This research enhances COOK (Cognitive Orthosis for cOoKing) by introducing a fuzzy logic-based adaptive framework overcoming static rule-based limitations, enabling context-aware cooking assistance adapting to individual cognitive impairments. We developed a four-context fuzzy inference system integrating stove operations, environmental conditions, cooking methods, and user profiles to replace COOK’s static rules. Dynamic coefficient adaptation personalizes risk assessment based on user cognitive profiles, with weights auto-adjusted using incident history and temporal factors. Our proposal was validated through simulation across 1.4 × 10^6 scenarios. The system demonstrated 76 R^2 = 0.759 ) with strong statistical significance (F-statistic = 175,000 , p < 0.001 ). Adaptive rule generation achieves an average confidence of 78

Human activity recognition plays a crucial role in enabling remote monitoring and assistive healthcare for older adults in smart home environments. However, traditional machine learning approaches often struggle to achieve high recognition accuracy without labeled data, while also preserving user privacy and protecting sensitive information. Although various studies have explored federated learning, effective models for human activity recognition using ambient sensors that simultaneously address privacy concerns and annotation scarcity remain limited. In this paper, we propose FedCSSL, a federated contrastive self-supervised learning framework based on the SimCLR architecture for human activity recognition using ambient sensor data. In our approach, each smart home (client) trains a local contrastive self-supervised model and periodically uploads its model weights to a central server. The server aggregates these weights using either the FedAvg or FedAdam algorithm and distributes the updated global model back to the clients for further training. We evaluated FedCSSL on four real-world smart homes within a residential community. The experiments involve three task scenarios and incremental scenarios to simulate real-world cold starts and knowledge transfer. The results demonstrate that FedCSSL effectively mitigates the annotation scarcity challenge by leveraging representations learned across smart homes without compromising privacy and security. The proposed framework has strong potential for broader application in smart home healthcare systems.

This paper explores the potential of Distributed Ledger Technologies (DLT) to support solidarity networks such as service exchange networks. Traditional centralized platforms often face challenges regarding privacy, security, and resource limitations. We propose a novel approach based on block typing, emphasizing modularity and cost-efficiency. This paper introduces Dynamic Block Typing, as an extension of Nano’s block-lattice structure, to support decentralized applications. Through the case study of the Accordrerie, a Quebec-based solidarity cooperative, we illustrate our proposal with a simple scenario of burnt-out light bulb replacement. This approach offers a scalable and flexible model for a wide range of applications, ensuring accessibility through lightweight and feeless technologies.

In this article, we focus on personal data management in service exchange networks, where members meet each other to share services based on their skills. Through the case study of the Accorderie (a Quebec solidarity cooperative), we propose an innovative protocol designed to reinforce the confidentiality of data relative to members’ addresses and service intervention locations. Using distributed ledger and peer-to-peer interaction, our proposal minimizes the Accorderie’s direct involvement while keeping its position as a trusted authority, allowing members to engage in direct interactions without reliance on a centralized platform. We present three versions of private set membership protocols specially designed to manage locations in the sharing economy. Finally, our findings suggest that decentralized solutions could be a relevant support for solidarity communities, in particular by enhancing member privacy and security, but also by facilitating and reducing maintenance costs.

Smart assistive technologies are increasingly being used to support independent living. Occupational therapists perceive such technologies as useful to helping their clients become more independent in their everyday activities, particularly individuals having sustained a traumatic brain injury. However, adapting and configuring these technologies to meet each person's individual needs remain challenging. In this study, occupational therapists and computer scientists collaborate to develop a recommendation system designed to assist in configuring a smart assistive technology that helps people having sustained a traumatic brain injury cook safely. The relevance of the recommendations made by this case-based recommendation system was first evaluated by comparing the recommended assistant options with the selections of an expert occupational therapist. Then, we questioned the final users, i.e. occupational therapists, about the usefulness of such a system. This first prototype demonstrated promising results, with about 80% correct recommendations for selection (with an average percentage ranging from 77.0% to 86.2% depending on the threshold set for recommended functionalities, using a database of 16 cases). Additionally, the system was found to be robust against an unbalanced database and has been estimated as useful and relevant by end-users, who remain the final decisionmakers. The latter also expressed concerns and suggested improvements, which were incorporated early in the development process to enhance acceptance and facilitate the tool's integration into clinical practice. As a result, the tools and methods used by the multidisciplinary team led to the successful development of a generic human-centered recommendation system that can be used to personalize smart assistive technologies to each unique person's needs.

