Search

Obstructive sleep apnea(OSA)in children is the most serious type of sleep disordered breathing. If left untreated, it can lead to serious complications, including malocclusion, which in turn can lead to the occurrence and development of OSA. The treatment of malocclusion in OSA children requires multidisciplinary cooperation. It is recommended that orthodontists pay close attention to the sleep breathing condition of children with OSA. The orthodontists should comprehensively consider children's age, growth and development stage, severity of OSA and malocclusion manifestations, and formulate personalized treatment plans to create favorable conditions for the healthy development of the dentofacial and even the whole body of children.

Policy search is an efficient learning method in the field of deep reinforcement learning (DRL), which is capable of solving large-scale problems with continuous state and action spaces and widely used in real-world problems. However, such method usually requires a large number of trajectory samples and extensive training time, and may suffer from poor generalization ability, making it difficult to generalize the learned policy model to seemingly small changes in the environment. In order to solve the above problems, this paper proposes a policy search DRL method based on latent space. Specifically, this paper extends the idea of state representation learning to action representation learning, i.e. learning a policy in the latent space of action representations, and then mapping the action representations to the real action space. With the introduction of representation learning models, this paper abandons the traditional end-to-end training manner in DRL and divides the whole task into two stages: large-scale representation model learning and the small-scale policy model learning, where unsupervised learning methods are employed to learn the representation models and policy search methods are used to learn the small-scale policy model. Large-scale representation models can ensure the capacity for generalization and expressiveness, while small-scale policy model can reduce the burden of policy learning, thus alleviating the issues of low sample utilization, low learning efficiency and weak generalization of action selection in DRL to some extent. Finally, the effectiveness of introducing the latent state and action representations is demonstrated by the intelligent control task CarRacing and Cheetah.

Objective·To investigate the effects of trimethylamine oxide (TMAO) on cardiac function in mice with hypertrophic cardiomyopathy (HCM) and its potential molecular mechanism.Methods·Mice with myosin heavy chain 6 (Myh6) c.1211G>A (p.R404Q+/-) point mutation were used as the animal model. According to dietary supplementation of TMAO and TMAO inhibitor iodomethylcholine (IMC), the wild type (WT) mice and HCM mice were divided into WT group, HCM group (HCM-1 group, HCM-2 group), WT+TMAO group, HCM+TMAO group and HCM+IMC group, respectively. Left ventricular fraction shortening (FS) and left ventricular posterior wall thickness (LVPW) were assessed by echocardiography in all mice. Enzyme-linked immunosorbent assay (ELISA) was used to detect the serum TMAO concentration of mice in the HCM-1 group and WT group. The regularity of myocardial cell arrangement of mice was evaluated by hematoxylin and eosin staining (HE staining). The proportion of myocardial fibrosis was evaluated by Masson staining. The activity of protein kinase A (PKA) in mouse myocardial tissue was detected by PKA kit. The expression of ryanodine receptor 2 (RyR2) and p-RyR2(S2808) in mouse myocardial tissue was detected by Western blotting.Results·The results of echocardiography showed that at 12 months of age, the FS of mice in the WT+TMAO group and HCM+TMAO group were lower than those in the corresponding WT group and HCM-1 group, respectively (P

Background Hypertensive disorders in pregnancy (HDPs) are the leading causes of maternal and perinatal death worldwide, and appropriate prenatal care is beneficial toward improve pregnancy outcomes in populations at high risk of preeclampsia. Objective To describe variations in regional manifestations and outcomes of HDPs among the Chinese population to aid in the development of region-specific perinatal management guidelines. Methods This is a retrospective study. The clinical data of patients with HDP in 3 different regions were collected to explore the characteristics of HDP patients in different regions of China. Results In Peking University First Hospital, a regional rescue center for prenatal care and delivery, the proportion of patients with high-risk factors for PE was considerably high; 37.8% were of advanced maternal age, and 14.7% were obese. Among the patients, at this hospital, the proportion of comorbidities (e.g., chronic hypertension) in HDP patients was higher than that in the other 2 specialized maternal and child health care hospitals. Conclusion Targeted prenatal care procedures should be established based on regional characteristics to improve the quality of perinatal health care and reduce the incidence of HDP.

Objective To analyze the clinical features and resuscitation outcome of adult patients with cardiac arrest (CA) in intensive care unit, and discuss the related factors affecting the success rate of cardiopulmonary resuscitation in adult patients with CA. Methods The clinical data of CA patients in the intensive care unit of the First Hospital of Jilin University from September 2019 to December 2020 were retrospectively analyzed. According to the outcome indicators, the patients were divided into return of spontaneous circulation (ROSC) group and non-ROSC group, survival-discharge group and death group. The relevant factors affecting resuscitation outcome were discussed. Results A total of 351 patients with CA were included in this study, including 206 males and 145 females; the median age was 63 years. There were 191 patients with cardiogenic CA and 267 patients with non-defibrillation rhythm (cardiac arrest and no electrical activity) at the onset of CA. After treatment with CPR, 152 patients had ROSC, of whom 42 survived and were discharged. Univariate Logistic regression analysis showed that coronary artery disease, non-defibrillation rhythm, no defibrillation rhythm during resuscitation, no electric defibrillation, and no emergency endotracheal intubation were the possible factors that reduced the incidence of ROSC. Young age, CPR duration ≤30 min, and small cumulative dose of epinephrine were the possible factors that increased the incidence of ROSC. Non-cardiogenic etiology, non-defibrillable rhythm, no electric defibrillation, and no defibrillable rhythm during resuscitation decreased the likelihood of survival and discharge, whereas CPR duration ≤30 min and a small cumulative dose of epinephrine increased the likelihood of survival and discharge. Multivariate Logistic regression analysis showed that CPR duration > 30 min was an independent risk factor for ROSC, and young age, emergency tracheal intubation, and small cumulative epinephrine dose were independent protective factors for ROSC. CPR duration > 30 min was an independent risk factor for survival and discharge, and cardiogenic etiology and presence of defibrillable rhythm during resuscitation were independent protective factors for survival and discharge. Conclusions CPR duration, cumulative epinephrine dose, CA initial rhythm, electric defibrillation, and emergence of defibrillable rhythm during resuscitation are associated factors for ROSC as well as survival and discharge, and CPR duration > 30 min is an independent risk factor. Clinical attention should be paid to these factors in order to improve the outcome of cardiopulmonary resuscitation.

