Search

This letter investigates a wireless communication system deploying distributed reconfigurable intelligent surfaces (RISs). Existing works have assumed that perfect timing synchronization is available among all the cooperative RISs. For practical considerations, we first study cooperative reflection design for multi-RIS-aided communication systems taking into account timing synchronization errors. We aim to minimize the mean-squared error of the recovered data in the presence of timing offsets subject to the unit modulus constraints imposed on the phase shifts of the RISs. In order to handle this sophisticated nonconvex problem, we develop a computationally-efficient algorithm based on the majorization-minimization framework where the RIS phase shift matrices and the timing offset equalizer are respectively developed in a semi-closed form and a closed form. Numerical examples validate the improved performance of our proposed design compared with various schemes.

To study the MRI characteristics of early extrapancreatic necrosis and compare them with those of peripancreatic fluid collections in acute pancreatitis (AP).This retrospective study enrolled 70 AP patients who had extrapancreatic collections visible on MRI within 1 week of onset. Extrapancreatic collections were divided into extrapancreatic necrosis and peripancreatic fluid collections based on follow-up MRI, CT, or pathology. The number and area of extrapancreatic collections, extrapancreatic inflammation on MRI (EPIM) score, MR severity index score and clinical characteristics were evaluated and compared between the two groups.Of the seventy AP patients, 32 (45.7%) had extrapancreatic necrosis, and 38 (54.3%) had peripancreatic fluid collections. The number and area of extrapancreatic collections, MR severity index score, EPIM score, and prevalence of associated hemorrhage were significantly higher in extrapancreatic necrosis patients than in those with peripancreatic fluid collections (p0.001). Among the single indicators, the accuracy of the area of extrapancreatic collections (AUC = 0.871) was comparable to that of the EPIM score for predicting extrapancreatic necrosis and was significantly higher than that of the other two indicators. The combination of all indicators showed the highest predictive accuracy (AUC = 0.949), and combinations of two or more indicators demonstrated significantly higher predictive accuracy for extrapancreatic necrosis than any single indicator (p0.05) except for the area of extrapancreatic collections (p0.05).The MRI characteristics have the potential to differentiate early extrapancreatic necrosis from peripancreatic fluid collections and help indicate extrapancreatic necrosis.

For the design and optimization of a tubular gas–liquid atomization mixer, the atomization and mixing characteristics of liquid jet breakup in the limited tube space is a key problem. In this study, the primary breakup process of liquid jet column was analyzed by high-speed camera, then the droplet size and velocity distribution of atomized droplets were measured by Phase-Doppler anemometry (PDA). The hydrodynamic characteristics of gas flow in tubular gas–liquid atomization mixer were analyzed by computational fluid dynamics (CFD) numerical simulation. The results indicate that the liquid flow rate has little effect on the atomization droplet size and atomization pressure drop, and the gas flow rate is the main influence parameter. Under all experimental gas flow conditions, the liquid jet column undergoes a primary breakup process, forming larger liquid blocks and droplets. When the gas flow rate (Qg) is less than 127 m3·h−1, the secondary breakup of large liquid blocks and droplets does not occur in venturi throat region. The Sauter mean diameter (SMD) of droplets measured at the outlet is more than 140 μm, and the distribution is uneven. When Qg > 127 m3·h−1, the large liquid blocks and droplets have secondary breakup process at the throat region. The SMD of droplets measured at the outlet is less than 140 μm, and the distribution is uniform. When 127

Background For patients with acute ischemic stroke undergoing endovascular mechanical thrombectomy with x-ray angiography, the use of adjuncts to maintain vessel patency, such as stents or antiplatelet medications, can increase risk of periprocedural complications. Criteria for using these adjuncts are not well defined. Purpose To evaluate use of MRI to guide critical decision making by using a combined biplane x-ray neuroangiography 3.0-T MRI suite during acute ischemic stroke intervention. Materials and Methods This retrospective observational study evaluated consecutive patients undergoing endovascular intervention for acute ischemic stroke between July 2019 and May 2020 who underwent either angiography with MRI or angiography alone. Cerebral tissue viability was assessed by using MRI as the reference standard. For statistical analysis, Fisher exact test and Student t test were used to compare groups. Results Of 47 patients undergoing acute stroke intervention, 12 patients (median age, 69 years; interquartile range, 60-77 years; nine men) underwent x-ray angiography with MRI whereas the remaining 35 patients (median age, 80 years; interquartile range, 68-86 years; 22 men) underwent angiography alone. MRI results influenced clinical decision making in one of three ways: whether or not to perform initial or additional mechanical thrombectomy, whether or not to place an intracranial stent, and administration of antithrombotic or blood pressure medications. In this initial experience, decision making during endovascular acute stroke intervention in the combined angiography-MRI suite was better informed at MRI, such that therapy was guided in real time by the viability of the at-risk cerebral tissue. Conclusion Integrating intraprocedural 3.0-T MRI into acute ischemic stroke treatment was feasible and guided decisions of whether or not to continue thrombectomy, to place stents, or to administer antithrombotic medication or provide blood pressure medications. © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Lev and Leslie-Mazwi in this issue.

Background: Coronavirus disease 2019 (COVID-19) has spread rapidly in China and globally. In order to control the spread of the epidemic, it is important to find an efficient diagnostic method. Objectives: The aim of this study was to assess the responses of antibodies during SARS-CoV-2 infection in relation to disease severity and to evaluate the association between the positive rate of antibody detection and nucleic acid test. Methods: Ninety patients with SARS-CoV-2 infection were recruited in this retrospective observational study. Demographic, clinical data, and SARS-CoV-2 IgM and IgG antibodies in serum specimens were detected at 4 and 6 weeks after diagnosis. Results: IgM and IgG antibody levels showed a decreased tendency, the titers at week 4 were higher than the titers at week 6: The positive rates of IgM at week 4 and 6 were 92.9% and 67.9%, respectively. The positive rates of IgG at week 4 and week 6 were 100%. No association was found between the positive rate of antibody detection at week 4 or 6 and that of nucleic acid test (P>0.05). No difference between the positive rate of antibodies against SARS-CoV-2 in severe and non-severe COVID-19 patients was observed. Conclusions: Antibody detection is an effective means in the diagnosis of COVID-19. The titer and positive rate of IgM are lower than those of IgG in the first six weeks after infection. Positive rate of antibodies was not different between severe and non-severe patients.

Objective A large amount of clinical data are stored in clinical notes that frequently contain spelling variations, typos, local practice-generated acronyms, synonyms, and informal words. Instead of relying on established but infrequently updated ontologies with keywords limited to formal language, we developed an artificial intelligence (AI) assistant (named “DeepSuggest”) that interactively offers suggestions to expand or pivot queries to help overcome these challenges. Methods We applied an unsupervised neural network (Word2Vec) to the clinical notes to build keyword contextual similarity matrix. With a user's input query, DeepSuggest generates a list of relevant keywords, including word variations (e.g., formal or informal forms, synonyms, abbreviations, and misspellings) and other relevant words (e.g., related diagnosis, medications, and procedures). Human intelligence is then used to further refine or pivot their query. Results DeepSuggest learns the semantic and linguistic relationships between the words from a large collection of local notes. Although DeepSuggest is only able to recall 0.54 of Systematized Nomenclature of Medicine (SNOMED) synonyms on average among the top 60 suggested terms, it covers the semantic relationship in our corpus for a larger number of raw concepts (6.3 million) than SNOMED ontology (24,921) and is able to retrieve terms that are not stored in existing ontologies. The precision for the top 60 suggested words averages at 0.72. Usability test resulted that DeepSuggest is able to achieve almost twice the recall on clinical notes compared with Epic (average of 5.6 notes retrieved by DeepSuggest compared with 2.6 by Epic). Conclusion DeepSuggest showed the ability to improve retrieval of relevant clinical notes when implemented on a local corpus by suggesting spelling variations, acronyms, and semantically related words. It is a promising tool in helping users to achieve a higher recall rate for clinical note searches and thus boosting productivity in clinical practice and research. DeepSuggest can supplement established ontologies for query expansion.

