groups. In a series of experiments on both synthetic and real-world text classification tasks, results demonstrate that Q-Diversity can consistently improve worst-case accuracy under different distributional shifts, outperforming state-of-the-art alternatives., Findings of ACL 2023

Objective: The objective of this study is to develop a deep-learning based detection and diagnosis technique for carotid atherosclerosis using a portable freehand 3D ultrasound (US) imaging system. Methods: A total of 127 3D carotid artery datasets were acquired using a portable 3D US imaging system. A U-Net segmentation network was firstly applied to extract the carotid artery on 2D transverse frame, then a novel 3D reconstruction algorithm using fast dot projection (FDP) method with position regularization was proposed to reconstruct the carotid artery volume. Furthermore, a convolutional neural network was used to classify the healthy case and diseased case qualitatively. 3D volume analysis including longitudinal reprojection algorithm and stenosis grade measurement algorithm was developed to obtain the clinical metrics quantitatively. Results: The proposed system achieved sensitivity of 0.714, specificity of 0.851 and accuracy of 0.803 respectively in diagnosis of carotid atherosclerosis. The automatically measured stenosis grade illustrated good correlation (r=0.762) with the experienced expert measurement. Conclusion: the developed technique based on 3D US imaging can be applied to the automatic diagnosis of carotid atherosclerosis. Significance: The proposed deep-learning based technique was specially designed for a portable 3D freehand US system, which can provide carotid atherosclerosis examination more conveniently and decrease the dependence on clinician's experience.

Significant advancements have been made in developing parametric models for digital humans, with various approaches concentrating on parts such as the human body, hand, or face. Nevertheless, connectors such as the neck have been overlooked in these models, with rich anatomical priors often unutilized. In this paper, we introduce HACK (Head-And-neCK), a novel parametric model for constructing the head and cervical region of digital humans. Our model seeks to disentangle the full spectrum of neck and larynx motions, facial expressions, and appearance variations, providing personalized and anatomically consistent controls, particularly for the neck regions. To build our HACK model, we acquire a comprehensive multi-modal dataset of the head and neck under various facial expressions. We employ a 3D ultrasound imaging scheme to extract the inner biomechanical structures, namely the precise 3D rotation information of the seven vertebrae of the cervical spine. We then adopt a multi-view photometric approach to capture the geometry and physically-based textures of diverse subjects, who exhibit a diverse range of static expressions as well as sequential head-and-neck movements. Using the multi-modal dataset, we train the parametric HACK model by separating the 3D head and neck depiction into various shape, pose, expression, and larynx blendshapes from the neutral expression and the rest skeletal pose. We adopt an anatomically-consistent skeletal design for the cervical region, and the expression is linked to facial action units for artist-friendly controls. HACK addresses the head and neck as a unified entity, offering more accurate and expressive controls, with a new level of realism, particularly for the neck regions. This approach has significant benefits for numerous applications and enables inter-correlation analysis between head and neck for fine-grained motion synthesis and transfer., Comment: Find HACK model on https://github.com/ZoneLikeWonderland/HACK-Model

The GPT-3.5 models have demonstrated impressive performance in various Natural Language Processing (NLP) tasks, showcasing their strong understanding and reasoning capabilities. However, their robustness and abilities to handle various complexities of the open world have yet to be explored, which is especially crucial in assessing the stability of models and is a key aspect of trustworthy AI. In this study, we perform a comprehensive experimental analysis of GPT-3.5, exploring its robustness using 21 datasets (about 116K test samples) with 66 text transformations from TextFlint that cover 9 popular Natural Language Understanding (NLU) tasks. Our findings indicate that while GPT-3.5 outperforms existing fine-tuned models on some tasks, it still encounters significant robustness degradation, such as its average performance dropping by up to 35.74\% and 43.59\% in natural language inference and sentiment analysis tasks, respectively. We also show that GPT-3.5 faces some specific robustness challenges, including robustness instability, prompt sensitivity, and number sensitivity. These insights are valuable for understanding its limitations and guiding future research in addressing these challenges to enhance GPT-3.5's overall performance and generalization abilities.

The principle of continual relation extraction~(CRE) involves adapting to emerging novel relations while preserving od knowledge. While current endeavors in CRE succeed in preserving old knowledge, they tend to fail when exposed to contaminated data streams. We assume this is attributed to their reliance on an artificial hypothesis that the data stream has no annotation errors, which hinders real-world applications for CRE. Considering the ubiquity of noisy labels in real-world datasets, in this paper, we formalize a more practical learning scenario, termed as \textit{noisy-CRE}. Building upon this challenging setting, we develop a noise-resistant contrastive framework named as \textbf{N}oise-guided \textbf{a}ttack in \textbf{C}ontrative \textbf{L}earning~(NaCL) to learn incremental corrupted relations. Compared to direct noise discarding or inaccessible noise relabeling, we present modifying the feature space to match the given noisy labels via attacking can better enrich contrastive representations. Extensive empirical validations highlight that NaCL can achieve consistent performance improvements with increasing noise rates, outperforming state-of-the-art baselines.

