Exploring the Preference of Corporations for Sponsorship Motives and the Impact of Sponsorship Motives on Sponsoring Intention in Post-Epidemic Era: Using Two Different Approaches-FPR and SEM

The aim of this study is to explore the preference of corporations for sponsorship motives and the impact of sponsorship motives on sponsoring intention in the post-epidemic era of COVID-19. Taking a Taiwanese company as a case study, a total of 60 expert groups comprising 300 respondents (the management of the sampled companies) were surveyed in the post-epidemic period, with data being collected from 60 sampled companies in February-May 2022. Data were analyzed by using two different functional approaches, including fuzzy preference relations (FPR) for the first survey (study 1) and structural equation model (SEM) for the second survey (study 2). Results reveal that corporate image is the most preferential motive of sponsorship and also demonstrates the most significant and positive influence on sponsoring intention. Meanwhile, the measured factor of performing corporate social responsibility (CSR) appears the most correlated with the construct of corporate image. Based on the results, the study can fully fill the gap between sponsorship motives and sponsoring intention in sponsorship knowledge. Additionally, the conjunction of FPR and SEM can also create methodological synergies, namely, enhancing complementary effects and achieving better holistic analysis. Findings also suggest that special attention should be paid to CSR, which plays a pivotal role in affecting the decision of corporations for sponsorship motives and sponsoring intention and, in a post-epidemic era, continuing to develop CSR actions to enhance corporate image can be the best strategy while facing internal and external challenges of implementing sustainable development (SD).

International Residents' Perspectives on Education and Challenges in Microsurgery Training

Objective Microsurgery remains an integral component of the surgical skillset and is essential for a diversity of reconstructive procedures. The apprenticeship also requires overcoming a steep learning curve, among many challenges. The method of microsurgical training differs depending on the countries' regions and resources of their health care system. Methods The Journal of Hand and Microsurgery leadership held an international webinar on June 19, 2021, consisting of a panel of residents from 10 countries and moderated by eminent panelists. This inaugural event aimed to share different experiences of microsurgery training on a global scale, identifying challenges to accessing and delivering training. Results Residents shared various structures and modes of microsurgical education worldwide. Areas of discussion also included microsurgical laboratory training, simulation training, knowledge sharing, burnout among trainees, and challenges for female residents in microsurgical training. Conclusion Microsurgical proficiency is attained through deliberate and continued practice, and there is a strong emphasis globally on training and guidance. However, much remains to be done to improve microsurgical training and start acting on the various challenges raised by residents.Copyright © 2022. Society of Indian Hand & Microsurgeons. All rights reserved.

PQSCT: Pseudo-Siamese BERT for Concept Tagging With Both Questions and Solutions

The global outbreak of the new coronavirus epidemic has promoted the development of intelligent education and the utilization of online learning systems. In order to provide students with intelligent services such as cognitive diagnosis and personalized exercises recommendation, a fundamental task is the concept tagging for exercises, which extracts knowledge index structures and knowledge representations for exercises. Unfortunately, to the best of our knowledge, existing tagging approaches based on exercise content either ignore multiple components of exercises, or ignore that exercises may contain multiple concepts. To this end, in this paper, we present a study of concept tagging. First, we propose an improved pre-trained BERT for concept tagging with both questions and solutions (QSCT). Specifically, we design a question-solution prediction task and apply the BERT encoder to combine questions and solutions, ultimately obtaining the final exercise representation through feature augmentation. Then, to further explore the relationship between questions and solutions, we extend the QSCT to a pseudo-siamese BERT for concept tagging with both questions and solutions (PQSCT). We optimize the feature fusion strategy, which integrates five different vector features from local and global into the final exercise representation. Finally, we conduct extensive experiments on real-world datasets, which clearly demonstrate the effectiveness of our proposed models for concept tagging. IEEE

Restoration of Humoral Response to Covid-19 Vaccination in Rituximab-Treated Patients