Objective This study sought to investigate the effect of local expression of galectin-3 in the development of stenotic arteriovenous fistula (AVF). Methods We collected stenotic venous tissues, adjacent nonstenotic venous tissues, and blood samples from end-stage renal disease (ESRD) patients with AVF stenosis, while normal venous tissues and blood samples were collected from ESRD patients before AVF creation as controls. Also blood samples were collected from ESRD patients with nonstenosis functional AVF. Galectin-3, proliferating cell nuclear antigen (PCNA), matrix metalloproteinase-9 (MMP-9), and α-SMA expression in the venous tissues were examined by immunohistochemistry, and the ERK1/2 pathway activity in the intima was accessed by western blot. Serum galectin-3 level was measured by ELISA. Thereafter, human pulmonary arterial smooth muscle cells (HPASMCs) were cultured in vitro, and the interaction between Galectin-3 and ERK1/2 pathway in HPASMCs was estimated by western blot. Results ESRD patients with stenotic AVF had a significant higher serum galectin-3 level than normal controls, and patients with non-stenotic functional AVF. The expression levels of galectin-3, phosphorylated ERK1/2, PCNA, MMP-9, and α-SMA in the stenotic venous tissues were higher than that in the normal venous tissues or the adjacent nonstenotic AVF venous tissues. Correlation analysis showed that the expression of galectin-3 of the neointima was positively correlated with PCNA and α-SMA in the stenotic AVF venous tissues. In HPASMCs, galectin-3 can increase the activity of phosphorylated ERK1/2 and promote the expression of α-SMA. Conclusion In the stenotic AVF of ESRD patients, expression of the galectin-3 was significantly increased, showing a positive relation with neointima development.

Existing text sentiment analysis methods mostly rely on a large number of language knowledge and sentiment resources. This paper proposes the Multi-channel convolution and bidirectional GRU multi-head attention capsule (AT-MC-BiGRU-Capsule), which uses vector neurons to replace scalar neurons to model text emotions, and uses capsules to characterize text emotions. In addition, traditional methods cannot extract the multi-level features of text sequence well. Multi-head attention can encode the dependencies between words, capture sentiment words in text, and using Convolutional Neural Network (CNN) and Bidirectional gated recurrent unit network (Bi-GRU) to extract local features and global semantic features of text respectively, the global average pooling layer is introduced to obtain the multi-level feature representation of the text sequence more comprehensively. This paper selects three English datasets and one Chinese dataset in the general corpus of sentiment classification to conduct experiments, and achieves better results than other baseline models.

Background and Purpose: The optimal recanalization strategy for acute ischemic stroke with large vessel occlusion continues to be an area of active interest. Network meta-analysis can provide insight when direct comparative evidence is lacking. Methods: A systematic review of the literature using PubMed, Embase, the Cochrane Central Register of Controlled Trials, and SinoMed was performed, and a search was conducted for clinical trials on ClinicalTrials.gov, the World Health Organization International Clinical Trials Registry Platform, and StrokeCenter.org. Four independent reviewers conducted the study selection, data abstraction, and quality assessments. Results: The literature review identified 17 trials including 3236 patients and 8 ongoing clinical trials. Sample sizes ranged from 7 to 656 participants. Intravenous thrombolysis (IVT) was the most common intervention, followed by IVT plus mechanical thrombectomy (MT), IVT plus intraarterial thrombolysis, intraarterial thrombolysis alone, and MT alone. In the pooled network meta-analysis, IVT+MT was associated with a higher rate of independent functioning. In contrast, IVT was ranked as the most ineffective treatment strategy with respect to neurological functions, while direct MT was ranked as the least safe intervention with respect to all-cause mortality. Also, irrespective of assessment tools, endovascular treatment plus IVT led to higher successful recanalization rate than thrombolysis alone. Conclusions: Compared with other recanalization treatments, IVT+MT seems to be the most effective strategy, without increasing detrimental effects, for thrombolysis-eligible patients with large vessel occlusion-acute ischemic stroke. To improve the current evidentiary basis for recanalization treatment, future trials and real-world studies are warranted and should use unified definitions of symptomatic intracranial hemorrhage and recanalization.

Background: Cardiac injury is common in severe coronavirus disease 2019 (COVID-19) and is associated with poor outcomes. We aimed to study predictors of in-hospital death, characteristics of arrhythmias and the effects of QT-prolonging therapy in patients with cardiac injury. Methods: We conducted a retrospective cohort study involving patients with severe COVID-19 who were admitted to Tongji Hospital in Wuhan, China, between Jan. 29 and Mar. 8, 2020. Among patients who had cardiac injury, which we defined as an elevated level of cardiac troponin I (cTnI), we identified demographic and clinical characteristics associated with mortality and need for invasive ventilation. Results: Among 1284 patients with severe COVID-19, 1159 had a cTnI level measured on admission to hospital, of whom 170 (14.7%) had results that showed cardiac injury. We found that mortality was markedly higher in patients with cardiac injury (71.2% v. 6.6%, p < 0.001). We determined that initial cTnI (per 10-fold increase, hazard ratio [HR] 1.32, 95% confidence interval [CI] 1.06–1.66) and peak cTnI level during illness (per 10-fold increase, HR 1.70, 95% CI 1.38–2.10) were associated with poor survival. Peak cTnI was also associated with the need for invasive ventilation (odds ratio 3.02, 95% CI 1.92–4.98). We found arrhythmias in 44 of the 170 patients with cardiac injury (25.9%), including 6 patients with ventricular tachycardia or fibrillation, all of whom died. We determined that patients who received QT-prolonging drugs had longer QTc intervals than those who did not receive them (difference in medians, 45 ms, p = 0.01), but such treatment was not independently associated with mortality (HR 1.04, 95% CI 0.69–1.57). Interpretation: We found that in patients with COVID-19 and cardiac injury, initial and peak cTnI levels were associated with poor survival, and peak cTnI was a predictor of need for invasive ventilation. Patients with COVID-19 warrant assessment for cardiac injury and monitoring, especially if therapy that can prolong repolarization is started. Trial registration: Chinese Clinical Trial Registry, No. ChiCTR2000031301.

Gentianella acuta (G. acuta), as a folk medicine, was used to treat heart disease by the Ewenki people in Inner Mongolia. However, the effect of G. acuta on acute myocardial infarction (AMI) is not clear. To explore the mechanisms of G. acuta on isoproterenol (ISO)-induced AMI, rats were administered G. acuta for 28 days, then injected intraperitoneally with ISO (85 mg/kg) on days 29 and 30. An electrocardiogram helped to evaluate the myocardial injury. Serum lactate dehydrogenase (LDH), creatinine kinase (CK) and aspartate aminotransferase (AST) levels were evaluated, and haematoxylin eosin, Masson's trichrome staining and terminal deoxynucleotidyl transferase dUTP nick end labelling (TUNEL) staining were used to detect myocardial histological changes. Radioimmunoassay was used to measure serum tumour necrosis factor alpha (TNFα) and interleukin (IL)-6. An enzyme-linked immunosorbent assay kit was used to analyse serum galectin-3 (Gal-3) levels. Immunohistochemistry, Western blotting and reverse transcription polymerase chain reaction were used to examine relevant molecular events. The results revealed that pre-treatment with G. acuta decreased the elevation in the ST segment; reduced serum LDH, CK and AST levels; alleviated cardiac structure disorder; and reduced inflammatory infiltration, abnormal collagen deposition and cardiomyocyte apoptosis that were induced by ISO. Furthermore, pre-treatment with G. acuta inhibited serum Gal-3 levels and Gal-3 expression in heart tissue, and also impeded TLR4/MyD88/NF-кB signalling activation, which ultimately prevented the expression of inflammatory cytokines. The study indicated that pre-treatment with G. acuta protects against ISO-induced AMI, and the protective role may be related to inhibiting Gal-3/TLR4/MyD88/NF-кB inflammatory signalling.