PURPOSE:Assistive Technologies for Cognition (ATCs), such as the Cognitive Orthosis for coOKing (COOK), offer support to individuals with traumatic brain injury by enhancing safety and independence. While the usability of COOK's client interface has been tested, the expert interface-used by occupational therapists to customize the interface to clients' needs-requires further study. We therefore aimed to: (1) Prioritize the modifications to be made to COOK's configuration interface; and (2) Describe the effectiveness and efficiency of the interface within a laboratory context. MATERIALS AND METHODS:A Human-Centered Design (HCD) approach was used. Fourteen occupational therapists and master's students in occupational therapy participated. A convergent mixed-methods design was used. Data was collected through laboratory testing of the interface. Qualitative data was analyzed using deductive thematic analysis [1], and quantitative data with descriptive statistics. RESULTS:Our study highlighted the importance of balancing effectiveness and efficiency during technology design. Participants emphasized that the configuration process should ensure both a quality that allows customization to meet client needs and an ease of use to minimize the time required for the configuration of COOK. Overall, a good usability of COOK's configuration interface was demonstrated both qualitatively and quantitatively. CONCLUSION:Future studies will be needed to implement the technology in clinical practice and gather further insights on this interface for the design of further iterative improvements to meet the needs of occupational therapists and their clients.

People who sustain a traumatic brain injury can benefit from the use of assistive technologies for cognition, like the Cognitive Orthosis for coOKing (COOK). However, such tools may require personalization for effective use. Although COOK includes multiple personalization options, occupational therapists face challenges with understanding how to configure it due to its complexity and their lack of training. This study thus aimed to: (a) Co-develop a configuration interface that could provide a better User eXperience (UX) for occupational therapists when configuring COOK; and (b) Document the anticipated UX of a mock-up of the configuration interface. First, a mock-up of COOK's configuration interface, named Config My COOK, was co-developed. Second, a qualitative descriptive research design was used to explore the perspective of occupational therapists from Quebec and Ontario using six online focus groups involving 15 participants. Inductive thematic analysis was conducted. Occupational therapists highlighted three features that had the potential to enhance UX, like the appealing look and perceived intuitivity of the configuration interface, support offered by the interface, and access to features to personalize COOK to clients' needs. However, four features had the potential to lessen it, including anticipated training requirements, time required for the configuration process, data confidentiality, and anticipated complexity of interacting with the interface. To counteract these obstacles, occupational therapists identified improvement suggestions which could be completed in the prototype phase. The human-centered design approach enabled us to design a configuration interface to personalize COOK to specific clients' needs, and document its mock-up's anticipated UX with occupational therapists.

BackgroundOlder adults with cognitive deficits face difficulties in recalling daily challenges and lack self-awareness, impeding home care clinicians from obtaining reliable information on functional decline and home care needs and possibly resulting in suboptimal service delivery. Activity of daily living (ADL) telemonitoring has emerged as a tool to optimize evaluation of ADL home care needs. Using ambient sensors, ADL telemonitoring gathers information about ADL behaviors such as preparing meals and sleeping. However, there is a significant gap in understanding on how ADL telemonitoring data can be integrated into clinical reasoning to better target home care services. ObjectiveThis paper aims to describe (1) how ADL telemonitoring data are used by clinicians to maintain care recipients with cognitive deficits at home and (2) the impact of ADL telemonitoring on home care service delivery. MethodsWe used an embedded mixed methods multiple-case study design to examine 3 health institutions located in the greater Montreal region in Quebec that offer public home care services. An ADL telemonitoring system—Innovative Easy Assistance System–Support for Older Adults’ Autonomy (Soutien à l’autonomie des personnes âgées in French)—was deployed within these 3 health institutions for 4 years. Subcases (care recipient, informal caregiver, and clinicians) were embedded within each case. For this paper, we used the data collected during interviews (45-60 min) with clinicians only. Quantitative metadata were also collected on each service provided to care recipients before and after the implementation of NEARS-SAPA to triangulate the qualitative data. ResultsWe analyzed 27 subcases comprising 29 clinicians who completed 57 postimplementation interviews concerning 147 telemonitoring reports. Data analysis showed a 4-step decision-making process used by clinicians: (1) extraction of relevant telemonitoring data, (2) comparison of telemonitoring data with other sources of information, (3) risk assessment of the care recipient’s ADL performance and ability to remain at home, and (4) maintenance or modification of the intervention plan. Quantitative data reporting the number of services received allowed the triangulation of qualitative data pertaining to step 4. Overall, the results suggest a stabilization in monthly services after the introduction of the ADL telemonitoring system, particularly in cases where the number of services were increasing before its implementation. This is consistent with qualitative data indicating that, in light of the telemonitoring data, most clinicians decided to maintain the current intervention plan rather than increase or reduce services. ConclusionsResults suggest that ADL telemonitoring contributed to service optimization on a case-by-case basis. ADL telemonitoring may have an important role in reassuring clinicians about their risk management and the appropriateness of service delivery, especially when questions remain regarding the relevance of services. Future studies may further explore the benefits of ADL telemonitoring for public health care systems with larger-scale implementation studies. International Registered Report Identifier (IRRID)RR2-10.2196/52284

Smart homes for ambient assisted living (SHAAL) can provide cognitive assistance and telemonitoring to foster independent living at home. However, building a personalized SHAAL is a complex and time-consuming co-construction process involving IT specialists, healthcare specialists, caregivers, and the elderly person herself. Not surprisingly, it is very difficult to access and integrate all these expertise at the same time in the same place. That is why we are developing a do-it-yourself (DIY) approach aiming at enabling a non-expert to build her own SHAAL. To get there, four phases need to be supported: design, installation, tests, operation, and maintenance. This paper focusses on the design phase. It presents an innovative prototype platform that leverages Augmented Reality (AR) and Artificial Intelligence (AI) to guide a non-expert through a user-friendly “do-it-yourself” (DIY) process to design in situ a SHAAL. This platform captures each category of expertise using templates and models. Thanks to ontologies and case-based reasoning, AR makes available this expertise and assists a user while she is designing in-situ her SHAAL. Two preliminary experiments involving expert and non-expert users showed promising results.