Background: Previous studies had demonstrated that patients with hematologic malignancies had suboptimal antibody response after receiving COVID-19 vaccines, especially among those having previously treated with anti- CD20 monoclonal antibodies. Method(s): Adult patients with non-Hodgkin's lymphoma or chronic lymphocytic leukemia (CLL) were enrolled before receiving the second dose of SARS-CoV-2 vaccine. Determinations of anti-SARS-CoV-2 spike and nucleocapsid IgG titers were performed every 1-3 months, after they received the second and the third dose of SARS-CoV-2 vaccine, respectively. Patients were excluded from analysis if they were diagnosed with COVID-19. All serum samples were tested for anti-nucleocapsid antibody and those tested positive were excluded from subsequent analyses. Result(s): A total of 85 participants were enrolled, including 42 (49.4%) with diffused large B-cell lymphoma, and 13 (15.3) with follicular lymphoma and 9 with CLL. 72 (84.7%) participants had received anti-CD20 monoclonal antibodies, with a median interval of 24 months between last anti-CD20 treatment and the second dose of vaccine, and 21 (24.7%) had HIV infection. Factors associated with failure to achieve an anti-spike IgG titer >141 BAU/ mL within 12 weeks after the second dose of vaccine included HIV infection (adjusted odds ratio [aOR], 0.14;95% CI, 0.04-0.51), active hematologic disease (aOR, 5.50;95% CI 1.42-21.32), receipt of anti-CD20 monoclonal antibodies (aOR, 6.65;95% CI 1.52-29.07), and receipt of two doses of homologous mRNA vaccination (aOR, 0.17;95% CI 0.05-0.56). In the participants having previously treated with anti-CD20 regimen, only 8.6% achieved an antibody response ( >141 BAU/mL) in the first year, while 78.3% achieved anti-spike IgG titer > 141 BAU/mL after two years post B-cell depleting treatment. After the third dose of SARS-CoV-2 vaccine, 53.6% achieved an antispike IgG titer > 141 BAU/mL in the first year post anti-CD20 treatment. Conclusion(s): Our study demonstrated that previous treatment with anti-CD20 monoclonal antibodies was associated a lower antibody response among patients with lymphoproliferative disorders receiving two doses of SARS-CoV-2 vaccine. While two doses of SARS-CoV-2 vaccines might not be sufficient even one year apart from the last dose of rituximab, a third dose of vaccine may boost anti-spike IgG particularly in the subset of recent exposure to rituximab. Anti-spike IgG determined 1-3 months after the second (A) / third (B) dose of COVID-19 vaccine, stratified by the interval between last anti-CD20 regimen and the second / third dose of COVID-19 vaccine. (Figure Presented).

Cost Advantages of Far East/Europe Trunk Route Deployment with Port Selection in East Asia

To cater to the gradually increasing sizes of ships, several traditional container ports in East Asia built deep-water wharves to attract shipping carriers to berth, a decision that is considered highly reasonable because it allows for shipping carriers to gain a cost advantage. For traditional Far East/Europe (F/E) trunk routes, shipping carriers must deploy vessels that are large enough at hub ports to maintain low transshipment costs. However, for a port to attract shipping carriers, it should be able to first meet the cargo demand of these carriers. The port would also need to improve the loading ratio to enjoy the cost advantage. Simultaneously, the port should leverage the loading and unloading efficiency of the terminal to gain a competitive advantage. Although the port congestion observed at the F/E trunk during COVID-19 was not as serious as that in North American ports, it was sufficient to affect the route deployment and port selection decisions of shipping carriers. Currently, because the size of container carriers is the most critical factor in the reduction of shipping costs, as demonstrated in this study, the upsizing trend of container ships is regarded as a highly relevant aspect in the deployment of trunk routes and the selection of hub ports. © 2023 National Taiwan Ocean University.

Stock Comovement and Financial Flexibility

We develop a dynamic model of corporate investment and financing, in which shocks to the value of collateralizable assets generate variation in firms' debt capacity. We show that the degree of similarity among firms' financial flexibility forecasts cross-sectional variation in return correlation. We test the implications of the model with firm-level data in two empirical analyses using i) an instrumental variable approach based on shocks to the value of collateralizable corporate assets and ii) the outbreak of the COVID-19 crisis as an event study. We find that firms in the same percentile of the cross-sectional distribution of financial flexibility have 62% higher correlation in stock-return residuals than firms 50 percentiles apart.

Impacts of COVID-19 Pandemic on Italian Electricity Demand and Markets

In this paper, the short-, medium-, and long-term effects of the COVID-19 pandemic on the Italian power system, particularly electricity consumption behavior and electricity market prices, are investigated by defining various metrics. The investigation reveals that COVID-19 lockdown caused a drop in load consumption and, consequently, a decrement in day-ahead market prices and an increase in ancillary service prices. © 2015 CSEE.