Audio-visual speech enhancement (AV-SE) aims to enhance degraded speech along with extra visual information such as lip videos, and has been shown to be more effective than audio-only speech enhancement. This paper proposes further incorporating ultrasound tongue images to improve lip-based AV-SE systems' performance. Knowledge distillation is employed at the training stage to address the challenge of acquiring ultrasound tongue images during inference, enabling an audio-lip speech enhancement student model to learn from a pre-trained audio-lip-tongue speech enhancement teacher model. Experimental results demonstrate significant improvements in the quality and intelligibility of the speech enhanced by the proposed method compared to the traditional audio-lip speech enhancement baselines. Further analysis using phone error rates (PER) of automatic speech recognition (ASR) shows that palatal and velar consonants benefit most from the introduction of ultrasound tongue images., Comment: To be published in InterSpeech 2023

Large language models have unlocked strong multi-task capabilities from reading instructive prompts. However, recent studies have shown that existing large models still have difficulty with information extraction tasks. For example, gpt-3.5-turbo achieved an F1 score of 18.22 on the Ontonotes dataset, which is significantly lower than the state-of-the-art performance. In this paper, we propose InstructUIE, a unified information extraction framework based on instruction tuning, which can uniformly model various information extraction tasks and capture the inter-task dependency. To validate the proposed method, we introduce IE INSTRUCTIONS, a benchmark of 32 diverse information extraction datasets in a unified text-to-text format with expert-written instructions. Experimental results demonstrate that our method achieves comparable performance to Bert in supervised settings and significantly outperforms the state-of-the-art and gpt3.5 in zero-shot settings.

Hyperspectral Image (HSI) classification based on deep learning has been an attractive area in recent years. However, as a kind of data-driven algorithm, the deep learning method usually requires numerous computational resources and high-quality labelled datasets, while the expenditures of high-performance computing and data annotation are expensive. In this paper, to reduce the dependence on massive calculation and labelled samples, we propose a deep Double-Channel dense network (DDCD) for Hyperspectral Image Classification. Specifically, we design a 3D Double-Channel dense layer to capture the local and global features of the input. And we propose a Linear Attention Mechanism that is approximate to dot-product attention with much less memory and computational costs. The number of parameters and the consumptions of calculation are observably less than contrapositive deep learning methods, which means DDCD owns simpler architecture and higher efficiency. A series of quantitative experiences on 6 widely used hyperspectral datasets show that the proposed DDCD obtains state-of-the-art performance, even though when the absence of labelled samples is severe.

Two previously undescribed 3(4), 8(9), 9(10)-trisecocycloartane triterpene dilactones with a featured eleven-membered ring, pseudoladilactones A and B (1-2), together with three known ones (4-5), were isolated from the 80 % ethanol extract of the cones of Pseudolarix amabilis. The structures of compounds 1 and 2 were determined on the basis of analysis of 1D- and 2D-NMR spectroscopic data. The absolute configuration of pseudoladilactones A (1) was elucidated by single crystal X-ray diffraction (CuKα). All isolates were tested for cytototoxicity against four human tumor cell lines A549, HCT116, SK-BR-3, and HepG2.

Adversarial training is one of the most powerful methods to improve the robustness of pre-trained language models (PLMs). However, this approach is typically more expensive than traditional fine-tuning because of the necessity to generate adversarial examples via gradient descent. Delving into the optimization process of adversarial training, we find that robust connectivity patterns emerge in the early training phase (typically $0.15\sim0.3$ epochs), far before parameters converge. Inspired by this finding, we dig out robust early-bird tickets (i.e., subnetworks) to develop an efficient adversarial training method: (1) searching for robust tickets with structured sparsity in the early stage; (2) fine-tuning robust tickets in the remaining time. To extract the robust tickets as early as possible, we design a ticket convergence metric to automatically terminate the searching process. Experiments show that the proposed efficient adversarial training method can achieve up to $7\times \sim 13 \times$ training speedups while maintaining comparable or even better robustness compared to the most competitive state-of-the-art adversarial training methods., EMNLP 2022

Semantic segmentation is a classic computer vision problem dedicated to labeling each pixel with its corresponding category. As a basic task for advanced tasks such as industrial quality inspection, remote sensing information extraction, medical diagnostic aid, and autonomous driving, semantic segmentation has been developed for a long time in combination with deep learning, and a lot of works have been accumulated. However, neither the classic FCN-based works nor the popular Transformer-based works have attained fine-grained localization of pixel labels, which remains the main challenge in this field. Recently, with the popularity of autonomous driving, the segmentation of road scenes has received increasing attention. Based on the cross-task consistency theory, we incorporate edge priors into semantic segmentation tasks to obtain better results. The main contribution is that we provide a model-agnostic method that improves the accuracy of semantic segmentation models with zero extra inference runtime overhead, verified on the datasets of road and non-road scenes. From our experimental results, our method can effectively improve semantic segmentation accuracy.

