Optimal adaptive nonpharmaceutical interventions to mitigate the outbreak of respiratory infections following the COVID-19 pandemic: a deep reinforcement learning study in Hong Kong, China.

BACKGROUND: Long-lasting nonpharmaceutical interventions (NPIs) suppressed the infection of COVID-19 but came at a substantial economic cost and the elevated risk of the outbreak of respiratory infectious diseases (RIDs) following the pandemic. Policymakers need data-driven evidence to guide the relaxation with adaptive NPIs that consider the risk of both COVID-19 and other RIDs outbreaks, as well as the available healthcare resources. METHODS: Combining the COVID-19 data of the sixth wave in Hong Kong between May 31, 2022 and August 28, 2022, 6-year epidemic data of other RIDs (2014-2019), and the healthcare resources data, we constructed compartment models to predict the epidemic curves of RIDs after the COVID-19-targeted NPIs. A deep reinforcement learning (DRL) model was developed to learn the optimal adaptive NPIs strategies to mitigate the outbreak of RIDs after COVID-19-targeted NPIs are lifted with minimal health and economic cost. The performance was validated by simulations of 1000 days starting August 29, 2022. We also extended the model to Beijing context. FINDINGS: Without any NPIs, Hong Kong experienced a major COVID-19 resurgence far exceeding the hospital bed capacity. Simulation results showed that the proposed DRL-based adaptive NPIs successfully suppressed the outbreak of COVID-19 and other RIDs to lower than capacity. DRL carefully controlled the epidemic curve to be close to the full capacity so that herd immunity can be reached in a relatively short period with minimal cost. DRL derived more stringent adaptive NPIs in Beijing. INTERPRETATION: DRL is a feasible method to identify the optimal adaptive NPIs that lead to minimal health and economic cost by facilitating gradual herd immunity of COVID-19 and mitigating the other RIDs outbreaks without overwhelming the hospitals. The insights can be extended to other countries/regions.

Integrating Relational Knowledge With Text Sequences for Script Event Prediction.

Script event prediction aims to infer subsequent events given an incomplete script. It requires a deep understanding of events, and can provide support for a variety of tasks. Existing models rarely consider the relational knowledge between events, they regard scripts as sequences or graphs, which cannot capture the relational information between events and the semantic information of script sequences jointly. To address this issue, we propose a new script form, relational event chain, that combines event chains and relational graphs. We also introduce a new model, relational-transformer, to learn embeddings based on this new script form. In particular, we first extract the relationship between events from an event knowledge graph to formalize scripts as relational event chains, then use the relational-transformer to calculate the likelihood of different candidate events, where the model learns event embeddings that encode both semantic and relational knowledge by combining transformers and graph neural networks (GNNs). Experimental results on both one-step inference and multistep inference tasks show that our model can outperform existing baselines, indicating the validity of encoding relational knowledge into event embeddings. The influence of using different model structures and different types of relational knowledge is analyzed as well.

Future trajectory of respiratory infections following the COVID-19 pandemic in Hong Kong.

The accumulation of susceptible populations for respiratory infectious diseases (RIDs) when COVID-19-targeted non-pharmaceutical interventions (NPIs) were in place might pose a greater risk of future RID outbreaks. We examined the timing and magnitude of RID resurgence after lifting COVID-19-targeted NPIs and assessed the burdens on the health system. We proposed the Threshold-based Control Method (TCM) to identify data-driven solutions to maintain the resilience of the health system by re-introducing NPIs when the number of severe infections reaches a threshold. There will be outbreaks of all RIDs with staggered peak times after lifting COVID-19-targeted NPIs. Such a large-scale resurgence of RID patients will impose a significant risk of overwhelming the health system. With a strict NPI strategy, a TCM-initiated threshold of 600 severe infections can ensure a sufficient supply of hospital beds for all hospitalized severely infected patients. The proposed TCM identifies effective dynamic NPIs, which facilitate future NPI relaxation policymaking.

Adaptively Weighted k-Tuple Metric Network for Kinship Verification.

