Aim-The purpose of this study was to assess the effect of the Burpee exercise on the endurance and short-term memory of adolescents aged 15-16 years. Methods-The experiment was performed in a coeducational school in Kirov (Russia). The four-month study involved 52 adolescents of both genders. During the study period, 30 physical education lessons were held in each class. Adolescents from the control group were involved in a typical program (also aimed at improving endurance), and adolescents from the experimental group additionally performed the Burpee exercise. Endurance in adolescents was assessed by means of an "all-out" Running 2000 m test, and short-term memory was assessed by means of the Jacobs test (tests were performed before and after the programs). Results-An analysis of variance revealed an interaction effect (F = 28.733, ηp2 = 0.578 and p < 0.001, and F = 104.353, ηp2 = 0.676 and p < 0.001 for the Running 2000 m test and the Jacobs test, respectively). The control group improved by 1.9% (p > 0.05) in the Running 2000 m and by 5.5% (p > 0.05) in the Jacobs test. In the experimental group, both improved significantly by 8.6% (p < 0.05) in the Running 2000 m test and by 26.0% (p < 0.05) in the Jacobs test. Conclusion-The Burpee exercise could be included in physical education classes to improve endurance and short-term memory in 15-16-year-old.

Research has shown that animation plays an important role in the development of children's executive function (EF), and the development of EF components, inhibitory control, working memory, and cognitive flexibility, is asynchronous. Thus, this study explores the developmental trajectories and animation features (fantasy and pacing) that influence each EF component, by examining 218 children aged 4-7. Pretest information, mainly the childhood EF inventory, was provided by parents: child's age, age of first exposure to animation, animation viewing time on weekdays and weekends, family income, and parents' education. The children in each age group were randomly divided into four groups to watch animations comprised of different animation features. After watching, their EF were measured by a day-night task, backward digit-span task, and flexible item-selection task. The results showed that the children's inhibitory control, working memory and cognitive flexibility levels all improved with age. Highly fantastical animations weakened children's performance on each subsequent EF task. Pacing had no effect on any of the components of children's EF. An interactive effect on inhibitory control was only found with fantasy in younger children; specifically, high-fantastical animations had a more pronounced short-lived weakening effect on inhibitory control in younger children (4-6 years) compared with low-fantastical animations. Future research should explore the long-term impact of content rather than the form of animation on younger children's EF.

Water level management is an important part of urban water system management. In flood season, the river should be controlled to ensure the ecological and landscape water level. In non-flood season, the water level should be lowered to ensure smooth drainage. In urban areas, the response of the river water level to rainfall and artificial regulation is relatively rapid and strong. Therefore, building a mathematical model to forecast the short-term trend of urban river water levels can provide a scientific basis for decision makers and is of great significance for the management of urban water systems. With a focus on the high uncertainty of urban river water level prediction, a real-time rolling forecast method for the short-term water levels of urban internal rivers and external rivers was constructed, based on long short-term memory (LSTM). Fuzhou City, China was used as the research area, and the forecast performance of LSTM was analyzed. The results confirm the feasibility of LSTM in real-time rolling forecasting of water levels. The absolute errors at different times in each forecast were compared, and the various characteristics and causes of the errors in the forecast process were analyzed. The forecast performance of LSTM under different rolling intervals and different forecast periods was compared, and the recommended values are provided as a reference for the construction of local operational forecast systems.

The main symptoms of schizophrenia are categorized as positive, negative, and cognitive. Cognitive impairments do not generally respond to antipsychotics. Cariprazine is a novel antipsychotic conceived with the idea that high affinity for D3 receptors may elicit a favorable response in the management of cognitive deficits. We evaluated the pro-cognitive properties of 14-day long pre-treatment with cariprazine (0.25, 0.5, and 1 mg/kg b.w. intraperitoneally) in experimental rodent models with scopolamine-induced memory impairment employing novel object recognition test (NORT), T-maze, Y-maze, and passive avoidance tasks (step-through and step-down). Statistical analysis was performed with One Way ANOVA. In NORT cariprazine increased the recognition index. In T-maze and Y-maze cariprazine increased the working memory index as well as the percentage of spontaneous alternation. Cariprazine improved learning and memory in both short-term and long-term memory retention tests in step-down and step-through tasks. Cariprazine improves learning, recognition, and spatial memory in rats with scopolamine-induced memory impairment. Cariprazine's beneficial effect on cognition is likely due to its affinity for D3 receptors, as well as agonism at 5-HT1A receptors. Most probably, the cognitive-enhancing properties of cariprazine are the result of integrated modulation in the amygdala, hippocampus, and prefrontal cortex.

