Credibility of vaccine-related content on Twitter during COVID-19 pandemic.

During national COVID-19 vaccine campaigns, people continuously engaged on Twitter to receive updates on the latest public health information, and to discuss and share their experiences. During this time, the spread of misinformation was widespread, which threatened the uptake of vaccines. It is therefore critical to understand the reasons behind vaccine misinformation and strategies to mitigate it. The current research aimed to understand the content of misleading tweets and the characteristics of their corresponding accounts. We performed a machine learning approach to identify misinformation in vaccine-related tweets, and calculated the demographic, engagement metrics and bot-like activities of corresponding accounts. We found critical periods where high amounts of misinformation coincided with important vaccine announcements, such as emergency approvals of vaccines. Moreover, we found Asian countries had a lower percentage of misinformation shared compared to Europe and North America. Our results showed accounts spreading misinformation had an overall 10% greater likelihood of bot activity and 15% more astroturf bot activity than accounts spreading accurate information. Furthermore, we found that accounts spreading misinformation had five times fewer followers and three times fewer verified badges than fact-sharing accounts. The findings of this study may help authorities to develop strategies to fight COVID-19 vaccine misinformation and improve vaccine uptake.

Classification of porcine reproductive and respiratory syndrome clinical impact in Ontario sow herds using machine learning approaches.

Since the early 1990s, porcine reproductive and respiratory syndrome (PRRS) virus outbreaks have been reported across various parts of North America, Europe, and Asia. The incursion of PRRS virus (PRRSV) in swine herds could result in various clinical manifestations, resulting in a substantial impact on the incidence of respiratory morbidity, reproductive loss, and mortality. Veterinary experts, among others, regularly analyze the PRRSV open reading frame-5 (ORF-5) for prognostic purposes to assess the risk of severe clinical outcomes. In this study, we explored if predictive modeling techniques could be used to identify the severity of typical clinical signs observed during PRRS outbreaks in sow herds. Our study aimed to evaluate four baseline machine learning (ML) algorithms: logistic regression (LR) with ridge and lasso regularization techniques, random forest (RF), k-nearest neighbor (KNN), and support vector machine (SVM), for the clinical impact classification of ORF-5 sequences and demographic data into high impact and low impact categories. First, baseline classifiers were evaluated using different input representations of ORF-5 nucleotides, amino acid sequences, and demographic data using a 10-fold cross-validation technique. Then, we designed a consensus voting ensemble approach to aggregate the different types of input representations for genetic and demographic data for classifying clinical impact. In this study, we observed that: (a) for abortion and pre-weaning mortality (PWM), different classifiers gained improvement over baseline accuracy, which showed the plausible presence of both genotypic-phenotypic and demographic-phenotypic relationships, (b) for sow mortality (SM), no baseline classifier successfully established such linkages using either genetic or demographic input data, (c) baseline classifiers showed good performance with a moderate variance of the performance metrics, due to high-class overlap and the small dataset size used for training, and (d) the use of consensus voting ensemble techniques helped to make the predictions more robust and stabilized the performance evaluation metrics, but overall accuracy did not substantially improve the diagnostic metrics over baseline classifiers.

Predictive analysis for pathogenicity classification of H5Nx avian influenza strains using machine learning techniques.