Vehicle recognition and monitoring are gaining importance in traffic management. However, due to the various sizes of cars, detection remains a difficulty, which directly impacts the accuracy of vehicle counts. The suggested vehicle recognition and counting method first extracts the highway road surface in the image and divides it into a distant regions. A newly developed segmentation strategy in the proposed vehicle identification and counting system first extracts a nd s eparates t he highway road Turface in the image into a distant region and a proximal area; the method is crucial for improving vehicle detection. The aforementioned locations are then sent to the YOLOv5m network to determine the vehicle’s kind and position. Finally, the ORB algorithm is utilized to create vehicle trajectories, which may be used to estimate the driving direction of the vehicle and determine the number of distinct cars. Several traffic surveillance recordings from various settings are utilized to validate the suggested meUhodoMoHZ. 5he experimental findings demonstrate that the suggested segmentation approach may give greater detection accuracy, particularly for the detection of little automobile things. In addition, the vehicle detection performance was significantly improved by 99.39 % of m AP compared to the Yolov5 basic. This work has broad practical implications for managing and controlling vehicle objects in traffic scenes.

IntroductionThe aim of this work was to determine the feasibility of using a deep learning approach to predict occult lymph node metastasis (OLM) based on preoperative FDG-PET/CT images in patients with clinical node-negative (cN0) lung adenocarcinoma.Materials and MethodsDataset 1 (for training and internal validation) included 376 consecutive patients with cN0 lung adenocarcinoma from our hospital between May 2012 and May 2021. Dataset 2 (for prospective test) used 58 consecutive patients with cN0 lung adenocarcinoma from June 2021 to February 2022 at the same center. Three deep learning models: PET alone, CT alone, and combined model, were developed for the prediction of OLM. The performance of the models was evaluated on internal validation and prospective test in terms of accuracy, sensitivity, specificity, and areas under the receiver operating characteristic curve (AUCs).ResultsThe combined model incorporating PET and CT showed the best performance, achieved an AUC of 0.81 [95% confidence interval (CI): 0.61, 1.00] in the prediction of OLM in internal validation set (n = 60) and an AUC of 0.87 (95% CI: 0.75, 0.99) in the prospective test set (n = 58). The model achieved 87.50% sensitivity, 80.00% specificity, and 81.00% accuracy in the internal validation set and achieved 75.00% sensitivity, 88.46% specificity, and 86.60% accuracy in the prospective test set.ConclusionThis study presented a deep learning approach to enable the prediction of occult nodal involvement based on the PET/CT images before surgery in cN0 lung adenocarcinoma, which would help clinicians select patients who would be suitable for sublobar resection.

Aim Emperipolesis, the cell-in-cell structure, is one of the pathological diagnostic standard of AIH. We have already identified that CD8+ T cells participated in the emperipolesis process in liver tissue of AIH patients. In this study, we aimed to clarify the characteristics and molecular mechanisms of emperipolesis in the patients with AIH in vitro and in mice with α-GalCer-induced acute hepatitis. Methods Peripheral blood mononuclear cell (PBMC) of patients with variant chronic liver diseases and healthy controls were co-cultured with hepatic cell line to induce emperipolesis in vitro. Confocal staining was performed to illustrate the cellular types and potential mechanisms of emperipolesis in AIH. In addition, the mouse model of α-GalCer-induced acute hepatitis which mimics human AIH in several features, was used to confirm the role of CD44/p-ERM/F-actin in the emperipolesis process in vivo. Results In co-culture system with PBMC and hepatic cell line, emperipolesis was observed most common in the patients with AIH. CD8+ T cells were the major cells participated in the emperipolesis process, and penetrated into hepatocytes actively via CD44/p-ERM/F-actin pathway. As a result, most CD8+ T cells engulfed in hepatocytes undergo self-apoptosis. In α-GalCer-induced acute hepatitis model, emperipolesis was observed around the inflammatory foci, and was inhibited by Ezrin phosphorylation inhibitor NSC668394. Similarly, activated murine CD8+ T cells penetrated into primary hepatocytes via CD44/p-ERM/F-actin pathway in vitro. Conclusions CD8+ T cells penetrated into hepatocytes actively via CD44/p-ERM/F-actin signaling pathway and undergo apoptosis, may be a compensative mechanism to attenuate overwhelming immune attack in AIH. This article is protected by copyright. All rights reserved.

