Search

With the increasing advance of computer and information technologies, numerous research papers have been published online as well as offline, and as new research fields have been continuingly created, users have a lot of trouble in finding and categorizing their interesting research papers. In order to overcome the limitations, this paper proposes a research paper classification system that can cluster research papers into the meaningful class in which papers are very likely to have similar subjects. The proposed system extracts representative keywords from the abstracts of each paper and topics by Latent Dirichlet allocation (LDA) scheme. Then, the K-means clustering algorithm is applied to classify the whole papers into research papers with similar subjects, based on the Term frequency-inverse document frequency (TF-IDF) values of each paper.

Background Many previous papers have investigated most-cited articles or most productive authors in academics, but few have studied most-cited authors. Two challenges are faced in doing so, one of which is that some different authors will have the same name in the bibliometric data, and the second is that coauthors’ contributions are different in the article byline. No study has dealt with the matter of duplicate names in bibliometric data. Although betweenness centrality (BC) is one of the most popular degrees of density in social network analysis (SNA), few have applied the BC algorithm to interpret a network’s characteristics. A quantitative scheme must be used for calculating weighted author credits and then applying the metrics in comparison. Objective This study aimed to apply the BC algorithm to examine possible identical names in a network and report the most-cited authors for a journal related to international mobile health (mHealth) research. Methods We obtained 676 abstracts from Medline based on the keywords “JMIR mHealth and uHealth” (Journal) on June 30, 2018. The author names, countries/areas, and author-defined keywords were recorded. The BCs were then calculated for the following: (1) the most-cited authors displayed on Google Maps; (2) the geographical distribution of countries/areas for the first author; and (3) the keywords dispersed by BC and related to article topics in comparison on citation indices. Pajek software was used to yield the BC for each entity (or node). Bibliometric indices, including h-, g-, and x-indexes, the mean of core articles on g(Ag)=sum (citations on g-core/publications on g-core), and author impact factor (AIF), were applied. Results We found that the most-cited author was Sherif M Badawy (from the United States), who had published six articles on JMIR mHealth and uHealth with high bibliometric indices (h=3; AIF=8.47; x=4.68; Ag=5.26). We also found that the two countries with the highest BC were the United States and the United Kingdom and that the two keyword clusters of mHealth and telemedicine earned the highest indices in comparison to other counterparts. All visual representations were successfully displayed on Google Maps. Conclusions The most cited authors were selected using the authorship-weighted scheme (AWS), and the keywords of mHealth and telemedicine were more highly cited than other counterparts. The results on Google Maps are novel and unique as knowledge concept maps for understanding the feature of a journal. The research approaches used in this study (ie, BC and AWS) can be applied to other bibliometric analyses in the future.

BackgroundPeriodic demographic health surveillance and surveys are the main sources of health information in developing countries. Conducting a survey requires extensive use of paper-pen and manual work and lengthy processes to generate the required information. Despite the rise of popularity in using electronic data collection systems to alleviate the problems, sufficient evidence is not available to support the use of electronic data capture (EDC) tools in interviewer-administered data collection processes. ObjectiveThis study aimed to compare data quality parameters in the data collected using mobile electronic and standard paper-based data capture tools in one of the health and demographic surveillance sites in northwest Ethiopia. MethodsA randomized controlled crossover health care information technology evaluation was conducted from May 10, 2016, to June 3, 2016, in a demographic and surveillance site. A total of 12 interviewers, as 2 individuals (one of them with a tablet computer and the other with a paper-based questionnaire) in 6 groups were assigned in the 6 towns of the surveillance premises. Data collectors switched the data collection method based on computer-generated random order. Data were cleaned using a MySQL program and transferred to SPSS (IBM SPSS Statistics for Windows, Version 24.0) and R statistical software (R version 3.4.3, the R Foundation for Statistical Computing Platform) for analysis. Descriptive and mixed ordinal logistic analyses were employed. The qualitative interview audio record from the system users was transcribed, coded, categorized, and linked to the International Organization for Standardization 9241-part 10 dialogue principles for system usability. The usability of this open data kit–based system was assessed using quantitative System Usability Scale (SUS) and matching of qualitative data with the isometric dialogue principles. ResultsFrom the submitted 1246 complete records of questionnaires in each tool, 41.89% (522/1246) of the paper and pen data capture (PPDC) and 30.89% (385/1246) of the EDC tool questionnaires had one or more types of data quality errors. The overall error rates were 1.67% and 0.60% for PPDC and EDC, respectively. The chances of more errors on the PPDC tool were multiplied by 1.015 for each additional question in the interview compared with EDC. The SUS score of the data collectors was 85.6. In the qualitative data response mapping, EDC had more positive suitability of task responses with few error tolerance characteristics. ConclusionsEDC possessed significantly better data quality and efficiency compared with PPDC, explained with fewer errors, instant data submission, and easy handling. The EDC proved to be a usable data collection tool in the rural study setting. Implementation organization needs to consider consistent power source, decent internet connection, standby technical support, and security assurance for the mobile device users for planning full-fledged implementation and integration of the system in the surveillance site.

The entire scientific and academic community has been mobilized to gain a better understanding of the COVID-19 disease and its impact on humanity. Most research related to COVID-19 needs to analyze large amounts of data in very little time. This urgency has made Big Data Analysis, and related questions around the privacy and security of the data, an extremely important part of research in the COVID-19 era. The White House OSTP has, for example, released a large dataset of papers related to COVID research from which the research community can extract knowledge and information. We show an example system with a machine learning-based knowledge extractor which draws out key medical information from COVID-19 related academic research papers. We represent this knowledge in a Knowledge Graph that uses the Unified Medical Language System (UMLS). However, publicly available studies rely on dataset that might have sensitive data. Extracting information from academic papers can potentially leak sensitive data, and protecting the security and privacy of this data is equally important. In this paper, we address the key challenges around the privacy and security of such information extraction and analysis systems. Policy regulations like HIPAA have updated the guidelines to access data, specifically, data related to COVID-19, securely. In the US, healthcare providers must also comply with the Office of Civil Rights (OCR) rules to protect data integrity in matters like plasma donation, media access to health care data, telehealth communications, etc. Privacy policies are typically short and unstructured HTML or PDF documents. We have created a framework to extract relevant knowledge from the health centers’ policy documents and also represent these as a knowledge graph. Our framework helps to understand the extent to which individual provider policies comply with regulations and define access control policies that enforce the regulation rules on data in the knowledge graph extracted from COVID-related papers. Along with being compliant, privacy policies must also be transparent and easily understood by the clients. We analyze the relative readability of healthcare privacy policies and discuss the impact. In this paper, we develop a framework for access control decisions that uses policy compliance information to securely retrieve COVID data. We show how policy compliance information can be used to restrict access to COVID-19 data and information extracted from research papers.

International audience; The task of tagging papers with semantic metadata in order to analyze their relatedness represents a good foundation for a paper recommender system. The analysis from this paper extends from previous research in order to create a graph of papers from a specific domain with the purpose of determining each article's importance within the considered corpus of papers. Moreover, as non-latent representations are powerful when used in conjunction with latent ones, our system retrieves semantically close words, not present in the paper, in order to improve the retrieval of papers. Our previous analyses used the semantic representation of papers in different semantic models with the purpose of creating visual graphs based on the semantic relatedness links between the abstracts. The current analysis takes a step forward by proposing a model that can suggest which papers are of the highest relevance, share similar concepts, and are semantically related with the initial query. Our study is performed using paper abstracts in the field of information technology extracted from the Web of Science citation index. The research includes a use case and its corresponding results by using interactive and exploratory network graph representations.

This paper provides tutorial introductions to Field-Programmable Gate Arrays (FPGAs) and the concept of hardware compilation - the translation of a high-level programming language directly into a hardware design. As an illustration, a simple stepper motor control program is presented. The research aims of the sensor validation programme are described, and the benefits of using hardware compilation techniques are presented. This leads on in the conclusion of the paper to a more general discussion of the interaction between research and technology, and in particular the influence of information technology upon control engineering.

