Abstracts Track 2026

Area 1 - Digital Health

Nr:	40
Title:	Development of a Decision Support System Regarding the Use of Digital Health Technologies among Older Adults: A Co-Design Study
Authors:	Marie-Pierre Gagnon, Sié Mathieu Aymar Romaric Da, Mohamed Amine Bouchlaghem and Anik Giguère
Abstract:	Background: Advanced age is a risk factor for the development of chronic diseases, often accompanied by a decline in physical and cognitive functions. Digital health technologies offer opportunities to improve access to care, support chronic disease management, promote independence, and strengthen social participation among older adults. These technologies include applications, as well as integrated systems and devices accessible via computers, smartphones, tablets, wearable devices, or environmental sensors. Despite their potential, the use of digital health technologies by older adults remains uneven. Older adults, informal caregivers, and healthcare professionals are often unaware of digital health technologies that could suit their needs, expectations and preferences. Objective: To describe the development process of a decision support system for digital health technologies based on a qualitative analysis resulting from a co-design workshop with older adults, caregivers, and professionals. Methods: The development of the decision support system includes three iterative steps: (1) understanding user needs and context, (2) developing and refining the prototype, and (3) observing users interacting with the prototype. This study covers the first two steps. The co-design workshop took place in a nursing home in Quebec (Canada) and gathered older adults, informal caregivers and healthcare professionals. A partner nurse working in the nursing home informed potential participants about the study, and those interested were invited to participate in the workshop. The workshop lasted two hours and was co-facilitated by the principal investigator and the partner nurse to reassure participants. The workshop was audio recorded, and we conducted a thematic analysis of the content. Results: A total of 11 participants took part in the workshop: 5 older adults, 2 informal caregivers, and 4 healthcare professionals. Five main themes emerged related to the needs, expectations, and preferences expressed by participants regarding the use and choice of digital health technologies: safe and reliable technologies (e.g., certification, endorsement); functional autonomy (e.g., support daily activities, technology dependency); acceptance (e.g., perceived benefits, attractivity); social support and communication (e.g., caregiver involvement, sharing function); information access and format (e.g., large fonts, audio function). These dimensions are key to ensure the usefulness and trustworthiness of the decision support system. Discussion: The co-design process informed the development of a decision support system using the Collaborative Platform for Authors of Healthcare Decision Aids (PADA) digital platform (currently in progress). Input from participants led to the inclusion of specific examples of digital health technologies and information about their availability and cost. We will validate the decision support system during a workshop with potential users (older adults, informal caregivers, healthcare providers), and a final version will be made publicly available. Conclusion: The co-design approach enabled users to be involved in developing a decision support system on digital health technologies informed by their needs, expectations, and preferences.

Nr:	47
Title:	Mobile-Based Unsupervised Assessment of Executive Functions among Healthy Older Adults
Authors:	Camille Laurent, Séverine Erhel and Corentin Gonthier
Abstract:	Remote cognitive assessments have become much more widespread in recent years : compared to traditional tools, they are useful to conduct testing with less constraints of place, time and frequency. The unsupervised nature of remote assessments makes themN/A more prone to interruptions and distractions, but some studies suggest that they are still feasible and can show good psychometric qualities, even among older adults (e.g., Dubbelman et al., 2023). However, prior feasibility studies have generally examined a single test or very broad assessments, preventing in-depth examination of test validity, e.g. in terms of nomological networks - relations with other cognitive constructs. For instance, the few studies interested in remote assessment of executive functioning have mostly focused on inhibition, eschewing other facets of executive control such as shifting, updating, or proactive and reactive control. The current study aimed to test the feasibility, reliability and validity of a mobile-based unsupervised evaluation of executive control among healthy older adults. After an initial questionnaire-based assessment of potential predictors of executive performance (daily cognitive failures, experience with mobile devices, metacognition...), participants completed four typical cognitive control tasks: flanker task (inhibition), n-back task (updating), number-letter task (shifting), and AX–CPT task (reactive and proactive control). The provisional sample includes 115 participants. Preliminary results show that the tasks demonstrate satisfying reliability based on internal consistency for response times and accuracy, and adequately reproduce the same average effects observed in laboratory tasks (e.g., interference on incongruent trials). Overall, this supports the feasibility of assessing inter-individual differences in executive functioning in older adults with mobile-based unsupervised assessments. In turn, this opens perspectives for research (e.g., remote longitudinal monitoring of older cohorts) and clinical practice (e.g., detecting early markers of Alzheimer’s type dementia). References : Dubbelman, M. A., Hall, T. C., Levesque, I. M., Mimmack, K. J., Sikkes, S. A. M., Fischer, S. H., Rentz, D. M., Sperling, R. A., Papp, K. V., Amariglio, R. E., & Marshall, G. A. (2023). Using a digital tool to detect early changes in everyday functioning in older adults: A pilot study of the Assessment of Smartphone Everyday Tasks (ASSET). Alzheimer’s & Dementia: Diagnosis, Assessment & Disease Monitoring, 15(4), e12506. https://doi.org/10.1002/dad2.12506

Area 2 - Telemedicine and Independent Living

Nr:	49
Title:	Semantic Tokenization and LLM-Based Interpretation: Bridging Raw Sensor Streams to Clinically Meaningful ADLs for Privacy-Preserving Home Monitoring
Authors:	Yong-Hyun Lim, Ho-Min Kang, Kang-Bok Seo and Soon-Ju Kang
Abstract:	Activities of daily living (ADL) assessment remains central to clinical decision-making in ageing and dementia care, yet routine practice still depends largely on questionnaires and episodic observations. This creates a persistent gap between real-world function and what is documented in clinical records. At the same time, smart-home sensing and AI-based activity recognition face well-known barriers: privacy concerns, annotation cost, domain shift across homes, weak explainability, and poor long-term usability in cognitively impaired populations. This position paper argues for a pragmatic middle path between rule-based systems and end-to-end black-box classifiers: semantic tokenization of ambient sensor streams followed by large language model (LLM) interpretation. In the proposed framework, continuous sensor signals are converted into human-interpretable semantic tokens (e.g., posture transition, nocturnal bathroom visit, kitchen appliance activation sequence), and an LLM integrates temporal and contextual information to infer clinically meaningful Basic ADL (B-ADL) and Instrumental ADL (I-ADL) events. We present the conceptual architecture, implementation status (simulation-stage prototype), and a translational validation agenda. Our claim is not that LLMs should replace clinicians or caregivers, but that semantically structured AI can improve privacy-preserving home monitoring while increasing explainability and communication value in ambient assisted living.