Recently, nitrate as one of the water pollutants has caused widespread concerns. The nitrite and ammonium of nitrate reduction products have also been regarded as contaminants. Thus, increasing the nitrate removal rate and improving the nitrogen selectivity simultaneously have attracted the attention of researchers. To remove the nitrate with high nitrogen selectivity from an aqueous solution, 2D g-C3N4 was intercalated into 3D TiO2/Ti3C2 to fabricate the TiO2/Ti3C2/g-C3N4 photocatalyst with 3D/2D ternary heterojunctions. 0.5 TiO2/Ti3C2/g-C3N4 photocatalyst exhibited the highest nitrate conversion and nitrogen selectivity, which reached 93.03% nitrate conversion and 96.62% nitrogen selectivity after 40 min of irradiation. Moreover, its photocatalytic performance remained stable after 5 cycles. It was observed that the high concentration of nitrite was advantageous for improving the nitrogen selectivity. In TiO2/Ti3C2/g-C3N4 composites, 3D Ti3C2 served not only as a template for preparing 2D g-C3N4, but also as a channel for the electron transfer between TiO2 and g-C3N4. Photogenerated electrons in the conduction band of TiO2 migrated to the valence band of g-C3N4 through Ti3C2. The corresponding holes remained on the valence band of TiO2 and were captured by formic acid in an aqueous solution to generate reducing COO˙−. Then, COO˙− in the solution and photoinduced electrons in the conduction band of g-C3N4 reduced nitrate to nontoxic and harmless nitrogen with high selectivity. Finally, the Z-scheme heterojunction was proposed over 0.5 TiO2/Ti3C2/g-C3N4 photocatalyst on working for the photocatalytic nitrate reduction in water.

Abstract Four new 3,4-secocycloartane triterpenoids, pseudolactones A–D (1–4), were isolated from the ethanol extract of the cones of Pseudol arixamabilis. Their structures were established by extensive 1D- and 2D-NMR experiments. The cones of P. arixamabilis are enriched in the ring-expanded or cleaved cycloartane triterpenoids. This work provides new insight into cycloartane triterpenoids from the cones of P. arixamabilis. Graphic Abstract

This study aims to evaluate the efficacy and safety of tanezumab administered as a fixed dosing regimen in patients with knee or hip osteoarthritis. Randomized controlled phase III trials (RCTs) that evaluated the efficacy and safety associated with tanezumab for knee or hip OA were systematically identified by searching electronic databases, including Cochrane Library, PubMed, and Embase, for 30 years from December 1990 up to July 2020. Ten relevant studies were included in our meta-analysis. The research was reported based on the preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to ensure the reliability and verity of results. Our study showed that the tanezumab groups were more effective than the placebo groups in terms of mean change from the baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) pain (P < .00001), WOMAC physical functional (P < .00001), and patient's global assessment (PGA) (P < .00001). Discontinued due to adverse events (AEs) (P < .00001), serious AEs (P = .02), abnormal peripheral sensations (both P < .00001), and peripheral neuropathy (P < .002) in tanezumab groups were significantly higher than that in control groups. Compared with placebo, tanezumab can effectively relieve pain and improve WOMAC physical function and patient's global assessment (PGA) in knee and hip osteoarthritis. Meanwhile, adverse events are transient in nature and generally well-tolerated. However, we still need large-sample, high-quality studies to investigate the long-term safety of tanezumab to give the conclusion.

Eggerthella lenta is part of the normal bacterial flora of the intestinal tract, but may cause life-threatening infections. E. lenta has been isolated from blood, abscesses, wounds, skin ulcers, and intra-abdominal infections. However, due to historical difficulties with laboratory identification by conventional biochemical methods, some cases of E. lenta infection have previously gone undiagnosed, especially in China where E. lenta infections are very rare. Recently, matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS), a new type of technology, has made it possible to identify E. lenta. We report a case of postoperative infection caused by E. lenta which was treated successfully using meropenem.

Sophorae tonkinensis Radix et Rhizoma (S. tonkinensis) has been recorded as a 'poisonous' Chinese herbal medicine in Chinese Pharmacopoeia 2020. The clinical reaction reports of S. tonkinensis indicated its neurotoxicity; however, there still exists dispute about its toxic substances. At present, no report is available on the blood and brain prototype research of S. tonkinensis. Most studies focused on alkaloids and less on other compounds. Moreover, the constituents absorbed into the blood and brain have been rarely investigated so far. This study established a rapid and efficient qualitative analysis method using UPLC-Q-TOF-MS

A novel and mild metal-free catalyzed radical-mediated cascade spirocyclization of

Postoperative surgical site infection is one of the serious postoperative complications of spine surgery, especially fungal infections. Late-stage surgical site fungal infections often lack typical clinical symptoms and have a variable clinical presentation. In this case, the patient was a senior patient with usually only tolerable pain and discomfort, which was detected 2 years after the first surgery. Such cases are even rarer for fungal bloodstream infections caused by delayed postoperative chronic fungal osteomyelitis and deserve further study for early identification and intervention to minimize harm.

Supplementary Material for The association between preoperative blood calcium and postoperative blood loss in patients undergoing heart valve replacement surgery by Mingke Li, Yawei Dai, Shuai Hu, Yansong Li, Dasheng Lu, Rui Zheng, and Kai Wang in Perfusion.