Keep it real: Assessing destination image congruence and its impact on tourist experience evaluations

Inaccurate promotional information about tourist destinations may result in tourists' negative evaluations. This study proposes a new approach to measure the congruence between projected and received images of a destination's attractions. Based on online textual data, this study investigates how image congruence influences tourists' evaluations of their destination experiences. Using promotional messages and reviews of attractions in Hainan, China obtained from a leading Chinese online travel agency (Ctrip) and a three-way fixed-effects regression model, this study demonstrates that image congruence positively affects tourists' appraisal of their destination experiences. External crises (e.g., the COVID-19 pandemic), the readability of promotional messages, and tourists' expertise moderate this relationship, reducing the positive impact of image congruence on tourist experience evaluation. This study bridges theoretical and empirical gaps in destination image (in)congruence research, informing tourism marketing agencies of effective promotional strategies in different contexts. © 2023

CODA-19: Using a non-expert crowd to annotate research aspects on 10,000+ abstracts in the COVID-19 open research dataset

This paper introduces CODA-19, a human-annotated dataset that codes the Background, Purpose, Method, Finding/Contribution, and Other sections of 10,966 English s in the COVID-19 Open Research Dataset. CODA-19 was created by 248 crowd workers from Amazon Mechanical Turk within 10 days, and achieved labeling quality comparable to that of experts. Each was annotated by nine different workers, and the final labels were acquired by majority vote. The inter-annotator agreement (Cohen's kappa) between the crowd and the biomedical expert (0.741) is comparable to inter-expert agreement (0.788). CODA-19's labels have an accuracy of 82.2% when compared to the biomedical expert's labels, while the accuracy between experts was 85.0%. Reliable human annotations help scientists access and integrate the rapidly accelerating coronavirus literature, and also serve as the battery of AI/NLP research, but obtaining expert annotations can be slow. We demonstrated that a non-expert crowd can be rapidly employed at scale to join the fight against COVID-19. © ACL 2020.All right reserved.

Effect of COVID-19 lockdown on the characterization and mixing state of carbonaceous particles in the urban atmosphere of Liaocheng, the North China Plain

To investigate the effect of COVID-19 control measures on aerosol chemistry, the chemical compositions, mixing states, and formation mechanisms of carbonaceous particles in the urban atmosphere of Liaocheng in the North China Plain (NCP) were compared before and during the pandemic using a single particle aerosol mass spectrometry (SPAMS). The results showed that the concentrations of five air pollutants including PM2.5, PM10, SO2, NO2, and CO decreased by 41.2%-71.5% during the pandemic compared to those before the pandemic, whereas O3 increased by 1.3 times during the pandemic because of the depressed titration of O3 and more favorable meteorological conditions. The count and percentage contribution of carbonaceous particles in the total detected particles were lower during the pandemic than those before the pandemic. The carbonaceous particles were dominated by elemental and organic carbon (ECOC, 35.9%), followed by elemental carbon-aged (EC-aged, 19.6%) and organic carbon-fresh (OCfresh, 13.5%) before the pandemic, while EC-aged (25.3%), ECOC (17.9%), and secondary ions-rich (SEC, 17.8%) became the predominant species during the pandemic. The carbonaceous particle sizes during the pandemic showed a broader distribution than that before the pandemic, due to the condensation and coagulation of carbonaceous particles in the aging processes. The relative aerosol acidity (Rra) was smaller before the pandemic than that during the pandemic, indicating the more acidic particle aerosol during the pandemic closely related to the secondary species and relative humidity (RH). More than 95.0% and 86.0% of carbonaceous particles in the whole period were internally mixed with nitrate and sulfate, implying that most of the carbonaceous particles were associated with secondary oxidation during their formation processes. The diurnal variations of oxalate particles and correlation analyses suggested that oxalate particles before the pandemic were derived from aqueous oxidation driven by RH and liquid water content (LWC), while oxalate particles during the pandemic were originated from O3dominated photochemical oxidation.(c) 2022 Chinese Society of Particuology and Institute of Process Engineering, Chinese Academy of Sciences. Published by Elsevier B.V. All rights reserved.

Hybrid Mode: A New Norm for Electrical Engineering Laboratory Education?