Merchants of e-commerce Websites expect recommender systems to entice more consumption which is highly correlated with the customers' purchasing propensity. However, most existing recommender systems focus on customers' general preference rather than purchasing propensity often governed by instant demands which we deem to be well conveyed by the questions asked by customers. A typical recommendation scenario is: Bob wants to buy a cell phone which can play the game PUBG. He is interested in HUAWEI P20 and asks “can PUBG run smoothly on this phone?” under it. Then our system will be triggered to recommend the most eligible cell phones to him. Intuitively, diverse user questions could probably be addressed in reviews written by other users who have similar concerns. To address this recommendation problem, we propose a novel Question-Driven Attentive Neural Network (QDANN) to assess the instant demands of questioners and the eligibility of products based on user generated reviews, and do recommendation accordingly. Without supervision, QDANN can well exploit reviews to achieve this goal. The attention mechanisms can be used to provide explanations for recommendations. We evaluate QDANN in three domains of Taobao. The results show the efficacy of our method and its superiority over baseline methods.

To evaluate the efficacy and safety of human amniotic membrane (HAM) allograft in treating chronic diabetic foot ulcers (DFUs), a comprehensive search of randomised controlled trials in MEDLINE, EMBASE, PubMed, CENTRAL and Web of Science was conducted to December 7, 2019. Two reviewers independently screened the studies, extracted data, and evaluated the quality of studies. The primary outcome was the proportion of complete healing. The secondary outcomes were mean time to complete healing and adverse events. Statistical analyses were performed using RevMan 5.3. We identified 257 articles, of which 7 articles (465 participants) were included in the meta‐analysis. The proportion of complete wound healing in HAM plus standard of care (SOC) group was 3.88 times as high as that in SOC alone (RR: 3.88 [95% CI: 2.34, 6.44]) at 6 weeks, and 2.01 times at 12 weeks (RR: 2.01 [95%CI: 1.45, 2.77]). The intervention group had a significantly shorter time to complete healing (MD: −30.33 days, [95% CI: −37.95, −22.72]). The number needed to treat within 6 weeks was 2.3 ([95% CI: 1.8, 3.1]). No significant difference was shown in adverse events. Results were consistent in a sensitivity analysis. Hence, HAM plus SOC is effective and safe in treating chronic DFUs.

Despite thousands of neuroprotectants demonstrating promise in preclinical trials, a neuroprotective therapeutic has yet to be approved for the treatment of acute brain injuries such as stroke or traumatic brain injury. Developing a more detailed understanding of models and populations demonstrating “neurological resilience” in spite of brain injury can give us important insights into new translational therapies. Resilience is the process of active adaptation to a stressor. In the context of neuroprotection, models of preconditioning and unique animal models of extreme physiology (such as hibernating species) reliably demonstrate resilience in the laboratory setting. In the clinical setting, resilience is observed in young patients and can be found in those with specific genetic polymorphisms. These important examples of resilience can help transform and extend the current neuroprotective framework from simply countering the injurious cascade into one that anticipates, monitors, and optimizes patients’ physiological responses from the time of injury throughout the process of recovery. This review summarizes the underpinnings of key adaptations common to models of resilience and how this understanding can be applied to new neuroprotective approaches. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (10.1007/s13311-020-00832-7) contains supplementary material, which is available to authorized users.

Acute necrotizing encephalopathy (ANE) is a rare and life-threatening disease. It is caused by a cytokine-mediated injury to the brain with characteristic hemorrhagic and edematous lesions involving the bilateral thalami, brainstem, and other subcortical structures. The disease is commonly associated with antecedent viral triggers such as influenza, parainfluenza, and more recently, SARS-CoV-2, with subsequent neurologic deterioration occurring within days to weeks. Here, we present a case of a pregnant adult woman who developed a hyperacute form of ANE, progressing to brain death within 36 hours of symptom onset. Her diagnosis was confirmed via brain imaging, CSF studies, and neurohistopathological analysis. This case highlights the importance of establishing an early diagnosis for this under-recognized disease, and also suggests an association between ANE and early pregnancy.

These days, cancer can still not be effectively cured because cancer cells readily develop resistance to anticancer drugs. Therefore, an effective combination of drugs with different mechanisms to prevent drug resistance has become a very important issue. Furthermore, the BH3‐only protein BNIP3 is involved in both apoptotic and autophagic cell death. In this study, lung cancer cells were treated with a chemotherapy drug alone or in combination to identify the role of BNIP3 and autophagy in combination chemotherapy for treating cancer. Our data revealed that various combinational treatments of two drugs could increase cancer cell death and cisplatin in combination with rapamycin or LBH589, which triggered the cell cycle arrest at the S phase. Cells with autophagosome and pEGFP‐LC3 puncta increased when treated with drugs. To confirm the role of autophagy, cancer cells were pre‐treated with the autophagy inhibitor 3‐methyladenine (3‐MA). 3‐MA sensitized cancer cells to chemotherapy drug treatments. These results suggest that autophagy may be responsible for cell survival in combination chemotherapy for lung cancer. Moreover, BNIP3 was induced and localized in mitochondria when cells were treated with drugs. The transfection of a dominant negative transmembrane deletion construct of BNIP3 (BNIP3ΔTM) and treatment of a reactive oxygen species (ROS) inhibitor suppressed chemo drug‐induced cell death. These results indicate that BNIP3 and ROS may be involved in combination chemo drug‐induced cell death. However, chemo drug‐induced autophagy may protect cancer cells from drug cytotoxicity. As a result, inhibiting autophagy may improve the effects of combination chemotherapy when treating lung cancer.

BACKGROUND This study aimed to investigate the presence of circulating tumor cells (CTCs) in patients with penile squamous cell carcinoma (PSCC). METHODS CTCs were isolated from 14 patients with PSCC, 6 patients with balanoposthitis, and 6 healthy individuals. CTCs were enriched based on cell surface markers and filtered through the IsoFlux device, followed by identification according to cell morphology and immunofluorescence studies. RESULTS CTCs were found in all PSCC blood samples, but not in balanoposthitis samples and samples from healthy individuals. Immunofluorescence studies confirmed the tumor origin. When the patients with PSCC were stratified according to metastatic inguinal lymph node status, a statistically significant difference was observed in the number of detected CTCs. CONCLUSION Our study showed that CTCs in PSCC may represent a valuable marker for differentiating PSCC from other tumors. Based on the correlation with some clinical parameters, CTC analysis is possibly relevant for non-invasive monitoring of disease progression and prognosis. The results also suggested a potential role of CTCs in preventing overtreatment, such as inguinal lymph node dissection.