Additional file 1: Table S1. Worst–Best/Best–Worst case analyses of the primary outcome. Table S2. Worst-possible/best-possible case analyses of the 72-h SOFA score. Table S3. Cox multivariate analysis of factors influencing 90-day mortality in patients. Fig. S1. Subgroup analysis of 90-day mortality. The forest map shows the grouped factors of the subgroup analysis, HR for 90-day mortality, 95% CI in each subgroup, and P value for interaction of treatment (intervention or placebo) and the factor. There were no significant interactions in any of the subgroups (P > 0.1 for all comparisons).

In this thesis, I evaluated the impact of pre-centrifugation sample management on the plasma metabolome, providing verification for the protocol used in a study where I applied metabolomics to better understand the role of body composition in fracture occurrence using a large population-based cohort. In paper 1, I applied untargeted liquid chromatography-mass spectrometry (LC-MS) metabolomics analysis to plasma samples (n = 471) from healthy donors to predict and evaluate the effect of pre-centrifugation temperature and delay time on metabolomics data using a combination of random forest and generalized linear models. I found that delayed treatment (up to 210 min) of blood samples can induce variation to the plasma metabolome at all three temperatures (4, 25 and 37 °C) and thus blood samples should be managed as expediently as possible, preferably at room temperature. Metabolite panels were constructed and validated to predict the pre-centrifugation management conditions of plasma samples. In paper 2, I used a full cohort of men and women (n = 44,366), and a subcohort of women (n = 5022) to investigate the associations between anthropometrics or body composition and fracture risk. I found that lower body mass index (BMI), prior loss of BMI and higher central obesity were associated with increased risks of fracture. The joint analysis revealed that prior gain of BMI was protective against fracture only in individuals with normal weight. The associations between lean mass index (LMI) or fat mass index (FMI) and fracture risk were heavily dependent on bone mineral density (BMD) in the study population. Based on findings of paper 2, a metabolome-wide association of LMI, FMI and BMD was conducted in the subcohort. Metabolites associated with at least one body composition outcome were then associated with hip fracture risk, resulting in the discovery of twenty-three metabolites of interest. Network analysis of these hip fracture-related metabolites highlighted the important role of lysoglycerophospholipids in the skeletomuscular-adipose crosstalk. Mendelian randomization analyses suggested several causal associations between these hip fracture-related metabolites and body composition measures or bone-related clinical biomarkers. In summary, metabolites can reflect perturbations induced by pre-centrifugation management of samples and thus provide valuable information of sample quality. To optimally prevent fractures at older age, low BMI, prior BMI loss and central obesity should be avoided. Metabolites reflecting differences in, and crosstalk of skeletomuscular-adipose tissues may contribute to a further understanding of fracture occurrence and discoveries of actionable targets for prevention.

Supplemental material, sj-docx-1-aop-10.1177_10600280221130280 for A Meta-Analysis on the Impact of Prenatal and Early Childhood Antimicrobial Use on Autism Spectrum Disorders by Li-guo Li, Hong-guang Fu, Yong-hong Zhao, Peng-ju Zhao, Qing-kai Meng, Rui-juan Zheng and En-yao Li in Annals of Pharmacotherapy

Natural language understanding (NLU) models tend to rely on spurious correlations (i.e., dataset bias) to achieve high performance on in-distribution datasets but poor performance on out-of-distribution ones. Most of the existing debiasing methods often identify and weaken these samples with biased features (i.e., superficial surface features that cause such spurious correlations). However, down-weighting these samples obstructs the model in learning from the non-biased parts of these samples. To tackle this challenge, in this paper, we propose to eliminate spurious correlations in a fine-grained manner from a feature space perspective. Specifically, we introduce Random Fourier Features and weighted re-sampling to decorrelate the dependencies between features to mitigate spurious correlations. After obtaining decorrelated features, we further design a mutual-information-based method to purify them, which forces the model to learn features that are more relevant to tasks. Extensive experiments on two well-studied NLU tasks demonstrate that our method is superior to other comparative approaches., Comment: Accepted as a long paper at COLING 2022 (Oral)

Three-dimensional (3D) ultrasound imaging technique has been applied for scoliosis assessment, but current assessment method only uses coronal projection image and cannot illustrate the 3D deformity and vertebra rotation. The vertebra detection is essential to reveal 3D spine information, but the detection task is challenging due to complex data and limited annotations. We propose VertMatch, a two-step framework to detect vertebral structures in 3D ultrasound volume by utilizing unlabeled data in semi-supervised manner. The first step is to detect the possible positions of structures on transverse slice globally, and then the local patches are cropped based on detected positions. The second step is to distinguish whether the patches contain real vertebral structures and screen the predicted positions from the first step. VertMatch develops three novel components for semi-supervised learning: for position detection in the first step, (1) anatomical prior is used to screen pseudo labels generated from confidence threshold method; (2) multi-slice consistency is used to utilize more unlabeled data by inputting multiple adjacent slices; (3) for patch identification in the second step, the categories are rebalanced in each batch to solve imbalance problem. Experimental results demonstrate that VertMatch can detect vertebra accurately in ultrasound volume and outperforms state-of-the-art methods. VertMatch is also validated in clinical application on forty ultrasound scans, and it can be a promising approach for 3D assessment of scoliosis., Comment: 15 pages, 8 figures

