Investigating Citizens' Acceptance of Contact Tracing Apps: Quantitative Study of the Role of Trust and Privacy.

BACKGROUND: The COVID-19 pandemic accelerated the need to understand citizen acceptance of health surveillance technologies such as contact tracing (CT) apps. Indeed, the success of these apps required widespread public acceptance and the alleviation of concerns about privacy, surveillance, and trust. OBJECTIVE: This study aims to examine the factors that foster a sense of trust and a perception of privacy in CT apps. Our study also investigates how trust and perceived privacy influence citizens' willingness to adopt, disclose personal data, and continue to use these apps. METHODS: Drawing on privacy calculus and procedural fairness theories, we developed a model of the antecedents and behavioral intentions related to trust and privacy perceptions. We used structural equation modeling to test our hypotheses on a data set collected at 2 time points (before and after the launch of a national CT app). The sample consisted of 405 Irish residents. RESULTS: Trust in CT apps was positively influenced by propensity to trust technology (ß=.074; P=.006), perceived need for surveillance (ß=.119; P<.001), and perceptions of government motives (ß=.671; P<.001) and negatively influenced by perceived invasion (ß=-.224; P<.001). Perceived privacy was positively influenced by trust (ß=.466; P<.001) and perceived control (ß=.451; P<.001) and negatively influenced by perceived invasion (ß=-.165; P<.001). Prelaunch intentions toward adoption were influenced by trust (ß=.590; P<.001) and perceived privacy (ß=.247; P<.001). Prelaunch intentions to disclose personal data to the app were also influenced by trust (ß=.215; P<.001) and perceived privacy (ß=.208; P<.001) as well as adoption intentions before the launch (ß=.550; P<.001). However, postlaunch intentions to use the app were directly influenced by prelaunch intentions (ß=.530; P<.001), but trust and perceived privacy only had an indirect influence. Finally, with regard to intentions to disclose after the launch, use intentions after the launch (ß=.665; P<.001) and trust (ß=.215; P<.001) had a direct influence, but perceived privacy only had an indirect influence. The proposed model explained 74.4% of variance in trust, 91% of variance in perceived privacy, 66.6% of variance in prelaunch adoption intentions, 45.9% of variance in postlaunch use intentions, and 83.9% and 79.4% of variance in willingness to disclose before the launch and after the launch, respectively. CONCLUSIONS: Positive perceptions of trust and privacy can be fostered through clear communication regarding the need and motives for CT apps, the level of control citizens maintain, and measures to limit invasive data practice. By engendering these positive beliefs before launch and reinforcing them after launch, citizens may be more likely to accept and use CT apps. These insights are important for the launch of future apps and technologies that require mass acceptance and information disclosure.

#Putkids1st: Health Professionals Using Social Media for Public Policy Advocacy-From Collective Action to Connective Action.

This study examines public policy advocacy by pediatricians and other health professionals in the hashtag community: #putkids1st. The study explores 4321 tweets that feature the hashtag, generated by 1231 unique users largely drawn from the American Association of Pediatricians and its members. The data are used to explore the structural dynamics of the hashtag community, the role of homophily, and to test a source-message framework to predict and recommendations to help improve engagement and retransmission of professional health advocacy messages.

Smart Streets as a Cyber-Physical Social Platform: A Conceptual Framework.

Streets perform a number of important functions and have a wide range of activities performed in them. There is a small but growing focus on streets as a more generalisable, atomised, and therefore more manageable unit of development and analysis than cities. Despite the public realm being one of the largest physical spaces on streets, the impact and potential of digitalisation projects on this realm is rarely considered. In this article, the smartness of a street is derived from the cyber-physical social infrastructure in the public realm, including data obtained from sensors, the interconnection between different services, technologies and social actors, intelligence derived from analysis of the data, and optimisation of operations within a street. This article conceptualises smart streets as basic units of urban space that leverage cyber-physical social infrastructure to provide and enable enhanced services to and between stakeholders, and through stakeholders' use of the street, generate data to optimise its services, capabilities, and value to stakeholders. A proposed conceptual framework is used to identify and explore how streets can be augmented and create value through cyber-physical social infrastructure and digital enhancements. We conclude with a discussion of future avenues of research.

A Brazilian classified data set for prognosis of tuberculosis, between January 2001 and April 2020.

After COVID-19, tuberculosis (TB) is the leading cause of death by an infectious disease in the world. This work presents a data set based on data collected from the Brazilian Information System for Notifiable Diseases (SINAN) for the period from January 2001 to April 2020 relating to patients diagnosed with tuberculosis in Brazil. The data from SINAN was pre-processed to generate a new data set with two distinct treatment outcome classes: CURED and DIED. The data set comprises 37 categorical attributes (including socio-demographic, clinical, and laboratory data) as well as the target class. There are 927,909 records of patients classified as CURED and 36,190 classified as DIED, totaling 964,099 records.

Prediction of malaria using deep learning models: A case study on city clusters in the state of Amazonas, Brazil, from 2003 to 2018.