BACKGROUND The effect of radical hysterectomy for patients with cervical cancer on voiding function remains controversial. The purpose of this study was to examine the association between radical hysterectomy for patients with cervical cancer and the odds of developing neurogenic bladder by using data from the National Health Insurance Research Database (NHIRD) in Taiwan. METHODS We identified 17 936 patients who underwent radical hysterectomy for cervical cancer between 2000 and 2013 among inpatients registered in the Longitudinal Health Insurance Database in Taiwan. Of the patients, those diagnosed as having cervical cancer without radical hysterectomy were selected and compared as a matched control group. Patients diagnosed as having cervical cancer before the index date, those with neurogenic bladder dysfunction before tracking, and those aged

Background: The coronavirus disease-19 (COVID-19) and its variants have increased rapidly worldwide since December 2019, with respiratory disease being a prominent complication. As such, optimizing evaluation methods and identifying factors predictive of disease progress remain critical. The purpose of the study was to assess late phase (≥3 weeks) pulmonary changes using intensity-based computed tomography (CT) scoring in COVID-19 patients and determine the clinical characteristics predicting lung abnormalities and recovery.Methods: We conducted a retrospective study on 42 patients (14 males, 28 females; age 65±10 years) with COVID-19. Only patients with at least 3 CT scans taken at least 3 weeks after initial symptom onset were included in the study. Two scoring methods were assessed: (1) area-based scoring (ABS) and (2) intensity-weighted scoring (IWS). Temporal changes in the average lung lesion were evaluated by the calculating the averaged area under the curve (AUC) of the CT score-time curve. Correlations between averaged AUCs and clinical characteristics were determined. Results: Using the ABS system, temporal changes in lung abnormalities during recovery were highly variable (P=0.934). By contrast, the IWS system detected more subtle changes in lung abnormalities during in COVID-19 patients, with consistent week-to-week relative reductions in IWS (P=0.025). Strong relationships were observed with D-dimer and C-reactive protein (CRP) levels on admission, with hazard ratios (HR)(95%CI) of 5.32 (1.25-22.6)(P=0.026) and 1.05 (1.10-1.09)(P=0.017), respectively. Conclusion: Our results suggest COVID-19-mediated pulmonary abnormalities persist well-beyond 3-weeks of symptom onset, with intensity-weighted rather than area-based scoring being more sensitive. Moreover, D-dimer and CRP levels were predictive of the recovery from the disease.

At present, cognitive diagnosis has become one of the important research issues of current intelligent education. Existing cognitive diagnosis models generally use a manually designed function to mine the students' practice process. These functions are usually liner and not enough to capture the complex relationship between students and exercises. In this paper, based on the Neural Cognitive Diagnosis (NeuralCD) framework, we propose an importance of knowledge point-based neural cognitive diagnosis model (IK-NeuralCD). It introduces the importance factor of knowledge point and uses the frequency of knowledge points examined to express the importance of knowledge point, which improves the degree of fitting of the complex relationship between students and exercises, thereby improving the diagnosis and prediction effect. Based on comparison between IK-NeuralCD and the existing typical model on the real dataset, this paper proves the effectiveness of the IK-NeuralCD model.

Online learning has become one of the important forms of current education. In the process of online learning, learners often have poor online learning effects due to their weak self-awareness and ability to regulate the cognitive process. Metacognition can help students recognize their own knowledge and skill levels, and monitor and adjust their own learning process, so as to achieve effective online learning. However, metacognition is not easy to observe and judge, how to model, analyze and evaluate learners’ metacognition, and make suggestions on teaching strategies, is a problem worthy of in-depth study. The learning behavior in the online learning process is basically a spontaneous behavior guided by metacognition, and its behavior sequence contains rich metacognitive monitoring and adjustment information. In this study, a metacognitive ability model based on the behavior sequences of online learning process is proposed. This model takes into account the impact of students’ cognitive level on behavior sequences, so as to fully extract the metacognitive features contained in the behavior sequences to realize the effective evaluation of metacognitive ability. Finally, the experiments were carried out on real data sets, and the results verified the effectiveness of the model.

• Background• It is not currently known whether free fatty acids can be used as predictors for the prognosis of patients with acute ST-segment elevation myocardial infarction. The results of many studies on the correlation between free fatty acids and the prognosis of patients with acute myocardial infarction are inconsistent. The purpose of this clinical study is to determine the predictive effect of different free fatty acids in the plasma on cardiovascular events within the 12 months after a patient suffers an acute ST-segment elevation myocardial infarction.• Methods• This study is a prospective, observational, single-center cohort study. After enrollment, all participants will receive standard treatment and will be tested for 31 kinds of plasma free fatty acids. The patients will be followed up 2, 6, and 12 months after discharge. The primary endpoint is the compound cardiovascular events from onset to 12 months, including the cumulative incidence of all-cause death, heart failure, or coronary revascularization events. According to the power and sample size calculation, it will be necessary to enroll 747 patients in this study. After follow-up, data will be collected and analyzed to explore the predictive effect of different free fatty acids in plasma on the prognosis of acute elevated ST-segment myocardial infarction, and the risk factors of the endpoint events within 12 months.• Discussion• This study will provide new insightful evidence of serum indicators indicating the 12-month cardiovascular events of acute ST-segment elevation myocardial infarction.Trial registration: This study was retrospectively registrated at Chinese Clinical Trail Registry (https://www.chictr.org.cn/) on 14th July,2021. Registration number: ChiCTR2100048731.

Reconfigurable intelligent surface (RIS) is a promising technology for realizing cost-and-energy efficient wideband communication in future wireless communications. We construct a prototype of the RIS communication which realizes a new radiofrequency (RF) chain-free architecture. This experimental prototype using RIS is tested at 5.8 GHz and achieves a 20 MHz data transmission efficiently. We also propose a deep neural network-based beam alignment for the RIS communication, which validated by experiment tests with the RIS system consisting of 16×16 meta-material reflecting elements in indoor environment. The provided results show that the proposed method can achieve beam alignment in experiment tests. It improves the performance of the RIS communication with reduced overhead.

To compare perioperative circulating tumor cells (CTC) in primary upper tract urothelial carcinoma (UTUC) patients who underwent hand-assisted retroperitoneoscopic nephroureterectomy (HANU) or robotic-assisted nephroureterectomy (RANU). A total of 29 patients received RANU (n = 10) or HANU (n = 19). Peripheral blood samples were collected before, 24 h after surgery (POh24) and on postoperative day 28 (POD28). The demographic and pathologic data are similar in both groups. RANU had a longer operative time (p = 0.031), less bleeding volume (p = 0.004), and comparable pain sore (p = 0.169). The mean CTC numbers before surgery (2.4 vs. 2.3, p = 0.482), POh24 (2.4 vs. 1.9, p = 0.668) and POD28 (0.5 vs. 0.6, p = 0.280) were not significant different among groups. The amount of CTCs in both groups decreased and reached similar level on POD28. No significant difference of overall and intravesical recurrence rate between the two approaches. In comparison to RANU, more surgical manipulation does not affect tumor cell translocation into the bloodstream in UTUC patients who received HANU. However, a longer follow-up would be needed for the final comparison of tumor recurrence.

Heart failure is induced by multiple pathological mechanisms, and current therapies are ineffective against heart failure with preserved ejection fraction (HFpEF). As there are limited animal models of HFpEF, its underlying mechanisms have not yet been elucidated. Here, we employed the descending aortic constriction (DAC) technique to induce chronic pressure overload in the left ventricles of Tibetan minipigs for 12 weeks. Cardiac function, pathological and cellular changes, fibrotic signaling activation, and gene expression profiles were explored. The left ventricles developed concentric hypertrophy from weeks 4 to 6 and transition to dilation starting in week 10. Notably, the left ventricular ejection fraction was maintained at >50% in the DAC group during the 12-week period. Pathological examination, biochemical analyses, and gene profile analysis revealed evidence of inflammation, fibrosis, cell death, and myofilament dephosphorylation in the myocardium of HFpEF model animals, together with gene expression shifts promoting cardiac remodeling and downregulating metabolic pathways. Furthermore, we noted the activation of several signaling proteins that impact cardiac fibrosis and remodeling, including transforming growth factor-β/SMAD family members 2/3, type I/III/V collagens, phosphatidylinositol 3-kinase, extracellular signal-regulated kinase, matrix metalloproteinases 2 and 9, tissue inhibitor of metalloproteinases 1 and 2, interleukins 6 and 1β, and inhibitor of κBα/nuclear factor-κB. Our findings demonstrate that this chronic pressure overload-induced porcine HFpEF model is a powerful tool to elucidate the mechanisms of this disease and translate preclinical findings.