Over the past decades, avian influenza (AI) outbreaks have been reported across different parts of the globe, resulting in large-scale economic and livestock loss and, in some cases raising concerns about their zoonotic potential. The virulence and pathogenicity of H5Nx (e.g., H5N1, H5N2) AI strains for poultry could be inferred through various approaches, and it has been frequently performed by detecting certain pathogenicity markers in their haemagglutinin (HA) gene. The utilization of predictive modeling methods represents a possible approach to exploring this genotypic-phenotypic relationship for assisting experts in determining the pathogenicity of circulating AI viruses. Therefore, the main objective of this study was to evaluate the predictive performance of different machine learning (ML) techniques for in-silico prediction of pathogenicity of H5Nx viruses in poultry, using complete genetic sequences of the HA gene. We annotated 2137 H5Nx HA gene sequences based on the presence of the polybasic HA cleavage site (HACS) with 46.33% and 53.67% of sequences previously identified as highly pathogenic (HP) and low pathogenic (LP), respectively. We compared the performance of different ML classifiers (e.g., logistic regression (LR) with the lasso and ridge regularization, random forest (RF), K-nearest neighbor (KNN), Naïve Bayes (NB), support vector machine (SVM), and convolutional neural network (CNN)) for pathogenicity classification of raw H5Nx nucleotide and protein sequences using a 10-fold cross-validation technique. We found that different ML techniques can be successfully used for the pathogenicity classification of H5 sequences with ∼99% classification accuracy. Our results indicate that for pathogenicity classification of (1) aligned deoxyribonucleic acid (DNA) and protein sequences, with NB classifier had the lowest accuracies of 98.41% (+/-0.89) and 98.31% (+/-1.06), respectively; (2) aligned DNA and protein sequences, with LR (L1/L2), KNN, SVM (radial basis function (RBF)) and CNN classifiers had the highest accuracies of 99.20% (+/-0.54) and 99.20% (+/-0.38), respectively; (3) unaligned DNA and protein sequences, with CNN's achieved accuracies of 98.54% (+/-0.68) and 99.20% (+/-0.50), respectively. ML methods show potential for regular classification of H5Nx virus pathogenicity for poultry species, particularly when sequences containing regular markers were frequently present in the training dataset.

Recommendations for ethical and responsible use of artificial intelligence in digital agriculture.

Artificial intelligence (AI) applications are an integral and emerging component of digital agriculture. AI can help ensure sustainable production in agriculture by enhancing agricultural operations and decision-making. Recommendations about soil condition and pesticides or automatic devices for milking and apple picking are examples of AI applications in digital agriculture. Although AI offers many benefits in farming, AI systems may raise ethical issues and risks that should be assessed and proactively managed. Poor design and configuration of intelligent systems may impose harm and unintended consequences on digital agriculture. Invasion of farmers' privacy, damaging animal welfare due to robotic technologies, and lack of accountability for issues resulting from the use of AI tools are only some examples of ethical challenges in digital agriculture. This paper examines the ethical challenges of the use of AI in agriculture in six categories including fairness, transparency, accountability, sustainability, privacy, and robustness. This study further provides recommendations for agriculture technology providers (ATPs) and policymakers on how to proactively mitigate ethical issues that may arise from the use of AI in farming. These recommendations cover a wide range of ethical considerations, such as addressing farmers' privacy concerns, ensuring reliable AI performance, enhancing sustainability in AI systems, and reducing AI bias.

Canadian COVID-19 Crisis Communication on Twitter: Mixed Methods Research Examining Tweets from Government, Politicians, and Public Health for Crisis Communication Guiding Principles and Tweet Engagement.

To foster trust on social media during a crisis, messages should implement key guiding principles, including call to action, clarity, conversational tone, compassion and empathy, correction of misinformation, and transparency. This study describes how crisis actors used guiding principles in COVID-19 tweets, and how the use of these guiding principles relates to tweet engagement. Original, English language tweets from 10 federal level government, politician, and public health Twitter accounts were collected between 11 March 2020 and 25 January 2021 (n = 6053). A 60% random sample was taken (n = 3633), and the tweets were analyzed for guiding principles. A tweet engagement score was calculated for each tweet and logistic regression analyses were conducted to model the relationship between guiding principles and tweet engagement. Overall, the use of guiding principles was low and inconsistent. Tweets that were written with compassion and empathy, or conversational tone were associated with greater odds of having higher tweet engagement. Across all guiding principles, tweets from politicians and public health were associated with greater odds of having higher tweet engagement. Using a combination of guiding principles was associated with greater odds of having higher tweet engagement. Crisis actors should consistently use relevant guiding principles in crisis communication messages to improve message engagement.

Trust and Engagement on Twitter During the Management of COVID-19 Pandemic: The Effect of Gender and Position.