Classroom design has important effects on the cognitive functions of students. However, this relationship has rarely been analysed in terms of gender. The aim of the present study, therefore, is to analyse the influence of different design variables (classroom geometry, wall colour, and artificial lighting) on university students' memories from a gender perspective. To do so, 100 university students performed a memory task while visualising different design configurations using a virtual reality setup. Key results show that certain parameters, such as 5.23 m classroom width, 10,500 Kelvin lighting colour temperature, or the blue hue on the walls influence men and women in a similar way, while a purple hue or walls with low colour saturation can generate significantly different behaviour, especially in cognitive processes such as short-term memory. In this study, the use of virtual reality proved to be a useful tool to explore the design effects of virtual learning environments, increasingly present due to training trends and catalysed by the 2020 pandemic. This is a turning point and an international novelty as it will enable the design of classrooms (both physical and virtual) that maximise the cognitive functions of learners, regardless of gender.

The COVID-19 pandemic is known to increase older adults' vulnerability to adverse outcomes. Alongside increased physical frailty, anxiety symptoms associated with the risk of SARS-CoV-2 contagion appear to represent its most prominent 'sequelae'. The attentional and linguistic resources required for decoding virus-related information may also influence the perceived threat of contagion. However, the possible role of neuropsychogeriatric factors on the latter dimension has never been assessed in a longitudinal study on the older population. To fill this gap, 50 healthy cognitively preserved older adults underwent a neuropsychological and physical frailty assessment before the pandemic (T0). Subsequently, they agreed to be interviewed and re-assessed during the lockdown (T1) and immediately after it (T2) through a longitudinal one-year study. Perceived threat of SARS-CoV-2 at T2 was predicted both by baseline anxiety and frailty scores, and by decreased performance in information processing speed and language comprehension tests. While confirming the joint role of frailty and anxiety, a moderation/interaction model showed that each of them was sufficient, at its highest level, to support the maximum degree of perceived threat of contagion. The contribution of neuropsychological factors to perceived threat of SARS-CoV-2 highlights their importance of tailoring information campaigns addressed to older people.

Exposure to lead-containing dusts is a global public health concern. This work addresses an important issue of whether eco-friendly water-based paints reduce the exposure potential of auto-repainting workers to metals. With this aim, metal levels in automobile paints and worker metal exposure were measured using both solvent- and water-based paints. The levels of metals, and particularly Pb, Cr (total), Fe, and Cu, in solvent-based paints varied greatly among colors and brands. Lead concentrations ranged from below the detection limit (~0.25 μg/g) to 107,928 μg/g (dry film) across all samples. In water-based paints, the concentrations of Pb and Cr (total) were generally two to three orders of magnitude lower, but the concentrations of Al and Cu exceeded those in some solvent-based paints. The personal short-term exposure of workers who applied water-based paints of popular colors, such as black and white, were generally low, with Pb levels of less than <4 µg/m³ and Cr (total) levels of less than 1 µg/m³. Conversely, mean short-term exposure to Pb during the painting of a yellow cab using solvent-based paints were 2028 µg/m³, which was ~14 times the Taiwan short-term permissible exposure limit, while the mean level of exposure to Cr (total) was 290 µg/m³, which was well below the exposure limit. This study demonstrates that water-based paints reduce the exposure potential to lead, and highlights the importance of source control in limiting the toxic metals in paints.

This study intends to identify the best model for predicting the incidence of hand, foot and mouth disease (HFMD) in Ningbo by comparing Autoregressive Integrated Moving Average (ARIMA) and Long Short-Term Memory Neural Network (LSTM) models combined and uncombined with exogenous meteorological variables.

Traffic accidents causing nonrecurrent congestion and road traffic injuries seriously affect public safety. It is helpful for traffic operation and management to predict the duration of traffic incidents. Most of the previous studies have been in a certain area with a single data source. This paper proposes a hybrid deep learning model based on multi-source incomplete data to predict the duration of countrywide traffic incidents in the U.S. The text data from the natural language description in the model were parsed by the latent Dirichlet allocation (LDA) topic model and input into the bidirectional long short-term memory (Bi-LSTM) and long short-term memory (LSTM) hybrid network together with sensor data for training. Compared with the four benchmark models and three state-of-the-art algorithms, the RMSE and MAE of the proposed method were the lowest. At the same time, the proposed model performed best for durations between 20 and 70 min. Finally, the data acquisition was defined as three phases, and a phased sequential prediction model was proposed under the condition of incomplete data. The results show that the model performance was better with the update of variables.