CONTEXT The COVID-19 pandemic has created an incredibly challenging period in which to deliver engineering laboratory exercises. Utilising available digital technologies, the authors converted traditional hands-on laboratory exercises to virtual labs and remote labs. Commencing in Semester 2, 2020, the authors' School has offered a hybrid teaching model which simultaneously delivers laboratory content to an on-campus cohort (who participate in traditional hands-on labs) and a remote-learning cohort (who participate via virtual and/or remote labs). While trying to ensure that the learning experience of both on-campus and remote-learning students were similar, and that teaching outcomes were maintained, the authors observed that the success in adaptation of existing course content to the hybrid teaching model differs between Units of Study (UoS). There is a challenge to understand the basis for these differences and how to optimally design teaching material and manage classes to achieve the best learning outcomes. PURPOSE The authors manage and coordinate operation and teaching across six electrical engineering teaching laboratories. This paper aims to report the degree of success of introducing hybrid laboratory education across twelve UoS. Specifically, based on student responses to a survey undertaken in Semester 1, 2021, the authors evaluate the effectiveness of the hybrid model by seeking to answer two questions: (1) Could the students be satisfied with the new hybrid model? (2) Could on-campus and remote-learning students have similar learning experiences? METHODS The study covers the School's teaching laboratory programs that span three broad teaching disciplines: power/energy, communications/photonics, and computer/digital electronics. They are organised in either mixed mode (both on-campus and remote cohorts undertake the same exercises) or parallel mode (cohorts complete different exercises that have common learning outcomes). Student survey data across twelve UoS are available, including responses about learning experience, tutor teaching, and additional comments. The method is mainly quantitatively statistical analysis, supplemented by qualitative study. OUTCOMES Overall, the hybrid lab program results in a satisfactory learning experience for students. This means that implementing electrical engineering laboratory teaching using a hybrid model is found to be both practical and applicable. However, students on-campus in the mixed mode and both cohorts in the parallel mode tended to adapt more successfully to the hybrid model than those remote students in the mixed mode. It prompts the educators to fine-tune the hybrid program to better accommodate the remote mixed mode students. CONCLUSIONS While the hybrid model can deliver effective laboratory education, the degree of success and student experience was found to vary between different cohorts. Further study is warranted to understand the factors behind these differences and to then explore more effective approaches to maximise the students' learning experience. This paper serves as a starting point for the community to discuss the new norm for engineering laboratory education. The pandemic has already had a transformational impact on the delivery of engineering education, and hybrid education may not be transient but instead a future steady state. Copyright © Huang, Chu & Jones, 2021.

Analysis on the accessibility of urban rescue in extreme weather

Schools are the key units for infectious disease prevention and control. Students are the key groups in prevention and control of infectious diseases. The establishment of links between schools and out campus hospitals is an important measure for epidemic prevention and control of Corona Virus Disease 2019 (COVID-19). How to use reasonable human resources, financial resources and material resources to make rational decisions and respond quickly to the best cost performance is the practical problem facing all schools. Therefore, the accessibility of Wuhan medical services to schools is analyzed through GIS, and the resilience of hospitals is analyzed by simulating short-term heavy rainfall weather. The results show that large hospitals in Wuhan are mostly concentrated in the city center, which can effectively resist the impact of sudden epidemic and disperse the risk to the periphery of the city. The buffer zone of large hospitals in Wuhan can effectively cover most schools and can respond quickly, but some hospitals have low accessibility in extreme weather. It is necessary to further improve the distribution of medical resources in various regions and improve emergency transportation planning. © 2022 ACM.

Early Warning and Monitoring of Coronavirus Disease 2019 Using Baidu Search Index and Baidu Information Index in Guangxi, China

Coronavirus disease 2019 (COVID-19) is an emerging infectious disease, and it is important to detect early and monitor the disease trend for policymakers to make informed decisions. We explored the predictive utility of Baidu Search Index and Baidu Information Index for early warning of COVID-19 and identified search keywords for further monitoring of epidemic trends in Guangxi. A time-series analysis and Spearman correlation between the daily number of cases and both the Baidu Search Index and Baidu Information Index were performed for seven keywords related to COVID-19 from January 8 to March 9, 2020. The time series showed that the temporal distributions of the search terms "coronavirus," "pneumonia" and "mask" in the Baidu Search Index were consistent and had 2 to 3 days' lead time to the reported cases;the correlation coefficients were higher than 0.81. The Baidu Search Index volume in 14 prefectures of Guangxi was closely related with the number of reported cases;it was not associated with the local GDP. The Baidu Information Index search terms "coronavirus" and "pneumonia" were used as frequently as 192,405.0 and 110,488.6 per million population, respectively, and they were also significantly associated with the number of reported cases (r(s) > 0.6), but they fluctuated more than for the Baidu Search Index and had 0 to 14 days' lag time to the reported cases. The Baidu Search Index with search terms "coronavirus," "pneumonia" and "mask" can be used for early warning and monitoring of the epidemic trend of COVID-19 in Guangxi, with 2 to 3 days' lead time.