Facial image-based kinship verification is a rapidly growing field in computer vision and biometrics. The key to determining whether a pair of facial images has a kin relation is to train a model that can enlarge the margin between the faces that have no kin relation while reducing the distance between faces that have a kin relation. Most existing approaches primarily exploit duplet (i.e., two input samples without cross pair) or triplet (i.e., single negative pair for each positive pair with low-order cross pair) information, omitting discriminative features from multiple negative pairs. These approaches suffer from weak generalizability, resulting in unsatisfactory performance. Inspired by human visual systems that incorporate both low-order and high-order cross-pair information from local and global perspectives, we propose to leverage high-order cross-pair features and develop a novel end-to-end deep learning model called the adaptively weighted k -tuple metric network (AW k -TMN). Our main contributions are three-fold. First, a novel cross-pair metric learning loss based on k -tuplet loss is introduced. It naturally captures both the low-order and high-order discriminative features from multiple negative pairs. Second, an adaptively weighted scheme is formulated to better highlight hard negative examples among multiple negative pairs, leading to enhanced performance. Third, the model utilizes multiple levels of convolutional features and jointly optimizes feature and metric learning to further exploit the low-order and high-order representational power. Extensive experimental results on three popular kinship verification datasets demonstrate the effectiveness of our proposed AW k -TMN approach compared with several state-of-the-art approaches. The source codes and models are released.1.

A Dissemination Model Based on Psychological Theories in Complex Social Networks.

Information spread on social media has been extensively studied through both model-driven theoretical research and data-driven case studies. Recent empirical studies have analyzed the differences and complexity of information dissemination, but theoretical explanations of its characteristics from a modeling perspective are underresearched. To capture the complex patterns of the information dissemination mechanism, we propose a resistant linear threshold (RLT) dissemination model based on psychological theories and empirical findings. In this article, we validate the RLT model on three types of networks and then quantify and compare the dissemination characteristics of the simulation results with those from the empirical results. In addition, we examine the factors affecting dissemination. Finally, we perform two case studies of the 2019 novel Corona Virus Disease (COVID-19)-related information dissemination. The dissemination characteristics derived by the simulations are consistent with the empirical research. These results demonstrate that the RLT model is able to capture the patterns of information dissemination on social media and thus provide model-driven insights into the interpretation of public opinion, rumor control, and marketing strategies on social media.

Role of Asymptomatic COVID-19 Cases in Viral Transmission: Findings From a Hierarchical Community Contact Network Model.

As part of ongoing efforts to contain the coronavirus disease (COVID-19) pandemic, understanding the role of asymptomatic patients in the transmission system is essential for infection control. However, the optimal approach to risk assessment and management of asymptomatic cases remains unclear. This study proposed a Susceptible, Exposed, Infectious, No symptoms, Hospitalized and reported, Recovered, Death (SEINRHD) epidemic propagation model. The model was constructed based on epidemiological characteristics of COVID-19 in China and accounting for the heterogeneity of social contact networks. The early community outbreaks in Wuhan were reconstructed and fitted with the actual data. We used this model to assess epidemic control measures for asymptomatic cases in three dimensions. The impact of asymptomatic cases on epidemic propagation was examined based on the effective reproduction number, abnormally high transmission events, and type and structure of transmission. Management of asymptomatic cases can help flatten the infection curve. Tracing 75% of the asymptomatic cases corresponds to a 32.5% overall reduction in new cases (compared with tracing no asymptomatic cases). Regardless of population-wide measures, household transmission is higher than other types of transmission, accounting for an estimated 50% of all cases. The magnitude of tracing of asymptomatic cases is more important than the timing; when all symptomatic patients were traced, tested, and isolated in a timely manner, the overall epidemic was not sensitive to the time of implementing the measures to trace asymptomatic patients. Disease control and prevention within families should be emphasized during an epidemic. Note to Practitioners-This article addresses the urgent need to assess the risk of another COVID-19 outbreak caused by asymptomatic cases and to find the optimal, most practical approach to asymptomatic case management. Previous studies mostly focused on the clinical and statistical characteristics of asymptomatic cases; few have evaluated the impact of asymptomatic case measures using mathematical modeling at the community scale. This study proposed a Susceptible, Exposed, Infectious, No symptoms, Hospitalized and reported, Recovered, Death (SEINRHD) propagation model based on local community structures and social contact networks, according to the development characteristics and trend of COVID-19 in a Chinese community. The conclusion provides theoretical support for emergency work of relevant departments in different periods of an epidemic. In the early stages of the epidemic, timely detection and isolation of symptomatic patients should be a priority. Where there are surplus resources for epidemic prevention, the authorities should consider increasing the proportion of asymptomatic patients being traced. Epidemic prevention measures among family members should be a primary focus of attention. This combination of strategies can help reduce the rate of viral transmission and result in extinguishing the epidemic.

