This work addresses the gaps in research and management of rare diseases in the Community of Portuguese-Speaking Countries. There is a lack of studies assessing rare disease-related digital and informational readiness, especially in African countries. International cooperation can lead to significant technological advances by measuring community members' digital maturity in managing health data, information systems, and medical technology. The main objectives include creating a collaborative network for rare diseases in the community, improving the registration and monitoring of patients, and promoting continuing education actions. From a scientific, technological, and innovation point of view, the project seeks to map the scenario using digital health tools, create strategies to establish and strengthen cooperation networks, define formative second opinion processes, and promote equitable access to digital health technologies to support decision-making. These actions will help to identify and fill gaps in governance processes, documentation, and practices, leading to realistic and equitable recommendations for better disease management in challenging contexts. Therefore, the cross-border collaboration is expected to promote a unified approach through the transnational sharing of scientific and clinical evidence, expanding the cooperation between team members and partner institutions.

Type 2 diabetic patients benefit significantly if the disease is well controlled through behavioral changes, namely adopting a healthy lifestyle. Currently, there is some evidence that technological strategies can help patient self-management. However, few studies have specifically targeted individuals who solely engage in automatic and personalized self-management practices. This study aims to synthesize the literature regarding personalized feedback recommendation systems to promote behavior change without health professionals’ direct intervention for the management of type 2 diabetes and to verify if the use of these systems improves health-related outcomes. A systematic review was performed from inception to April 13, 2021, based on a search conducted in six databases. According to the defined search expression, studies addressing type 2 diabetic patients and recommendation systems were included. In total, 2186 papers were initially identified, but only 22 met the specific inclusion criteria after screening. Discrepancies in the selection of studies were discussed in consensus meetings. To assess the quality of the articles, two tools were employed according to the types of articles retrieved. Selected papers were summarized regarding specific characteristics such as clinical and technological outcomes. Studies incorporating a recommendation system into their technological solution showed a positive effect on the evaluated outcomes, except for those with longer duration, where the effect was not statistically significant. Although most studies did not report the type of system used, expert systems (rule-based) were found to be the most prevalent among those that did. As behaviors are often difficult to change quickly, it is recommended that future studies extend their follow-up timeframe. Nevertheless, the data obtained suggest that technological solutions incorporating a recommendation system show potential to improve health-related outcomes and demonstrated good usability.

Type 2 diabetes is an increasingly prevalent disease and patients do not always manage the disease properly. Therefore, creating tools that help diabetics self-manage their condition over time is of the utmost importance. Technological tools that include features meeting this population's needs, with an integrated personalized feedback system, in an environment of gamified incentives, may be the way for developing a sustained app's usage. This work aims to present a still in development, novel approach to managing diabetes type 2 with a focus on data analysis from user-inputs, gamification, and personalized coaching, with an accessible user interface. Thus, we will present some functionalities, in particular the theoretical and practical concept that is behind this development, namely the Transtheoretical model of behavior change to create different profiles, to provide customized feedback according to the user's behavioral stage, and also to ascertain if it induces any improvement at the behavioral level.

BACKGROUND: Unexpected variability across healthcare datasets may indicate data quality issues and thereby affect the credibility of these data for reutilization. No gold-standard reference dataset or methods for variability assessment are usually available for these datasets. In this study, we aim to describe the process of discovering data quality implications by applying a set of methods for assessing variability between sources and over time in a large hospital database. METHODS: We described and applied a set of multisource and temporal variability assessment methods in a large Portuguese hospitalization database, in which variation in condition-specific hospitalization ratios derived from clinically coded data were assessed between hospitals (sources) and over time. We identified condition-specific admissions using the Clinical Classification Software (CCS), developed by the Agency of Health Care Research and Quality. A Statistical Process Control (SPC) approach based on funnel plots of condition-specific standardized hospitalization ratios (SHR) was used to assess multisource variability, whereas temporal heat maps and Information-Geometric Temporal (IGT) plots were used to assess temporal variability by displaying temporal abrupt changes in data distributions. Results were presented for the 15 most common inpatient conditions (CCS) in Portugal. MAIN FINDINGS: Funnel plot assessment allowed the detection of several outlying hospitals whose SHRs were much lower or higher than expected. Adjusting SHR for hospital characteristics, beyond age and sex, considerably affected the degree of multisource variability for most diseases. Overall, probability distributions changed over time for most diseases, although heterogeneously. Abrupt temporal changes in data distributions for acute myocardial infarction and congestive heart failure coincided with the periods comprising the transition to the International Classification of Diseases, 10th revision, Clinical Modification, whereas changes in the Diagnosis-Related Groups software seem to have driven changes in data distributions for both acute myocardial infarction and liveborn admissions. The analysis of heat maps also allowed the detection of several discontinuities at hospital level over time, in some cases also coinciding with the aforementioned factors. CONCLUSIONS: This paper described the successful application of a set of reproducible, generalizable and systematic methods for variability assessment, including visualization tools that can be useful for detecting abnormal patterns in healthcare data, also addressing some limitations of common approaches. The presented method for multisource variability assessment is based on SPC, which is an advantage considering the lack of gold standard for such process. Properly controlling for hospital characteristics and differences in case-mix for estimating SHR is critical for isolating data quality-related variability among data sources. The use of IGT plots provides an advantage over common methods for temporal variability assessment due its suitability for multitype and multimodal data, which are common characteristics of healthcare data. The novelty of this work is the use of a set of methods to discover new data quality insights in healthcare data.