SARS-CoV-2-correlated ASGR1 is a novel potential marker for the treatment and identification of multiple human cancers

Objectives: Cancer patients are reported to be more susceptible to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), and the COVID-19 (the Corona Virus Disease 2019) patients with cancer suffer from certain serious complications. ASGR1 has been recently identified as a novel receptor of SARS-CoV-2 in human cells;however, there are limited studies on ASGR1 in various human cancers. Method(s): This study utilized a comprehensive analysis of COVID-19-related ASGR1 in multiple human cancers based on 18,589 multi-center samples. Using Wilcoxon rank-sum analysis, a difference in ASGR1 expression between cancer and control tissues was detected. Cox regression analysis, Kaplan-Meier curves, and receiver operating characteristic curves were utilized to determine the correlation between ASGR1 expression and the clinical parameters of cancer patients. The immune relevance and potential mechanisms of ASGR1 in various cancers were also investigated. Result(s): Abnormal ASGR1 mRNA expression was observed in 16 of 20 different cancers (e.g., it was upregulated in colon adenocarcinoma but downregulated in cholangiocarcinoma;P < 0.05). ASGR1 was related to prognosis, e.g., overall survival, in 14 cancers (P < 0.05), such as adrenocortical carcinoma. The gene was also found to be a potential marker that can be utilized to distinguish eleven cancers from controls with moderate to high accuracy (e.g., the area under the curve for cholangiocarcinoma = 1.000). ASGR1 expression was related to DNA methyltransferases, mismatch repair genes, immune checkpoints, levels of tumor mutational burden, microsatellite instability, neoantigen count, and immune infiltration levels in certain cancers (P < 0.05). The gene plays a role in multiple cancers by affecting four signaling pathways, such as cytokine-cytokine receptor interaction. Cancer patients with high ASGR1 expression are sensitive to 25 drugs, including ulixertinib. Conclusion(s): SARS-CoV-2-correlated ASGR1 is a novel marker that can be used for treating and identifying multiple human cancers. Copyright © 2022 E-Century Publishing Corporation. All rights reserved.

Stock Comovement and Financial Flexibility

We develop a dynamic model of corporate investment and financing, in which shocks to the value of collateralizable assets generate variation in firms' debt capacity. We show that the degree of similarity among firms' financial flexibility forecasts cross-sectional variation in return correlation. We test the implications of the model with firm-level data in two empirical analyses using (i) an instrumental variable approach based on shocks to the value of collateralizable corporate assets and (ii) the outbreak of the COVID-19 crisis as an event study. We find that firms in the same percentile of the cross-sectional distribution of financial flexibility have 62% higher correlation in stock-return residuals than firms 50 percentiles apart. © 2022 Cambridge University Press. All rights reserved.

A Systematic Review of Multimodal Approaches to Online Misinformation Detection

During the COVID-19 pandemic, the spread of pandemic-related misinformation on social media has had a significantly adverse impact on society. The sources of such misinformation usually use not only well-tailored text but also eye-catching images to establish their credibility. In this paper, we present an overview of current efforts on the task of detecting online COVID-19 conspiracy theory and misinformation. We perform a review of multimedia misinformation datasets related to the topic and an exploratory study on the state-of-the-art approaches towards these tasks. These approaches fuse textual analysis with modeling of images, propagation graphs, user reputation and fact-checking to build a comprehensive multimodal understanding of online misinformation. Our analysis indicates that using modalities in addition to text has a significant improvement on the performance of detecting misinformation, and out of the modalities presented, modeling user reputation and graph with social data are the most effective approaches. We conclude that a dataset that unifies all modalities is needed, and we present several promising directions for future research. © 2022 IEEE.