In orthodontic and orthognathic practice, cephalometric analysis is an integral tool throughout the clinical process. However, as landmark identification is still unautomated, both the conventional and semiautomated approaches are open to considerable subjectivity and could be time-consuming for inexperienced clinicians. Deep learning (DL), a state-of-the-art artificial intelligence (AI) technique, is highly effective in image recognition. In recent years, many studies have focused on the application of DL in cephalometric analysis, including automated landmark detection, automated diagnosis, cervical vertebral maturation stage determination, adenoid hypertrophy analysis and upper airway identification. Studies show that DL can effectively improve the efficiency of cephalometric analysis. In most studies, the accuracy of DL can reach more than 80%, and its difference from the gold standard is clinically acceptable, demonstrating good potential for future applications. However, most studies are limited to landmark detection, and the broadness and richness of the training dataset are limited. Future studies should broaden the research scope, improve the algorithm, elevate the richness of the datasets, and combine DL with other AI algorithms to improve its accuracy, stability and generalizability.

The sterile line is the basis of crop heterosis utilization. To broaden the sources of male sterility in tobacco, the Ntms1 (Nicotiana tabacum L. ms1) gene was cloned from the tobacco variety K326 by homologous cloning based on the Cams1 (Capsicum annuum L. ms1) gene sequence of male-sterility genes in pepper. The protein structure and physicochemical properties of the two genes were determined by bioinformatics analysis, and the function of the Ntms1 gene was verified by the CRISPR/Cas9 system. The results showed that the sequences of Ntms1 and Cams1 were 85.25% similar, and plant homeodomains were found in both genes; the physical and chemical properties were also very similar. It is speculated that the Ntms1 gene had the same function as the Cams1 gene in controlling male sterility. Compared to the wild-type plants, the filaments of the Ntms1 knockout mutant plants were shorter, and the stamen was shorter than the pistil. The anthers did not develop fully and had few viable pollen grains; the tapetum and the anther wall had developed abnormally, and the anther chamber was severely squeezed. The malondialdehyde content in the mutant plants was significantly higher than that in the wild-type plants, while self-fertility was significantly lower in the mutant plants. The results showed that the Ntms1 gene plays an important role in regulating fertility in tobacco.

Deep reinforcement learning (DRL) as an important learning paradigm in the field of machine learning, has received increasing attentions after AlphaGo defeats the human. DRL interacts with the environment by trials and errors, and obtains the optimal policy by maximizing the cumulative reward. Reinforcement learning can be divided into two categories: model-free reinforcement learning and model-based reinforcement learning. The tra-ining process of model-free reinforcement learning needs a large number of samples. It is difficult for model-free reinforcement learning to get good performance when the sampling budget is limited, and a large number of samples cannot be collected. However, model-based reinforcement learning can reduce the real sample demand and improve the data efficiency through making full use of the environment model. This paper focuses on the field of model-based reinforcement learning, introduces its research status, investigates its classical algorithms, and discusses?future development trend and application prospect.

The co-occurrence of cadmium (Cd) pollution and selenium (Se) deficiency commonly exists in global soils, especially in China. As a result, there is great interest in developing practical agronomic strategies to simultaneously achieve Cd remediation and Se mobilization in paddy soils, thereby enhancing food quality/safety. To this end, we conducted a field-plot trial on soils having high geological background levels of Cd (0.67 mg kg-1) and Se (0.50 mg kg-1). We explored 12 contrasting fertilizers (urea, potassium sulfate (K2SO4), calcium-magnesium-phosphate (CMP)), amendments (manure and biochar) and their combinations on Cd/Se bioavailability. Soil pH, total organic carbon (TOC), soil available Cd/Se, Cd/Se fractions and Cd/Se accumulation in different rice components were determined. No significant differences existed in mean grain yield among treatments. Results showed that application of urea and K2SO4 decreased soil pH, whereas the CMP fertilizer and biochar treatments increased soil pH. There were no significant changes in TOC concentrations. Three treatments (CMP, manure, biochar) significantly decreased soil available Cd, whereas no treatment affected soil available Se at the maturity stage. Four treatments (CMP, manure, biochar and manure＋urea＋CMP＋K2SO4) achieved our dual goal of Cd reduction and Se enrichment in rice grain. Structural equation modeling (SEM) demonstrated that soil available Cd and root Cd were negatively affected by pH and organic matter (OM), whereas soil available Se was positively affected by pH. Moreover, redundancy analysis (RDA) showed strong positive correlations between soil available Cd, exchangeable Cd and reducible Cd with grain Cd concentration, as well as between pH and soil available Se with grain Se concentration. Further, there was a strong negative correlation between residual Cd/Se (non-available fraction) and grain Cd/Se concentrations. Overall, this study identified the primary factors affecting Cd/Se bioavailability, thereby providing new guidance for achieving safe production of Se-enriched rice through fertilizer/amendment management of Cd-enriched soils.