During the COVID-19 pandemic, health and political leaders have attempted to update citizens using Twitter. Here, we examined the difference between environments that social media has provided for male/female or health/political leaders to interact with people during the COVID-19 pandemic. The comparison was made based on the content of posts and public responses to those posts as well as user-level and post-level metrics. Our findings suggest that although health officers and female leaders generated more contents on Twitter, political leaders and male authorities were more active in building networks. Offensive language was used more frequently toward males than females and toward political leaders than health leaders. The public also used more appreciation keywords toward health leaders than politicians, while more judgmental and economy-related keywords were used toward politicians. Overall, depending on the gender and position of leaders, Twitter provided them with different environments to communicate and manage the pandemic.

Prediction of COVID-19 Waves Using Social Media and Google Search: A Case Study of the US and Canada.

The ongoing COVID-19 pandemic has posed a severe threat to public health worldwide. In this study, we aimed to evaluate several digital data streams as early warning signals of COVID-19 outbreaks in Canada, the US and their provinces and states. Two types of terms including symptoms and preventive measures were used to filter Twitter and Google Trends data. We visualized and correlated the trends for each source of data against confirmed cases for all provinces and states. Subsequently, we attempted to find anomalies in indicator time-series to understand the lag between the warning signals and real-word outbreak waves. For Canada, we were able to detect a maximum of 83% of initial waves 1 week earlier using Google searches on symptoms. We divided states in the US into two categories: category I if they experienced an initial wave and category II if the states have not experienced the initial wave of the outbreak. For the first category, we found that tweets related to symptoms showed the best prediction performance by predicting 100% of first waves about 2-6 days earlier than other data streams. We were able to only detect up to 6% of second waves in category I. On the other hand, 78% of second waves in states of category II were predictable 1-2 weeks in advance. In addition, we discovered that the most important symptoms in providing early warnings are fever and cough in the US. As the COVID-19 pandemic continues to spread around the world, the work presented here is an initial effort for future COVID-19 outbreaks.

An analysis of COVID-19 vaccine sentiments and opinions on Twitter.

OBJECTIVE: We identified public sentiments and opinions toward the COVID-19 vaccines based on the content of Twitter. MATERIALS AND METHODS: We retrieved 4,552,652 publicly available tweets posted within the timeline of January 2020 to January 2021. Following extraction, we identified vaccine sentiments and opinions of tweets and compared their progression by time, geographical distribution, main themes, keywords, posts engagement metrics and accounts characteristics. RESULTS: We found a slight difference in the prevalence of positive and negative sentiments, with positive being the dominant polarity and having higher engagements. The amount of discussion on vaccine rejection and hesitancy was more than interest in vaccines during the course of the study, but the pattern was different in various countries. We found the accounts producing vaccine opposition content were partly Twitter bots or political activists while well-known individuals and organizations generated the content in favour of vaccination. CONCLUSION: Understanding sentiments and opinions toward vaccination using Twitter may help public health agencies to increase positive messaging and eliminate opposing messages in order to enhance vaccine uptake.

A framework for the risk prediction of avian influenza occurrence: An Indonesian case study.

Avian influenza viruses can cause economically devastating diseases in poultry and have the potential for zoonotic transmission. To mitigate the consequences of avian influenza, disease prediction systems have become increasingly important. In this study, we have proposed a framework for the prediction of the occurrence and spread of avian influenza events in a geographical area. The application of the proposed framework was examined in an Indonesian case study. An extensive list of historical data sources containing disease predictors and target variables was used to build spatiotemporal and transactional datasets. To combine disparate sources, data rows were scaled to a temporal scale of 1-week and a spatial scale of 1-degree × 1-degree cells. Given the constructed datasets, underlying patterns in the form of rules explaining the risk of occurrence and spread of avian influenza were discovered. The created rules were combined and ordered based on their importance and then stored in a knowledge base. The results suggested that the proposed framework could act as a tool to gain a broad understanding of the drivers of avian influenza epidemics and may facilitate the prediction of future disease events.

A decision support framework for prediction of avian influenza.

