Jose A. Gonzalez, Lam A. Cheah, Phil D. Green, James M. Gilbert, Stephen R. Ell, Roger K. Moore, Ed Holdsworth
314 - 321
By speech articulator movement and training a transformation to audio we can restore the power of speech to someone who has lost their larynx. We sense changes in magnetic field caused by movements of small magnets attached to the lips and tongue. The sensor transformation uses recurrent neural networks.
Stuart Cunningham, Phil Green, Heidi Christensen, José Joaquín Atria, André Coy, Massimiliano Malavasi, Lorenzo Desideri, Frank Rudzicz
322 - 329
The CloudCAST platform provides a series of speech recognition services that can be integrated into assistive technology applications. The platform and the services provided by the public API are described. Several exemplar applications have been developed to demonstrate the platform to potential developers and users.
Speech impairment like dysarthria poses a major risk to participation in society due to reduced speech intelligibility. This paper focuses on the ongoing development of the technology-based pilot training system, ISi-Speech, for treatment of dysarthria incorporating automatic speech recognition, a psychological model of motivation and evidence-based exercises.
The paper discusses the collection and analysis of a three-month language sample from an adult using a low-tech augmentative and alternative communication (AAC) system transcribed to analyse the frequency of words and letters to make suggestions for more efficiency.
This article proposes a natural language-based approach to accessibility of charts. Formal underpinnings are used to semantically annotate the constituent elements of a vector graphic to support accessing and modifying the content by natural language queries in a web interface. A prototype which lets blind users perform a number of analytical activities on bar charts is introduced.
Simon Judge, Nicola Randall, Yvonne Lynch, Stuart Meredith, Liz Moulam, Janice Murray, Juliet Goldbart
347 - 350
A systematic review of the language and communication characteristics of communication aids considered in identifying the appropriate aid for a child is introduced. The aim is to improve the decision-making around the provision of symbol communication aids to children.
The use of Assistive Technology with people with profound and multiple disabilities is a specialised subfield. Drawing on its authors' experiences and observations at a UK charitable foundation that offers specialised educational, residential and therapeutic facilities, this paper describes the distinct challenges that present themselves when applying AT to address the needs of this often-overlooked sector of the community. It is hoped that this will help raise awareness and contribute to improving the overall quality of AT provision in this area.
E.A. Draffan, Mike Wald, Nadine Zeinoun, David Banes
355 - 362
The use of an online voting system has been developed to enable democratic choices of newly designed symbols to support speech, language and literacy skills in a localisation situation. The system works for those using and supporting Augmentative and Alternative Communication (AAC) symbols on electronic systems by the provision of simplified scales of acceptance and adapted grids. The methodology and results highlighted the importance of user participation at the outset and concrete examples of symbol adaptations that were found necessary to ensure higher levels of user satisfaction. Design changes included appropriate local dress codes, linguistic nuances, social settings, the built environment and religious sensitivities.
To promote successful use of Assistive Technology (AT) supporting Augmentative and Alternative Communication (AAC) and compensating for dyslexia, the last steps of their provision, delivery and instruction, use, maintenance and evaluation, were optimised. In co-creation with all stakeholders based on a list of requirements an integral method and tools were developed.
This paper outlines the clinical doctoral research plan (funded by HEE/NIHR Integrated Clinical Academic Programme) to engage the wide range of people who use alternative and augmentative communication (AAC) to understand the perspectives of users in order to develop a Patient Reported Outcome Measure (PROM) to evaluate equipment and services.
Jacob A. Andrews, Arlene J. Astell, Laura J.E. Brown, Robert F. Harrison, Mark S. Hawley
374 - 380
Under-diagnosis of depression and anxiety is common in older adults. This project took a mixed methods approach to explore the application of machine learning and technology for early detection of these conditions. Mood measures collected with digital technologies were used to predict depression and anxiety status according to the Geriatric Depression Scale (GDS) and the Hospital Anxiety and Depression Scale (HADS). Interactive group activities and interviews were used to explore views of older adults and healthcare professionals on this approach respectively. The results show good potential for using a machine learning approach with mood data to predict later depression, though prospective results are preliminary. Qualitative findings highlight motivators and barriers to use of mental health technologies, as well as usability issues. If consideration is given to these issues, this approach could allow alerts to be provided to healthcare staff to draw attention to service users who may go on to experience depression.
This study examines the potential application of an EMG technology as an assistive control method for individuals with severe physical disabilities. The usability and acceptability of a prototype EMG switch was evaluated with health care professionals, researchers and end users using participatory approaches.
Mohammed Saleh, Reza Saatchi, Fiona Lecky, Derek Burke
385 - 388
A system to determine the likelihood of survival for trauma injury patients is being developed. It uses a fuzzy logic approach that can model complex processes without reliance on sophisticated mathematical formulations and may have the potential to be more accurate than the existing approaches. The outline operation of the system that is currently in a prototype stage is described.
Forceful restraint of psychiatric patients is lawful only in cases of violent uncontrolled behavior. The methods used to limit physical freedom are mainly mechanical means of confinement. The study presents a novel “Personal Protective System” designed to limit patients' violent actions but allow them free non- violent normal functioning.
In the field of deep learning, this paper presents the design of a wearable computer vision system for visually impaired users. The Assistive Technology solution exploits a powerful single board computer and smart glasses with a camera in order to allow its user to explore the objects within his surrounding environment, while it employs Google TensorFlow machine learning framework in order to real time classify the acquired stills. Therefore the proposed aid can increase the awareness of the explored environment and it interacts with its user by means of audio messages.