Impacts of Export Restrictions on the Global Personal Protective Equipment Trade Network During COVID-19.

The COVID-19 pandemic has caused a dramatic surge in demand for personal protective equipment (PPE) worldwide. Many countries have imposed export restrictions on PPE to ensure the sufficient domestic supply. The surging demand and export restrictions cause shortage contagions on the global PPE trade network. Here, an integrated network model is developed, which integrates a metapopulation model and a threshold model, to investigate the shortage contagion patterns. The metapopulation model captures disease contagion across countries. The threshold model captures the shortage contagion on the global PPE trade network. Due to the Pareto distribution in global exports, the shortage contagion pattern is mainly determined by the export restriction policies of the top exporters. Export restrictions exacerbate the shortages of PPE and cause the shortage contagion to transmit even faster than the disease contagion. To some extent, export restrictions can provide benefits for self-sufficient countries, at the sacrifice of immediate economic shocks at not-self-sufficient countries. With export restrictions, a large amount of PPE is hoarded instead of being distributed to where it is most needed, particularly at the early stage. Cooperation between countries plays an essential role in preventing global shortages of PPE regardless of the production level. Except for promoting global cooperation, governments and international organizations should take actions to reduce supply chain barriers and work together to increase global PPE production.

Correction: COVID-19 Vaccine Tweets After Vaccine Rollout: Sentiment-Based Topic Modeling.

[This corrects the article DOI: 10.2196/31726.].

Equitable access to COVID-19 vaccines makes a life-saving difference to all countries.

Despite broad agreement on the negative consequences of vaccine inequity, the distribution of COVID-19 vaccines is imbalanced. Access to vaccines in high-income countries (HICs) is far greater than in low- and middle-income countries (LMICs). As a result, there continue to be high rates of COVID-19 infections and deaths in LMICs. In addition, recent mutant COVID-19 outbreaks may counteract advances in epidemic control and economic recovery in HICs. To explore the consequences of vaccine (in)equity in the face of evolving COVID-19 strains, we examine vaccine allocation strategies using a multistrain metapopulation model. Our results show that vaccine inequity provides only limited and short-term benefits to HICs. Sharper disparities in vaccine allocation between HICs and LMICs lead to earlier and larger outbreaks of new waves. Equitable vaccine allocation strategies, in contrast, substantially curb the spread of new strains. For HICs, making immediate and generous vaccine donations to LMICs is a practical pathway to protect everyone.

Reconstruction of the Transmission Chain of COVID-19 Outbreak in Beijing's Xinfadi Market, China.

OBJECTIVES: The aim of the study was to reconstruct the complete transmission chain of the COVID-19 outbreak in Beijing's Xinfadi Market using data from epidemiological investigations, which contributes to reflecting transmission dynamics and transmission risk factors. METHODS: We set up a transmission model, and the model parameters are estimated from the survey data via Markov chain Monte Carlo sampling. Bayesian data augmentation approaches are used to account for uncertainty in the source of infection, unobserved onset, and infection dates. RESULTS: The rate of transmission of COVID-19 within households is 9.2%. Older people are more susceptible to infection. The accuracy of our reconstructed transmission chain was 67.26%. In the gathering place of this outbreak, the Beef and Mutton Trading Hall of Xinfadi market, most of the transmission occurs within 20 m, only 19.61% of the transmission occurs over a wider area (>20 m), with an overall average transmission distance of 13.00 m. The deepest transmission generation is 9. In this outbreak, there were 2 abnormally high transmission events. CONCLUSIONS: The statistical method of reconstruction of transmission trees from incomplete epidemic data provides a valuable tool to help understand the complex transmission factors and provides a practical guideline for investigating the characteristics of the development of epidemics and the formulation of control measures.