Based on the existing disadvantages of traditional teaching mode of Satellite Meteorology course in Chengdu University of Information Technology and the new teaching mode overview of “Internet Plus flipped class”, this paper is mainly to construct flipped class teaching mode of Satellite Meteorology under the background of “Internet Plus” in Chengdu University of Information Technology. Through the Internet, it also tries to integrate this teaching mode into flipped class before class, in class, after class, which means to provide a new thought for the teaching of this course in colleges and universities.

Information technology has changed the way health care is delivered. Electronic health records which are prevalently deployed to replace or supplement paper documentations have made distributed information access at various points of care and work activity achievable with the use of mobile information devices. Our particular concern is with nurse's information flow, where nurse's notes and observations taken at the point of care feed into the electronic record. In these cases, digital technology has not yet entirely replaced paper and pen, because the latter still provide greater ease and flexibility of use when compared to current digital technologies. Even when mobile digital technology is available, clinicians still prefer creating handwritten notes, and then later manually transposing them into the digital medium. Within this context, we created a prototype that integrated digital paper with electronic health charts to retain the benefits of paper and pen, as well as digital medium. A focus group evaluation of this prototype demonstrated promise and potential for its value in a medical environment.

Cloud computing is emerging as a powerful architecture to perform large-scale and complex computing. It extends the information technology (IT) capability by providing ondemand access to computer resources for dedicated use. The information security and privacy are the major concerns over the cloud from user perspective. This paper surveys and evaluates the architecture, data security and privacy issues in cloud computing like data confidentiality, integrity, authentication, trust, service level agreements and regulatory issues. The objective of this paper is to review comprehensively the current challenges of data security and privacy being faced by cloud computing and critically analyze these issues.

Rhetorical sentence classification is an interesting approach for making extractive summaries but this technique still needs to be developed because the performance of automatic rhetorical sentence classification is still poor. Rhetorical sentences are sentences that contain rhetorical words or phrases. Rhetorical sentences not only appear in the contents of a paper but also in the title. In this study, features related to section class and title class that have been proposed in a previous research were further developed. Our method uses different techniques to reach automatic section class extraction for which we introduce new, format-based features. Furthermore, we propose automatic rhetoric phrase extraction from the title. The corpus we used was a collection of technical-experimental scientific papers. Our method uses the Support Vector Machine (SVM) algorithm and the Naive Bayesian algorithm for classification. The four categories used were: Problem, Method, Data, and Result . It was hypothesized that these features would be able to improve classification accuracy compared to previous methods. The F-measure for these categories reached up to 14%.

The method for filtering information from large volumes of text is called Information Extraction. It is a limited task than understanding the full text. In full text understanding, we express in an explicit fashion about all the information in a given text. But, in Information Extraction, we delimit in advance, as part of the specification of the task and the semantic range of the result. Only extractive summarization method is considered and developed for the study. In this article a model for summarization from large documents using a novel approach has been proposed by considering one of the South Indian regional languages (Kannada). It deals with a single document summarization based on statistical approach. The purpose of summary of an article is to facilitate the quick and accurate identification of the topic of the published document. The objective is to save prospective readers’ time and effort in finding the useful information in a given huge article. Various analyses of results were also discussed by comparing it with the English language.

When faced with the tasks of reading and writing a complex technical paper, many nonnative scientists and engineers who have a solid background in English grammar and vocabulary lack an adequate knowledge of commonly used structural patterns at the discourse level. In this paper, we propose a novel computer software tool that can assist these people in the understanding and construction of technical papers, by automatically identifying the structure of writing in different fields and disciplines. The system is tested using research article abstracts and is shown to be a fast, accurate, and useful aid in the reading and writing process.

Advances in information technology have simplified many processes in our lives. However, in many cases trust issues arise when new technology is introduced, and voting is one prominent example. To increase voters' trust, current e-voting systems provide paper audit trails (PATs) which enable automatic tally and/or manual audit of the election result. PATs may contain only the encrypted vote or the plaintext vote in human-readable and/or machine-readable format. Previous studies report voter privacy concerns with PATs containing additional information (e.g. QR-Codes) other than the human-readable plaintext vote. However, omitting such PATs negatively influences security and/or efficiency. Hence, to address these concerns we applied the coping and threat appraisal principles of the protection motivation theory in the communication process. We evaluated them in separate surveys focused on the EasyVote system [15]. Results show that the coping appraisal is more promising than the threat appraisal approach. While our findings provide novel directions on addressing privacy concerns in the e-voting context, corresponding limitations need to be considered for future user studies.

With the increased use of Internet and information technol- ogy all over the world, there is an increased amount of criminal activities that involve computing and digital data. These digital crimes (e-crimes) impose new challenges on prevention, detection, investigation, and pros- ecution of the corresponding offences. Computer forensics (also known as cyberforensics) is an emerging research area that applies computer in- vestigation and analysis techniques to help detection of these crimes and gathering of digital evidence suitable for presentation in courts. This new area combines the knowledge of information technology, forensics science, and law and gives rise to a number of interesting and challeng- ing problems related to computer security and cryptography that are yet to be solved. In this paper, we present and discuss some of these prob- lems together with two successful cases of computer forensics technology developed in Hong Kong that enable the law enforcement departments to detect and investigate digital crimes more efficiently and effectively. We believe that computer forensics research is an important area in ap- plying security and computer knowledge to build a better society.

It is well known that undiscovered errors in a requirements specification is extremely expensive to be fixed when discovered in the software maintenance phase. Errors in the requirement phase can be reduced through the validation and verification of the requirements specification. Many logic-based requirements specification languages have been developed to achieve these goals. However, the execution and reasoning of a logic-based requirements specification can be very slow. An effective way to improve their performance is to execute and reason the logic-based requirements specification in parallel. In this paper, we present a hybrid model to facilitate the parallel execution of a logic-based requirements specification language. A logic-based specification is first applied by a data dependency analysis technique which can find all the mode combinations that exist within a specification clause. This mode information is used to support a novel hybrid parallel execution model, which combines both top-down and bottom-up evaluation strategies. This new execution model can find the failure in the deepest node of the search tree at the early stage of the evaluation, thus this new execution model can reduce the total number of nodes searched in the tree, the total processes needed to be generated, and the total communication channels needed in the search process. A simulator has been implemented to analyze the execution behavior of the new model. Experiments show significant improvement based on several criteria.

The First International Conference on Information Systems Technology and Management was held in New Delhi in March, 2007. The conference was organized jointly by the University of Florida, USA and the Institute of Management and Technology, Ghaziabad, India. Forty-seven papers, divided into three tracks, were presented at the conference. The tracks were: Information Systems, Information Technology and Information Management. For this special issue, devoted to ICISTM-07, nine papers were submitted for consideration, of which two have been accepted for publication. The first paper, titled, ‘Reactivity Based Model to Study Online Auctions Dynamics,’ by Pereira et al. addresses bidder behavior issues in online auction environments. The paper develops a new characterization model for online auctions. In this model, they take into account the change in a bidder’s behavior motivated by events during the course of an auction. The new model is applied to actual datasets of eBay auctions to understand specific online auction negotiation patterns. The model has implications for the design of more complete automatic agents. The second paper titled ‘Computing Full Disjunction Using COJO’ by Kumar et al. addresses the issue of data integration in a relational database management environment. The authors propose a ‘Correct Outer Join Ordering’ strategy which results in full disjunction of relations resulting in a better database design compared to existing strategies.

I am happy and honored to publish a special issue on Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment, and Management (HNICEM) in the JACIII. The demand for the application of HNICEM is rapidly growing. We have received many papers focused in the fields of Manufacturing, Academe, and Business during the first international conference on HNICEM. This special issue includes 11 papers carefully and evenly selected from these fields. I strongly believed that the topics in this issue encompass technologies that will play a significant role in the development of HNICEM. I am very grateful to Prof. Toshio Fukuda and Prof. Kaoru Hirota, the Chief editors of JACIII for inviting me as Guest editor in this Journal. I would like to express my warmest thanks and gratitude to the distinguished authors that have contributed to this special issue, for their outstanding contributions and cooperation. I am hoping that the JACIII will continually provide papers regarding progress of new technologies particularly in the fields of HNICEM.

