Classification of COVID-19 and non-COVID-19 lung computed tomography images using machine learning

A novel coronavirus disease 2019 (COVID-19) was first reported in Wuhan, China in late December 2019. In March 2020, World Health Organization (WHO) declared this sudden epidemic as a global pandemic. It is highly contagious and can cause serious lung inflammation. The typical symptoms are fever, cough, shortness of breath, headache and sore throat. Till 23 August 2021, a total of more than 211 million cases of COVID-19 have been reported to WHO worldwide, with a total of more than 4.4 million of deaths. Hence early detection is crucial to control the spread. Currently, the key diagnosis method is the reverse transcription polymerase chain reaction (RT-PCR) test using swab samples. However, it is subject to certain limitations, such as low sensitivity and shortage of kits. To address these issues, lung computed tomography (CT) scan can be the alternative as it is fast, easy, and proven to be sensitive in detecting COVID-19 cases. This study presents an automated method to differentiate the COVID-19 CT images from the Non-COVID-19 images using different convolutional neural networks (CNN) through three stages procedures. In the first stage, the dataset which consists of 746 images of COVID-19 and Non-COVID-19 was split into 3 parts for training, validation, and testing, respectively. The training and validation data were then applied with different augmentation techniques to increase the dataset, while the testing data remained with no augmentation. In stage 2, 10 different pretrained CNNs were initialized to train and classify the binary class. In stage 3, gradient descent class activation mapping (GradCAM) was used for abnormality localization. The best performance was achieved by ResNet152, ResNeXt, GoogleNet, and DenseNet201 with the highest overall accuracy of 98.51%. ResNet152, GoogleNet, and DenseNet201 had achieved a sensitivity of 100%, and specificity of 97.06%, whereas ResNeXt had achieved a sensitivity of 96.97%, and specificity of 100%. © 2023 Elsevier Inc. All rights reserved.

FROM RESEARCH TO PRACTICE: bridging the “ knowledge-action” gap

This chapter discusses the inadequate application of the available scientific knowledge to the improvement of human health in urban areas, leading to what we define as a “knowledge-action” gap. The current state of efforts to implement the “Nature as Health” concept is reviewed, and recommendations are made to adopt an evidence-based design process as a framework for addressing the knowledge-action gap and the science-action gap. While Chapter 39 provided case studies from near the Pacific Rim, this chapter details the cases in Taiwan that bridged the knowledge-action gap in the relationship between nature and public health. Chapter 40 discussed several research tools, methods, and interdisciplinary concepts that might lend light to future studies. This chapter will discuss more detailed information of the HealthCloud app and its application, which provides psychological questionnaires and monitors heart rate and environmental information and could be a useful tool for data collection on the changing behaviors and patterns of humans experiencing nature, especially during the COVID-19 pandemic. A “Landscape and Health Information Note” application could be used to connect health data and environmental information as a feedback system for users. The chapter concludes with the following questions: what is the future relationship between human and environment interaction? Given the rapid advancement of technology and the COVID-19 pandemic, how can we apply this technology and the strategies of landscape design to fill the knowledge-action gap? Finally, how can we respond to sustainable development goals (SDGs)? © 2022 selection and editorial matter, Yizhao Yang and Anne Taufen;individual chapters, the contributors.

Clinical severity of COVID-19 with omicron variant predominance in relation to vaccination status, age, comorbidities- a single center in Selangor, Malaysia.