Social Media Images as an Emerging Tool to Monitor Adherence to COVID-19 Public Health Guidelines: Content Analysis.

BACKGROUND: Innovative surveillance methods are needed to assess adherence to COVID-19 recommendations, especially methods that can provide near real-time or highly geographically targeted data. Use of location-based social media image data (eg, Instagram images) is one possible approach that could be explored to address this problem. OBJECTIVE: We seek to evaluate whether publicly available near real-time social media images might be used to monitor COVID-19 health policy adherence. METHODS: We collected a sample of 43,487 Instagram images in New York from February 7 to April 11, 2020, from the following location hashtags: #Centralpark (n=20,937), #Brooklyn Bridge (n=14,875), and #Timesquare (n=7675). After manually reviewing images for accuracy, we counted and recorded the frequency of valid daily posts at each of these hashtag locations over time, as well as rated and counted whether the individuals in the pictures at these location hashtags were social distancing (ie, whether the individuals in the images appeared to be distanced from others vs next to or touching each other). We analyzed the number of images posted over time and the correlation between trends among hashtag locations. RESULTS: We found a statistically significant decline in the number of posts over time across all regions, with an approximate decline of 17% across each site (P<.001). We found a positive correlation between hashtags (#Centralpark and #Brooklynbridge: r=0.40; #BrooklynBridge and #Timesquare: r=0.41; and #Timesquare and #Centralpark: r=0.33; P<.001 for all correlations). The logistic regression analysis showed a mild statistically significant increase in the proportion of posts over time with people appearing to be social distancing at Central Park (P=.004) and Brooklyn Bridge (P=.02) but not for Times Square (P=.16). CONCLUSIONS: Results suggest the potential of using location-based social media image data as a method for surveillance of COVID-19 health policy adherence. Future studies should further explore the implementation and ethical issues associated with this approach.

COVID-19 Vaccine Tweets After Vaccine Rollout: Sentiment-Based Topic Modeling.

BACKGROUND: COVID-19 vaccines are one of the most effective preventive strategies for containing the pandemic. Having a better understanding of the public's conceptions of COVID-19 vaccines may aid in the effort to promptly and thoroughly vaccinate the community. However, because no empirical research has yet fully explored the public's vaccine awareness through sentiment-based topic modeling, little is known about the evolution of public attitude since the rollout of COVID-19 vaccines. OBJECTIVE: In this study, we specifically focused on tweets about COVID-19 vaccines (Pfizer, Moderna, AstraZeneca, and Johnson & Johnson) after vaccines became publicly available. We aimed to explore the overall sentiments and topics of tweets about COVID-19 vaccines, as well as how such sentiments and main concerns evolved. METHODS: We collected 1,122,139 tweets related to COVID-19 vaccines from December 14, 2020, to April 30, 2021, using Twitter's application programming interface. We removed retweets and duplicate tweets to avoid data redundancy, which resulted in 857,128 tweets. We then applied sentiment-based topic modeling by using the compound score to determine sentiment polarity and the coherence score to determine the optimal topic number for different sentiment polarity categories. Finally, we calculated the topic distribution to illustrate the topic evolution of main concerns. RESULTS: Overall, 398,661 (46.51%) were positive, 204,084 (23.81%) were negative, 245,976 (28.70%) were neutral, 6899 (0.80%) were highly positive, and 1508 (0.18%) were highly negative sentiments. The main topics of positive and highly positive tweets were planning for getting vaccination (251,979/405,560, 62.13%), getting vaccination (76,029/405,560, 18.75%), and vaccine information and knowledge (21,127/405,560, 5.21%). The main concerns in negative and highly negative tweets were vaccine hesitancy (115,206/205,592, 56.04%), extreme side effects of the vaccines (19,690/205,592, 9.58%), and vaccine supply and rollout (17,154/205,592, 8.34%). During the study period, negative sentiment trends were stable, while positive sentiments could be easily influenced. Topic heatmap visualization demonstrated how main concerns changed during the current widespread vaccination campaign. CONCLUSIONS: To the best of our knowledge, this is the first study to evaluate public COVID-19 vaccine awareness and awareness trends on social media with automated sentiment-based topic modeling after vaccine rollout. Our results can help policymakers and research communities track public attitudes toward COVID-19 vaccines and help them make decisions to promote the vaccination campaign.