Recently, virus diseases, such as SARS-CoV, MERS-CoV, and COVID-19, continue to emerge and pose a severe public health problem. These diseases threaten the lives of many people and cause serious social and economic losses. Recent developments in information technology (IT) and connectivity have led to the emergence of Internet of Things (IoT) and Artificial Intelligence (AI) applications in many industries. These industries, where IoT and AI together are making significant impacts, are the healthcare and the diagnosis department. In addition, by actively communicating with smart devices and various biometric sensors, it is expanding its application fields to telemedicine, healthcare, and disease prevention. Even though existing IoT and AI technologies can enhance disease detection, monitoring, and quarantine, their impact is very limited because they are not integrated or applied rapidly to the emergence of a sudden epidemic. Especially in the situation where infectious diseases are rapidly spreading, the conventional methods fail to prevent large-scale infections and block global spreads through prediction, resulting in great loss of lives. Therefore, in this paper, various sources of infection information with local limitations are collected through virus disease information collector, and AI analysis and severity matching are performed through AI broker. Finally, through the Integrated Disease Control Center, risk alerts are issued, proliferation block letters are sent, and post-response services are provided quickly. Suppose we further develop the proposed integrated virus disease control model. In that case, it will be possible to proactively detect and warn of risk factors in response to infectious diseases that are rapidly spreading worldwide and strengthen measures to prevent spreading of infection in no time.

Machine learning has been the corner stone in analysing and extracting information from data and often a problem of missing values is encountered. Missing values occur because of various factors like missing completely at random, missing at random or missing not at random. All these may result from system malfunction during data collection or human error during data pre-processing. Nevertheless, it is important to deal with missing values before analysing data since ignoring or omitting missing values may result in biased or misinformed analysis. In literature there have been several proposals for handling missing values. In this paper, we aggregate some of the literature on missing data particularly focusing on machine learning techniques. We also give insight on how the machine learning approaches work by highlighting the key features of missing values imputation techniques, how they perform, their limitations and the kind of data they are most suitable for. We propose and evaluate two methods, the k nearest neighbor and an iterative imputation method (missForest) based on the random forest algorithm. Evaluation is performed on the Iris and novel power plant fan data with induced missing values at missingness rate of 5% to 20%. We show that both missForest and the k nearest neighbor can successfully handle missing values and offer some possible future research direction.

Natural Language Processing (NLP) is one of the most captivating applications of Deep Learning. In this survey, we consider how the Data Augmentation training strategy can aid in its development. We begin with the major motifs of Data Augmentation summarized into strengthening local decision boundaries, brute force training, causality and counterfactual examples, and the distinction between meaning and form. We follow these motifs with a concrete list of augmentation frameworks that have been developed for text data. Deep Learning generally struggles with the measurement of generalization and characterization of overfitting. We highlight studies that cover how augmentations can construct test sets for generalization. NLP is at an early stage in applying Data Augmentation compared to Computer Vision. We highlight the key differences and promising ideas that have yet to be tested in NLP. For the sake of practical implementation, we describe tools that facilitate Data Augmentation such as the use of consistency regularization, controllers, and offline and online augmentation pipelines, to preview a few. Finally, we discuss interesting topics around Data Augmentation in NLP such as task-specific augmentations, the use of prior knowledge in self-supervised learning versus Data Augmentation, intersections with transfer and multi-task learning, and ideas for AI-GAs (AI-Generating Algorithms). We hope this paper inspires further research interest in Text Data Augmentation.

Motivation The microbes that live in an environment can be identified from the combined genomic material, also referred to as the metagenome. Sequencing a metagenome can result in large volumes of sequencing reads. A promising approach to reduce the size of metagenomic datasets is by clustering reads into groups based on their overlaps. Clustering reads are valuable to facilitate downstream analyses, including computationally intensive strain-aware assembly. As current read clustering approaches cannot handle the large datasets arising from high-throughput metagenome sequencing, a novel read clustering approach is needed. In this article, we propose OGRE, an Overlap Graph-based Read clustEring procedure for high-throughput sequencing data, with a focus on shotgun metagenomes. Results We show that for small datasets OGRE outperforms other read binners in terms of the number of species included in a cluster, also referred to as cluster purity, and the fraction of all reads that is placed in one of the clusters. Furthermore, OGRE is able to process metagenomic datasets that are too large for other read binners into clusters with high cluster purity. Conclusion OGRE is the only method that can successfully cluster reads in species-specific clusters for large metagenomic datasets without running into computation time- or memory issues. Availabilityand implementation Code is made available on Github (https://github.com/Marleen1/OGRE). Supplementary information Supplementary data are available at Bioinformatics online.

BackgroundDigital phenotyping and machine learning are currently being used to augment or even replace traditional analytic procedures in many domains, including health care. Given the heavy reliance on smartphones and mobile devices around the world, this readily available source of data is an important and highly underutilized source that has the potential to improve mental health risk prediction and prevention and advance mental health globally.ObjectiveThis study aimed to apply machine learning in an acute mental health setting for suicide risk prediction. This study uses a nascent approach, adding to existing knowledge by using data collected through a smartphone in place of clinical data, which have typically been collected from health care records.MethodsWe created a smartphone app called Strength Within Me, which was linked to Fitbit, Apple Health kit, and Facebook, to collect salient clinical information such as sleep behavior and mood, step frequency and count, and engagement patterns with the phone from a cohort of inpatients with acute mental health (n=66). In addition, clinical research interviews were used to assess mood, sleep, and suicide risk. Multiple machine learning algorithms were tested to determine the best fit.ResultsK-nearest neighbors (KNN; k=2) with uniform weighting and the Euclidean distance metric emerged as the most promising algorithm, with 68% mean accuracy (averaged over 10,000 simulations of splitting the training and testing data via 10-fold cross-validation) and an average area under the curve of 0.65. We applied a combined 5×2 F test to test the model performance of KNN against the baseline classifier that guesses training majority, random forest, support vector machine and logistic regression, and achieved F statistics of 10.7 (P=.009) and 17.6 (P=.003) for training majority and random forest, respectively, rejecting the null of performance being the same. Therefore, we have taken the first steps in prototyping a system that could continuously and accurately assess the risk of suicide via mobile devices.ConclusionsPredicting for suicidality is an underaddressed area of research to which this paper makes a useful contribution. This is part of the first generation of studies to suggest that it is feasible to utilize smartphone-generated user input and passive sensor data to generate a risk algorithm among inpatients at suicide risk. The model reveals fair concordance between phone-derived and research-generated clinical data, and with iterative development, it has the potential for accurate discriminant risk prediction. However, although full automation and independence of clinical judgment or input would be a worthy development for those individuals who are less likely to access specialist mental health services, and for providing a timely response in a crisis situation, the ethical and legal implications of such advances in the field of psychiatry need to be acknowledged.

Background In recent years, there has been an exponential growth of mobile health (mHealth)–related apps. This has occurred in a somewhat unsupervised manner. Therefore, having a set of criteria that could be used by all stakeholders to guide the development process and the assessment of the quality of the apps is of most importance. Objective The aim of this paper is to study the validity of the Mobile App Development and Assessment Guide (MAG), a guide recently created to help stakeholders develop and assess mobile health apps. Methods To conduct a validation process of the MAG, we used the Delphi method to reach a consensus among participating stakeholders. We identified 158 potential participants: 45 patients as potential end users, 41 health care professionals, and 72 developers. We sent participants an online survey and asked them to rate how important they considered each item in the guide to be on a scale from 0 to 10. Two rounds were enough to reach consensus. Results In the first round, almost one-third (n=42) of those invited participated, and half of those (n=24) also participated in the second round. Most items in the guide were found to be important to a quality mHealth-related app; a total of 48 criteria were established as important. “Privacy,” “security,” and “usability” were the categories that included most of the important criteria. Conclusions The data supports the validity of the MAG. In addition, the findings identified the criteria that stakeholders consider to be most important. The MAG will help advance the field by providing developers, health care professionals, and end users with a valid guide so that they can develop and identify mHealth-related apps that are of quality.