INTRODUCTION: Recently, the rapid surge of reported COVID-19 cases attributed to the Omicron variant of severe acute respiratory syndrome coronavirus (SARS-CoV-2) created an immediate concern across nations. Local information pertaining to the new variant of concern (VOC) is lacking. We aimed to determine the clinical characteristics of COVID-19 during a period of Omicron prevalence among patients hospitalised from February 1 to 21, 2022 at Sungai Buloh Hospital and to estimate the risks of disease progression presumably caused by this variant in association with gender, age, comorbidity, and vaccination status. MATERIALS AND METHODS: In this retrospective, singlecentered, retrospective cohort study, all hospitalised adults with laboratory-confirmed COVID-19, aged 18 and above, were recruited from February 1 to 21, 2022. Clinical characteristics, investigations, and outcomes were assessed. RESULTS: A total of 2279 patients aged 18 years and above with laboratory-proven COVID-19 were recruited and analysed, excluding 32 patients owing to incomplete data. Majority of the study population had a mean age of 41.8 ± 17.7, was female-predominant (1329/2279, 58.6%), had completed a primary series of vaccination with a booster (1103/2279, 48.4%), and had no underlying medical conditions (1529/2279, 67.4%). The risk of COVID-19-related disease progression was significantly lower in hospitalised patients under the age of 50 who were female, had no comorbidity, and had completed two doses of the primary series with or without a booster. [respectively, OR 7.94 (95% CI 6.16, 10.23); 1.68 (1.34, 2,12); 2.44 (1.85, 3.22); 2.56 (1.65, 3.97), p< 0.001]. CONCLUSION: During the period of Omicron prevalence, a favourable outcome of COVID-19 was strongly associated with female gender, age below 50, a comorbidity-free condition, and having completed immunization. With this new observation, it could help improve public health planning and clinical management in response to the emergence of the latest VOC.

COVID-19 Racial Discrimination and Mental Health of Korean Americans: Role of Ethnic Identity and Coping Strategy

The onset of COVID-19 has resulted in higher rates of racial discrimination toward Asian American and Pacific Islanders, including Korean Americans. This study used moderated mediation analyses to examine the relationship between COVID-19-related racial discrimination and anxiety, depression, and life satisfaction among Korean Americans (KA;N = 270) and explored the buffering effect of ethnic identity and coping strategies. Experiences of pandemic-related racial discrimination were linked to the severity of anxiety and depression levels among KA, which resulted in lower levels of life satisfaction. Coping strategies moderated the link between depression and life satisfaction but not between anxiety and life satisfaction. Ethnic identity exacerbated the relationship between racial discrimination and levels of anxiety and depression. The significance of these findings highlights the important role of coping strategy and ethnic identity in mental health among KA during the pandemic. Based on these findings, implications for professional counselors are outlined.

Implementation of E-Learning in New Taipei City During COVID-19

During the COVID-19 pandemic, educational institutions around the world have closed, affecting more than 60% of students and causing massive disruption to the education system. Taiwan is no exception. For this sudden and dramatic change, teachers, students, and parents all confront significant challenges. In order to make specific suggestions for improvement, the study aims to explore the current state of e-learning in Taiwan and to understand the difficulties faced by teachers, parents, and students. The study conducted in-depth interviews with 20 teachers, 12 parents, and 24 students. The research results show that in New Taipei City, there are barely any online courses for grades 1 and 2, blended learning is mainly for grades 3 and 4, and synchronous e-learning is designed for grades 5 and 6. The main challenges in adopting e-learning in primary schools include, as follows: (i) Teachers, parents, and students are unfamiliar with the user interface of the e-learning platform. (ii) Insufficient hardware and software equipment at home. (iii) Teachers cannot take care of special students. (iv) Communication between teachers and parents is not smooth. (v) Difficulty in assessing learning progress online. (vi) Students are easily distracted from their studies. Based on the above research results, the researchers put forward specific suggestions for future online teaching practices. © 2022 by the authors;licensee Modestum. All Rights Reserved.