This research explores the impact of receiving feedback through a Personalised Self-Managed Rehabilitation System (PSMrS) for home-based post-stroke rehabilitation on the users' self-efficacy; more specifically, mastery experiences and the interpretation of biomechanical data. Embedded within a realistic evaluation methodological approach, exploring the promotion of self-efficacy from the utilisation of computer-based technology to facilitate post-stroke upper-limb rehabilitation in the home included; semi-structured interviews, quantitative user data (activity and usage), observations and field notes. Data revealed that self-efficacy was linked with obtaining positive knowledge of results feedback. Encouragingly, this also transferred to functional activities such as, confidence to carry out kitchen tasks and bathroom personal activities. Findings suggest the PSMrS was able to provide key sources of self-efficacy by providing feedback which translated key biomechanical data to the users. Users could interpret and understand their performance, gain a sense of mastery and build their confidence which in some instances led to increased confidence to carry out functional activities. However, outcome expectations and socio-structural factors impacted on the self-efficacy associated with the use of the system. Increasing the understanding of how these factors promote or inhibit self-management and self-efficacy is therefore crucial to the successful adoption of technology solutions and promotion of self-efficacy.
This paper describes the initial results of work to create a recommender system to match technology products to people with I/DD by applying machine learning to a large volume of data about: people with I/DD; the technology products they use; and the outcomes they aim to achieve with technology.
Waltraud Ernst, Gerhard Nussbaum, Veronika M. Berger, Zoltan Major
413 - 420
Mouthsticks are quite an old kind of assistive technology (AT) but nevertheless they are up to now the Swiss army knives among AT. Unfortunately the popularity of mouthsticks massively decreased during the 1990s with the result that knowledge about how to produce good mouthsticks got lost and that there are hardly any adaptable mouthsticks available on the market. This paper discusses the development of a personalized mouthstick with the involvement of end users – people with severe physical disabilities - and occupational therapists as experts of everyday use. A participatory approach was chosen. The results of the analysis of a standardized questionnaire, group discussions and a collaborative workshop with IT-designers, polymer engineers, end users, occupational therapists and gender and diversity researchers are presented and discussed. This proved the necessity of the development of a personalized mouthstick.
Generating innovations – including Assistive Technology products or services – requires expertise in project planning and management. University faculty and small businesses ventures often lack such expertise for development and production tasks, yet governments allocate most funding to these sectors. They need help to achieve results. Four evidence-based models now exist to guide investigators intending to generate AT innovations in: 1) standards/guidelines; 2) tools/ instruments; 3) freeware or 4) commercial products.
Kirsten Rassmus-Gröhn, Charlotte Magnusson, Bitte Rydeman, Gary Randall, Sarah Belson
429 - 436
Ideation means to generate ideas, and when involving non-designers in these activities they need to be informed about the scope of the possibilities without limiting their imagination. This is a general challenge, which becomes particularly important when it comes to advanced technology ideation together with participants that may not have in-depth knowledge of technological designs and solutions. In this study, we supported the ideation process by presenting a kit of magic objects (consisting of cards and physical props) to stroke survivors participating in a co-design workshop carried out within the STARR EU project. The kit was seen to generally work well, but improvements are suggested for the introduction, the design of the cards and the number of objects used.
Veronika M. Berger, Stephan Pölzer, Gerhard Nussbaum, Waltraud Ernst, Zoltan Major
437 - 444
To increase the independence of people with reduced hand/arm functionality, a process to generate personalizable mouth sticks was developed based on the participatory design principle. In a web tool, anybody can choose the geometry and the materials of their mouth piece, stick and tip. Manufacturing techniques (e.g. 3D printing) and materials used in the process are discussed and evaluated.
Guy Collier, Nicola Kayes, Stephen Reay, Amanda Bill
445 - 448
This paper will present a design anthropological perspective on an ongoing project called ‘Living Well with Mild Cognitive Impairment (MCI)’. The project explores how people with MCI (and their families) manage and respond to changes in their memory and thinking. One of the primary aims of this project is to design an online resource that will support people to ‘Live Well’ within the context of possible cognitive decline. The resource was originally proposed to function as a kind of online community, where users could both share and learn about home-grown strategies for managing the cognitive changes associated with MCI in everyday life. Much of this project has been guided by the methodological approach of design anthropology, which encourages project researchers and stakeholders to critically examine underlying assumptions and conceptual frameworks, which in this case revolve around the disputed MCI category. In this paper we will provide some background to the Living Well project before highlighting a number of key insights attained from design anthropology.
METI/AMED are conducting a project on the development and deployment of robotic devices for nursing care to enhance the autonomy of elderly persons and assist care givers. An evaluation protocol is presented and the devices developed in the project are introduced. The devices consist of transfer assist devices (wearable/non-wearable), walking assist devices (outdoor/indoor), safety surveillance sensors (nursing home/private home), bath lift and toilet assist.
In Japan, several types of robotic rollators and walking trolleys have been developed with financial assistance from the Japanese government, and some of them have been commercialized in 2015 and 2016. In this paper, the specifications and features of these devices are presented. Moreover, the physically assistive effects of these devices in elderly users are verified and discussed using the results of fundamental function tests and verification experiments.
This paper briefly introduces the development of transfer assistive robots in terms of development support. In addition, some products that have been developed are reviewed. Our study reveals there are two development styles with completely opposite directions, namely: complex to simple; and simple to complex.