Background Health education emerged as an important intervention for improving the awareness and self-management abilities of chronic disease patients. The development of information technologies has changed the form of patient educational materials from traditional paper materials to electronic materials. To date, the amount of patient educational materials on the internet is tremendous, with variable quality, which makes it hard to identify the most valuable materials by individuals lacking medical backgrounds. Objective The aim of this study was to develop a health recommender system to provide appropriate educational materials for chronic disease patients in China and evaluate the effect of this system. Methods A knowledge-based recommender system was implemented using ontology and several natural language processing (NLP) techniques. The development process was divided into 3 stages. In stage 1, an ontology was constructed to describe patient characteristics contained in the data. In stage 2, an algorithm was designed and implemented to generate recommendations based on the ontology. Patient data and educational materials were mapped to the ontology and converted into vectors of the same length, and then recommendations were generated according to similarity between these vectors. In stage 3, the ontology and algorithm were incorporated into an mHealth system for practical use. Keyword extraction algorithms and pretrained word embeddings were used to preprocess educational materials. Three strategies were proposed to improve the performance of keyword extraction. System evaluation was based on a manually assembled test collection for 50 patients and 100 educational documents. Recommendation performance was assessed using the macro precision of top-ranked documents and the overall mean average precision (MAP). Results The constructed ontology contained 40 classes, 31 object properties, 67 data properties, and 32 individuals. A total of 80 SWRL rules were defined to implement the semantic logic of mapping patient original data to the ontology vector space. The recommender system was implemented as a separate Web service connected with patients' smartphones. According to the evaluation results, our system can achieve a macro precision up to 0.970 for the top 1 recommendation and an overall MAP score up to 0.628. Conclusions This study demonstrated that a knowledge-based health recommender system has the potential to accurately recommend educational materials to chronic disease patients. Traditional NLP techniques combined with improvement strategies for specific language and domain proved to be effective for improving system performance. One direction for future work is to explore the effect of such systems from the perspective of patients in a practical setting.

Background Maximal oxygen consumption (VO2max) is one of the most predictive biometrics for cardiovascular health and overall mortality. However, VO2max is rarely measured in large-scale research studies or routine clinical care because of the high cost, participant burden, and requirement for specialized equipment and staff. Objective To overcome the limitations of clinical VO2max measurement, we aim to develop a digital VO2max estimation protocol that can be self-administered remotely using only the sensors within a smartphone. We also aim to validate this measure within a broadly representative population across a spectrum of smartphone devices. Methods Two smartphone-based VO2max estimation protocols were developed: a 12-minute run test (12-MRT) based on distance measured by GPS and a 3-minute step test (3-MST) based on heart rate recovery measured by a camera. In a 101-person cohort, balanced across age deciles and sex, participants completed a gold standard treadmill-based VO2max measurement, two silver standard clinical protocols, and the smartphone-based 12-MRT and 3-MST protocols in the clinic and at home. In a separate 120-participant cohort, the video-based heart rate measurement underlying the 3-MST was measured for accuracy in individuals across the spectrum skin tones while using 8 different smartphones ranging in cost from US $99 to US $999. Results When compared with gold standard VO2max testing, Lin concordance was pc=0.66 for 12-MRT and pc=0.61 for 3-MST. However, in remote settings, the 12-MRT was significantly less concordant with the gold standard (pc=0.25) compared with the 3-MST (pc=0.61), although both had high test-retest reliability (12-MRT intraclass correlation coefficient=0.88; 3-MST intraclass correlation coefficient=0.86). On the basis of the finding that 3-MST concordance was generalizable to remote settings whereas 12-MRT was not, the video-based heart rate measure within the 3-MST was selected for further investigation. Heart rate measurements in any of the combinations of the six Fitzpatrick skin tones and 8 smartphones resulted in a concordance of pc≥0.81. Performance did not correlate with device cost, with all phones selling under US $200 performing better than pc>0.92. Conclusions These findings demonstrate the importance of validating mobile health measures in the real world across a diverse cohort and spectrum of hardware. The 3-MST protocol, termed as heart snapshot, measured VO2max with similar accuracy to supervised in-clinic tests such as the Tecumseh (pc=0.94) protocol, while also generalizing to remote and unsupervised measurements. Heart snapshot measurements demonstrated fidelity across demographic variation in age and sex, across diverse skin pigmentation, and between various iOS and Android phone configurations. This software is freely available for all validation data and analysis code.

Background With the growing adult population using electronic hearing devices such as cochlear implants or hearing aids, there is an increasing worldwide need for auditory training (AT) to promote optimal device use. However, financial resources and scheduling conflicts make clinical AT infeasible. Objective To address this gap between need and accessibility, we primarily aimed to develop a mobile health (mHealth) app called Speech Banana for AT. The app would be substantially more affordable and portable than clinical AT; would deliver a validated training model that is reflective of modern techniques; and would track users’ progress in speech comprehension, providing greater continuity between periodic in-person visits. To improve international availability, our secondary aim was to implement the English language training model into Korean as a proof of concept for worldwide usability. Methods A problem- and objective-centered Design Science Research Methodology approach was adopted to develop the Speech Banana app. A review of previous literature and computer-based learning programs outlined current AT gaps, whereas interviews with speech pathologists and users clarified the features that were addressed in the app. Past and present users were invited to evaluate the app via community forums and the System Usability Scale. Results Speech Banana has been implemented in English and Korean languages for iPad and web use. The app comprises 38 lessons, which include analytic exercises pairing visual and auditory stimuli, and synthetic quizzes presenting auditory stimuli only. During quizzes, users type the sentence heard, and the app provides visual feedback on performance. Users may select a male or female speaker and the volume of background noise, allowing for training with a range of frequencies and signal-to-noise ratios. There were more than 3200 downloads of the English iPad app and almost 100 downloads of the Korean app; more than 100 users registered for the web apps. The English app received a System Usability Scale rating of “good” from 6 users, and the Korean app received a rating of “OK” from 16 users. Conclusions Speech Banana offers AT accessibility with a validated curriculum, allowing users to develop speech comprehension skills with the aid of a mobile device. This mHealth app holds potential as a supplement to clinical AT, particularly in this era of global telemedicine.

Background The use of wearables facilitates data collection at a previously unobtainable scale, enabling the construction of complex predictive models with the potential to improve health. However, the highly personal nature of these data requires strong privacy protection against data breaches and the use of data in a way that users do not intend. One method to protect user privacy while taking advantage of sharing data across users is federated learning, a technique that allows a machine learning model to be trained using data from all users while only storing a user’s data on that user’s device. By keeping data on users’ devices, federated learning protects users’ private data from data leaks and breaches on the researcher’s central server and provides users with more control over how and when their data are used. However, there are few rigorous studies on the effectiveness of federated learning in the mobile health (mHealth) domain. Objective We review federated learning and assess whether it can be useful in the mHealth field, especially for addressing common mHealth challenges such as privacy concerns and user heterogeneity. The aims of this study are to describe federated learning in an mHealth context, apply a simulation of federated learning to an mHealth data set, and compare the performance of federated learning with the performance of other predictive models. Methods We applied a simulation of federated learning to predict the affective state of 15 subjects using physiological and motion data collected from a chest-worn device for approximately 36 minutes. We compared the results from this federated model with those from a centralized or server model and with the results from training individual models for each subject. Results In a 3-class classification problem using physiological and motion data to predict whether the subject was undertaking a neutral, amusing, or stressful task, the federated model achieved 92.8% accuracy on average, the server model achieved 93.2% accuracy on average, and the individual model achieved 90.2% accuracy on average. Conclusions Our findings support the potential for using federated learning in mHealth. The results showed that the federated model performed better than a model trained separately on each individual and nearly as well as the server model. As federated learning offers more privacy than a server model, it may be a valuable option for designing sensitive data collection methods.