Data science approaches to confronting the COVID-19 pandemic: a narrative review.

During the COVID-19 pandemic, more than ever, data science has become a powerful weapon in combating an infectious disease epidemic and arguably any future infectious disease epidemic. Computer scientists, data scientists, physicists and mathematicians have joined public health professionals and virologists to confront the largest pandemic in the century by capitalizing on the large-scale 'big data' generated and harnessed for combating the COVID-19 pandemic. In this paper, we review the newly born data science approaches to confronting COVID-19, including the estimation of epidemiological parameters, digital contact tracing, diagnosis, policy-making, resource allocation, risk assessment, mental health surveillance, social media analytics, drug repurposing and drug development. We compare the new approaches with conventional epidemiological studies, discuss lessons we learned from the COVID-19 pandemic, and highlight opportunities and challenges of data science approaches to confronting future infectious disease epidemics. This article is part of the theme issue 'Data science approaches to infectious disease surveillance'.

Sustainable targeted interventions to mitigate the COVID-19 pandemic: A big data-driven modeling study in Hong Kong.

Nonpharmaceutical interventions (NPIs) for contact suppression have been widely used worldwide, which impose harmful burdens on the well-being of populations and the local economy. The evaluation of alternative NPIs is needed to confront the pandemic with less disruption. By harnessing human mobility data, we develop an agent-based model that can evaluate the efficacies of NPIs with individualized mobility simulations. Based on the model, we propose data-driven targeted interventions to mitigate the COVID-19 pandemic in Hong Kong without city-wide NPIs. We develop a data-driven agent-based model for 7.55×106 Hong Kong residents to evaluate the efficacies of various NPIs in the first 80 days of the initial outbreak. The entire territory of Hong Kong has been split into 4905 500×500m2 grids. The model can simulate detailed agent interactions based on the demographics data, public facilities and functional buildings, transportation systems, and travel patterns. The general daily human mobility patterns are adopted from Google's Community Mobility Report. The scenario without any NPIs is set as the baseline. By simulating the epidemic progression and human movement at the individual level, we propose model-driven targeted interventions which focus on the surgical testing and quarantine of only a small portion of regions instead of enforcing NPIs in the whole city. The effectiveness of common NPIs and the proposed targeted interventions are evaluated by 100 extensive simulations. The proposed model can inform targeted interventions, which are able to effectively contain the COVID-19 outbreak with much lower disruption of the city. It represents a promising approach to sustainable NPIs to help us revive the economy of the city and the world.

Demand-driven spreading patterns of African swine fever in China.

African swine fever (ASF) is a highly contagious hemorrhagic viral disease of domestic and wild pigs. ASF has led to major economic losses and adverse impacts on livelihoods of stakeholders involved in the pork food system in many European and Asian countries. While the epidemiology of ASF virus (ASFV) is fairly well understood, there is neither any effective treatment nor vaccine. In this paper, we propose a novel method to model the spread of ASFV in China by integrating the data of pork import/export, transportation networks, and pork distribution centers. We first empirically analyze the overall spatiotemporal patterns of ASFV spread and conduct extensive experiments to evaluate the efficacy of a number of geographic distance measures. These empirical analyses of ASFV spread within China indicate that the first occurrence of ASFV has not been purely dependent on the geographical distance from existing infected regions. Instead, the pork supply-demand patterns have played an important role. Predictions based on a new distance measure achieve better performance in predicting ASFV spread among Chinese provinces and thus have the potential to enable the design of more effective control interventions.

