The world’s population is ageing, which puts a strain on healthcare systems, since older age is usually associated to a larger incidence of chronic diseases. An active lifestyle is important for people to maintain good health at any age, but is especially relevant for older people. Since older people, after retiring, typically spend more time at home, often alone, encouraging them to maintain or increase their level of physical activity is a challenge, especially in their domestic environment. To address this challenge, it is imperative to be able to monitor and quantify people’s physical activity during daily life at home. In this context, and in the scope of a project with the industry, we propose Aktiv@Home, a system for unobtrusive monitoring of an older person’s physical activity at home. This system also allows relevant activity-related information to be shared with health professionals or (in)formal caregivers. As a proof-of-concept, a prototype of the system was implemented, which gathers activity-related information from a given person by relying on two radars installed in the environment. An initial validation of the prototype’s capability of tracking the person’s position, while walking, was carried out using radar data collected from five volunteers. The obtained results show that the proposed solution can be used to track a single person and obtain relevant activity information. Moreover, the implemented prototype serves as a good basis for developing a system that monitors older people living alone, which can help enhance their health and well-being, as well as their autonomy, allowing them to stay longer at their own homes.
Complications affecting the lower limbs are considered as the most common debilitating manifestations of Type 2 Diabetes (T2D), accounting for significant preventable morbidity and mortality by predisposing the foot to neuropathy, infections, and ulcerations, which could lead to lower extremity amputations. Effective management of the diabetic foot (DF) is critical to prevent the formation of ulcers, as the disease increases their susceptibility to infection, due to compromised/impaired sweat gland function causing dry, crack-prone skin. This study presents the design and development of novel humidity sensors for diabetic foot care, using Fiber Bragg Grating (FBG) sensors. To determine optimal suitability for insole incorporation, the sensors were integrated into various materials, including Dragon skin 10 and agar/chitosan mixtures at concentrations of 1:1, 1:3, and 3:1, respectively to evaluate sensitivity and responsiveness/recovery characteristics to humidity changes. Experimental findings demonstrated that agar/chitosan mixtures at 1:3 and 1:1 concentration are optimal for monitoring humidity changes due to their rapid response and recovery. This underscores the potential of these sensors to enhance wound management in the DF by improving moisture control, hence aiding in accelerating the healing processes toward the prevention of limb loss.
The anterior cruciate ligament (ACL) is the most commonly injured ligament in the knee and ACL rupture often necessitates surgical intervention and rehabilitation. This pilot study investigated the immediate impact of a single prehabilitation physiotherapy session on gait patterns in a 25-year-old male with a right knee ACL injury. Utilizing inertial measurement units (IMUs) and force plates (FPs), kinematic and kinetic data were collected pre- and post-therapy. The results demonstrated significant improvements in gait symmetry and stability post-therapy, with a marked reduction in knee angle variability. A notable decrease in mediolateral force on the uninjured leg was also observed, indicating enhanced balance and reduced compensatory movements. These findings highlight the potential benefits of incorporating prehabilitation physiotherapy to optimize knee function and gait stability in ACL injury patients.
Ear-EEG sensing is on the horizon for ear-worn devices, such as hearing instruments, and it is anticipated to revolutionize cognitive state monitoring outside the clinical setting, owing to its unobtrusive nature. Like scalp-EEG, ear-EEG is prone to non-neural physiological artifacts, such as ocular, muscular, and cardiac. Although artifacts are typically treated as non-desirable signals intended for rejection, they could serve as a source of information about the individual, thus advancing ear-EEG to ear-ExG. As there is still a scarcity of literature within this area, using the cEEGrid around-the-ear electrode array, we systematically recorded the jaw artifact in a series of experimental scenarios engaging the muscles involved in mastication and speech production. Within the scope of preliminary data analysis, we evaluated the statistical significance of a state-of-art-based feature set in terms of chewingspeaking classification by means of a Wilcoxon rank sum test. The resulting.. -values indicated that the features with.. < 0.05 are typically those expressing the chewing rate and the increase in amplitude, as well as a shift towards higher frequencies, that takes place within the characteristic chewing "bursts." By fully utilizing the research potential of the obtained dataset and with the prospect of ear-ExG integration into hearing devices, this work aspires to promote healthy ageing and enhance accessibility in technology through novel human-computer interfaces.
