A Closer Look at Spatial-Slice Features Learning for COVID-19 Detection (preprint)

Conventional Computed Tomography (CT) imaging recognition faces two significant challenges: (1) There is often considerable variability in the resolution and size of each CT scan, necessitating strict requirements for the input size and adaptability of models. (2) CT-scan contains large number of out-of-distribution (OOD) slices. The crucial features may only be present in specific spatial regions and slices of the entire CT scan. How can we effectively figure out where these are located? To deal with this, we introduce an enhanced Spatial-Slice Feature Learning (SSFL++) framework specifically designed for CT scan. It aim to filter out a OOD data within whole CT scan, enabling our to select crucial spatial-slice for analysis by reducing 70% redundancy totally. Meanwhile, we proposed Kernel-Density-based slice Sampling (KDS) method to improve the stability when training and inference stage, therefore speeding up the rate of convergence and boosting performance. As a result, the experiments demonstrate the promising performance of our model using a simple EfficientNet-2D (E2D) model, even with only 1% of the training data. The efficacy of our approach has been validated on the COVID-19-CT-DB datasets provided by the DEF-AI-MIA workshop, in conjunction with CVPR 2024. Our source code will be made available.

Simple 2D Convolutional Neural Network-based Approach for COVID-19 Detection (preprint)

This study explores the use of deep learning techniques for analyzing lung Computed Tomography (CT) images. Classic deep learning approaches face challenges with varying slice counts and resolutions in CT images, a diversity arising from the utilization of assorted scanning equipment. Typically, predictions are made on single slices which are then combined for a comprehensive outcome. Yet, this method does not incorporate learning features specific to each slice, leading to a compromise in effectiveness. To address these challenges, we propose an advanced Spatial-Slice Feature Learning (SSFL++) framework specifically tailored for CT scans. It aims to filter out out-of-distribution (OOD) data within the entire CT scan, allowing us to select essential spatial-slice features for analysis by reducing data redundancy by 70\%. Additionally, we introduce a Kernel-Density-based slice Sampling (KDS) method to enhance stability during training and inference phases, thereby accelerating convergence and enhancing overall performance. Remarkably, our experiments reveal that our model achieves promising results with a simple EfficientNet-2D (E2D) model. The effectiveness of our approach is confirmed on the COVID-19-CT-DB datasets provided by the DEF-AI-MIA workshop.

Exploring the impacts of FDA temporary food labeling flexibility among the food allergy community during the Covid-19 pandemic

In an attempt to improve access to food distribution systems during the Covid-19 pandemic, the United States Food and Drug Administration (FDA) issued temporary flexibility in food labeling requirements. This requirement affects the Top 8 allergens and could lead to accidental allergen ingestion. The purpose of this study was to utilize a qualitative method to collect and analyze comments to FDA that were available via Allergic Living's website from May 22, 2020, to June 18, 2020, following the release of the food labeling change. In total, 3,033 comments were analyzed using the thematic analysis six-phase framework approach. Various themes were extracted relating to the emotional responses of the new labeling including fear, safety concerns, and trust of certain food product brands among the food allergy community. This research has implications for a broad spectrum of the U.S. food industry including restaurants, grocery stores, specialty food associations, and consumers. Stakeholders and experts in the food allergy community should be consulted when a policy change is occurring, especially during a pandemic affecting access to safe food.

Strong Baseline and Bag of Tricks for COVID-19 Detection of CT Scans (preprint)

This paper investigates the application of deep learning models for lung Computed Tomography (CT) image analysis. Traditional deep learning frameworks encounter compatibility issues due to variations in slice numbers and resolutions in CT images, which stem from the use of different machines. Commonly, individual slices are predicted and subsequently merged to obtain the final result; however, this approach lacks slice-wise feature learning and consequently results in decreased performance. We propose a novel slice selection method for each CT dataset to address this limitation, effectively filtering out uncertain slices and enhancing the model's performance. Furthermore, we introduce a spatial-slice feature learning (SSFL) technique\cite{hsu2022} that employs a conventional and efficient backbone model for slice feature training, followed by extracting one-dimensional data from the trained model for COVID and non-COVID classification using a dedicated classification model. Leveraging these experimental steps, we integrate one-dimensional features with multiple slices for channel merging and employ a 2D convolutional neural network (CNN) model for classification. In addition to the aforementioned methods, we explore various high-performance classification models, ultimately achieving promising results.