JMIR mhealth and uhealth 9(1), e19346 (2021). doi:10.2196/19346, Published by JMIR Publications, Toronto

Background In affective exergames, game difficulty is dynamically adjusted to match the user’s physical and psychological state. Such an adjustment is commonly made based on a combination of performance measures (eg, in-game scores) and physiological measurements, which provide insight into the player’s psychological state. However, although many prototypes of affective games have been presented and many studies have shown that physiological measurements allow more accurate classification of the player’s psychological state than performance measures, few studies have examined whether dynamic difficulty adjustment (DDA) based on physiological measurements (which requires additional sensors) results in a better user experience than performance-based DDA or manual difficulty adjustment. Objective This study aims to compare five DDA methods in an affective exergame: manual (player-controlled), random, performance-based, personality-performance–based, and physiology-personality-performance–based (all-data). Methods A total of 50 participants (N=50) were divided into five groups, corresponding to the five DDA methods. They played an exergame version of Pong for 18 minutes, starting at a medium difficulty; every 2 minutes, two game difficulty parameters (ball speed and paddle size) were adjusted using the participant’s assigned DDA method. The DDA rules for the performance-based, personality-performance–based, and all-data groups were developed based on data from a previous open-loop study. Seven physiological responses were recorded throughout the sessions, and participants self-reported their preferred changes to difficulty every 2 minutes. After playing the game, participants reported their in-game experience using two questionnaires: the Intrinsic Motivation Inventory and the Flow Experience Measure. Results Although the all-data method resulted in the most accurate changes to ball speed and paddle size (defined as the percentage match between DDA choice and participants’ preference), no significant differences between DDA methods were found on the Intrinsic Motivation Inventory and Flow Experience Measure. When the data from all four automated DDA methods were pooled together, the accuracy of changes in ball speed was significantly correlated with players’ enjoyment (r=0.38) and pressure (r=0.43). Conclusions Although our study is limited by the use of a between-subjects design and may not generalize to other exergame designs, the results do not currently support the inclusion of physiological measurements in affective exergames, as they did not result in an improved user experience. As the accuracy of difficulty changes is correlated with user experience, the results support the development of more effective DDA methods. However, they show that the inclusion of physiological measurements does not guarantee a better user experience even if it yields promising results in offline cross-validation.

Background A mobile app generates passive data, such as GPS data traces, without any direct involvement from the user. These passive data have transformed the manner of traditional assessments that require active participation from the user. Passive data collection is one of the most important core techniques for mobile health development because it may promote user retention, which is a unique characteristic of a software medical device. Objective The primary aim of this study was to quantify user retention for the “Staff Hours” app using survival analysis. The secondary aim was to compare user retention between passive data and active data, as well as factors associated with the survival rates of user retention. Methods We developed an app called “Staff Hours” to automatically calculate users’ work hours through GPS data (passive data). “Staff Hours” not only continuously collects these passive data but also sends an 11-item mental health survey to users monthly (active data). We applied survival analysis to compare user retention in the collection of passive and active data among 342 office workers from the “Staff Hours” database. We also compared user retention on Android and iOS platforms and examined the moderators of user retention. Results A total of 342 volunteers (224 men; mean age 33.8 years, SD 7.0 years) were included in this study. Passive data had higher user retention than active data (P=.011). In addition, user retention for passive data collected via Android devices was higher than that for iOS devices (P=.015). Trainee physicians had higher user retention for the collection of active data than trainees from other occupations, whereas no significant differences between these two groups were observed for the collection of passive data (P=.700). Conclusions Our findings demonstrated that passive data collected via Android devices had the best user retention for this app that records GPS-based work hours.

Background Mobile apps are widely used in health professions, which increases the need for simple methods to determine the quality of apps. In particular, teachers need the ability to curate high-quality mobile apps for student learning. Objective This study aims to systematically search for and evaluate the quality of clinical skills mobile apps as learning tools. The quality of apps meeting the specified criteria was evaluated using two measures—the widely used Mobile App Rating Scale (MARS), which measures general app quality, and the Mobile App Rubric for Learning (MARuL), a recently developed instrument that measures the value of apps for student learning—to assess whether MARuL is more effective than MARS in identifying high-quality apps for learning. Methods Two mobile app stores were systematically searched using clinical skills terms commonly found in medical education and apps meeting the criteria identified using an approach based on PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. A total of 9 apps were identified during the screening process. The apps were rated independently by 2 reviewers using MARS and MARuL. Results The intraclass correlation coefficients (ICCs) for the 2 raters using MARS and MARuL were the same (MARS ICC [two-way]=0.68; P Conclusions This systematic search and rating of clinical skills apps for learning found that the quality of apps was highly variable. However, 2 apps—Geeky Medics-OSCE revision and OSCE PASS: Medical Revision—rated highly for both versions and with both quality measures. MARS and MARuL showed similar abilities to differentiate the quality of the 9 apps. However, MARuL’s incorporation of teaching and learning elements as part of a multidimensional measure of quality may make it more appropriate for use with apps focused on teaching and learning, whereas MARS’s more general rating of quality may be more appropriate for health apps targeting a general health audience. Ratings of the 9 apps by both measures also highlighted the variable quality of clinical skills mobile apps for learning.

BackgroundStress levels among college students have been on the rise for the last few decades. Currently, rates of reported stress among college students are at an all-time high. Traditionally, the dominant way to assess stress levels has been through pen-and-paper surveys. ObjectiveThe aim of this study is to use passive sensing data collected via mobile phones to obtain a rich and potentially less-biased source of data that can be used to help better understand stressors in the college experience. MethodsWe used a mobile sensing app, StudentLife, in tandem with a pictorial mobile phone–based measure of stress, the Mobile Photographic Stress Meter, to investigate the situations and contexts that are more likely to precipitate stress. ResultsUsing recently developed methods for handling high-dimensional longitudinal data, penalized generalized estimating equations, we identified a set of mobile sensing features (absolute values of beta >0.001 and robust z>1.96) across the domains of social activity, movement, location, and ambient noise that were predictive of student stress levels. ConclusionsBy combining recent statistical methods and mobile phone sensing, we have been able to study stressors in the college experience in a way that is more objective, detailed, and less intrusive than past research. Future work can leverage information gained from passive sensing and use that to develop real-time, targeted interventions for students experiencing a stressful time.

Background Workplace bullying has been measured in many studies to investigate its effects on mental health issues. However, none have used web-based computerized adaptive testing (CAT) with bully classifications and convolutional neural networks (CNN) for reporting the extent of individual bullying in the workplace. Objective This study aims to build a model using CNN to develop an app for automatic detection and classification of nurse bullying-levels, incorporated with online Rasch computerized adaptive testing, to help assess nurse bullying at an earlier stage. Methods We recruited 960 nurses working in a Taiwan Ch-Mei hospital group to fill out the 22-item Negative Acts Questionnaire-Revised (NAQ-R) in August 2012. The k-mean and the CNN were used as unsupervised and supervised learnings, respectively, for: (1) dividing nurses into three classes (n=918, 29, and 13 with suspicious mild, moderate, and severe extent of being bullied, respectively); and (2) building a bully prediction model to estimate 69 different parameters. Finally, data were separated into training and testing sets in a proportion of 70:30, where the former was used to predict the latter. We calculated the sensitivity, specificity, and receiver operating characteristic curve (area under the curve [AUC]), along with the accuracy across studies for comparison. An app predicting the respondent bullying-level was developed, involving the model’s 69 estimated parameters and the online Rasch CAT module as a website assessment. Results We observed that: (1) the 22-item model yields higher accuracy rates for three categories, with an accuracy of 94% for the total 960 cases, and accuracies of 99% (AUC 0.99; 95% CI 0.99-1.00) and 83% (AUC 0.94; 95% CI 0.82-0.99) for the lower and upper groups (cutoff points at 49 and 66 points) based on the 947 cases and 42 cases, respectively; and (2) the 700-case training set, with 95% accuracy, predicts the 260-case testing set reaching an accuracy of 97. Thus, a NAQ-R app for nurses that predicts bullying-level was successfully developed and demonstrated in this study. Conclusions The 22-item CNN model, combined with the Rasch online CAT, is recommended for improving the accuracy of the nurse NAQ-R assessment. An app developed for helping nurses self-assess workplace bullying at an early stage is required for application in the future.