A New TPACK Training Model for Tackling the Ongoing Challenges of COVID-19

This study investigated the effects of integrating the “CloudClassRoom” (CCR) and the DEmo-CO-design/teach-feedback-DEbriefing (DECODE) model to improve pre-service teachers’ online technological pedagogical and content knowledge (TPACK). The DECODE model includes four stages: Teacher’s DEmonstrations, Students CO-train in using CloudClassRoom, Students CO-design a CloudClassRoom-integrated course, Students CO-teach, and finally DE-brief what they have learned through the stages mentioned above. This model integrates teacher-student experiences, teaching-learning processes, and technology-embedded systems to promote collaborative and active learning, information and resources sharing, and creative communication. A self-evaluating questionnaire with open-ended questions evaluated participants’ technological pedagogical and content knowledge outcomes. CloudClassRoom significantly increases technology-related knowledge considering the current social distancing measures provoked by COVID-19. The findings show that DECODE with CloudClassRoom provides an integrated process for improving pre-service teachers’ technological pedagogical and content knowledge, assisting pre-service teachers in designing educational technology-integrated courses. © 2022 by the authors. Licensee MDPI, Basel, Switzerland.

Exploring the relationship between landscape features and brain activation using vision AI

This research aims to investigate landscape features depicted by machine learning correlated with brain activation on the emotional response. Functional magnetic resonance imaging (fMRI) was used to scan participants’ brains while viewing various types of environmental images. The analysis focuses on emotion-related brain activation which is related to emotion and mental health. By using the Google Vision AI, this study tried to identify labeled visual elements of the images by AI feature detection (REST and RPC API) to understand what environmental features potentially predict emotion response based on the neuroimaging data. The study explored the use of feature detection and fMRI data to construct research on the visual landscape assessment field. This method is expected to be more accurate and objectively reveal the relationship between human cognitive health and environmental factors. While the COVID-19 pandemic has accelerated human demands for new technology, it has also given rise to new possible applications of artificial intelligence, this research corresponds to the use of AI and the neuro-activation that could become a new resource of many decision-making grounds. © 2021 International Society for Horticultural Science. All rights reserved.

Promoting students' learning achievement and self-efficacy: A mobile chatbot approach for nursing training

The aims of nursing training include not only mastering skills but also fostering the competence to make decisions for problem solving. In prenatal education, cultivating nurses' knowledge and competence of vaccine administration is a crucial issue for protecting pregnant women and newborns from infection. Therefore, obstetric vaccination knowledge has become a basic and essential training program for nursing students. However, most of these training programs are given via the lecture-based teaching approach with skills practice, providing students with few opportunities to think deeply about the relevant issues owing to the lack of interaction and context. This could have a negative impact on their learning effectiveness and clinical judgment. To address this problem, a mobile chatbot-based learning approach is proposed in this study to enable students to learn and think deeply in the contexts of handling obstetric vaccine cases via interacting with the chatbot. In order to verify the effectiveness of the proposed approach, an experiment was implemented. Two classes of 36 students from a university in northern Taiwan were recruited as participants. One class was the experimental group learning with the proposed approach, while the other class was the control group learning with the conventional approach (ie, giving lectures to explain the instructional content and training cases). The results indicate that applying a mobile chatbot for learning can enhance nursing students' learning achievement and self-efficacy. In addition, based on the analysis of the interview results, students generally believed that learning through the mobile chatbot was able to promote their self-efficacy as well as their learning engagement and performance. Practitioner notes What is already known about this topic Issues relevant to AI technology in education have been extensively discussed and explored around the world. Among the various AI systems, the potential of chatbots has been highlighted by researchers owing to the user-friendly interface developed using the natural language processing (NLP) technology. Few studies using AI chatbots in professional training have been conducted. What this paper adds In this study, a mobile chatbot was used in a nursing training program to enhance students' learning achievement and self-efficacy for handling vaccine cases. The mobile chatbot significantly improved the students' learning achievement and self-efficacy in comparison with the conventional learning approach in the vaccine training program. From the interview results, it was found that the students generally believed that the mobile chatbot was able to promote their self-efficacy as well as learning engagement and performances in the vaccine training program. Implications for practice and/or policy Mobile chatbots have great potential for professional training owing to their convenient and user-friendly features. It would be worth applying mobile chatbots as well as other NLP-based applications to other professional training programs in the future. © 2021 British Educational Research Association.