Many studies have used sensors attached to adults in order to collect signals by which one can carry out analyses to predict falls. In addition, there are research studies in which videos and photographs were used to extract and analyze body posture and body kinematics. The present study proposes an integrated approach consisting of body kinematics and machine learning. The model data consist of video recordings collected in the UP-Fall Detection dataset experiment. Three models based on long-short-term memory (LSTM) network-4p-SAFE, 5p-SAFE, and 6p-SAFE for four, five, and six parameters-were developed in this work. The parameters needed for these models consist of some coordinates and angles extracted from videos. These models are easy to apply to the sequential images collected by ordinary cameras, which are installed everywhere, especially on aged-care premises. The accuracy of predictions was as good as 98%. Finally, the authors discuss that, by applying these models, the health and wellness of adults and elderlies will be considerably promoted.

Comfort is a significant factor that affects passengers' choice of autonomous vehicles. The comfort of an autonomous vehicle is largely determined by its control algorithm. Therefore, if the comfort of passengers can be predicted based on factors that affect comfort and the control algorithm can be adjusted, it can be beneficial to improve the comfort of autonomous vehicles. In view of this, in the present study, a human-driven experiment was carried out to simulate the typical driving state of a future autonomous vehicle. In the experiment, vehicle motion parameters and the comfort evaluation results of passengers with different physiological characteristics were collected. A single-factor analysis method and binary logistic regression analysis model were used to determine the factors that affect the evaluation results of passenger comfort. A passenger comfort prediction model was established based on the bidirectional long short-term memory network model. The results demonstrate that the accuracy of the passenger comfort prediction model reached 84%, which can provide a theoretical basis for the adjustment of the control algorithm and path trajectory of autonomous vehicles.

Atrial fibrillation (AF) is a common arrhythmia that can lead to stroke, heart failure, and premature death. Manual screening of AF on electrocardiography (ECG) is time-consuming and prone to errors. To overcome these limitations, computer-aided diagnosis systems are developed using artificial intelligence techniques for automated detection of AF. Various machine learning and deep learning (DL) techniques have been developed for the automated detection of AF. In this review, we focused on the automated AF detection models developed using DL techniques. Twenty-four relevant articles published in international journals were reviewed. DL models based on deep neural network, convolutional neural network (CNN), recurrent neural network, long short-term memory, and hybrid structures were discussed. Our analysis showed that the majority of the studies used CNN models, which yielded the highest detection performance using ECG and heart rate variability signals. Details of the ECG databases used in the studies, performance metrics of the various models deployed, associated advantages and limitations, as well as proposed future work were summarized and discussed. This review paper serves as a useful resource for the researchers interested in developing innovative computer-assisted ECG-based DL approaches for AF detection.

Predicting the diffusion rule of toxic gas plays a distinctly important role in emergency capability assessment and rescue work. Among diffusion prediction models, the traditional artificial neural network has exhibited excellent performance not only in prediction accuracy but also in calculation time. Nevertheless, with the continuous development of deep learning and data science, some new prediction models based on deep learning algorithms have been shown to be more advantageous because their structure can better discover internal laws and external connections between input data and output data. The long short-term memory (LSTM) network is a kind of deep learning neural network that has demonstrated outstanding achievements in many prediction fields. This paper applies the LSTM network directly to the prediction of toxic gas diffusion and uses the Project Prairie Grass dataset to conduct experiments. Compared with the Gaussian diffusion model, support vector machine (SVM) model, and back propagation (BP) network model, the LSTM model of deep learning has higher prediction accuracy (especially for the prediction at the point of high concentration values) while avoiding the occurrence of negative concentration values and overfitting problems found in traditional artificial neural network models.

Late preterm children born between 340/7 and 366/7 weeks' gestation account for ≈70% of prematurely born infants. There is growing concern about this population at risk of mild neurodevelopmental problems, learning disabilities and lower academic performance. Following the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) statement, this paper analyzes recent published evidence from 16selected studies involving late preterm children and control group assessments at preschool and/or school age, mainly focusing on cognitive functioning, language learning and academic achievement. The review identifies the assessment tools used in these studies (standardized tests, parental questionnaires and laboratory tasks) and the areas being evaluated from preschool (age 3 years) to primary school levels. Results reveal the presence of mild difficulties, pointing to suboptimal outcomes in areas such as executive function, short term verbal memory, literacy skills, attention and processing speed. Some difficulties are transient, but others persist, possibly compromising academic achievement, as suggested by the few studies reporting on higher risk for poor school performance. Given the increasing number of late preterm children in our society the review highlights the need to implement screening strategies to facilitate early risk detection and minimize the negative effects of this morbidity in childhood.

