RehbeCa is a platform for mobile cervical rehabilitation that emerged with the goal to provide a treatment of nonspecific neck pain at home (or elsewhere). The platform includes a mobile application addressed to patients with an exergame that integrates a head-tracker interface, which allows the monitoring and analysis of the fulfilment of the neck therapeutic exercises prescribed by physiotherapists. Among all the potential patients of the mobile application, there are elder patients. As the chronological age has been considered as the main barrier to access digital technology, the evaluation of the developed mobile application should consider the group of the elder. We present a user study with 36 participants to evaluate the usability of the mobile application, including 11 elder participants. The results are presented in terms of the System Usability Scale (SUS).

Self-efficacy (SE) is important for task motivation in rehabilitation patients. We propose a method for measuring SE using anticipatory gaze. Experiments were conducted with on-screen tasks with multiple difficulty levels, and gaze and manipulation data were collected. The four anticipatory gaze-related features and the subjective SE were compared, and the highest correlation coefficient between the gaze feature and SE was 0.573. We also construct a regression model using machine-learning methods and verified results. The mean absolute error of the model was 11.1, suggesting the possibility of measuring SE using gaze information. We plan to apply this method to real-world rehabilitation.

This study provides a comprehensive exploration of deep learning architectures for human activity recognition (HAR), focusing on hybrid models (leveraging Convolutional Neural Networks (CNN) with Long- and Short-Term Memory (LSTM)) and a range of alternative deep learning framework. The main goal is to evaluate the performance and effectiveness of the hybrid CNN-LSTM model compared to independent models such as Gated Recurrent Units (GRU), Recurrent Neural Networks (RNN), and traditional CNN architectures. By examining multiple models, this study aims to elucidate the advantages and disadvantages of each approach to accurately identify and classify human activities. The study examines the nuanced capabilities of each model, exploring their respective abilities to capture the spatial and temporal dependencies inherent in activity data. Through rigorous comparative analysis, this study provides insights into the effectiveness of hybrid CNN-LSTM models compared to other popular deep learning architectures, paving the way for advancements in HAR systems.

The UN Convention advocates for disabled individuals' rights to live independently and be included in the community. Despite global commitments, barriers like inaccessible housing and insufficient support prevail. Italy's Puglia region addresses these through the Pro.V.I. initiative (Progetto Vita Indipendente), marking progress towards transforming rights into tangible life improvements.

This study evaluates Pro.V.I.'s impact on persons with disabilities in Puglia using the Individually Prioritised Problem Assessment (IPPA) to analyze data from 40 beneficiaries before and after the initiative. Findings indicate significant enhancements in autonomy, social participation, and overall quality of life. Assistive technologies and personalized support boosted self-determination and societal integration, while advances in home autonomy and technology access were crucial in realizing personal goals.

Results underscore a strong correlation between regional efforts for independent living and life quality improvements for people with disabilities. This case study contributes to discussions on social policies supporting autonomy, highlighting the wider implications for enhancing global independent living conditions for the disabled. Pro.V.I's success demonstrates how collective commitment can overcome barriers, setting a precedent for future disability rights and social inclusion advancements.