A Survey of Human-Computer Interaction Technology in Intelligent Home for Children's Learning

The development of the COVID-19 epidemic has increased the home learning time of children. More researchers began to pay attention to children's learning in home. This survey reviewed the frontier and classic cases in the field of interactive design of children's home learning in the past five years, analyzed tangible user interface, augmented reality, and multimodal interaction in human-computer interaction of children's home learning. This paper reviewed the application of interactive system in children's learning and points out its positive side in development of ability, process of learning, habits of learning, and environment of learning of children. Through analysis, we advise that it is necessary to create home learning applications, link smart home systems, and build an interactive learning environment for smart home learning environment design. Finally, we point out the technical and ethical problems existing in the current research, proposes that intelligent perception, emotion recognition, and expression technologies should be introduced in the future, and looks forward to the development of this field. © 2023 Institute of Computing Technology. All rights reserved.

A deep learning approach for automatic echocardiographic right ventricular strain measurements using a limited dataset

Background: Speckle tracking echocardiography provides quantification of myocardial deformation and is useful in the assessment of myocardial function. Right ventricular (RV) strain has been suggested as a sensitive tool for diagnosing cardiomyopathies and assessing long term patient outcomes for patients with pulmonary hypertension, severe tricuspid regurgitation and COVID-19 infection. Recent advances in deep learning (DL) have made promising advances in automating the labour-intensive delineation of regions of interest (ROIs). However, compared to echocardiograms with left ventricular (LV) strain, RV strain data is scarce, making DL models difficult to train. Purpose(s): To investigate whether annotated LV strain data could be beneficial in training a DL model for automatic RV strain when using a limited RV dataset. Method(s): The dataset consisted of anonymized still frames from 141 echocardiograms of the RV in the RV-focused 4 chamber view with corresponding cardiologist-defined ROI. Exams included healthy subjects and patients with heart failure, valvular disease, and conduction abnormalities. ROIs and still images were extracted at the mid-systole, and then quality assessed by an experienced cardiologist as high, medium, or low. The dataset was randomly split into 68%/17%/15% sets for training, validation, and testing. A convolutional neural network for image segmentation (UNet) with a residual neural network (ResNet50) encoder was used, with a combination of binary cross entropy and Dice loss functions. Augmentation, predefined ImageNet weights and pre-training were also employed. For pre-training, 715 still images in the apical 4 chamber view with LV defined ROIs were used, both in their original and horizontally flipped view. Predicted ROIs were reintroduced into commercially available echocardiogram analysis software to automatically calculate longitudinal strain (LS) values. Result(s): The model pre-trained with the flipped LV images achieved the highest performance with a mean absolute difference of 1.26 percentage points (95% confidence interval (CI): 0.62-1.89 percentage points) between manually measured and DL-assisted LS. Median absolute LS difference was 0.85 (95% CI: 0.28-1.57) percentage points. A Bland-Altman plot revealed two outliers and no obvious trends. In comparison, the mean and median absolute LS differences for the model without pre-training were 1.87 (95% CI: 0.73-3.00) and 1.09 (95% CI: 0.56-1.63) percentage points, respectively. Conclusion(s): The current study demonstrates that DL-assisted, automated RV strain measurement is feasible even with a small dataset, and that performance can be increased by using images annotated for LV strain. While the majority of the predicted RV strain results were within the typical range of intra- and interobserver variability, a few outliers were observed. These outliers could possibly be avoided with the use of larger datasets.

[Clinical update in non-invasive positive pressure ventilation for respiratory failure in 2022].

In this article, we searched the research literatures related to clinical investigation of non-invasive positive pressure ventilation (NPPV) in acute respiratory failure(ARF)/chronic respiratory failure(CRF) between 1st October 2021 and 30th September 2022 through Medline, and reviewed the important advances. Three prospective randomized controlled studies related to the efficacy and safety of NPPV and/or high-flow nasal cannula oxygen therapy (HFNC) on patients with COVID-19 with ARF were reported, showing that NPPV (including continuous positive airway pressure and bilevel positive airway pressure) was able to reduce the intubation rate, but the efficacy of HFNC was contradictory. In addition, progress has been made in outcome prediction models for ARF treated with NPPV, NPPV-related cardiac arrest, and the impact of human-machine interface on NPPV treatment outcomes. The effects of NPPV as preoxygenation method before intubation was reported to be able to reduce severe desaturation during intubation, especially in obese population. The use of NPPV in extubated patients resulting in reduced reintubation rate was also studied. With regard to long-term home application of NPPV, five indicators of successful initiation were proposed, but the success rate was low in clinical practice. Some reports showed that psychological support could improve the adherence to NPPV. The results of these studies contributed to the rational selection and optimal application of NPPV in clinical practice.

Effects of covid-19 emergency response levels on air quality in the guangdong-hong kong-macao greater bay area, china

The COVID-19 pandemic has affected air quality due to extreme changes in human behavior. We assessed the air quality response to different emergency levels during different COVID-19 periods and the naught period in the Guangdong-Hong Kong-Macao Greater Bay Area (GBA). We obtained the following conclusions: (1) The measures implemented to prevent and control of COVID-19 directly impacted ambient air pollutants. The air quality index and the concentrations of NO2, PM2.5, PM10, and CO in the GBA for 1–19 January 2020 declined 19.4%, 16.7%, 27.5%, 15.8%, and 25.7%, respectively, compared to the same time period in 2016–2019. (2) The reduction in air pollution was strongly associated with the first-level emergency response during this pandemic. The AQI, NO2, PM2.5, PM10, O3, and CO in the GBA decreased by 37.4%, 47.0%, 40.5%, 44.8%, 6.7%, and 24.1%, respectively. We found no statistically significant difference in the concentrations of different pollutants, except for NO2, during the second-and third-level responses. (3) The higher the emergency response level, the greater the NO2 pollutants reduction. The NO2 concentration was reduced by 47.0%, 25.5%, and 12.1% at emergency response levels 1, 2, and 3, respectively. The results highlight the importance of understanding the role of emergency response in air quality, and provide reference for authorities to formulate more scientific and reasonable emergency responses to epidemic prevention and control. © The Author(s).