Subjective well-being is a comprehensive psychological indicator for measuring quality of life. Studies have found that emotional measurement methods and measurement accuracy are important for well-being-related research. Academic emotion is an emotion description in the field of education. The subjective well-being of learners in an online learning environment can be studied by analyzing academic emotions. However, in a large-scale online learning environment, it is extremely challenging to classify learners' academic emotions quickly and accurately for specific comment aspects. This study used literature analysis and data pre-analysis to build a dimensional classification system of academic emotion aspects for students' comments in an online learning environment, as well as to develop an aspect-oriented academic emotion automatic recognition method, including an aspect-oriented convolutional neural network (A-CNN) and an academic emotion classification algorithm based on the long short-term memory with attention mechanism (LSTM-ATT) and the attention mechanism. The experiments showed that this model can provide quick and effective identification. The A-CNN model accuracy on the test set was 89%, and the LSTM-ATT model accuracy on the test set was 71%. This research provides a new method for the measurement of large-scale online academic emotions, as well as support for research related to students' well-being in online learning environments.

Anti-vaccination attitudes have been an issue since the development of the first vaccines. The increasing use of social media as a source of health information may contribute to vaccine hesitancy due to anti-vaccination content widely available on social media, including Twitter. Being able to identify anti-vaccination tweets could provide useful information for formulating strategies to reduce anti-vaccination sentiments among different groups. This study aims to evaluate the performance of different natural language processing models to identify anti-vaccination tweets that were published during the COVID-19 pandemic. We compared the performance of the bidirectional encoder representations from transformers (BERT) and the bidirectional long short-term memory networks with pre-trained GLoVe embeddings (Bi-LSTM) with classic machine learning methods including support vector machine (SVM) and naïve Bayes (NB). The results show that performance on the test set of the BERT model was: accuracy = 91.6%, precision = 93.4%, recall = 97.6%, F1 score = 95.5%, and AUC = 84.7%. Bi-LSTM model performance showed: accuracy = 89.8%, precision = 44.0%, recall = 47.2%, F1 score = 45.5%, and AUC = 85.8%. SVM with linear kernel performed at: accuracy = 92.3%, Precision = 19.5%, Recall = 78.6%, F1 score = 31.2%, and AUC = 85.6%. Complement NB demonstrated: accuracy = 88.8%, precision = 23.0%, recall = 32.8%, F1 score = 27.1%, and AUC = 62.7%. In conclusion, the BERT models outperformed the Bi-LSTM, SVM, and NB models in this task. Moreover, the BERT model achieved excellent performance and can be used to identify anti-vaccination tweets in future studies.

Electronic medical records are an integral part of medical texts. Entity recognition of electronic medical records has triggered many studies that propose many entity extraction methods. In this paper, an entity extraction model is proposed to extract entities from Chinese Electronic Medical Records (CEMR). In the input layer of the model, we use word embedding and dictionary features embedding as input vectors, where word embedding consists of a character representation and a word representation. Then, the input vectors are fed to the bidirectional long short-term memory to capture contextual features. Finally, a conditional random field is employed to capture dependencies between neighboring tags. We performed experiments on body classification task, and the F1 values reached 90.65%. We also performed experiments on anatomic region recognition task, and the F1 values reached 93.89%. On both tasks, our model had higher performance than state-of-the-art models, such as Bi-LSTM-CRF, Bi-LSTM-Attention, and Vote. Through experiments, our model has a good effect when dealing with small frequency entities and unknown entities; with a small training dataset, our method showed 2-4% improvement on F1 value compared to the basic Bi-LSTM-CRF models. Additionally, on anatomic region recognition task, besides using our proposed entity extraction model, 12 rules we designed and domain dictionary were adopted. Then, in this task, the weighted F1 value of the three specific entities extraction reached 84.36%.