Relationship between angiotensin-converting enzyme inhibitors or angiotensin receptor blockers and COVID-19 incidence or severe disease.

BACKGROUND: Angiotensin-converting enzyme inhibitors (ACEIs) or angiotensin receptor blockers (ARBs) may be associated with higher susceptibility of COVID-19 infection and adverse outcomes. We compared ACEI/ARB use and COVID-19 positivity in a case-control design, and severity in COVID-19 positive patients. METHODS: Consecutive patients who attended Hong Kong's public hospitals or outpatient clinics between 1 January and 28 July 2020 for COVID-19 real time-PCR (RT-PCR) tests were included. Baseline demographics, past comorbidities, laboratory tests and use of different medications were compared between COVID-19 positive and negative patients. Severe endpoints for COVID-19 positive patients were 28-day mortality, need for intensive care admission or intubation. RESULTS: This study included 213 788 patients (COVID-19 positive: n = 2774 patients; negative: n = 211 014). In total, 162 COVID-19 positive patients (5.83%) met the severity outcome. The use of ACEI/ARB was significantly higher amongst cases than controls (n = 156/2774, 5.62 vs. n = 6708/211014, 3.17%; P < 0.0001). Significant univariate predictors of COVID-19 positivity and severe COVID-19 disease were older age, higher Charlson score, comorbidities, use of ACEI/ARB, antidiabetic, lipid-lowering, anticoagulant and antiplatelet drugs and laboratory tests (odds ratio >1, P < 0.05). The relationship between the use of ACEI/ARB and COVID-19 positivity or severe disease remained significant after multivariable adjustment. No significant differences in COVID-19 positivity or disease severity between ACEI and ARB use were observed (P > 0.05). CONCLUSION: There was a significant relationship between ACEI/ARB use and COVID-19 positivity and severe disease after adjusting for significant confounders.

Automated Generation of Personalized Shock Wave Lithotripsy Protocols: Treatment Planning Using Deep Learning.

BACKGROUND: Though shock wave lithotripsy (SWL) has developed to be one of the most common treatment approaches for nephrolithiasis in recent decades, its treatment planning is often a trial-and-error process based on physicians' subjective judgement. Physicians' inexperience with this modality can lead to low-quality treatment and unnecessary risks to patients. OBJECTIVE: To improve the quality and consistency of shock wave lithotripsy treatment, we aimed to develop a deep learning model for generating the next treatment step by previous steps and preoperative patient characteristics and to produce personalized SWL treatment plans in a step-by-step protocol based on the deep learning model. METHODS: We developed a deep learning model to generate the optimal power level, shock rate, and number of shocks in the next step, given previous treatment steps encoded by long short-term memory neural networks and preoperative patient characteristics. We constructed a next-step data set (N=8583) from top practices of renal SWL treatments recorded in the International Stone Registry. Then, we trained the deep learning model and baseline models (linear regression, logistic regression, random forest, and support vector machine) with 90% of the samples and validated them with the remaining samples. RESULTS: The deep learning models for generating the next treatment steps outperformed the baseline models (accuracy = 98.8%, F1 = 98.0% for power levels; accuracy = 98.1%, F1 = 96.0% for shock rates; root mean squared error = 207, mean absolute error = 121 for numbers of shocks). The hypothesis testing showed no significant difference between steps generated by our model and the top practices (P=.480 for power levels; P=.782 for shock rates; P=.727 for numbers of shocks). CONCLUSIONS: The high performance of our deep learning approach shows its treatment planning capability on par with top physicians. To the best of our knowledge, our framework is the first effort to implement automated planning of SWL treatment via deep learning. It is a promising technique in assisting treatment planning and physician training at low cost.