Analysis of Physical Fitness Differences among College Students before and after COVID-19

COVID-19 has not only had a profound impact on social and economic development, but also changed social lifestyles and values. Due to the prevention and control of the epidemic, people's demand for physical exercise has been limited to a certain extent, while the epidemic has increased the requirements for individual immunity, forming a contradiction. In this paper, the fitness tests of college students before and after COVID-19 were taken as samples, and the differences of physical tests in 2019 and 2021 were compared by t-test and non-parametric test. The study found that after the epidemic, the body shape of the college students deteriorated slightly, and the physical fitness tests significantly lower after gaining weight. In general, the decline of college students' physical fitness is not obvious after the epidemic, but it is a common phenomenon. Then put forward countermeasures and suggestions of optimizing physical exercise. © 2022 WCSE. All Rights Reserved.

Detecting COVID-19 Conspiracy Theories with Transformers and TF-IDF

The sharing of fake news and conspiracy theories on social media has wide-spread negative effects. By designing and applying different machine learning models, researchers have made progress in detecting fake news from text. However, existing research places a heavy emphasis on general, common-sense fake news, while in reality fake news often involves rapidly changing topics and domain-specific vocabulary. In this paper, we present our methods and results for three fake news detection tasks at MediaEval benchmark 2021 that specifically involve COVID-19 related topics. We experiment with a group of text-based models including Support Vector Machines, Random Forest, BERT, and RoBERTa. We find that a pre-trained transformer yields the best validation results, but a randomly initialized transformer with smart design can also be trained to reach accuracies close to that of the pre-trained transformer. Copyright 2021 for this paper by its authors.

Change in breast cancer risk perception related to perceived COVID-19 risk: A WISDOM sub-population analysis

Background: Obstacles to access to care due to the perceptions of the risk posed by COVID-19 have led to unprecedented disruptions in cancer care. Yet, little is understood about whether perceived COVID- 19 risk influences perceptions of cancer risk. We examined how COVID-19 risk perception was associated with perceptions of breast cancer risk over one year of the pandemic among women enrolled in the WISDOM study, a PCORI-funded pragmatic trial testing risk-based cancer screening that began before the pandemic. Methods: We conducted four longitudinal surveys among the 13,002 women enrolled in the WISDOM study from May - December 2020. Responses from 8,285 women are eligible for inclusion in this analysis leading to a total sample size of 16,859 survey responses. Surveys were conducted online and asked women's perceived lifetime chance of developing breast cancer (0- 100%). COVID-19 risk perception was reported on a 5-point scale from Very Low to Very High. We computed the difference between breast cancer risk perception at each COVID-19 survey to pre-COVID breast cancer risk perception, measured as a secondary aim of the WISDOM study, and compared that to COVID-19 risk perception at each time point. Results: Across the four survey waves, most women perceived low COVID risk: 29% very low, 42% moderately low, 23% neither high nor low and 6% high or very high. Overall, breast cancer risk perception declined for those with very low COVID-19 risk perception and rose for women in the highest levels of COVID-19 risk perception. However, changes in breast cancer risk perception associated with COVID risk perception were small. For example, in survey wave 4, breast cancer risk change was -2.4% very low, -1.4% low, 2.5% not high or low and 3.1% high or very high. (Table). Conclusions: Among women participating in a pragmatic trial testing riskbased cancer screening, COVID risk perception had a small relationship with change in breast cancer risk perception. Change in breast cancer risk perception paralleled COVID-19 risk perception. This calls for exploration of the underpinnings of these risk changes and may have implications for changes in cancer screening behavior related to COVID-19.

condLSTM-Q: A novel deep learning model for predicting COVID-19 mortality in fine geographical scale

Background: Modern machine learning-based models have not been harnessed to their total capacity for disease trend predictions prior to the COVID-19 pandemic. This work is the first use of the conditional RNN model in predicting disease trends that we know of during development that complemented classical epidemiological approaches. Methods: We developed the long short-term memory networks with quantile output (condLSTM-Q) model for making quantile predictions on COVID-19 death tolls. Results: We verified that the condLSTM-Q was accurately predicting fine-scale, county-level daily deaths with a two-week window. The model’s performance was robust and comparable to, if not slightly better than well-known, publicly available models. This provides unique opportunities for investigating trends within the states and interactions between counties along state borders. In addition, by analyzing the importance of the categorical data, one could learn which features are risk factors that affect the death trend and provide handles for officials to ameliorate the risks. Conclusion: The condLSTM-Q model performed robustly, provided fine-scale, county-level predictions of daily deaths with a two-week window. Given the scalability and generalizability of neural network models, this model could incorporate additional data sources with ease and could be further developed to generate other valuable predictions such as new cases or hospitalizations intuitively. © The Author (s) 2022. Published by Higher Education Press.