With the increasing interest of several accessibility researchers in the development of technological tools to better support people with intellectual disabilities in their day-to-day lives, it is becoming more and more important to understand the point of view of those directly involved. Although the number of studies conducted with the participation of users with intellectual disabilities has increased in recent years, there are still many questions we need to ask, starting from the preferences and challenges experienced by this group of individuals. For this reason, the present exploratory study aims to bring to light, through semi-structured interviews, the preferences and difficulties that two target groups - clients and staff members from a support centre in Italy - currently have regarding the use and utility of technology within their daily life. In this work, we describe the methodologies and insights emerging from the thematic analysis of the collected data from the interviews, enabling us to better understand and inform the design of more accessible technology for all.
An inclusive society actively seeks the equitable and respectful participation of all its members, regardless of their differences. This concept goes beyond tolerating diversity; it involves valuing and respecting each individual, ensuring everyone has equal access to information and opportunities, and actively participating in all aspects of life. To have a full inclusive society, we have to measure and monitor our impact in the environment, specifically in the oceans and marine life. This paper addresses this challenge by proposing a framework that leverages data aggregation and advanced machine learning techniques for Fine-Grained Visual Classification of marine species. Our methodology employs the Swin Transformer architecture, enhanced with the Fine-Grained Visual Classification Plug-in Module, to process and classify diverse marine datasets. We aggregated multiple marine datasets, preprocessed them to eliminate invalid entries, and trained our model on the refined dataset. Our findings demonstrate that dataset aggregation significantly enhances model accuracy and robustness, especially for large-scale models. Notably, the aggregated data model achieved 94.75% overall accuracy on a dataset comprising 2,548 classes and 391,374 images, compared to 85.93% on individual datasets like WildFish++.
In a world where innovations are made daily, how people interact with and use technology is becoming increasingly important. It is relevant to investigate how new technologies are used and accepted in this environment. Acceptance assessment has been done in research on creating information systems, considering the UTAUT model and the data analysis technique PLS-SEM. Although this digital environment is pertinent to society in general, it is even more appropriate when interacting with tourists because it may give personalised goods and services. In order to assess acceptance in terms of guest insights proportioned by technologies to recommend personalised services and products by evaluating an application that recommends products and services personalised, considering guests' intelligence, this article analyses the use and acceptance of a technological application whose characteristics meet the aforementioned.
Serious Games (SGs) have the potential to provide clinical care and enhance patients’ quality of life, while incorporating an element of entertainment. As part of the iPROLEPSIS Horizon Europe project, we introduce two Sensorimotor Art games designed as SGs to assist Psoriatic Arthritis (PsA) patients in managing their symptoms. These SGs provide a platform for self-expression and pain relief through engaging in rhythmic puzzle activities. The co-design process for the proposed Sensorimotor Art games followed an agile methodology, involving 14 experts (including clinicians, researchers, and game developers), to gather feedback on game requirements, storyboards, and mechanics. From the thematic analysis of the transcribed discussion, four main themes emerged, namely: Clinical value (Theme 1), Motor skills and adaptation (Theme 2), Creative engagement (Theme 3), and Feedback and future directions (Theme 4). Clinical value underlines the therapeutic benefits of Sensorimotor Art games for pain distraction and emotional well-being. Motor skills and adaptation focus on hand/finger use and adapting games for older patients. Creative engagement emphasizes fostering creativity, a positive environment, and goal achievement. Finally, Feedback and future directions highlight the importance of continuous feedback and occupational therapists’ involvement in coming sessions. The development of Sensorimotor Art games lays the groundwork for digital interventions to alleviate psychological distress and improve fine motor symptoms in PsA patients. At the same time, these SGs can also provide insights to healthcare providers and policymakers for developing future digital solutions tailored to PsA patients.