Online Shopping has become a part of our daily routine, but it still cannot offer intuitive experience as store shopping. Nowadays, most e-commerce Websites offer a Question Answering (QA) system that allows users to consult other users who have purchased the product. However, users still need to wait patiently for others’ replies. In this paper, we investigate how to provide a quick response to the asker by plausible answer identification from product reviews. By analyzing the similarity and discrepancy between explicit answers and reviews that can be answers, a novel multi-task deep learning method with carefully designed attention mechanisms is developed. The method can well exploit large amounts of user generated QA data and a few manually labeled review data to address the problem. Experiments on data collected from Amazon demonstrate its effectiveness and superiority over competitive baselines.

Background Computed tomography (CT) or MR images may cause the severity of early acute pancreatitis (AP) to be underestimated. As an innovative image analysis method, radiomics may have potential clinical value in early prediction of AP severity. Purpose To develop a contrast-enhanced (CE) MRI-based radiomics model for the early prediction of AP severity. Study type Retrospective. Subjects A total of 259 early AP patients were divided into two cohorts, a training cohort (99 nonsevere, 81 severe), and a validation cohort (43 nonsevere, 36 severe). Field strength/sequence 3.0T, T1 -weighted CE-MRI. Assessment Radiomics features were extracted from the portal venous-phase images. The "Boruta" algorithm was used for feature selection and a support vector machine model was established with optimal features. The MR severity index (MRSI), the Acute Physiology and Chronic Health Evaluation (APACHE) II, and the bedside index for severity in acute pancreatitis (BISAP) were calculated to predict the severity of AP. Statistical tests Independent t-test, Mann-Whitney U-test, chi-square test, Fisher's exact tests, Boruta algorithm, receiver operating characteristic analysis, DeLong test. Results Eleven potential features were chosen to develop the radiomics model. In the training cohort, the area under the curve (AUC) of the radiomics model, APACHE II, BISAP, and MRSI were 0.917, 0.750, 0.744, and 0.749, and the P value of AUC comparisons between the radiomics model and scoring systems were all less than 0.001. In the validation cohort, the AUC of the radiomics model, APACHE II, BISAP, and MRSI were 0.848, 0.725, 0.708, and 0.719, respectively, and the P value of AUC comparisons were 0.96 (radiomics vs. APACHE II), 0.40 (radiomics vs. BISAP), and 0.46 (radiomics vs. MRSI). Data conclusion The radiomics model had good performance in the early prediction of AP severity. Level of evidence 3 Technical Efficacy Stage: 2 J. Magn. Reson. Imaging 2020;51:397-406.

LiVPO4F has been suggested as an excellent electrode for Li-ion battery due to its various advantages. However, the LiVPO4F-based materials reported so far showed poor rate property and required further improvement. Herein, we describe a simple method to synthesize the boron-doped graphene-decorated LiVPO4F@C (BG-LiVPO4F@C) composite for the first time. This designed nanostructure can provide short diffusion distance for Li+ and offer enhanced electronic conductivity for the transport of electron, leading to superior Li+ storage performance. The obtained electrode delivers high capacity of 125.2 mAh g−1 at 10C. Besides, it exhibits excellent cyclic stability with a high capacity retention of about 92.5% after 300 cycles at 20C. Thus, the as-synthesized BG-LiVPO4F@C cathode is a promising candidate for Li-ion battery.

© The Author(s) 2019. Published by Baishideng Publishing Group Inc. All rights reserved. BACKGROUND Cirrhosis is a chronic late stage liver disease associated with hepatitis viruses, alcoholism, and metabolic disorders, such as Wilson disease (WD). There are no clear markers or clinical features that define cirrhosis originating from these disparate origins. We hypothesized that cirrhosis is not one disease and cirrhosis of different etiology may have differential clinical hepatic features. AIM To delineate the liver features between WD-associated cirrhosis and hepatitis Bassociated cirrhosis in the Chinese population. METHODS In this observational study, we reviewed the medical data of consecutive inpatients who had WD-associated cirrhosis or hepatitis B-associated cirrhosis from January 2010 to August 2018, and excluded patients who had carcinoma, severe heart or pulmonary diseases, or other liver diseases. According to the etiology of cirrhosis, patients were divided into two groups: WD-associated cirrhosis group (60 patients) and hepatitis B-associated cirrhosis group (56 patients). The liver fibrosis degree, liver function indices, and portal hypertension features of these patients were compared between the two groups. RESULTS No inter-group differences were observed in the diagnostic liver fibrosis markers, however, clinical features clearly defined the origin of cirrhosis. WD-associated cirrhosis patients (16-29 years) had lower levels of alanine transaminase, aspartate transaminase, and bilirubin, lower prothrombin time, lower incidence of hepatic encephalopathy, and lower portal vein diameter (P < 0.05), compared to cirrhosis resulting from hepatitis B in older patients (45-62 years). Importantly, they had decreased risks of progression from Child-Pugh grade A to B (odds ratio = 0.046, 95% confidence interval: 0.006-0.387, P = 0.005) and of ascites (odds ratio = 0.08, 95% confidence interval: 0.01-0.48, P = 0.005). Conversely, WDassociated cirrhosis patients had a higher risk of splenomegaly (odds ratio = 4.15, 95% confidence interval: 1.38-12.45, P = 0.011). CONCLUSION WD-associated cirrhosis presents a higher risk of splenomegaly associated with leukopenia and thrombocytopenia, although revealing milder liver dysfunction and portal hypertension symptoms, which recommends WD patients to be monitored for associated complications.

BACKGROUND/PURPOSE: Blood type has become an increasingly recognized risk factor for coagulopathy. We explored the association between blood type and hematoma expansion (HE) after intracerebral hemorrhage (ICH). METHODS: Spontaneous ICH patients prospectively enrolled in an ongoing ICH cohort study at Columbia University Irving Medical Center from 2009 to 2016 were evaluated. Primary ICH patients with admission blood type testing were evaluated for HE differences, defined as > 33% relative HE. The association of blood type with radiographic HE outcomes was assessed using multivariable logistic regression models. The association of blood type and poor clinical outcomes using modified Rankin Scale (mRS 4–6) was additionally explored. RESULTS: Of 272 ICH patients with blood type data and neuroimaging available to determine HE, there were 146 (54%) type-O, 82 (30%) type-A, 34 (13%) type-B, and 10 (3%) type-AB patients. No significant baseline demographic, clinical, or radiographic differences were noted between blood types. Type-B blood was associated with more HE compared to other blood types (OR 2.82; 95% CI 1.23–6.45) after adjusting for known covariates of HE (anticoagulant use, time to admission computed tomography scan, and baseline hematoma volume). No associations with blood type and poor 3 month mRS were identified, but these analyses were limited secondary to our smaller cohort. CONCLUSIONS: There may be differences in HE after ICH in patients with different blood types. Further work is required to replicate these findings and identify the pathophysiologic mechanisms behind coagulopathy between blood types after ICH.