BackgroundApproximately 90% of global cervical cancer (CC) is mostly found in low- and middle-income countries. In most cases, CC can be detected early through routine screening programs, including a cytology-based test. However, it is logistically difficult to offer this program in low-resource settings due to limited resources and infrastructure, and few trained experts. A visual inspection following the application of acetic acid (VIA) has been widely promoted and is routinely recommended as a viable form of CC screening in resource-constrained countries. Digital images of the cervix have been acquired during VIA procedure with better quality assurance and visualization, leading to higher diagnostic accuracy and reduction of the variability of detection rate. However, a colposcope is bulky, expensive, electricity-dependent, and needs routine maintenance, and to confirm the grade of abnormality through its images, a specialist must be present. Recently, smartphone-based imaging systems have made a significant impact on the practice of medicine by offering a cost-effective, rapid, and noninvasive method of evaluation. Furthermore, computer-aided analyses, including image processing–based methods and machine learning techniques, have also shown great potential for a high impact on medicinal evaluations.ObjectiveIn this study, we demonstrate a new quantitative CC screening technique and implement a machine learning algorithm for smartphone-based endoscopic VIA. We also evaluated the diagnostic performance and practicability of the approach based on the results compared to the gold standard and from physicians’ interpretation.MethodsA smartphone-based endoscope system was developed and applied to the VIA screening. A total of 20 patients were recruited for this study to evaluate the system. Overall, five were healthy, and 15 were patients who had shown a low to high grade of cervical intraepithelial neoplasia (CIN) from both colposcopy and cytology tests. Endoscopic VIA images were obtained before a loop electrosurgical excision procedure for patients with abnormal tissues, and their histology tissues were collected. Endoscopic VIA images were assessed by four expert physicians relative to the gold standard of histopathology. Also, VIA features were extracted from multiple steps of image processing techniques to find the differences between abnormal (CIN2+) and normal (≤CIN1). By using the extracted features, the performance of different machine learning classifiers, such as k-nearest neighbors (KNN), support vector machine, and decision tree (DT), were compared to find the best algorithm for VIA. After determining the best performing classifying model, it was used to evaluate the screening performance of VIA.ResultsAn average accuracy of 78%, with a Cohen kappa of 0.571, was observed for the evaluation of the system by four physicians. Through image processing, 240 sliced images were obtained from the cervicogram at each clock position, and five features of VIA were extracted. Among the three models, KNN showed the best performance for finding VIA within holdout 10-fold cross-validation, with an accuracy of 78.3%, area under the curve of 0.807, a specificity of 80.3%, and a sensitivity of 75.0%, respectively. The trained model performed using an unprovided data set resulted in an accuracy of 80.8%, specificity of 84.1%, and sensitivity of 71.9%. Predictions were visualized with intuitive color labels, indicating the normal/abnormal tissue using a circular clock-type segmentation. Calculating the overlapped abnormal tissues between the gold standard and predicted value, the KNN model overperformed the average assessments of physicians for finding VIA.ConclusionsWe explored the potential of the smartphone-based endoscopic VIA as an evaluation technique and used the cervicogram to evaluate normal/abnormal tissue using machine learning techniques. The results of this study demonstrate its potential as a screening tool in low-resource settings.

Background: Data collection in Sub-Saharan Africa has traditionally been paper-based. However, the popularization of Android mobile devices and data capture software has brought paperless data management within reach. We used Open Data Kit (ODK) technology on Android mobile devices during a household survey in the Niger Delta region of Nigeria. Objective: The aim of this study was to describe the pros and cons of deploying ODK for data management. Methods: A descriptive cross-sectional household survey was carried out by 6 data collectors between April and May 2016. Data were obtained from 1706 persons in 601 households across 6 communities in 3 states in the Niger Delta. The use of Android mobile devices and ODK technology involved form building, testing, collection, aggregation, and download for data analysis. The median duration for data collection per household and per individual was 25.7 and 9.3 min, respectively. Results: Data entries per device ranged from 33 (33/1706, 1.93%) to 482 (482/1706, 28.25%) individuals between 9 (9/601, 1.5%) and 122 (122/601, 20.3%) households. The most entries (470) were made by data collector 5. Only 2 respondents had data entry errors (2/1706, 0.12%). However, 73 (73/601, 12.1%) households had inaccurate date and time entries for when data collection started and ended. The cost of deploying ODK was estimated at US $206.7 in comparison with the estimated cost of US $466.7 for paper-based data management. Conclusions: We found the use of mobile data capture technology to be efficient and cost-effective. As Internet services improve in Africa, we advocate their use as effective tools for health information management. [JMIR Mhealth Uhealth 2017;5(11):e171]

Communication among patients, colleagues, and staff in healthcare has changed dramatically in the last decade. Digital technology and social media sites have allowed instantaneous access to information. The potential for information technology to improve access to healthcare, enhance the quality, and lower the cost is significant. Text messaging, tweeting, chatting, and blogging are rapidly replacing e-mail as the preferred means of communication in healthcare. This review will highlight how digital technology is changing the way surgeons communicate with colleagues and patients as well as provide some guidance as to how to avoid some of the pitfalls and problems that this form of communication can bring.

Background: Mobile health (mHealth) apps show a growing importance for patients and health care professionals. Apps in this category are diverse. Some display important information (ie, drug interactions), whereas others help patients to keep track of their health. However, insufficient transport security can lead to confidentiality issues for patients and medical professionals, as well as safety issues regarding data integrity. mHealth apps should therefore deploy intensified vigilance to protect their data and integrity. This paper analyzes the state of security in mHealth apps. Objective: The objectives of this study were as follows: (1) identification of relevant transport issues in mHealth apps, (2) development of a platform for test purposes, and (3) recommendation of practices to mitigate them. Methods: Security characteristics relevant to the transport security of mHealth apps were assessed, presented, and discussed. These characteristics were used in the development of a prototypical platform facilitating streamlined tests of apps. For the tests, six lists of the 10 most downloaded free apps from three countries and two stores were selected. As some apps were part of these top 10 lists in more than one country, 53 unique apps were tested. Results: Out of the 53 apps tested from three European App Stores for Android and iOS, 21/53 (40%) showed critical results. All 21 apps failed to guarantee the integrity of data displayed. A total of 18 apps leaked private data or were observable in a way that compromised confidentiality between apps and their servers; 17 apps used unprotected connections; and two apps failed to validate certificates correctly. None of the apps tested utilized certificate pinning. Many apps employed analytics or ad providers, undermining user privacy. Conclusions: The tests show that many mHealth apps do not apply sufficient transport security measures. The most common security issue was the use of any kind of unprotected connection. Some apps used secure connections only for selected tasks, leaving all other traffic vulnerable. [JMIR Mhealth Uhealth 2017;5(10):e147]

Background With a wide range of use cases in both research and clinical domains, collecting continuous mobile health (mHealth) streaming data from multiple sources in a secure, highly scalable, and extensible platform is of high interest to the open source mHealth community. The European Union Innovative Medicines Initiative Remote Assessment of Disease and Relapse-Central Nervous System (RADAR-CNS) program is an exemplary project with the requirements to support the collection of high-resolution data at scale; as such, the Remote Assessment of Disease and Relapse (RADAR)-base platform is designed to meet these needs and additionally facilitate a new generation of mHealth projects in this nascent field. Objective Wide-bandwidth networks, smartphone penetrance, and wearable sensors offer new possibilities for collecting near-real-time high-resolution datasets from large numbers of participants. The aim of this study was to build a platform that would cater for large-scale data collection for remote monitoring initiatives. Key criteria are around scalability, extensibility, security, and privacy. Methods RADAR-base is developed as a modular application; the backend is built on a backbone of the highly successful Confluent/Apache Kafka framework for streaming data. To facilitate scaling and ease of deployment, we use Docker containers to package the components of the platform. RADAR-base provides 2 main mobile apps for data collection, a Passive App and an Active App. Other third-Party Apps and sensors are easily integrated into the platform. Management user interfaces to support data collection and enrolment are also provided. Results General principles of the platform components and design of RADAR-base are presented here, with examples of the types of data currently being collected from devices used in RADAR-CNS projects: Multiple Sclerosis, Epilepsy, and Depression cohorts. Conclusions RADAR-base is a fully functional, remote data collection platform built around Confluent/Apache Kafka and provides off-the-shelf components for projects interested in collecting mHealth datasets at scale.

BackgroundThe use of fitness trackers as tools of self-management to promote physical activity is increasing. However, the content of fitness trackers remains unexplored. ObjectiveThe aim of this study was to use the Behavior Change Technique Taxonomy v1 (BCTTv1) to examine if swim-proof fitness trackers below Aus $150 (US$ 105) incorporate behavior change techniques (BCTs) that relate to self-management strategies to increase physical activity and reduce sedentary behavior and to determine if content of the fitness trackers correspond to physical activity guidelines. MethodsA total of two raters used the BCTTv1 to code 6 fitness trackers that met the inclusion criteria. The inclusion criteria were the ability to track activity, be swim proof, be compatible with Android and Apple operating systems, and cost below Aus $150. ResultsAll fitness trackers contained BCTs known to promote physical activity, with the most frequently used BCTs overlapping with self-management strategies, including goal setting, self-monitoring, and feedback on behavior. Fitbit Flex 2 (Fitbit Inc) contained the most BCTs at 20. Huawei Band 2 Pro (Huawei Technologies) and Misfit Shine 2 (Fossil Group) contained the least BCTs at 11. ConclusionsFitness trackers contain evidence-based BCTs that overlap with self-management strategies, which have been shown to increase physical activity and reduce sedentary behavior. Fitness trackers offer the prospect for physical activity interventions that are cost-effective and easily accessed by a wide population.

