Johannes Tröger

Automated speech-based screening for alzheimer's disease in a care service scenario

This paper describes a benchmark study for a lightweight and low-cost dementia screening tool. The tool is easy to administer, requires no additional experimentation material, and automatically evaluates and indicates potential subjects with dementia. The protocol foresees that subjects answer four distinct tasks, three of which are ordinary questions and one is a counting prompt. In our care use case, older people are assessed remotely via the tool, potentially even via telephone or within a daily care service routine. The assessment results are subsequently sent to professionals who initiate further steps. A machine learning classifier was trained on the French Dem@Care corpus. Solely utilizing vocal features, the classifier reaches 89% accuracy. Implications for the use case and further steps are discussed.

Oil in the Machine: Technical Support for a Human-Centred Service System for Public Transport

The Mobia project is a 3-year R&D project focussing on making public transport more accessible for elderly persons and persons with mobility impairments. We research and develop of a combination of a human service system consisting of on the one hand mobility helpers, that is, humans available for helping passengers with things they cannot do or are uncomfortable with, such as, entering and exiting the carriers. On the other hand, we develop a technological backbone containing a coordination system and user interfaces for passengers, mobility helpers and the Mobia call center with the goal of coordinating the point of service. The project applies a user-centred design methodology in which qualitative (N = 29) and quantitative (N = 250) interviews have been used to identify barriers in public transport, develop personae and scenarios.

Fully Automatic Speech-Based Analysis of the Semantic Verbal Fluency Task

Background: Semantic verbal fluency (SVF) tests are routinely used in screening for mild cognitive impairment (MCI). In this task, participants name as many items as possible of a semantic category under a time constraint. Clinicians measure task performance manually by summing the number of correct words and errors. More fine-grained variables add valuable information to clinical assessment, but are time-consuming. Therefore, the aim of this study is to investigate whether automatic analysis of the SVF could provide these as accurate as manual and thus, support qualitative screening of neurocognitive impairment. Methods: SVF data were collected from 95 older people with MCI (n = 47), Alzheimer’s or related dementias (ADRD; n = 24), and healthy controls (HC; n = 24). All data were annotated manually and automatically with clusters and switches. The obtained metrics were validated using a classifier to distinguish HC, MCI, and ADRD. Results: Automatically extracted clusters and switches were highly correlated (r = 0.9) with manually established values, and performed as well on the classification task separating HC from persons with ADRD (area under curve [AUC] = 0.939) and MCI (AUC = 0.758). Conclusion: The results show that it is possible to automate fine-grained analyses of SVF data for the assessment of cognitive decline.

Board Games and Regulars’ Tables — Extending User Centred Design in the Mobia Project

To allow persons with mobility issues to remain mobile, the three-year project Mobia set out to develop a technology-supported human-based service within public transport: mobility guides provide a helping hand for passengers with mobility issues. The Mobia system can be positioned within the field of AAL and consists of multiple user interfaces and actors to coordinate and realise the point of service. This paper introduces two extension to the User-Centred Design methodology which were successfully applied during the course of the project. The first is the development of a large board game like demonstrator that allows for simulation of realistic scenarios within the scenario during which observation of user behaviour can take place. The second – regulars’ table – is a monthly face2face meeting between developers, passengers and mobility guides serving as a platform for exchanging experiences, testing and discussing new ideas. Results from the Mobia project are discussed.

Telephone-based Dementia Screening I: Automated Semantic Verbal Fluency Assessment

Dementia has a large economic impact on our society as cost-effective population-wide screening for early signs of dementia is still an unsolved medical supply resource problem. A solution should be fast, require a minimum of external material, and automatically indicate potential persons at risk of cognitive decline. Despite encouraging results, leveraging pervasive sensing technologies for automatic dementia screening, there are still two main issues: significant hardware costs or installation efforts and the challenge of effective pattern recognition. Conversely, automatic speech recognition (ASR) and speech analysis have reached sufficient maturity and allow for low-tech remote telephone-based screening scenarios. Therefore, we examine the technologic feasibility of automatically assessing a neuropsychological test---Semantic Verbal Fluency (SVF)--via a telephone-based solution. We investigate its suitability for inclusion into an automated dementia frontline screening and global risk assessment, based on concise telephone-sampled speech, ASR and machine learning classification. Results are encouraging showing an area under the curve (AUC) of 0.85. We observe a relatively low word error rate of 33% despite phone-quality speech samples and a mean age of 77 years of the participants. The automated classification pipeline performs equally well compared to the classifier trained on manual transcriptions of the same speech data. Our results indicate SVF as a prime candidate for inclusion into an automated telephone-screening system.