Hierarchical structures with tailored macro/micro-porosity architecture play an important role in bone tissue regeneration. In 3D printing process, the printing ink formulation will influence on the ceramic macro and micro porous architectures. In this paper, HA powders of nano-sized grains (NP) with diameters of 30–50 nm, air jet milling powders (AP) with diameters of 10–30 μm, and spherical powder (SP) with diameters of 10–50 μm, were used as the initial printing materials in the printing ink formulations. The viscosity and the rheological behavior of printing inks were studied. The microstructure and morphology of the printed scaffold were observed and the mechanical properties of different types of scaffolds were tested. The results showed that the initial printing materials would influence on printing performance, both of the AP and SP inks may print porous scaffolds successfully. However, NP printing inks exhibited dramatic shrinkage and it is not suitable for 3D printing of bioceramics. The printing ink formulation also have effects on the ceramic macro and micro porous architectures and mechanical properties. The maximum compressive strength of SPS were 5.5 MPa, 3.2 MPa and 0.9 MPa with porosities of 60%, 70% and 80%, respectively. As the macroporosity decreases, the mechanical properties of the material would drop dramatically. With the same porosities, the compressive strength of APS were slightly higher than that of the SPS specimens.

Objective: The objective of the study was to investigate the relationship of amino acid metabolism with hypertriglyceridemia in diabetic patients under statins free of prior cardiovascular diseases.Methods: Two independent cross-sectional hospital based cohorts, i.e., Liaoning Medical University First Affiliated Hospital (LMUFAH, n = 146) and the Second Affiliated Hospital of Dalian Medical University (SAHDMU, n = 294) were included in the current analysis. Hypertriglyceridemia was defined as triglyceride ≥1.7 mmol/L, and well-controlled LDL-C was defined as Results: Patients, 86 and 106, were with hypertriglyceridemia in LMUFAH and SAHDMU, respectively. We observed that elevated alanine, asparagine, leucine, and valine were consistently associated with increased hypertriglyceridemia in both cohorts. In fixed-effect pooled analysis, the OR (95% CI) per SD increase was 1.71 (1.32–2.20) for alanine, 1.62 (1.20–2.19) for asparagine, 1.64 (1.22–2.20) for leucine, and 1.62 (1.22–2.13) for valine (all P values ranged from 0.0018 to Conclusion: Among 23 amino acids, alanine, asparagine, leucine, and valine were positively associated with increased residual risk of hypertriglyceridemia in diabetic patients with statin treatment.

BACKGROUNDWe recently described mortality of cardiac injury in COVID-19 patients. Admission activation of immune, thrombotic biomarkers and their ability to predict cardiacinjury and mortality patterns in COVID-19 is unknown.METHODSThis retrospective cohort study included 170 COVID-19 patients with cardiac injury at admission to Tongji Hospital in Wuhan from January 29–March 8, 2020. Temporal evolution of inflammatory cytokines, coagulation markers, clinical, treatment and mortality were analyzed.RESULTSOf 170 patients, 60 (35.3%) died early (CONCLUSIONSIn COVID-19 patients with cardiac injury, admission IL-6 and D-dimer predicted subsequent elevation of cTnI and early death, highlighting the need for early inflammatory cytokine-based risk stratification in patients with cardiac injury.Condensed AbstractCOVID-19 with cardiac injury is associated with worse survival. Admission activation of immune, thrombotic biomarkers and their ability to predict cardiac injury and mortality patterns in COVID-19 is unknown. This study proved that cardiac injury in these patients is closely related to the activation of immunological and thrombotic pathways and can be predicted by admission biomarkers of these pathways. This study supports the strategy of biomarker-guided, point-of-care therapy that warrants further studies in a randomized manner to develop anti-immune and anti-thrombotic treatment regimens in severe COVID-19 patients with cardiac injury.

Background The ablation targets of atrial fibrillation (AF) are adjacent to bronchi and pulmonary arteries (PAs). We used computed tomography (CT) to evaluate the anatomical correlation between left atrium (LA)-pulmonary vein (PV) and adjacent structures. Methods Data were collected from 126 consecutive patients using coronary artery CT angiography. The LA roof was divided into three layers and nine points. The minimal spatial distances from the nine points and four PV orifices to the adjacent bronchi and PAs were measured. The distances from the PV orifices to the nearest contact points of the PVs, bronchi, and PAs were measured. Results The anterior points of the LA roof were farther to the bronchi than the middle or posterior points. The distances from the nine points to the PAs were shorter than those to the bronchi (5.19 ± 3.33 mm vs 8.62 ± 3.07 mm; P P Conclusions The right superior PV orifices were closer to the bronchi and PAs than other PV orifices. The ablation at the mid-posterior LA roof had a higher possibility to damage bronchi. CT is a feasible method to assess the anatomical adjacency in vivo, which might provide guidance for AF ablation.

Objectives: To assess the late phase CT changes of COVID-19 patients, and figure out factors predicting lung abnormality in late phase.Methods: We conducted a retrospective study on 42 patients (14 males, 28 females; age 65±10 years) with COVID-19 admitted between February 7, 2020 and March 27, 2020. Only patients with at least 3 CT scans taken at least 3 weeks after initial symptom onset were included in the study. CT images were analyzed by 2 independent radiologists using different scoring: (1) area-based scoring (ABS); and (2) intensity-weighted scoring (IWS). Temporal changes in the average lung lesion were evaluated by averaged area under the curve (AUC) of the CT score-time curve. Correlations between averaged AUCs and clinical characteristics were determined. Results: Temporal changes in lung abnormalities during recovery (weeks 3 through 8) of CT findings using the ABS system were variable (P=0.934). By contrast, the IWS system detected more subtle changes in lung abnormalities during the late phase of recovery in COVID-19 patients, with consistent week-to-week relative reductions in IWS (P=0.025). In assessing the correlation between averaged AUCs and clinical characteristics, strong relationships were observed with D-dimer and C-reactive protein (CRP) levels on admission, with hazard ratios (HR)(95%CI) of 5.32 (1.25-22.6)(P=0.026) and 1.05 (1.10-1.09)(P=0.017), respectively. Conclusion: Our results suggest an intensity-weighted rather than area-based scoring system is more sensitive to detect subtle temporal CT changes in COVID-19, with D-dimer and CRP levels on admission being predictive of the time course of late phase recovery from the disease.

Texts convey sophisticated knowledge. However, texts also convey sensitive information. Despite the success of general-purpose language models and domain-specific mechanisms with differential privacy (DP), existing text sanitization mechanisms still provide low utility, as cursed by the high-dimensional text representation. The companion issue of utilizing sanitized texts for downstream analytics is also under-explored. This paper takes a direct approach to text sanitization. Our insight is to consider both sensitivity and similarity via our new local DP notion. The sanitized texts also contribute to our sanitization-aware pretraining and fine-tuning, enabling privacy-preserving natural language processing over the BERT language model with promising utility. Surprisingly, the high utility does not boost up the success rate of inference attacks., ACL-ICJNLP'21 Findings; The first two authors contributed equally

Contemporary careers are changing and face many challenges. This creates a need for innovative cross-cultural and multidisciplinary research. In this chapter, twenty-three participants in a European/South Korean research mega-project provide an overview of their diverse experiences of trans-national, trans-sectoral, and trans-generational work. First, the project’s architects explore the context and rationale in five countries, selection of partners, securing funding and the underpinning principles of ‘strategic entrepreneurship’ and ‘multi-sociation’. Key workstream leaders then outline the project content, including diplomacy in research design, reflecting gender, ensuring impact, capturing learning, communicating to various audiences. Major aspects of Implementation are described: quality management, managing academics and risk, transnational and trans-sectorial working (between academia and industry), novel methodologies and finally the career implications for PhD students involved. These are honest and pragmatic reflections on the way to best practice.