A Multi-Dimensional Nomological Network of Mobile Payment Continuance

Despite the high utilization of mobile payment during the COVID-19 pandemic, this situation may change in the post-pandemic era. Therefore, great value can be derived from determining the significant antecedents of mobile payment continuance intention. This study looks to do so by introducing a Multi-Dimensional Nomological Network of Mobile Payment Continuance. A two-stage Partial Least Square-Structural Equation Modeling and Artificial Neural Network was utilized for the data analysis. The results provided empirical support to establish the overall nomological network. In addition, more than 70% of the variance in continuance intention was captured. Overall, this study provides practitioners with detailed insights to develop strategies for sustainable utilization and academics with a dynamic framework to look into users' mobile payment continuance intention.

Emergence and Persistent Dominance of Omicron BA.2.3.7 Variant in Community Outbreaks in Taiwan (preprint)

Using high-throughput sequencing of the SARS-CoV-2 genome, we have analyzed 2405 PCR-positive swab samples from 2339 individuals and identified the Omicron BA.2.3.7 variant as a major lineage of the recent community outbreak in Taiwan. Since April 2022, a series of new waves of Omicron cases have surfaced in Taiwan and spread throughout the island. We conducted genomic surveillance with a high success rate and have submitted 1966 full-length Omicron sequences to GISAID. This has permitted identification of signature amino acid changes (ORF1a:L631F, S:K97E, N:M322I) in 1584 (80.6%) of the translated SARS-CoV-2 sequences. The newly established BA.2.3.7 lineage, which is relatively common in Asian countries, is now persistently present in Taiwan. By June 2022 this dominant strain had established a substantial existence in the sequence pool, resulting in additional mutations. The rapid spread and expansion of the Omicron BA.2.3.7 lineage in Asia has had an important socioeconomic impact on health policy.

Embracing mobile shopping: what matters most in the midst of a pandemic?

Purpose>The COVID-19 pandemic has brought about significant changes to the lives of many people. One of which is the accelerated digitalization in the commerce sector. Hence, this study looks to determine the antecedents that affect users' mobile shopping actual use during a pandemic.Design/methodology/approach>The research model was founded on the integration of the Protection Motivation Theory and Transactional Theory of Stress. This is in addition to further extending the integrated research model with other constructs to develop a multi-dimensional framework that accounted for the health, personal, technological and social dimensions. The data was collected from users of mobile shopping through an online survey which was then analyzed via Partial Least Squares–Structural Equation Modeling.Findings>This study provides empirical support to establish the major role of COVID-19 pandemic in affecting the actual usage of mobile shopping. Furthermore, the determinants of actual usage of mobile shopping were found to be multi-dimensional. In particular, the Importance-Performance Map Analysis revealed that emphasis should be place on perceived vulnerability.Originality/value>As other studies focused on users' intentions, this study looked into the factors that influence the actual usage of mobile shopping. Furthermore, this study emphasizes on the contextualization of time and situation in addition to a multi-dimensional approach toward the subject matter. Overall, this study offers practical insights to stakeholders to better tailor their mobile shopping services in times of a pandemic and advances the literature on actual usage of mobile shopping.

Risk factors associated with household transmission of SARS-CoV-2 in Negeri Sembilan, Malaysia.