Expert systems are designed to address a broad range of problems in a wide variety of areas. Different methodologies have been applied and different tools have been developed to support the creation of expert systems for specific domains, both for traditional application areas and also for emergent topics. In this special issue, a series of articles covering some of the sub-areas of expert systems supported by computer systems are presented. Specifically, this special issue comprises five research papers addressing topics such as authoring tools for multisensory virtual reality applications, computer-interpretable guidelines (CIGs), analysis of user interfaces, intelligent tutoring systems, and a sentiment analysis approach for authorship identification. These manuscripts are extended versions of selected papers from WorldCIST'18, the 6th World Conference on Information Systems and Technologies, held in Naples, Italy, in 2018. This conference was an important forum for researchers and professionals to exchange ideas and experiences involving research problems and corporate solutions related to information systems and technologies in different contexts. Coelho, Melo, Barbosa, Martins, Teixeira, and Bessa, “Authoring Tools for Creating 360 Multisensory Videos - Evaluation of Different Interfaces” (EXSY12418). In this paper, the authors propose an authoring tool for multisensory virtual reality applications with three different authoring interfaces (desktop, immersive, and tangible interface) for the creation of multisensory 360 videos, with the advantage of having a live preview of the multisensory content that is being produced. An evaluation of the three authoring interfaces is presented, having into account gender, system usability, presence, satisfaction, and effectiveness, that is, time to accomplish tasks, number of errors, and number of help requests. Results point out that immersive and tangible interfaces have higher levels of satisfaction than desktop interface as it allows more freedom of interaction, while desktop interface has the lowest time to accomplish the tasks due to user's familiarity with keyboard and mouse. Silva, Oliveira, Gonçalves, and Novais, “Enhancing Decision Making By Providing A Unified System For CIG Management” (EXSY12412). The need to integrate clinical practice guidelines into daily clinical practice requires computer systems capable of operationalizing their knowledge and providing an improved experience in their performance. This paper presents a comprehensive architecture for the implementation of CIGs, including components that allow the creation and manipulation of knowledge elements of clinical practice guidelines, the execution of CIGs with the temporal verification of clinical tasks and the identification and drug conflict resolution. The approach provides a step-by-step assistant for healthcare professionals in the form of an agenda that detects drug interactions when they are prescribed simultaneously and applies a mitigation algorithm to select possible, conflict-free alternatives. Pastushenko, Hynek, and Hruška, “Evaluation of User Interface Design Metrics using Generator of Realistic-Looking Dashboard Samples” (EXSY12434). The analysis of user interfaces using quantitative metrics is a direct way to quickly measure the usability of the interface and various other design aspects such as the adequacy of the page layout or the selected colours. However, the development and evaluation of objective metrics corresponding to the user's perception generally requires a sufficiently large training set of user interface samples. This paper describes the problem of generating the realistic-looking interfaces used for the design and evaluation of quantitative user interface metrics and guidelines. Authors describe a workflow for preparing such samples and demonstrates the applicability of the generator in the production of dashboard samples that are used to evaluate existing metrics of interface aesthetics, showing the possibility of their improvement. Jiménez, Juárez-Ramírez, Castillo, Ramírez-Noriega, and Quezada, “The use of a Weighted Affective Lexicon in a Tutoring System to Improve Student Motivation to Learn” (EXSY12483). This study analyses the use of a weighted affective lexicon in tutoring systems to improve students' motivation. Several educational systems have integrated text-based affective feedback, but analyses of the educational lexicon from the perspective of affectivity are uncommon. Providing affective support in these systems is important because it can improve the student's motivation to learn. Particularly, this study constructed and evaluated an educational lexicon in Spanish. An affective lexicon was integrated in a tutoring system and a review of the impacts of the system on the motivation of its potential users was presented. Authors argue that results of the study will support the development and adoption of intelligent tutoring systems, with benefit for prospective users. Martins, Almeida, Henriques, and Novais, “A sentiment analysis approach to increase authorship identification” (EXSY12469). This paper presents a process to analyse the emotion contained in social media messages (such as Facebook) to identify the author's emotional profile and use it to improve the ability to predict the author of the message. Writing style is considered the way in which an author expresses his thoughts, influenced by the characteristics of the language, period, school or nation, and often writing style can identify the author. Nowadays, the discussion on authorship identification is more relevant due to the considerable amount of fake news spread in social media, in which it is difficult to identify who is the author of a text and even a simple quote can impact an author's public image, particularly when these texts or quotes are from politicians. In this study authors used pre-processing techniques, lexicon-based approaches, and machine learning to achieve a good authorship identification when considering the emotional profile on the writing style. The approach led to an increased ability to identify the author of a text by considering only the author's emotional profile, detected previously from prior texts.

Process mining can be used to extract healthcare processes related information from event logs by performing analysis exploiting the information recorded in it. We report a process mining analysis made to an event log containing traces on user activity recorded by a Virtual Electronic Patient Record (VEPR) system of a Central Hospital. A set of technical analyses were performed. Results from the discovery and characterization of global behavior and from a time series analysis on observed user tasks are reported. Process mining was applied successfully to discover, characterize and analyze user behavior recorded from VEPR. Worth noting the execution of tasks profile observed after log out, revealing significant security problems.