Long story generation (LSG) is one of the coveted goals in natural language processing. Different from most text generation tasks, LSG requires to output a long story of rich content based on a much shorter text input, and often suffers from information sparsity. In this paper, we propose \emph{TopNet} to alleviate this problem, by leveraging the recent advances in neural topic modeling to obtain high-quality skeleton words to complement the short input. In particular, instead of directly generating a story, we first learn to map the short text input to a low-dimensional topic distribution (which is pre-assigned by a topic model). Based on this latent topic distribution, we can use the reconstruction decoder of the topic model to sample a sequence of inter-related words as a skeleton for the story. Experiments on two benchmark datasets show that our proposed framework is highly effective in skeleton word selection and significantly outperforms the state-of-the-art models in both automatic evaluation and human evaluation., Comment: KDD2021, 9 pages

Background: Wilms' tumor (nephroblastoma) is the most common solid primary renal malignancy in children but extremely rare in adult population. We report the clinical presentation and therapeutic results of our adult Wilms' tumor patients and review the previous literatures about diagnosis and treatments. Methods: A retrospective review was carried out to investigate the patients with adult Wilms' tumor at the Tri-Service General Hospital during a 35-year period (1980–2014). Seven patients ≥16 years old (one female and six males; median age at the time of diagnosis of 24 years; range: 21–48 years) were staged according to the National Wilms' Tumor Study staging system as follows: three patients had Stage II disease, two patients had Stage III disease, and two patients had Stage IV disease. All statistical values were represented as mean ± standard deviation. All tests were two-sided, with P < 0.05 considered statistically significant. All statistical analyses were carried out using SPSS software (IBM, Armonk, NY, USA). Results: The most common clinical presentation is flank pain (57%) followed by palpable mass (43%). Only one patient presented with gross hematuria. Four patients underwent renal arteriography, which all disclosed irregular, tortuous vessel pattern with faint tumor stain. All patients underwent radical nephrectomy (one – hand-assisted laparoscopic nephrectomy and six – open transperitoneal nephrectomy). Two patients developed lung metastases after radical surgery. Six patients received multimodal therapy, including surgery, chemotherapy, and radiation therapy. The mean disease-free survival is 81 months. The 3- and 5-year survival rates are 83% and 71% separately. Conclusions: Selective renal arteriography could provide a valuable diagnostic hint in adult Wilms' tumor. Using multimodality therapy, improved prognosis could be achieved.

A vast amount of vital clinical data is available within unstructured texts such as discharge summaries and procedure notes in Electronic Medical Records (EMRs). Automatically transforming such unstructured data into structured units is crucial for effective data analysis in the field of clinical informatics. Recognizing phrases that reveal important medical information in a concise and thorough manner is a fundamental step in this process. Existing systems that are built for opendomain texts are designed to detect mostly non-medical phrases, while tools designed specifically for extracting concepts from clinical texts are not scalable to large corpora and often leave out essential context surrounding those detected clinical concepts. We address these issues by proposing a framework, CliniPhrase, which adapts domain-specific deep neural network based language models (such as ClinicalBERT) to effectively and efficiently extract high-quality phrases from clinical documents with a limited amount of training data. Experimental results on the MIMIC-III dataset show that our method can outperform the current state-of-the-art techniques by up to 18% in terms of F 1 measure while being very efficient (up to 48 times faster).11Our source code, pre-trained models and documentations are available online at: https://github.com/kaushikmani/PhraseMiningLM

CONTEXTE: Les atteintes cardiaques sont fréquentes dans les cas graves de maladie à coronavirus 2019 (COVID-19) et sont associées à un mauvais pronostic. Notre étude portait sur les facteurs prédictifs de mortalité intrahospitalière, les caractéristiques de l’arythmie et les effets des traitements qui allongent l’intervalle QT chez les patients ayant une atteinte cardiaque. MÉTHODES: Nous avons fait une étude de cohorte rétrospective des cas graves de COVID-19 admis à l’hôpital Tongji, à Wuhan, en Chine, entre le 29 janvier et le 8 mars 2020. En examinant ceux qui avaient une atteinte cardiaque, définie ici comme un taux élevé de troponine I cardiaque (TnIc), nous avons déterminé les caractéristiques biologiques et cliniques associées à la mortalité et au besoin de ventilation invasive. RÉSULTATS: Parmi les 1284 cas graves de COVID-19, 1159 avaient au dossier un taux de TnIc mesuré à l’admission, qui pour 170 (14,7 %) participants indiquait une atteinte cardiaque. Les patients ayant une atteinte cardiaque avaient un taux de mortalité nettement plus élevé que les autres patients (71,2 % c. 6,6 %; p < 0,001). Nous avons constaté que le taux de TnIc initial (pour chaque augmentation d’un facteur 10, rapport de risque [HR] 1,32, intervalle de confiance [IC] à 95 % 1,06–1,66) et le taux de TnIc maximal atteint au cours de la maladie (pour chaque augmentation d’un facteur 10, HR 1,70, IC à 95 % 1,38–2,10) étaient associés à de minces chances de survie. Le taux de TnIc maximal était aussi associé au besoin de ventilation invasive (rapport de cotes 3,02, IC à 95 % 1,92–4,98). Sur les 170 patients ayant une atteinte cardiaque, 44 (25,9 %) présentaient une arythmie. Les 6 qui souffraient de tachycardie ou de fibrillation ventriculaires sont morts. Nous avons remarqué que les patients qui recevaient des médicaments allongeant l’intervalle QT avaient un intervalle QTc plus long que ceux qui n’en recevaient pas (différence entre les médianes 45 ms; p = 0,01), mais que ce traitement n’était pas directement lié à la mortalité (HR 1,04, IC à 95 % 0,69–1,57). INTERPRÉTATION: Chez les patients ayant la COVID-19 et une atteinte cardiaque, les taux initial et maximal de TnIc sont associés à de minces chances de survie, et le taux maximal est un facteur prédictif du besoin de ventilation invasive. Les malades de la COVID-19 doivent subir un dépistage des atteintes cardiaques et être surveillés, surtout si on leur fait suivre un traitement qui peut prolonger la repolarisation. Enregistrement de l’essai : Registre des essais cliniques chinois, n° ChiCTR2000031301.

Background Geriatric patients with coronavirus disease (COVID‐19) are at high risk of developing cardiac injury. Identifying the factors that affect high‐sensitivity cardiac troponin I may indicate the cause of cardiac injury in elderly patients, and this could hopefully assist in protecting heart function in this patient population. Methods One hundred and eighty inpatients who were admitted for COVID‐19 were screened. Patients older than 60 years were included in this study, and the clinical characteristics and laboratory results of the cohort were analyzed. The correlation between cardiac injury and clinical/laboratory variables was statistically analyzed, and further logistic regression was performed to determine how these variables influence cardiac injury in geriatric patients. Results Age (p, Our study explored potential risk factors related to cardiac injury in elderly patients with COVID‐19. The results suggest that age and inflammatory factors influence cardiac injury in elderly patients. Interfering with inflammation in this patient population may potentially confer cardiac protection.