AIM: Households are a significant venue for the transmission of SARS-CoV-2. We conducted a study to characterise the transmission dynamics and identify risk factors for household transmission of SARS-CoV-2 in Negeri Sembilan, Malaysia. METHODS: This retrospective observational study included 185 families of paediatric COVID-19 cases from 1 February 2020 to 31 December 2020. We identified the index case for each household and gathered the socio-demographic, epidemiological investigation results and risk factors for household transmission from medical case records. The secondary attack rate was calculated, and logistic regression analyses were used to identify risk factors associated with secondary household transmission of SARS-CoV-2. RESULTS: Of the 848 household contacts, 466 acquired secondary infections, resulting in a secondary attack rate of 55%. The median age of the secondary cases was 12 years. Female household contacts and household contacts who slept in the same room with the index case were significantly associated with increased risk for COVID-19. Other independent risk factors associated with higher transmission risk in the household included an index case who was symptomatic, a household index case aged greater than 18 years and a male household index case. CONCLUSIONS: High rates of household transmission of COVID-19 were found, indicating households were a major setting of transmission of SARS-CoV-2. Our data provide insight into the risk factors for household transmission of SARS-CoV-2 in Malaysia.

Clinical and epidemiological characteristics of children with COVID-19 in Negeri Sembilan, Malaysia.

OBJECTIVES: To describe the clinical and epidemiological characteristics of children with coronavirus disease 2019 (COVID-19) in the state of Negeri Sembilan, Malaysia in the setting of mandatory hospital isolation and quarantine for all confirmed cases. METHODS: A multi-centre, retrospective observational study was performed among children aged ≤12 years with laboratory-proven COVID-19 between 1 February and 31 December 2020. RESULTS: In total, 261 children (48.7% males, 51.3% females) were included in this study. The median age was 6 years [interquartile range (IQR) 3-10 years]. One hundred and fifty-one children (57.9%) were asymptomatic on presentation. Among the symptomatic cases, fever was the most common presenting symptom. Two hundred and forty-one (92.3%) cases were close contacts of infected household or extended family members. Twenty-one (8.4%) cases had abnormal radiological findings. All cases were discharged alive without requiring supplemental oxygen therapy or any specific treatment during hospitalization. The median duration of hospitalization was 7 days (IQR 6-10 days). One (2.1%) of the uninfected guardians accompanying a child in quarantine tested positive for severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) upon discharge. CONCLUSIONS: COVID-19 in children was associated with mild symptoms and a good prognosis. Familial clustering was an important epidemiologic feature in the outbreak in Negeri Sembilan. The risk of transmission of SARS-CoV-2 from children to guardians in hospital isolation was minimal despite close proximity.

The Relationship between Endurance Involvement and Travel Behavior in Camping and the Moderating Effect of Place Attachment

Camping tourism is one of the fastest growing segments of the tourism industry. Global trends in camping show that this type of recreational activity has begun to emerge in Asia. Meeting the expectations of potential and current tourists in a camping destination will lead to the accomplishment of competitive advantage. The endurance involvement and place attachment play important roles in understanding camping behaviors. However, limited research identified place attachment as a moderator of the influence of endurance involvement on camping travel behavior. To fill this gap, based on place attachment theory, our study developed a conceptual model that postulates endurance involvement and place attachment as predictors of camping tourism. 216 self-administered questionnaires were retuned. Confirmatory factor analysis was used to test the psychometric properties of the measures, while regression analysis was employed to assess the research hypotheses. The study results show that endurance involvement in camping affects camping travel behavior, whereas camping place dependence, camping place identity, and camping affective attachment moderate the effect of camping endurance involvement on camping travel behavior. Camping place attachment strengthens the relationship between camping endurance involvement and camping travel behavior. To maintain the sustainable development of camping tourism, campsite authorities should attach more importance to strengthening campers’ emotional bonds and concentrate on how to encourage them to participate in tourism activities. The findings have several theoretical and managerial implications for camping tourism marketing and the development of campground.