BackgroundWith the rise of mobile technology, an increasing number of people use mobile-based social media to access health information. Many scholars have explored the nature of health information on social media; however, the impact of such information on people was understudied. ObjectiveThis study aimed to examine the nature and impact of health information on mobile-based social media. Specifically, we investigated how the levels of threat and efficacy of breast cancer prevention information affect individuals’ engagement with the information, such as readings and likes. MethodsBreast cancer prevention articles posted on a Chinese mobile-based social media platform (ie, WeChat Subscription Account [WeChat SA]) from January 1 to December 31, 2017, were extracted using the Python Web Crawler. We used content analysis and analysis of covariance to analyze our data. ResultsThe results revealed that the vast majority of titles and main bodies of the articles involved one of the extended parallel process model components: threat or efficacy. ConclusionsBreast cancer prevention information on WeChat SA was well designed. Both threat and efficacy significantly affected the number of readings, whereas only efficacy had a significant effect on the number of likes. Moreover, breast cancer prevention information that contained both high levels of threat and efficacy gained the largest number of readings and likes.

Background Medical smartphone apps and mobile health devices are rapidly entering mainstream use because of the rising number of smartphone users. Consequently, a large amount of consumer-generated data is being collected. Technological advances in innovative sensory systems have enabled data connectivity and aggregation to become cornerstones in developing workable solutions for remote monitoring systems in clinical practice. However, few systems are currently available to handle such data, especially for clinical use. Objective The aim of this study was to develop and implement the digital health research platform for mobile health (DHARMA) that combines data saved in different formats from a variety of sources into a single integrated digital platform suitable for mobile remote monitoring studies. Methods DHARMA comprises a smartphone app, a Web-based platform, and custom middleware and has been developed to collect, store, process, and visualize data from different vendor-specific sensors. The middleware is a component-based system with independent building blocks for user authentication, study and patient administration, data handling, questionnaire management, patient files, and reporting. Results A prototype version of the research platform has been tested and deployed in multiple clinical studies. In this study, we used the platform for the follow-up of pregnant women at risk of developing pre-eclampsia. The patients’ blood pressure, weight, and activity were semi-automatically captured at home using different devices. DHARMA automatically collected and stored data from each source and enabled data processing for the end users in terms of study-specific parameters, thresholds, and visualization. Conclusions The increasing use of mobile health apps and connected medical devices is leading to a large amount of data for collection. There has been limited investment in handling and aggregating data from different sources for use in academic and clinical research focusing on remote monitoring studies. In this study, we created a modular mobile health research platform to collect and integrate data from a variety of third-party devices in several patient populations. The functionality of the platform was demonstrated in a real-life setting among women with high-risk pregnancies.

Background: Mental health disorders affect multiple aspects of patients’ lives, including mood, cognition, and behavior. eHealth and mobile health (mHealth) technologies enable rich sets of information to be collected noninvasively, representing a promising opportunity to construct behavioral markers of mental health. Combining such data with self-reported information about psychological symptoms may provide a more comprehensive and contextualized view of a patient’s mental state than questionnaire data alone. However, mobile sensed data are usually noisy and incomplete, with significant amounts of missing observations. Therefore, recognizing the clinical potential of mHealth tools depends critically on developing methods to cope with such data issues. Objective: This study aims to present a machine learning–based approach for emotional state prediction that uses passively collected data from mobile phones and wearable devices and self-reported emotions. The proposed methods must cope with high-dimensional and heterogeneous time-series data with a large percentage of missing observations. Methods: Passively sensed behavior and self-reported emotional state data from a cohort of 943 individuals (outpatients recruited from community clinics) were available for analysis. All patients had at least 30 days’ worth of naturally occurring behavior observations, including information about physical activity, geolocation, sleep, and smartphone app use. These regularly sampled but frequently missing and heterogeneous time series were analyzed with the following probabilistic latent variable models for data averaging and feature extraction: mixture model (MM) and hidden Markov model (HMM). The extracted features were then combined with a classifier to predict emotional state. A variety of classical machine learning methods and recurrent neural networks were compared. Finally, a personalized Bayesian model was proposed to improve performance by considering the individual differences in the data and applying a different classifier bias term for each patient. Results: Probabilistic generative models proved to be good preprocessing and feature extractor tools for data with large percentages of missing observations. Models that took into account the posterior probabilities of the MM and HMM latent states outperformed those that did not by more than 20%, suggesting that the underlying behavioral patterns identified were meaningful for individuals’ overall emotional state. The best performing generalized models achieved a 0.81 area under the curve of the receiver operating characteristic and 0.71 area under the precision-recall curve when predicting self-reported emotional valence from behavior in held-out test data. Moreover, the proposed personalized models demonstrated that accounting for individual differences through a simple hierarchical model can substantially improve emotional state prediction performance without relying on previous days’ data. Conclusions: These findings demonstrate the feasibility of designing machine learning models for predicting emotional states from mobile sensing data capable of dealing with heterogeneous data with large numbers of missing observations. Such models may represent valuable tools for clinicians to monitor patients’ mood states. This project has received funding from the European Union's Horizon 2020 Research and Innovation Program under the Marie Sklodowska-Curie grant agreement number 813533. This work was partly supported by the Spanish government (Ministerio de Ciencia e Innovación) under grants TEC2017-92552-EXP and RTI2018-099655-B-100; the Comunidad de Madrid under grants IND2017/TIC-7618, IND2018/TIC-9649, IND2020/TIC-17372, and Y2018/TCS-4705; the BBVA Foundation under the Domain Alignment and Data Wrangling with Deep Generative Models (Deep-DARWiN) project; and the European Union (European Regional Development Fund and the European Research Council) through the European Union's Horizon 2020 Research and Innovation Program under grant 714161. The authors thank Enrique Baca-Garcia for providing demographic and clinical data and assisting in interpreting and summarizing the data.

Background Emotional state in everyday life is an essential indicator of health and well-being. However, daily assessment of emotional states largely depends on active self-reports, which are often inconvenient and prone to incomplete information. Automated detection of emotional states and transitions on a daily basis could be an effective solution to this problem. However, the relationship between emotional transitions and everyday context remains to be unexplored. Objective This study aims to explore the relationship between contextual information and emotional transitions and states to evaluate the feasibility of detecting emotional transitions and states from daily contextual information using machine learning (ML) techniques. Methods This study was conducted on the data of 18 individuals from a publicly available data set called ExtraSensory. Contextual and sensor data were collected using smartphone and smartwatch sensors in a free-living condition, where the number of days for each person varied from 3 to 9. Sensors included an accelerometer, a gyroscope, a compass, location services, a microphone, a phone state indicator, light, temperature, and a barometer. The users self-reported approximately 49 discrete emotions at different intervals via a smartphone app throughout the data collection period. We mapped the 49 reported discrete emotions to the 3 dimensions of the pleasure, arousal, and dominance model and considered 6 emotional states: discordant, pleased, dissuaded, aroused, submissive, and dominant. We built general and personalized models for detecting emotional transitions and states every 5 min. The transition detection problem is a binary classification problem that detects whether a person’s emotional state has changed over time, whereas state detection is a multiclass classification problem. In both cases, a wide range of supervised ML algorithms were leveraged, in addition to data preprocessing, feature selection, and data imbalance handling techniques. Finally, an assessment was conducted to shed light on the association between everyday context and emotional states. Results This study obtained promising results for emotional state and transition detection. The best area under the receiver operating characteristic (AUROC) curve for emotional state detection reached 60.55% in the general models and an average of 96.33% across personalized models. Despite the highly imbalanced data, the best AUROC curve for emotional transition detection reached 90.5% in the general models and an average of 88.73% across personalized models. In general, feature analyses show that spatiotemporal context, phone state, and motion-related information are the most informative factors for emotional state and transition detection. Our assessment showed that lifestyle has an impact on the predictability of emotion. Conclusions Our results demonstrate a strong association of daily context with emotional states and transitions as well as the feasibility of detecting emotional states and transitions using data from smartphone and smartwatch sensors.