BackgroundIn the early stages of the coronavirus disease pandemic, the anti-malarial drug hydroxychloroquine (HCQ) and the antibiotic drug azithromycin (AZM) were widely used as emerging treatments. However, controversial cardiac toxicity results obtained from clinical trials and epidemic studies suggest that the cardiotoxicity of these two drugs should be re-evaluated. In the present study, we aimed to assess the impact of a short course of AZM or HCQ + AZM combination treatment on ECG and cardiac function in healthy guinea pigs.MethodsThirty-two male guinea pigs were randomly divided into four groups: control; AZM; HCQ; and HCQ + AZM groups. At 3, 6, and 9 days after treatment, electrocardiograms (ECGs) and echocardiographic techniques were used to determine important ECG parameters and cardiac functional parameters of the left ventricle (including posterior wall thickness, end systolic/end diastolic volume, ejection fraction, and fractional shortening).ResultsAlthough AZM decreased the heart rates of guinea pigs on day 9 (under anesthetized conditions), HCQ + AZM decreased heart rates on days 3, 6, and 9. The corrected QT intervals of guinea pigs after AZM and HCQ + AZM treatments were significantly increased, compared with CON and HCQ treatment respectively, on days 3, 6, and 9. However, QRS complex durations were not significantly different between the groups. AZM significantly decreased left ventricular ejection fraction (LVEF) and left ventricular fraction shortening (LVFS) on days 3, 6, and 9, whereas HCQ + AZM only decreased LVEF and LVFS on day 9. Posterior wall thickness and of the left ventricle in the diastolic and systolic states were not significantly different between these groups. In addition, compared with CON, AZM and HCQ decreased the EDV. And, in comparison with HCQ treatment, HCQ + AZM treatment increased ESV on day 9.ConclusionsAccording to our study, AZM significantly prolongs the QT interval and damages cardiac function. Moreover, HCQ + AZM treatment increased the risk of cardiac dysfunction compared with HCQ treatment.

Learning to capture text-table alignment is essential for tasks like text-to-SQL. A model needs to correctly recognize natural language references to columns and values and to ground them in the given database schema. In this paper, we present a novel weakly supervised Structure-Grounded pretraining framework (StruG) for text-to-SQL that can effectively learn to capture text-table alignment based on a parallel text-table corpus. We identify a set of novel prediction tasks: column grounding, value grounding and column-value mapping, and leverage them to pretrain a text-table encoder. Additionally, to evaluate different methods under more realistic text-table alignment settings, we create a new evaluation set Spider-Realistic based on Spider dev set with explicit mentions of column names removed, and adopt eight existing text-to-SQL datasets for cross-database evaluation. STRUG brings significant improvement over BERT-LARGE in all settings. Compared with existing pretraining methods such as GRAPPA, STRUG achieves similar performance on Spider, and outperforms all baselines on more realistic sets. The Spider-Realistic dataset is available at https://doi.org/10.5281/zenodo.5205322., Accepted to NAACL 2021. The Spider-Realistic dataset is available at https://doi.org/10.5281/zenodo.5205322

Chitosan is a biodegradable and biocompatible natural polysaccharide that has a wide range of applications in the field of dentistry due to its functional versatility and ease of access. Recent studies find that chitosan and its derivatives can be embedded in materials for dental adhesives, barrier membranes, bone replacement, tissue regeneration, and antimicrobial agent to better manage oral diseases. In this paper, we provide a comprehensive overview on the preparation, applications, and major breakthroughs of chitosan biomaterials. Furthermore, incorporation of chitosan additives for the modification and improvement of dental materials has been discussed in depth to promote more advanced chitosan-related research in the future.

Background Variability in the anatomy and orientation of the triangle of Koch (TK) complicates ablation procedures involving the atrioventricular (AV) node. We used CT angiography (CTA) to assess the anatomical TK orientation, the CS ostium direction, and the relationship between the two, and we validated an individualized CS-guided projection during ablation procedures. Methods In 104 patients without structural heart disease undergoing computed tomography (CT) angiography, TK orientations were determined in relation to the coronary sinus ostium (CSo) as well as two standard right anterior oblique (RAO) projection angles (30o and 45o) commonly used in ablation procedures. Results A CS-guided RAO projection (RAOCS) was shown to best track the orientation of the TK compared to RAO30° and 45°, with TK orientation strongly correlating with the CSo direction (r = 0.86, P Conclusion In hearts with a normal structure, the CSo direction offers a reliable predictor of the TK orientation which can be used to guide the projection of the TK during ablation procedures.

Myocardial fibrosis (MF) plays a key role in the development and progression of heart failure (HF) with limited effective therapies. Galectin-3 (Gal-3) is a biomarker associated with fibrosis and inflammation in patients with HF. The Gal-3 inhibitor modified citrus pectin (MCP) protects against cardiac dysfunction, though the underlying mechanism remains unclear. The aim of this study was to investigate the effect and mechanism of MCP on MF using an isoproterenol (ISO)-induced rat model of HF. Cardiac function was analyzed by echocardiography and electrocardiography. Histopathological changes in the heart tissue were assessed by hematoxylin-eosin and Masson trichrome staining. The mRNA and protein expression levels of signaling molecules and pro-inflammatory cytokines were monitored by immunohistochemistry, western blot, qRT-PCR and ELISA analyses. The results demonstrated that MCP ameliorated cardiac dysfunction, decreased myocardial injury and reduced collagen deposition. Furthermore, MCP downregulated the expression of Gal-3, TLR4 and MyD88, thereby inhibiting NF-κB-p65 activation. MCP also decreased the expression of IL-1β, IL-18 and TNF-α, which have been implicated in the pathogenesis of HF. These inhibitory effects were observed on day 15 and continued until day 22. Taken together, these results suggest that MCP ameliorates cardiac dysfunction through inhibiting inflammation and MF. These effects may be through downregulating Gal-3 expression and suppressing activation of the TLR4/MyD88/NF-κB signaling pathway. The present study supports the use of Gal-3 as a therapeutic target for the treatment of MF after myocardial infarction.

Despite the widely successful applications, bootstrapping and fine-tuning semantic parsers are still a tedious process with challenges such as costly data annotation and privacy risks. In this paper, we suggest an alternative, human-in-the-loop methodology for learning semantic parsers directly from users. A semantic parser should be introspective of its uncertainties and prompt for user demonstration when uncertain. In doing so it also gets to imitate the user behavior and continue improving itself autonomously with the hope that eventually it may become as good as the user in interpreting their questions. To combat the sparsity of demonstration, we propose a novel annotation-efficient imitation learning algorithm, which iteratively collects new datasets by mixing demonstrated states and confident predictions and re-trains the semantic parser in a Dataset Aggregation fashion (Ross et al., 2011). We provide a theoretical analysis of its cost bound and also empirically demonstrate its promising performance on the text-to-SQL problem. Code will be available at https://github.com/sunlab-osu/MISP., Accepted to EMNLP 2020. 21 pages, 6 figures

Machine reading comprehension has made great progress in recent years owing to large-scale annotated datasets. In the clinical domain, however, creating such datasets is quite difficult due to the domain expertise required for annotation. Recently, Pampari et al. (EMNLP'18) tackled this issue by using expert-annotated question templates and existing i2b2 annotations to create emrQA, the first large-scale dataset for question answering (QA) based on clinical notes. In this paper, we provide an in-depth analysis of this dataset and the clinical reading comprehension (CliniRC) task. From our qualitative analysis, we find that (i) emrQA answers are often incomplete, and (ii) emrQA questions are often answerable without using domain knowledge. From our quantitative experiments, surprising results include that (iii) using a small sampled subset (5%-20%), we can obtain roughly equal performance compared to the model trained on the entire dataset, (iv) this performance is close to human expert's performance, and (v) BERT models do not beat the best performing base model. Following our analysis of the emrQA, we further explore two desired aspects of CliniRC systems: the ability to utilize clinical domain knowledge and to generalize to unseen questions and contexts. We argue that both should be considered when creating future datasets., Accepted by ACL 2020