We present a novel solution for automatic task allocation in multi-device environments, where configured robots compete for task assignment when announcing tasks, minimizing manual intervention. To this end, we propose the specification of a task assignment system and a task-oriented programming method aimed at automating processes and optimizing resource utilization in multiple controller environments. The proposed solution with its market-based algorithm and developed architecture improves the adaptability, scalability and overall efficiency of the system. The research discussion extends to broader implications that are consistent with the overall goal of improving robot capabilities in various deployment scenarios.
Emotions play a crucial role in shaping human behavior and interactions in various contexts. While emotions can enhance an individual's effectiveness, they can also hinder performance. However, our understanding and prediction of human behavior are often limited to individuals' own perceptions of their thoughts and emotions. The development of technology, specifically automatic facial recognition of emotions (FER), has the potential to advance our ability to predict behavior. This technology allows researchers to objectively measure an individual's internal emotional state. To ensure its effectiveness, standardization in protocols across scientific fields is necessary. This involves creating and analyzing diverse databases to train machine learning models more effectively and improving the algorithms used. The objective of this literature review is to explore the applications of FER technology and evaluate the effectiveness of its results in recent research. Although the results have been satisfactory, researchers have used various data collection methods, which may complicate replication in future studies. To ensure reproducibility, it is important to develop an experimental protocol that clearly outlines the data collection process. Additionally, employing the self-report method can strengthen the findings obtained from the software. Considering the unique characteristics of individuals, it is advisable to establish predetermined values before analyzing and generalizing the results obtained from the software.
Autism Spectrum Disorder (ASD) affects individuals in diverse ways, making personalized therapeutic approaches crucial. In this context, we propose a personalized mobile application designed for music-based therapy tailored to people with ASD. This adaptive piano app can be customized to suit the individual abilities of each user. The paper is structured as follows: The introduction provides context on autism and the importance of personalized therapy. The background section reviews related studies on music-based therapy. The methodology section introduces "Professor Piano," our adaptive and adaptable music therapy application. The results and discussion section explores the challenges encountered during development and presents the findings from a heuristic evaluation conducted by experts. Finally, the conclusion summarizes the main insights and implications of the study.
Investigating the effect of Tibetan Singing Bowls (TSB) meditation on physiological features including heart rate is limited. This study focused on applying multiscale modified diffusion entropy analysis (MSMDEA) on heart-rate fluctuations exhibiting 1/f noise. MSMDEA identifies the contribution of crucial events within a 1/f-noise time series that differs from the well-known Fractional Brownian Motion (FBM) 1/f-noise. Crucial events (`) are defined as the waiting time distribution characterized by an inverse power law probability density function with the temporal complexity coefficient ` at 2< ` <3 computed from the MSMDEA complexity scaling parameter X. Eight healthy participants underwent a 50-minute session of TSB meditation with their heart rate signals measured using a Polar Ignite wristwatch. MSMDEA was applied across 20time scales on the signals from the first ten minutes of the session as baseline and compared to the TSB meditation of 40 minutes. Correlation rate analysis was also performed, showing increased synchronization properties between subjects associated with TSB compared to baseline. Results indicate a shift to higher complexity for all scales in the TSB meditation group (p<0.05), suggesting that TSB leads to an increase in 1/f-noise associated with crucial events and has a strong influence on the sympathovagal balance associated with 1/f-noise and consequently on improving stress and anxiety. Results also show the superiority of complexity analysis using crucial events in determining and characterizing the differences in HR signals before and during meditative states compared to a traditional HRV analysis using the standard time and frequency HRV metrics.