Clinical Outcomes of Patients With and Without HIV Hospitalised with COVID-19 in England During the Early Stages of the Pandemic: A Matched Retrospective Multicentre Analysis (RECEDE-C19 Study) (preprint)

Background: It is unclear from epidemiological data for COVID-19 infections, whether people living with HIV (PLWH) have a different outcomes compared to the general population. We conducted a multi-centre, retrospective matched cohort study of SARS-CoV-2 PCR-positive hospital inpatients analysed by HIV-status.Methods: HIV-negative patients were matched to PLWH admitted to hospital before 31 st May 2020, with a 3:1 ratio by: hospital site, SARS-CoV-2 test date +/- 7 days, age +/- 5 years, gender, and index of multiple deprivation decile (IMDD) +/- 1. The primary objective was clinical improvement (≥2-point improvement on a 7-point ordinal scale) or hospital discharge by day 28, whichever was earlier.Results: 68 PLWH and 181 HIV-negative comparators were included. After adjustment for ethnicity, frailty, baseline hypoxia, duration of symptoms prior to baseline, body mass index categories, and comorbidities (hypertension, chronic cardiac disease, chronic lung disease, active malignancy, diabetes, and chronic renal disease), the effect size of HIV status was not associated with time to clinical improvement or discharge from hospital (aHR 0.70, 95%CI 0.43, 1.17; p=0.18), despite unadjusted hazards of PLWH achieving the primary outcome being 43% lower (p=0.005). Baseline frailty (aHR=0.79; 95%CI 0.65, 0.95; p=0.011), malignancy (aHR=0.37; 95%CI 0.17, 0.82; p=0.014) remained associated with poorer outcomes. PLWH were more likely of black and minority ethnicities (75.0% vs 48.6%, p=0.0002), higher median clinical frailty score (3 IQR 2-5 vs 2 IQR 1-4, p=0.0069), higher proportion of active malignancy (14.4% vs 9.9%, p=0.29). Median body mass index (BMI) was lower amongst PLWH (27.7 IQR 23.9-32.3 vs 29.4 IQR 24.7-34.3, p=0.19). Median CD4 count of PLWH was 352cells/µL (IQR 235-619) and 95.7% had suppressed viral loads <200copies/mL, 63/68 (92.3%) were taking antiretroviral therapy.Conclusions: Differences in clinical outcomes of COVID-19 hospitalisations in PLWH may be due to other important factors including increased frailty and comorbidities such as malignancies, rather than HIV-status alone.Funding Statement: This study has not received any funding sources.Declaration of Interests: MJL has received grants and honoraria from Gilead Sciences and Viiv Healthcare not related to this work. SF has received research grants to her institution from NIH, MRC, BMGF. JT has received support for virtual conference registration from ViiV Healthcare and research grants from the Medical Research Council and the British HIV Association not related to this work. CvH has received educational grants, conference support and advisory board fees from ViiV Healthcare, Gilead Sciences, MSC not related to this work. MP reports grants and personal fees from Gilead Sciences and personal fees from QIAGEN, outside the submitted work. MP is supported by a NIHR Development and Skills Enhancement Award (NIHR301192) and in receipt of funding from UKRI / MRC (MR/V027549/1). He acknowledges the support from UKRI, the NIHR Leicester BRC and NIHR ARC East Midlands. No other competing interests, financial relationships with any organisations that might have an interest in the submitted work, or other relationships or activities that could appear to have influenced the submitted work have been reported by other authors.Ethics Approval Statement: Ethical approval was granted by the UK Health Research Authority (REC reference 20/HRA/2278).

Sequence variation among SARS-CoV-2 isolates in Taiwan (preprint)

Taiwan experienced two waves of imported cases of coronavirus disease 2019 (COVID-19), first from China in January to late February, followed by those from other countries starting in early March. Additionally, several cases could not be traced to any imported cases and were suspected as sporadic local transmission. Twelve full viral genomes were determined in this study by Illumina sequencing either from virus isolates or directly from specimens, among which 5 originated from clustered infections. Phylogenetic tree analysis revealed that these sequences were in different clades, indicating that no major strain has been circulating in Taiwan. A deletion in open reading frame 8 was found in one isolate. Only a 4-nucleotide difference was observed among the 5 genomes from clustered infections.