BACKGROUND: Malaria is curable. Nonetheless, over 229 million cases of malaria were recorded in 2019, along with 409,000 deaths. Although over 42 million Brazilians are at risk of contracting malaria, 99% percent of all malaria cases in Brazil are located in or around the Amazon rainforest. Despite declining cases and deaths, malaria remains a major public health issue in Brazil. Accurate spatiotemporal prediction of malaria propagation may enable improved resource allocation to support efforts to eradicate the disease. METHODS: In response to calls for novel research on malaria elimination strategies that suit local conditions, in this study, we propose machine learning (ML) and deep learning (DL) models to predict the probability of malaria cases in the state of Amazonas. Using a dataset of approximately 6 million records (January 2003 to December 2018), we applied k-means clustering to group cities based on their similarity of malaria incidence. We evaluated random forest, long-short term memory (LSTM) and dated recurrent unit (GRU) models and compared their performance. RESULTS: The LSTM architecture achieved better performance in clusters with less variability in the number of cases, whereas the GRU presents better results in clusters with high variability. Although Diebold-Mariano testing suggested that both the LSTM and GRU performed comparably, GRU can be trained significantly faster, which could prove advantageous in practice. CONCLUSIONS: All models showed satisfactory accuracy and strong performance in predicting new cases of malaria, and each could serve as a supplemental tool to support regional policies and strategies.

Arboviral disease record data - Dengue and Chikungunya, Brazil, 2013-2020.

One of the main categories of Neglected Tropical Diseases (NTDs) are arboviruses, of which Dengue and Chikungunya are the most common. Arboviruses mainly affect tropical countries. Brazil has the largest absolute number of cases in Latin America. This work presents a unified data set with clinical, sociodemographic, and laboratorial data on confirmed patients of Dengue and Chikungunya, as well as patients ruled out of infection from these diseases. The data is based on case notification data submitted to the Brazilian Information System for Notifiable Diseases, from Portuguese Sistema de Informação de Agravo de Notificação (SINAN), from 2013 to 2020. The original data set comprised 13,421,230 records and 118 attributes. Following a pre-processing process, a final data set of 7,632,542 records and 56 attributes was generated. The data presented in this work will assist researchers in investigating antecedents of arbovirus emergence and transmission more generally, and Dengue and Chikungunya in particular. Furthermore, it can be used to train and test machine learning models for differential diagnosis and multi-class classification.

Corrigendum to "A dataset for accounting, finance and economics research on US data breaches" [Data in Brief 35 (2021) 1-6/ 106924].

[This corrects the article DOI: 10.1016/j.dib.2021.106924.].

Machine learning and deep learning techniques to support clinical diagnosis of arboviral diseases: A systematic review.

BACKGROUND: Neglected tropical diseases (NTDs) primarily affect the poorest populations, often living in remote, rural areas, urban slums or conflict zones. Arboviruses are a significant NTD category spread by mosquitoes. Dengue, Chikungunya, and Zika are three arboviruses that affect a large proportion of the population in Latin and South America. The clinical diagnosis of these arboviral diseases is a difficult task due to the concurrent circulation of several arboviruses which present similar symptoms, inaccurate serologic tests resulting from cross-reaction and co-infection with other arboviruses. OBJECTIVE: The goal of this paper is to present evidence on the state of the art of studies investigating the automatic classification of arboviral diseases to support clinical diagnosis based on Machine Learning (ML) and Deep Learning (DL) models. METHOD: We carried out a Systematic Literature Review (SLR) in which Google Scholar was searched to identify key papers on the topic. From an initial 963 records (956 from string-based search and seven from a single backward snowballing procedure), only 15 relevant papers were identified. RESULTS: Results show that current research is focused on the binary classification of Dengue, primarily using tree-based ML algorithms. Only one paper was identified using DL. Five papers presented solutions for multi-class problems, covering Dengue (and its variants) and Chikungunya. No papers were identified that investigated models to differentiate between Dengue, Chikungunya, and Zika. CONCLUSIONS: The use of an efficient clinical decision support system for arboviral diseases can improve the quality of the entire clinical process, thus increasing the accuracy of the diagnosis and the associated treatment. It should help physicians in their decision-making process and, consequently, improve the use of resources and the patient's quality of life.

Prediction of malaria using deep learning models: A case study on city clusters in the state of Amazonas, Brazil, from 2003 to 2018

ABSTRACT Background: Malaria is curable. Nonetheless, over 229 million cases of malaria were recorded in 2019, along with 409,000 deaths. Although over 42 million Brazilians are at risk of contracting malaria, 99% percent of all malaria cases in Brazil are located in or around the Amazon rainforest. Despite declining cases and deaths, malaria remains a major public health issue in Brazil. Accurate spatiotemporal prediction of malaria propagation may enable improved resource allocation to support efforts to eradicate the disease. Methods: In response to calls for novel research on malaria elimination strategies that suit local conditions, in this study, we propose machine learning (ML) and deep learning (DL) models to predict the probability of malaria cases in the state of Amazonas. Using a dataset of approximately 6 million records (January 2003 to December 2018), we applied k-means clustering to group cities based on their similarity of malaria incidence. We evaluated random forest, long-short term memory (LSTM) and dated recurrent unit (GRU) models and compared their performance. Results: The LSTM architecture achieved better performance in clusters with less variability in the number of cases, whereas the GRU presents better results in clusters with high variability. Although Diebold-Mariano testing suggested that both the LSTM and GRU performed comparably, GRU can be trained significantly faster, which could prove advantageous in practice. Conclusions: All models showed satisfactory accuracy and strong performance in predicting new cases of malaria, and each could serve as a supplemental tool to support regional policies and strategies.