Additional file 1: Supplementary Materials and Methods. Table S1. Demographic and clinical characteristics of the subjects who participated in RNA-Seq and proteogenomic analysis. Table S2. Five differentially expressed circRNAs and significant fold changes. Table S3. Demographic and clinical characteristics of the subjects in FISH analysis. Table S4. Demographic and clinical characteristics of subjects with colorectal cancer and cancer-free controls. Table S5. Characteristics of patients with other types of cancer. Table S6. Demographic and clinical characteristics of the colorectal cancer patients before and after operation. Table S7. Characterization of circLPAR1-binding proteins. Table S8. The correlation between protein and its parental gene expression and prediction of RNA binding ability. Table S9. Sequences of circLPAR1 used for FISH analysis in the study. Table S10. Sequences of primers used for RT-qPCR in the study. Figure S1. The expression of circRNAs detected by RT-qPCR. Figure S2. The expression of circLPAR1 in CRC cells treated with actinomycin D and colorectal tumors. Figure S3. Characterization of exosomes derived from human plasma. Figure S4. The expression of traditional clinical biomarkers in plasma. Figure S5. ROC curve analysis of exosomal circLPAR1 and traditional clinical biomarkers in human plasma. Figure S6. Fluorescence observation of colorectal cancer cells after incubation with PKH67-labelled exosomes or non-exosomes. Figure S7. Fluorescence observation of colorectal cancer cells incubation with PKH67-labelled exosomes pretreated with cytochalasin D. Figure S8. The expression of circLPAR1 in exosomes derived from colorectal cancer cells. Figure S9. The effect of exosomal circLPAR1 on invasion and migration abilities of HCT116 cells. Figure S10. The effect of circLPAR1 overexpression on colorectal cancer cellular phenotypes. Figure S11. The cellular localization of circLPAR1 in colorectal cancer cells. Figure S12. MS2-CP-Flag was pulled down by anti-flag and evaluated by Western blot. Figure S13. The proteomic analysis showing 23 proteins levels based on 25 paired colorectal cancer tissues. Figure S14. Mass spectrometry analysis identification of eIF3h. Figure S15. The effect of BRD4 regulation by circLPAR1 on colorectal cancer cellular phenotype. Figure S16. The enrichment of miR-762 in the complex with MS2-CP-Flag. Figure S17. Effect of exosomal circLPAR1 on colorectal cancer progression in vivo.

G-C3N4 was supported on the surface of foamed ceramic, and the g-C3N4@foamed ceramic was packed in the photocatalytic reactor in a layered manner. Effects of urea and H2O2 concentration on NO conversion were investigated. Pulse experiments were carried out to investigate the change of NO conversion with time under different concentrations of H2O2. The contribution of each route that NO converted and the selectivities of products were calculated. Results showed that the photo-reduction of NO accounted for 2%, and the photo-oxidation of NO accounted for 14%, and the rest was absorbed by the humidifier. Products include ammonia (NH + 4-N), nitrogen (N2), nitrite (NO- 2-N) and nitrate (NO- 3-N), of which N2 accounted for 9%.

Humans have been suffering from vitiligo for a long time. Target vitiligo drugs have yet been approved. Activation of Wnt/β-catenin signalling has potential in the therapeutic use of vitiligo, so exploring new drugs that specifically directly activate Wnt is worthwhile to obtain new anti-vitiligo agents. In this work, two portions design and synthesis were put into effect. firstly, 17 phenanthridine derivatives with C-4 substitutes were designed and synthesized, which compounds 4, 6, 12, 13 served as H-acceptor with protein showed enhance melanogenesis activity; Secondly, 7 hybrid new scaffolds of compounds were designed and synthesized, scaffold hopping compound 36 that aromatic benzene was replaced pyrazole on ring C showed enhance melanogenesis and tyrosinase activity; The last and most important, a comprehensive optimization and SARs of compound 36 were carried out, compounds 41 and 43 shared phenolic hydroxyl or 3-methyl-pyridine substitutes at C-7 position remarkably improved the capacity of melanogenesis and tyrosinase activity. Compound 43 were identified as new anti-vitiligo agents that specifically activate the Wnt/β-catenin signalling pathway by targeting Axin. Structure-activity relationship analysis implied that H-acceptor substitutions at the C-4 position and phenolic hydroxyl or pyridine substitutions at the C-7 position would improve the activities of the compounds. These findings reveal a new therapeutic strategy for vitiligo, and compounds 41 and 43 may represent potential compounds for vitiligo treatment.