Background SMS text messaging programs are increasingly being used for secondary prevention, and have been shown to be effective in a number of health conditions including cardiovascular disease. SMS text messaging programs have the potential to increase the reach of an intervention, at a reduced cost, to larger numbers of people who may not access traditional programs. However, patients regularly reply to the SMS text messages, leading to additional staffing requirements to monitor and moderate the patients’ SMS text messaging replies. This additional staff requirement directly impacts the cost-effectiveness and scalability of SMS text messaging interventions. Objective This study aimed to test the feasibility and accuracy of developing a machine learning (ML) program to triage SMS text messaging replies (ie, identify which SMS text messaging replies require a health professional review). Methods SMS text messaging replies received from 2 clinical trials were manually coded (1) into “Is staff review required?” (binary response of yes/no); and then (2) into 12 general categories. Five ML models (Naïve Bayes, OneVsRest, Random Forest Decision Trees, Gradient Boosted Trees, and Multilayer Perceptron) and an ensemble model were tested. For each model run, data were randomly allocated into training set (2183/3118, 70.01%) and test set (935/3118, 29.98%). Accuracy for the yes/no classification was calculated using area under the receiver operating characteristics curve (AUC), false positives, and false negatives. Accuracy for classification into 12 categories was compared using multiclass classification evaluators. Results A manual review of 3118 SMS text messaging replies showed that 22.00% (686/3118) required staff review. For determining need for staff review, the Multilayer Perceptron model had highest accuracy (AUC 0.86; 4.85% false negatives; and 4.63% false positives); with addition of heuristics (specified keywords) fewer false negatives were identified (3.19%), with small increase in false positives (7.66%) and AUC 0.79. Application of this model would result in 26.7% of SMS text messaging replies requiring review (true + false positives). The ensemble model produced the lowest false negatives (1.43%) at the expense of higher false positives (16.19%). OneVsRest was the most accurate (72.3%) for the 12-category classification. Conclusions The ML program has high sensitivity for identifying the SMS text messaging replies requiring staff input; however, future research is required to validate the models against larger data sets. Incorporation of an ML program to review SMS text messaging replies could significantly reduce staff workload, as staff would not have to review all incoming SMS text messages. This could lead to substantial improvements in cost-effectiveness, scalability, and capacity of SMS text messaging–based interventions.

BackgroundMobile devices are increasingly used to collect location-based information from individuals about their physical activities, dietary intake, environmental exposures, and mental well-being. Such research, which typically uses wearable devices or mobile phones to track location, benefits from the growing availability of fine-grained data regarding human mobility. However, little is known about the comparative geospatial accuracy of such devices. ObjectiveIn this study, we compared the data quality of location information collected from two mobile devices that determine location in different ways—a global positioning system (GPS) watch and a mobile phone with Google’s Location History feature enabled. MethodsA total of 21 chronically ill participants carried both devices, which generated digital traces of locations, for 28 days. A mobile phone–based brief ecological momentary assessment (EMA) survey asked participants to manually report their location at 4 random times throughout each day. Participants also took part in qualitative interviews and completed surveys twice during the study period in which they reviewed recent mobile phone and watch trace data to compare the devices’ trace data with their memory of their activities on those days. Trace data from the devices were compared on the basis of (1) missing data days, (2) reasons for missing data, (3) distance between the route data collected for matching day and the associated EMA survey locations, and (4) activity space total area and density surfaces. ResultsThe watch resulted in a much higher proportion of missing data days (P

BackgroundPersonalized blood glucose (BG) prediction for diabetes patients is an important goal that is pursued by many researchers worldwide. Despite many proposals, only a few projects are dedicated to the development of complete recommender system infrastructures that incorporate BG prediction algorithms for diabetes patients. The development and implementation of such a system aided by mobile technology is of particular interest to patients with gestational diabetes mellitus (GDM), especially considering the significant importance of quickly achieving adequate BG control for these patients in a short period (ie, during pregnancy) and a typically higher acceptance rate for mobile health (mHealth) solutions for short- to midterm usage. ObjectiveThis study was conducted with the objective of developing infrastructure comprising data processing algorithms, BG prediction models, and an appropriate mobile app for patients’ electronic record management to guide BG prediction-based personalized recommendations for patients with GDM. MethodsA mobile app for electronic diary management was developed along with data exchange and continuous BG signal processing software. Both components were coupled to obtain the necessary data for use in the personalized BG prediction system. Necessary data on meals, BG measurements, and other events were collected via the implemented mobile app and continuous glucose monitoring (CGM) system processing software. These data were used to tune and evaluate the BG prediction model, which included an algorithm for dynamic coefficients tuning. In the clinical study, 62 participants (GDM: n=49; control: n=13) took part in a 1-week monitoring trial during which they used the mobile app to track their meals and self-measurements of BG and CGM system for continuous BG monitoring. The data on 909 food intakes and corresponding postprandial BG curves as well as the set of patients’ characteristics (eg, glycated hemoglobin, body mass index [BMI], age, and lifestyle parameters) were selected as inputs for the BG prediction models. ResultsThe prediction results by the models for BG levels 1 hour after food intake were root mean square error=0.87 mmol/L, mean absolute error=0.69 mmol/L, and mean absolute percentage error=12.8%, which correspond to an adequate prediction accuracy for BG control decisions. ConclusionsThe mobile app for the collection and processing of relevant data, appropriate software for CGM system signals processing, and BG prediction models were developed for a recommender system. The developed system may help improve BG control in patients with GDM; this will be the subject of evaluation in a subsequent study.

Background: The emergence of smartphones, wearable sensor technologies, and smart homes allows the nonintrusive collection of activity data. Thus, health-related events, such as activities of daily living (ADLs; eg, mobility patterns, feeding, sleeping, ...) can be captured without patients' active participation. We designed a system to detect changes in the mobility patterns based on the smartphone's native sensors and advanced machine learning and signal processing techniques. Objective: The principal objective of this work is to assess the feasibility of detecting mobility pattern changes in a sample of outpatients with depression using the smartphone's sensors. The proposed method processed the data acquired by the smartphone using an unsupervised detection technique. Methods: In this study, 38 outpatients from the Hospital Fundacion Jimenez Diaz Psychiatry Department (Madrid, Spain) participated. The Evidence-Based Behavior (eB(2)) app was downloaded by patients on the day of recruitment and configured with the assistance of a physician. The app captured the following data: inertial sensors, physical activity, phone calls and message logs, app usage, nearby Bluetooth and Wi-Fi connections, and location. We applied a change-point detection technique to location data on a sample of 9 outpatients recruited between April 6, 2017 and December 14, 2017. The change-point detection was based only on location information, but the eB(2) platform allowed for an easy integration of additional data. The app remained running in the background on patients' smartphone during the study participation. Results: The principal outcome measure was the identification of mobility pattern changes based on an unsupervised detection technique applied to the smartphone's native sensors data. Here, results from 5 patients' records are presented as a case series. The eB(2) system detected specific mobility pattern changes according to the patients' activity, which may be used as indicators This study was partly supported by Ministerio de Economía of Spain under project: AID (TEC2014-62194-EXP) and aMBITION (TEC2017-92552-EXP), the Ministerio de Economía of Spain jointly with the European Commission under projects ADVENTURE (TEC2015-69868-C2-1-R) and CAIMAN (TEC2017-86921-C2-2-R), the Comunidad de Madrid under project CASI-CAM-CM (S2013/ICE-2845), Fundación Jiménez Díaz Hospital, Instituto de Salud Carlos III (PI16/01852), Delegación del Gobierno para el Plan Nacional de Drogas (20151073), American Foundation for Suicide Prevention (LSRG-1-005-16), the French Embassy in Madrid, Spain, the fondation de l’Avenir, and the Fondation de France. The work of PMM has been supported by FPI grant BES-2016-077626. Publicado