Is Co-creation Superior to User Centred Design? Preliminary Results from User Interface Design for Inclusive Public Transport.

Mobility is a basic need which is especially critical for older people and persons with physical impairments. Within the scope of a long-term effort, we have established a human-based, technology-aided support service that helps mobility-impaired passengers to use public transport by engaging so-called mobility guides. A mobility guide is a person that supports passengers, for instance door-to-door. In order to interact with the system, e.g. ordering trips and guides, passengers can use a smartphone app or a webpage. Following a user-centered approach, we conducted monthly meetups – regulars’ tables – in order to maintain contact and passengers’ engagement with the goal of eliciting feedback about the UIs’ usability and to jointly discuss new features. However, both quality and quantity of feedback as well as engagement reduced over time. In this paper, we describe and discuss the positive effect of replacing the regulars’ tables with co-creation based meetings with a small dedicated group of passengers – co-developers. Results of the first co-developer workshops indicate substantial improvements in factors like engagement and quality of feedback resulting in concrete enhancements of the UIs.

AUTOMATIC SPEECH ANALYSIS FOR THE DETECTION OF EMOTIONAL DISTURBANCES IN PERSONS WITH DEMENTIA

Compared to scoring 1, scoring 5 on the ADFlag scale conveyed a 19.66-fold increase (15.98-24.18; Chisq<0.0001) in the odds for incipient AD diagnostic by A+T+N+ criteria, and scoring 4 a 2.96-fold increase (2.62-3.33; Chisq<0.0001). Patients scoring 5 on the ADFlag scale were all part of the Alzheimer’s pathophysiologic continuum (A+). Together, patients scoring either 4 or 5 represented 73% of all A+ patients, including 63.63% of the A+T+N+ or A+T+N(incipient AD). Moreover, we found that an ADFlag score of 4 at baseline significantly increased the odds for conversion 4.17-fold compared to patients scoring 1 (3.61-4.80; Chisq<0.0001) and 8.34-fold compared to patients scoring 3 (6.37-10.91; Chisq<0.0001). A generalized regression predictive model confirmed the association between conversion to AD and scoring 4 on the ADFlag scale. Conclusions: ADFlag has significant prognostic value and it’s use as a stratification tool to prescreen aMCI patients into preventive or therapeutic trials may enhance their chance of success.

DiDiER - digitized services in dietary counselling for people with increased health risks related to malnutrition and food allergies

The goal of the DiDiER project is to verifiably improve services in the field of dietary counselling. This will be achieved by digitising information to increase counselling intensity and to improve workflows for the service provider. The project will develop an IT-based support system for dietary counselling, covering two use cases, facilitation and support of the work of nutritionists in ambulatory allergological nutrition counselling and of nutritionists involved in the care of geriatric patients, especially of those with frailty. One of the projects significant features is that the users sensitive data remain under his or her personal control at all times.

Technology Experience Café—Enabling Technology–Driven Social Innovation for an Ageing Society

Effective technology innovation process management in the context of active healthy ageing has the potential to improve older adults’ quality of life, allowing them to maintain their independence and age in their own homes for longer. But as older adults significantly differ from the general population in technology use and its impact on their quality of life, tools are needed that (1) involve this target group into the innovation process, as well as (2) capture the diverse needs of technology for various stakeholders involved in this process. This paper presents the framework called Technology Experience Cafe (TEC), developed within the European project SIforAGE, answering exactly this need. Detailed information on the methodology and its implementation in five sites, in four different countries across Europe, focusing on participating stakeholders, general design of the TEC, and used evaluation tools, is provided. Preliminary results show, that (1) the target group’s perception of the TEC as a framework was thoroughly positive and TECs had a positive impact on older adults’ technology related attitudes and (2) that stakeholders’ benefits affiliated with their involvement in the TECs are manifold. Implications and limitations are discussed.