PM2.5 not only harms physical health but also has negative impacts on the public's wellbeing and cognitive and behavioral patterns. However, traditional air quality assessments may fail to provide comprehensive, real-time monitoring of air quality because of the sparse distribution of air quality monitoring stations. Overcoming some key limitations of traditional surface monitoring data, Web-based social media platforms, such as Twitter, Weibo, and Facebook, provide a promising tool and novel perspective for environmental monitoring, prediction, and evaluation. This study aims to investigate the relationship between PM2.5 levels and people's emotional intensity by observing social media postings. This study defines the "emotional intensity" indicator, which is measured by the number of negative posts on Weibo, based on Weibo data related to haze from 2016 and 2017. This study estimates sentiment polarity using a recurrent neural networks model based on LSTM (Long Short-Term Memory) and verifies the correlation between high PM2.5 levels and negative posts on Weibo using a Pearson correlation coefficient and multiple linear regression model. This study makes the following observations: (1) Taking the two-year data as an example, this study recorded the significant influence of PM2.5 levels on netizens' posting behavior. (2) Air quality, meteorological factors, the seasons, and other factors have a strong influence on netizens' emotional intensity. (3) From a quantitative viewpoint, the level of PM2.5 varies by 1 unit, and the number of negative Weibo posts fluctuates by 1.0168 units. Thus, it can be concluded that netizens' emotional intensity is significantly positively affected by levels of PM2.5. The high correlation between PM2.5 levels and emotional intensity and the sensitivity of social media data shows that social media data can be used to provide a new perspective on the assessment of air quality.

The contribution of this paper is twofold. First, a new data driven approach for predicting the Covid-19 pandemic dynamics is introduced. The second contribution consists in reporting and discussing the results that were obtained with this approach for the Brazilian states, with predictions starting as of 4 May 2020. As a preliminary study, we first used an Long Short Term Memory for Data Training-SAE (LSTM-SAE) network model. Although this first approach led to somewhat disappointing results, it served as a good baseline for testing other ANN types. Subsequently, in order to identify relevant countries and regions to be used for training ANN models, we conduct a clustering of the world's regions where the pandemic is at an advanced stage. This clustering is based on manually engineered features representing a country's response to the early spread of the pandemic, and the different clusters obtained are used to select the relevant countries for training the models. The final models retained are Modified Auto-Encoder networks, that are trained on these clusters and learn to predict future data for Brazilian states. These predictions are used to estimate important statistics about the disease, such as peaks and number of confirmed cases. Finally, curve fitting is carried out to find the distribution that best fits the outputs of the MAE, and to refine the estimates of the peaks of the pandemic. Predicted numbers reach a total of more than one million infected Brazilians, distributed among the different states, with São Paulo leading with about 150 thousand confirmed cases predicted. The results indicate that the pandemic is still growing in Brazil, with most states peaks of infection estimated in the second half of May 2020. The estimated end of the pandemics (97% of cases reaching an outcome) spread between June and the end of August 2020, depending on the states.

Named Entity Recognition (NER) in the healthcare domain involves identifying and categorizing disease, drugs, and symptoms for biosurveillance, extracting their related properties and activities, and identifying adverse drug events appearing in texts. These tasks are important challenges in healthcare. Analyzing user messages in social media networks such as Twitter can provide opportunities to detect and manage public health events. Twitter provides a broad range of short messages that contain interesting information for information extraction. In this paper, we present a Health-Related Named Entity Recognition (HNER) task using healthcare-domain ontology that can recognize health-related entities from large numbers of user messages from Twitter. For this task, we employ a deep learning architecture which is based on a recurrent neural network (RNN) with little feature engineering. To achieve our goal, we collected a large number of Twitter messages containing health-related information, and detected biomedical entities from the Unified Medical Language System (UMLS). A bidirectional long short-term memory (BiLSTM) model learned rich context information, and a convolutional neural network (CNN) was used to produce character-level features. The conditional random field (CRF) model predicted a sequence of labels that corresponded to a sequence of inputs, and the Viterbi algorithm was used to detect health-related entities from Twitter messages. We provide comprehensive results giving valuable insights for identifying medical entities in Twitter for various applications. The BiLSTM-CRF model achieved a precision of 93.99%, recall of 73.31%, and F1-score of 81.77% for disease or syndrome HNER; a precision of 90.83%, recall of 81.98%, and F1-score of 87.52% for sign or symptom HNER; and a precision of 94.85%, recall of 73.47%, and F1-score of 84.51% for pharmacologic substance named entities. The ontology-based manual annotation results show that it is possible to perform high-quality annotation despite the complexity of medical terminology and the lack of context in tweets.