(1) Background: Throughout the history of medical and psychology practice, specialists have worked to improve the quality of treatment and rehabilitation, which has led to the emergence of concepts such as serious games. These tools focus on different areas of intervention procedures, one of which is to help people with intellectual disability (ID). Individuals with ID have problems with executive functions (EFs), which are related to adaptive functioning. Recent studies showed that serious games positively impact cognitive, social, and communication skills in people with ID. The purpose of this study is to analyze the solutions that have been found in EF training for adults with ID in recent years, evaluating them with a number of key parameters and identifying the features and possible problems in the further development of our system. (2) Methods: A review was conducted starting with 573 articles in English related to serious games and selected from studies that had been published since 2015. Finally, 10 were examined in detail as they focused on EFs in adults with ID. They were searched in seven major databases (“Association for Computing Machinery” (ACM), IEEE Xplore database, DBLP computer science bibliography, Google Scholar, PubMed, SCOPUS, and PsycInfo). (3) Results: It was determined that the most frequent EFs referred to in the studies analyzed were planning and decision-making, followed by working memory and social cognition, behavioral regulation, flexibility, and inhibition capacity. The basic approach to the creation of support systems was also analyzed in terms of technical and program execution. The trend results’ analysis evidenced improvements in EFs, even though they were not significant. This comprehensive technique enabled the identification of the main features and aspects to be taken into account for further development of our system.

This article proposes an example of a multiplatform interactive serious game, which is an additional tool and assistant used in the rehabilitation of patients with musculoskeletal system problems. In medicine, any actions and procedures aimed at helping the rehabilitation of patients should entail the most comfortable, but at the same time, effective approach. Regardless of how these actions are orientated, whether for rehabilitation following surgery, fractures, any problems with the musculoskeletal system, or just support for the elderly, rehabilitation methods undoubtedly have good goals, although often the process itself can cause all kinds of discomfort and aversion among patients. This paper presents an interactive platform which enables a slightly different approach to be applied in terms of routine rehabilitation activities and this will help make the process more exciting. The main feature of the system is that it works in several ways: for normal everyday use at home, or for more in-depth observation of various biological parameters, such as heart rate, temperature, and so on. The basic component of the system is the real-time tracking system of the body position, which constitutes both a way to control the game (controller) and a means to analyze the player’s activity. As for the closer control of rehabilitation, the platform also provides the opportunity for medical personnel to monitor the player in real time, with all the data obtained from the game being used for subsequent analysis and comparison. Following several laboratory tests and feedback analysis, the progress indicators are quite encouraging in terms of greater patient interest in this kind of interaction, and effectiveness of the developed platform is also on average about 30–50% compared to conventional exercises, which makes it more attractive in terms of patient support.

This paper presents a comparably simpler yet effective deep learning approach for sentiment analysis of Twitter topics. We automatically collected positive and negative tweets and labeled them manually, and thus created a new dataset. We then leveraged BiGRU model with an ensemble approach for the binary classification of tweets. Our finalized BiGRU model offered an accuracy of 84.8% as well as an averaged F1-measure of 84.8%(±0.3). Moreover, the ensemble approach, using an averaged prediction of 5-fold strategy, provided the accuracy of 86.3% along with the averaged F1-measure of 86.3%(±0.05). Consequently, the ensemble approach offered better performance even on a smaller dataset used in this study.