This integrative review assesses the influence of motorcycle ergonomics on discomfort and riding posture, as well as the factors that should be studied for improved human-motorcycle interaction. The PICO search strategy was employed, following the guiding question "How can ergonomics influence motorcyclists' discomfort and riding posture?", and limiting the search to the last 10 years, in order to ensure the analysis of updated data. The eligibility assessment of the articles followed the PRISMA protocol. After a comprehensive review, a total of nine articles were included in the analysis. When examining the body regions where participants reported experiencing the greatest discomfort while riding, a notable consistency emerged across the various studies. These regions included the neck, shoulders, back, hands and wrists, and buttocks. The available results showed that several ergonomic factors, such as body posture, are influenced by the anthropometric characteristics of the rider, exposure to vibration, and the design of the motorcycle itself. These factors collectively contribute to levels of discomfort and musculoskeletal disorders. It was determined that ergonomic adaptations to the motorcycle are necessary, with the initial modifications being relatively minor, such as a lumbar support. In the idealized ergonomic design, the motorcycle is shaped to fit each individual, thereby providing a more pleasant and healthy ride.
This paper presents a novel approach to generating dyslexia-friendly text using neural text generation techniques. We propose a framework that leverages transformer-based language models, specifically GPT and T5, and incorporates syllable and morphological analysis to enhance the readability and comprehension of text for dyslexic readers. Our approach involves fine-tuning the language models on a curated dataset of dyslexia-friendly text, validated through human assessments and feedback from individuals with dyslexia. We conduct a two-phase experiment with 14 undergraduate students with dyslexia to evaluate the effectiveness of our generated text. The results demonstrate improvements in reading time for participants presented with the refined dyslexia-friendly passages, while also highlighting the importance of individual preferences and text engagement. Furthermore, we provide insights into the specific challenges faced by dyslexic readers and propose targeted approaches to address these issues. This research contributes to the advancement of text accessibility by automating the process of converting standard text into dyslexia-friendly formats. The insights gained from this study inform the design of dyslexia-friendly materials and emphasize the importance of a holistic approach to text accessibility. Our framework has the potential to increase the availability of dyslexia-friendly content and support individuals with dyslexia in accessing written information.
The ageing population in developed countries poses significant challenges for both the healthcare systems and the well-being of older people themselves. A key concern is maintaining independence and quality of life. To address these challenges, we propose a Virtual Assistant that provides support in the context of the kitchen, focusing on step-by-step cooking guidance and pantry management, aiming at mitigating the challenges related to food preparation, allowing older people to maintain their autonomy and independence. Adopting a user-centred design approach, we conducted a first development iteration involving an initial user survey, definition of representative Personas and usage Scenarios, extraction of system requirements and definition of the system architecture, implementation of a first prototype, and evaluation with target end-users. The results of the evaluation showed there is still room for improvements, with the participants reporting some difficulty in carrying out the defined tasks (mean score of 3 out of 5 for two out three tasks) and poor usability (although very close to OK mean score of 51 in the System Usability Scale). Nonetheless, the prototype serves as a proof-of-concept, being a good basis for the next development iterations. The obtained results further provide very valuable and useful information, not only regarding specific difficulties (e.g., in initiating the assistant), but also on the most appreciated features (e.g., recipe guidance) and suggestions for future additions (e.g., having a timer when following a recipe). This feedback will help enhancing the envisioned solution, which has the potential of empowering older individuals to prepare meals and manage their food supplies independently and, consequently, improving their overall well-being.
This research investigates the use of artificial intelligence algorithms to identify behavioural patterns in computer use, with the aim of detecting trends that help to flag cases of depression by analysing the human-computer interaction records of these users, thereby increasing the quality of the data for early detection of these situations. Following design science methodology, a case study will be conducted using an existing mental health screening questionnaire, integrating an artificial intelligence layer to map mouse and keyboard interactions, followed by machine learning analysis of the records. The results of the machine learning assisted questionnaires will be compared with the results of the questionnaires without the mapping. If there is a significant difference, this model could be useful for making predictions about emotional states, contributing to the field of artificial intelligence and helping to prevent depression, which is the focus of the research, although the aim is to look at mental health in a global way.