For years, avian influenza has influenced economies and human health around the world. The emergence and spread of avian influenza virus have been uncertain and sudden. The virus is likely to spread through several pathways such as poultry transportation and wild bird migration. The complicated and global spread of avian influenza calls for surveillance tools for timely and reliable prediction of disease events. These tools can increase situational awareness and lead to faster reaction to events. Here, we aimed to design and evaluate a decision support framework that aids decision makers by answering their questions regarding the future risk of events at various geographical scales. Risk patterns were driven from pre-built components and combined in a knowledge base. Subsequently, questions were answered by direct queries on the knowledge base or through a built-in algorithm. The evaluation of the system in detecting events resulted in average sensitivity and specificity of 69.70% and 85.50%, respectively. The presented framework here can support health care authorities by providing them with an opportunity for early control of emergency situations.

The Assessment of Twitter's Potential for Outbreak Detection: Avian Influenza Case Study.

Social media services such as Twitter are valuable sources of information for surveillance systems. A digital syndromic surveillance system has several advantages including its ability to overcome the problem of time delay in traditional surveillance systems. Despite the progress made with using digital syndromic surveillance systems, the possibility of tracking avian influenza (AI) using online sources has not been fully explored. In this study, a Twitter-based data analysis framework was developed to automatically monitor avian influenza outbreaks in a real-time manner. The framework was implemented to find worrisome posts and alerting news on Twitter, filter irrelevant ones, and detect the onset of outbreaks in several countries. The system collected and analyzed over 209,000 posts discussing avian influenza on Twitter from July 2017 to November 2018. We examined the potential of Twitter data to represent the date, severity and virus type of official reports. Furthermore, we investigated whether filtering irrelevant tweets can positively impact the performance of the system. The proposed approach was empirically evaluated using a real-world outbreak-reporting source. We found that 75% of real-world outbreak notifications of AI were identifiable from Twitter. This shows the capability of the system to serve as a complementary approach to official AI reporting methods. Moreover, we observed that one-third of outbreak notifications were reported on Twitter earlier than official reports. This feature could augment traditional surveillance systems and provide a possibility of early detection of outbreaks. This study could potentially provide a first stepping stone for building digital disease outbreak warning systems to assist epidemiologists and animal health professionals in making relevant decisions.

Forecasting herd-level porcine epidemic diarrhea (PED) frequency trends in Ontario (Canada).

Porcine Epidemic Diarrhea Virus (PEDV) emerged in North America in 2013. The first case of PEDV in Canada was identified on an Ontario farm in January 2014. Surveillance was instrumental in identifying the initial case and in minimizing the spread of the virus to other farms. With recent advances in predictive analytics showing promise for health and disease forecasting, the primary objective of this study was to apply machine learning predictive methods (random forest, artificial neural networks, and classification and regression trees) to provincial PEDV incidence data, and in so doing determine their accuracy for predicting future PEDV trends. Trend was defined as the cumulative number of new cases over a four-week interval, and consisted of four levels (zero, low, medium and high). Provincial PEDV incidence and prevalence estimates from an industry database, as well as temperature, humidity, and precipitation data, were combined to create the forecast dataset. With 10-fold cross validation performed on the entire dataset, the overall accuracy was 0.68 (95% CI: 0.60 - 0.75), 0.57 (95% CI: 0.49 - 0.64), and 0.55 (0.47 - 0.63) for the random forest, artificial neural network, and classification and regression tree models, respectively. Based on the cross-validation approach to evaluating predictive accuracy, the random forest model provided the best prediction.

A review of knowledge discovery process in control and mitigation of avian influenza.