Zheng RP, Ma DK, Li Z, Zhang HF. Cancer Manag Res. 2020;12:6165–6175. The Editor and Publisher of Cancer Management and Research wish to issue an Expression of Concern for the published article. Following publication, concerns were raised about the scientific integrity of the articles content. We reached out to the authors requesting they supply information that would confirm the integrity of the data. However, the authors did not provide the requested files and could not satisfy our requirements to confirm the authenticity of some types of data. Therefore, as we continue to work through the issues raised, we advise readers to interpret the information presented in the article with due caution. The authors have been notified about this Expression of Concern. Read the original article

Hydrodynamic mixing is one of the important environment factors in determining phytoplankton community compositions. Here the influences of continuous hydrodynamic mixing on abundance, morphology, and dominance of Microcystis were investigated in microcosm and lab experiments. Our research results showed that Cyanophyta contributed 57.16% to the total biomass in control, but Chlorophyta was the dominant group in continuous hydrodynamic mixing (CHM) group, contributing 76.54% to the total biomass in the microcosm experiment. The average number of Microcystis in control was 1.95 folds in CHM group. However, the mean abundance of Scenedesmus quadricauda and Pseudanabaena limnetica in CHM was 2.47 and 2.97 folds in control. In the lab experiment, the average number of Microcystis flos-aquae in control was 2.97 folds in CHM group. The mean size of M. flos-aquae colony in control (34.65 μm) group were significantly bigger than that in the CHM (26.78 μm) group. This research results demonstrated that continuous hydrodynamic mixing weakened the dominance of Microcystis, but was beneficial for the others algae (S. quadricauda and P. limnetica) and is helpful in understanding the effect of hydrodynamic mixing on Microcystis blooms in freshwater ecosystem.

A practical method is developed for the synthesis of oxazolidinone derivatives, an important class of heterocyclic compounds. The effect of bases and solvents on this cyclization reaction is discussed and a simple new base–solvent system (triethylamine in toluene) is found to be the most effective. The reaction conditions developed here are mild and no by-products are observed. Moreover, using optimized conditions, a number of differently substituted propargylic alcohols are cyclized to the corresponding N-substituted-4-methylene-oxazolidinones in yields of up to 99%.

Objectives Collagen triple helix repeat containing-1 (CTHRC1) is a highly conserved extracellular matrix glycoprotein that is overexpressed in two murine models of cholestatic liver fibrosis. Elevated CTHRC1 has been found to attenuate liver fibrosis in these murine models, thus we aimed to study the expression of CTHRC1 in patients with cholestatic liver diseases and its correlation with hepatic conditions. Methods Ninety patients with chronic liver disease, including 48 had primary biliary cholangitis (PBC), 18 had primary sclerosing cholangitis (PSC) and 24 had chronic hepatitis B (CHB), together with five healthy controls (HC), were recruited to this study. Participants' liver sections were analyzed using immunohistochemistry. Serum CTHRC1 levels in another cohort of 59 patients with PBC and 10 age-matched HC were detected by enzyme-linked immunosorbent assay. Results CTHRC1 protein was primarily expressed in activated hepatic stellate cells (HSC). CTHRC1 expression was significantly increased in the PBC and PSC groups, compared with the HC and CHB groups. Importantly, the hepatic fibrosis stage of the PBC group was positively correlated with hepatic CTHRC1 expression (r = 0.425, P = 0.003). Meanwhile, there were significant correlations between serum CTHRC1 levels and both the degrees of hepatic inflammation and fibrosis stage in the PBC group (r = 0.300, P = 0.022; r = 0.321, P = 0.012). Conclusion CTHRC1 may play a role in hepatic fibrogenesis in PBC and that serum CTHRC1 may be a potential novel noninvasive biomarker in the assessment of liver fibrosis and inflammation.

The melanocortin-3 receptor (MC3R) is known to be involved in regulation of energy homeostasis, regulating feed efficiency and nutrient partitioning in mammals. Its physiological roles in non-mammalian vertebrates, especially economically important aquaculture species, are not well understood. Channel catfish (Ictalurus punctatus) is the main freshwater aquaculture species in North America. In this study, we characterized the channel catfish MC3R. The mc3r of channel catfish encoded a putative protein (ipMC3R) of 367 amino acids. We transfected HEK293T cells with ipMC3R plasmid for functional studies. Five agonists, including adrenocorticotropin, α-melanocyte stimulating hormone (α-MSH), β-MSH, [Nle4, D-Phe7]-α-MSH, and D-Trp8-γ-MSH, were used in the pharmacological studies. Our results showed that ipMC3R bound β-MSH with higher affinity and D-Trp8-γ-MSH with lower affinity compared with human MC3R. All agonists could stimulate ipMC3R and increase intracellular cAMP production with sub-nanomolar potencies. The extracellular signal-regulated kinases 1 and 2 (ERK1/2) activation could also be triggered by ipMC3R. The ipMC3R exhibited constitutive activities in both cAMP and ERK1/2 pathways, and Agouti-related protein served as an inverse agonist at ipMC3R, potently inhibiting the high basal cAMP level. Moreover, we showed that melanocortin receptor accessory protein 2 (MRAP2) preferentially modulated ipMC3R in cAMP production rather than ERK1/2 activation. Our study will assist further investigation of the physiological roles of the ipMC3R, especially in energy homeostasis, in channel catfish.