The Internet of Things (IoT) has been widely implemented for objects, uniquely identified, to become accessible through the internet. Several communication protocol technologies were studied and applied to interconnect objects using the internet. Nowadays, one of the most used is Low Power Wide Area Network (LPWAN) implemented over Narrow Band-Internet of Things (NB-IoT) or Long Range Wide Area Network (LoRaWAN) platforms. In this paper, a LoRaWAN architecture and infrastructure implementation is addressed to secure data and communications protecting Network Servers and communication between gateways and the demilitarized zone (DMZ), using several secure techniques and infrastructure virtualization software for containers.
Inclusion encompasses the creation of accessible and inclusive environments for individuals with disabilities. This research focuses on identifying the challenges encountered by individuals with visual impairments when utilizing e-government web services on mobile devices. The primary aim is to gain comprehensive insights into prevailing practices and assistive technologies employed, while shedding light on the issues confronted by visually impaired users during their interaction with the Kuwait e-government portal. The research methodology encompasses usability testing with participants having visual impairments, employing the System Usability Scale, automatic accessibility evaluation tool assessments, and expert accessibility evaluations. The findings indicate the presence of multiple accessibility barriers on the Kuwaiti e-government website using a mobile, impeding visually impaired users from effectively accessing its services while employing screen reader technology. Issues identified include the absence of alternative text and insufficient color contrast. Additionally, this research presents a compilation of challenges and limitations faced by Arabic-speaking visual impairment users when utilizing the Kuwait e-government portal on mobile devices. Furthermore, a set of recommendations and best practices is provided for government webmasters, designers, developers, educational institutions, and policymakers, aimed at improving the usability and accessibility of e-government portals.
In today’s digital age, video content is prevalent, serving as a primary source of information, education, and entertainment. However, the Deaf and Hard of Hearing (DHH) community often faces significant challenges in accessing video content due to the inadequacy of automatic speech recognition (ASR) systems in providing accurate and reliable captions. This paper addresses the urgent need to improve video caption quality by leveraging Large Language Models (LLMs). We present a comprehensive study that explores the integration of LLMs to enhance the accuracy and context-awareness of captions generated by ASR systems. Our methodology involves a novel pipeline that corrects ASR-generated captions using advanced LLMs. It explicitly focuses on models like GPT-3.5 and Llama2-13B due to their robust performance in language comprehension and generation tasks. We introduce a dataset representative of real-world challenges the DHH community faces to evaluate our proposed pipeline. Our results indicate that LLM-enhanced captions significantly improve accuracy, as evidenced by a notably lower Word Error Rate (WER) achieved by ChatGPT-3.5 (WER: 9.75%) compared to the original ASR captions (WER: 23.07%), ChatGPT-3.5 shows an approximate 57.72% improvement in WER compared to the original ASR captions.
Technology-assisted learning has become integral to the educational sector due to its versatility. Despite the availability of numerous technologies, there remains a significant gap in sign language learning tools utilized for foreign or second language learning for hearing children. Additionally, preferred learning style, pivotal for technology-assisted learning, often goes unnoticed. This research aims to explore the feasibility of delivering sign language education to elementary schoolchildren through technological platforms. To this end, a web-based sign language learning platform was planned, developed, and tested with 28 schoolchildren between the ages of 6 and 7 from Finland and India. Subsequently, the Perceptual Learning Style Questionnaire (PLSPQ) was utilized to assess participants’ preferred learning styles. The analysis from the user studies of two distinctive cultural contexts was useful in identifying the effects of cultural factors on the perceptions of sign language and learning preferences. The results indicate high level of engagement and superior learning outcomes among participants from the studies, with a notable preference for collaborative and teacher-instructed activities in technology-assisted learning environments. Thus, effectiveness of facilitating technology-assisted sign language education for young learners could be established.