In the last several decades, avian influenza virus has caused numerous outbreaks around the world. These outbreaks pose a significant threat to the poultry industry and also to public health. When an avian influenza (AI) outbreak occurs, it is critical to make informed decisions about the potential risks, impact, and control measures. To this end, many modeling approaches have been proposed to acquire knowledge from different sources of data and perspectives to enhance decision making. Although some of these approaches have shown to be effective, they do not follow the process of knowledge discovery in databases (KDD). KDD is an iterative process, consisting of five steps, that aims at extracting unknown and useful information from the data. The present review attempts to survey AI modeling methods in the context of KDD process. We first divide the modeling techniques used in AI into two main categories: data-intensive modeling and small-data modeling. We then investigate the existing gaps in the literature and suggest several potential directions and techniques for future studies. Overall, this review provides insights into the control of AI in terms of the risk of introduction and spread of the virus.

Detecting and Predicting Emerging Disease in Poultry With the Implementation of New Technologies and Big Data: A Focus on Avian Influenza Virus.

Future demands for food will place agricultural systems under pressure to increase production. Poultry is accepted as a good source of protein and the poultry industry will be forced to intensify production in many countries, leading to greater numbers of farms that house birds at elevated densities. Increasing farmed poultry can facilitate enhanced transmission of infectious pathogens among birds, such as avian influenza virus among others, which have the potential to induce widespread mortality in poultry and cause considerable economic losses. Additionally, the capability of some emerging poultry pathogens to cause zoonotic human infection will be increased as greater numbers of poultry operations could increase human contact with poultry pathogens. In order to combat the increased risk of spread of infectious disease in poultry due to intensified systems of production, rapid detection and diagnosis is paramount. In this review, multiple technologies that can facilitate accurate and rapid detection and diagnosis of poultry diseases are highlighted from the literature, with a focus on technologies developed specifically for avian influenza virus diagnosis. Rapid detection and diagnostic technologies allow for responses to be made sooner when disease is detected, decreasing further bird transmission and associated costs. Additionally, systems of rapid disease detection produce data that can be utilized in decision support systems that can predict when and where disease is likely to emerge in poultry. Other sources of data can be included in predictive models, and in this review two highly relevant sources, internet based-data and environmental data, are discussed. Additionally, big data and big data analytics, which will be required in order to integrate voluminous and variable data into predictive models that function in near real-time are also highlighted. Implementing new technologies in the commercial setting will be faced with many challenges, as will designing and operating predictive models for poultry disease emergence. The associated challenges are summarized in this review. Intensified systems of poultry production will require new technologies for detection and diagnosis of infectious disease. This review sets out to summarize them, while providing advantages and limitations of different types of technologies being researched.

Corrigendum: Detecting and Predicting Emerging Disease in Poultry With the Implementation of New Technologies and Big Data: A Focus on Avian Influenza Virus.

[This corrects the article DOI: 10.3389/fvets.2018.00263.].

Cytokine gene expression in chicken cecal tonsils following treatment with probiotics and Salmonella infection.

Probiotics are currently employed for control of pathogens and enhancement of immune response in chickens. In this study, we investigated the underlying immunological mechanisms of the action of probiotics against colonization of the chicken intestine by Salmonella enterica subsp. enterica serovar Typhimurium (Salmonella serovar Typhimurium). Birds received probiotics by oral gavage on day 1 of age and, subsequently, received Salmonella serovar Typhimurium on day 2 of age. Cecal tonsils were removed on days 1, 3 and 5 post-infection (p.i.), RNA was extracted and subjected to real-time quantitative RT-PCR for measurement of interleukin (IL)-6, IL-10, IL-12 and interferon (IFN)-gamma gene expression. There was no significant difference in IL-6 and IL-10 gene expression in cecal tonsils of chickens belonging to various treatment groups. Salmonella serovar Typhimurium infection resulted in a significant increase in IL-12 expression in cecal tonsils on days 1 and 5p.i. However, when chickens were treated with probiotics prior to experimental infection with Salmonella, the level of IL-12 expression was similar to that observed in uninfected control chickens. Treatment of birds with probiotics resulted in a significant decrease in IFN-gamma gene expression in cecal tonsils of chickens infected with Salmonella compared to the Salmonella-infected birds not treated with probiotics. These findings reveal that repression of IL-12 and IFN-gamma expression is associated with probiotic-mediated reduction in intestinal colonization with Salmonella serovar Typhimurium.