A dataset for accounting, finance and economics research on US data breaches.

This data article describes a dataset of data breaches in US listed firms over a ten-year period. Data breaches represent major events that pose serious challenges to organisations. The number of incidents has been on the increase over the last decade and this has attracted the interest of the media, consumers and regulators. While there is a well-established literature on cybersecurity in Computer Science and Information Systems journals, studies exploring the economic and business impacts of data breaches represent a relatively recent phenomenon. There is a nascent but fast-growing literature in accounting, finance and economics that focuses on the financial impacts of data breaches and this dataset provides a useful resource for future studies in this space. By providing data on the company identifier, the type of breach, the dates of breach disclosure, and relates these dates to the company's fiscal year, the dataset can be merged quickly with existing accounting and finance datasets. The dataset includes data on 506 incidents over a ten-year period thereby enabling cross-sectional and longitudinal analyses.

Artificial Mercosur license plates dataset.

Mercosur (a.k.a. Mercosul) is a trade bloc comprising five South American countries. In 2018, a unified Mercosur license plate model was rolled out. Access to large volumes of ground truth Mercosur license plates with sufficient presentation variety is a significant challenge for training supervised models for license plate detection (LPD) in automatic license plate recognition (ALPR) systems. To address this problem, a Mercosur license plate generator was developed to generate artificial license plate images meeting the new standard with sufficient variety for ALPR training purposes. This includes images with variation due to occlusions and environmental conditions. An embedded system was developed for detecting legacy license plates in images of real scenarios and overwriting these with artificially generated Mercosur license plates. This data set comprises 3,829 images of vehicles with synthetic license plates that meet the new Mercosur standard in real scenarios, and equivalent number of text files containing label information for the images, all organized in a CSV file with compiled image file paths and associated labels.

Sorting the Healthy Diet Signal from the Social Media Expert Noise: Preliminary Evidence from the Healthy Diet Discourse on Twitter.

Over 2.8 million people die each year from being overweight or obese, a largely preventable disease. Social media has fundamentally changed the way we communicate, collaborate, consume, and create content. The ease with which content can be shared has resulted in a rapid increase in the number of individuals or organisations that seek to influence opinion and the volume of content that they generate. The nutrition and diet domain is not immune to this phenomenon. Unfortunately, from a public health perspective, many of these 'influencers' may be poorly qualified in order to provide nutritional or dietary guidance, and advice given may be without accepted scientific evidence and contrary to public health policy. In this preliminary study, we analyse the 'healthy diet' discourse on Twitter. While using a multi-component analytical approach, we analyse more than 1.2 million English language tweets over a 16-month period in order to identify and characterise the influential actors and discover topics of interest in the discourse. Our analysis suggests that the discourse is dominated by non-health professionals. There is widespread use of bots that pollute the discourse and seek to create a false equivalence on the efficacy of a particular nutritional strategy or diet. Topic modelling suggests a significant focus on diet, nutrition, exercise, weight, disease, and quality of life. Public health policy makers and professional nutritionists need to consider what interventions can be taken in order to counteract the influence of non-professional and bad actors on social media.

#StayHome: Monitoring and benchmarking social isolation trends in Caruaru and the Região Metropolitana do Recife during the COVID-19 pandemic.

This technical report presents information related to the Social Isolation Index (SII) of the city of Caruaru, Pernambuco, Brazil. The data was provided by In Loco, a technology startup that has collected the movement of around 60 million Brazilians through cell phone location.

Data set for automatic detection of online misogynistic speech.

The data set is composed of 2285 definitions posted on the Urban Dictionary platform from 1999 to May 2016. The data was classified as misogynistic and non-misogynistic by three independent researchers with domain knowledge. The data set is available in public repository in a table containing two columns: the text-based definition from Urban Dictionary and its respective classification (1 for misogynistic and 0 for non-misogynistic).

Accelerometer-Based Human Fall Detection Using Convolutional Neural Networks.

Human falls are a global public health issue resulting in over 37.3 million severe injuries and 646,000 deaths yearly. Falls result in direct financial cost to health systems and indirectly to society productivity. Unsurprisingly, human fall detection and prevention are a major focus of health research. In this article, we consider deep learning for fall detection in an IoT and fog computing environment. We propose a Convolutional Neural Network composed of three convolutional layers, two maxpool, and three fully-connected layers as our deep learning model. We evaluate its performance using three open data sets and against extant research. Our approach for resolving dimensionality and modelling simplicity issues is outlined. Accuracy, precision, sensitivity, specificity, and the Matthews Correlation Coefficient are used to evaluate performance. The best results are achieved when using data augmentation during the training process. The paper concludes with a discussion of challenges and future directions for research in this domain.