The embedding method of fiber Bragg grating sensors (FBGs) has a significant impact on the measurability and reliability, as a result of the monitoring of FBGs depending on the strain transmission from the measured object. In this study, structural adhesive was used to join the FBGs with steel wire, and the relevant measurability and reliability of the adhesive-bonded FBGs with steel wire were investigated by loading & unloading cyclic test and fatigue test in the application of FBGs to monitor bridge cable force. It was found that there was a linear relationship between the load and the wavelength for both the loading and the unloading phases, and the data line from the loading process was parallel to that of the unloading process. In addition, 200 million fatigue cycles of steel wire decreased slightly the wavelength of FBGs when the steel wire was applied the same load. The differences in load vs. wavelength between loading and unloading phases, and with and without fatigue tests were likely attributed by the mechanical properties of steel wire, namely, the load-displacement curves of steel wire were responsible for the change in load vs. wavelength. Nevertheless, our experimental results indicated that the FBGs bonded steel wire using structural adhesive had an exemplary measurability and reliability, which will find more application in monitoring bridge cables.

Emperipolesis is a pathological feature for the diagnosis of autoimmune hepatitis (AIH). We have previously found that CD8The peripheral blood mononuclear cells (PBMC) of patients with various chronic liver diseases and healthy controls were co-cultured with hepatic cell lines to induce emperipolesis in vitro. Confocal staining was performed to illustrate the cellular types and potential mechanisms of emperipolesis in AIH. In addition, a murine model of α-GalCer-induced acute hepatitis that mimics human AIH was used to confirm the role of CD44/p-ERM/F-actin in the emperipolesis process in vivo.In the co-cultured system of PBMC and hepatic cell line, emperipolesis was observed most commonly in patients with AIH. The main cells participating in emperipolesis were CD8CD8

Objective: The spinous process angle (SPA) is one of the essential parameters to denote three-dimensional (3-D) deformity of spine. We propose an automatic segmentation method based on Stacked Hourglass Network (SHN) to detect the spinous processes (SP) on ultrasound (US) spine images and to measure the SPAs of clinical scoliotic subjects. Methods: The network was trained to detect vertebral SP and laminae as five landmarks on 1200 ultrasound transverse images and validated on 100 images. All the processed transverse images with highlighted SP and laminae were reconstructed into a 3D image volume, and the SPAs were measured on the projected coronal images. The trained network was tested on 400 images by calculating the percentage of correct keypoints (PCK); and the SPA measurements were evaluated on 50 scoliotic subjects by comparing the results from US images and radiographs. Results: The trained network achieved a high average PCK (86.8%) on the test datasets, particularly the PCK of SP detection was 90.3%. The SPAs measured from US and radiographic methods showed good correlation (r>0.85), and the mean absolute differences (MAD) between two modalities were 3.3{\deg}, which was less than the clinical acceptance error (5{\deg}). Conclusion: The vertebral features can be accurately segmented on US spine images using SHN, and the measurement results of SPA from US data was comparable to the gold standard from radiography., Comment: 9 pages,5 figures

Osteopenia is indicated as a common phenomenon in patients who have scoliosis. Quantitative ultrasound (QUS) has been used to assess skeletal status for decades, and recently ultrasound imaging using reflection signals from vertebra were as well applied to measure spinal curvatures on children with scoliosis. The objectives of this study are to develop a new method which can robustly extract a parameter from ultrasound spinal data for estimating bone quality of scoliotic patients and to investigate the potential for the parameter on predicting curve progression. The frequency amplitude index (FAI) was calculated based on the spectrum of the original radio frequency (RF) signals reflected from the tissue-vertebra interface. The correlation between FAI and reflection coefficient was validated using decalcified bovine bone samples in vitro, and the FAIs of scoliotic subjects were investigated in vivo referring to BMI, Cobb angles and curve progression status. The results showed that the intra-rater measures were highly reliable between different trials (ICC=0.997). The FAI value was strongly correlated to the reflection coefficient of bone tissue ($R^{2}=0.824$), and the lower FAI indicated the higher risk of curve progression for the non-mild cases. This preliminary study reported that the FAI method can provide a feasible and promising approach to assess bone quality and monitor curve progression of the patients who have AIS., Comment: typos corrected, positions of figures' captions corrected