Ambient Assistance uses the Internet of Things (IoT) technologies to monitor elderly people at home, improving their quality of life through wearable sensors, audio-video technology, and ubiquitous computing. Social isolation, a critical issue for seniors, poses a greater risk to mortality than obesity or physical inactivity, threatening the health and independence of adults over 65. While commercial solutions exist, they are costly and often disregard user context and caregiver needs. This article presents a system that prioritizes user autonomy by notifying the elderly adult before caregivers. It supports various devices and provides a platform for exploring wearable sensors for the elderly, with the aim of reducing social isolation by reassuring loved ones when the user is in a safe area. The system offers caregivers contextual alerts, real-time visualization, and two-way communication when the elderly person leaves a designated safe zone. Our study consisted of two phases: laboratory tests to validate system components and interfaces, followed by a 2-year longitudinal study with an elderly participant and their caregiver, analyzing long-term usage patterns and real-world effectiveness. Our technological findings demonstrate the system's robustness and efficiency. We achieved an 80% uptime over the 2-year period, with an average notification delay of less than 2 s. The system successfully integrated with various IoT devices, including smartwatches, door sensors, and motion detectors. Key challenges included minimizing false alarms, which were reduced to less than 1% using MapReduce-based data processing algorithms, and ensuring seamless operation in diverse network conditions. Although promising, the limitations of the study require further research with a larger sample size. This project contributes to the field of ambient assistance by offering a distributed lambda architecture and a technologically advanced approach to combating social isolation among the elderly. Balances the needs of both seniors and their caregivers while prioritizing system reliability, adaptability, and security.

Deep learning models have significantly contributed to recognizing older adults' daily activities for telemonitoring and assistance. However, recognizing human activities in real-world smart homes over the long term presents substantial challenges. Obtaining the ground truth is time-consuming and costly, yet it is crucial for training and improving deep learning models. Inspired by the impressive performance of self-supervised learning models, this paper utilizes a model based on the SimCLR framework and a self-attention mechanism for downstream human activity recognition. The model leverages the limited and intermittent labeled activities collected by the Label Older Adults' Daily Activities (LOADA) application, which was deployed and used to acquire activity labels in the real-world, uncontrolled smart homes of three young people and two older adults for over one month. The experimental results demonstrate significant performance in activity recognition, employing semi-supervised learning with limited labels, and transfer learning scenarios where representations learned from one smart home are transferred to another. This research could inspire other human activity recognition community researchers to overcome labeling challenges for monitoring older adults in real-world scenarios.

Deep learning models have gained prominence in human activity recognition using ambient sensors, particularly for telemonitoring older adults’ daily activities in real-world scenarios. However, collecting large volumes of annotated sensor data presents a formidable challenge, given the time-consuming and costly nature of traditional manual annotation methods, especially for extensive projects. In response to this challenge, we propose a novel AttCLHAR model rooted in the self-supervised learning framework SimCLR and augmented with a self-attention mechanism. This model is designed for human activity recognition utilizing ambient sensor data, tailored explicitly for scenarios with limited or no annotations. AttCLHAR encompasses unsupervised pre-training and fine-tuning phases, sharing a common encoder module with two convolutional layers and a long short-term memory (LSTM) layer. The output is further connected to a self-attention layer, allowing the model to selectively focus on different input sequence segments. The incorporation of sharpness-aware minimization (SAM) aims to enhance model generalization by penalizing loss sharpness. The pre-training phase focuses on learning representative features from abundant unlabeled data, capturing both spatial and temporal dependencies in the sensor data. It facilitates the extraction of informative features for subsequent fine-tuning tasks. We extensively evaluated the AttCLHAR model using three CASAS smart home datasets (Aruba-1, Aruba-2, and Milan). We compared its performance against the SimCLR framework, SimCLR with SAM, and SimCLR with the self-attention layer. The experimental results demonstrate the superior performance of our approach, especially in semi-supervised and transfer learning scenarios. It outperforms existing models, marking a significant advancement in using self-supervised learning to extract valuable insights from unlabeled ambient sensor data in real-world environments.

Within large and growing human communities where interactions occur, trust is a key factor to consider. Computational trust models have then been widely studied since the 2000s targeting items ratings (e.g. in e-commerce) or M2M (e.g. in IoT network). Among these models, EigenTrust is today one of the most popular and studied ones. It provides a global reputation calculation and is efficient in distributed networks, but not fully satisfactory for human interactions. On the opposite, the Bi-lattice model is well suited for human networks interactions such as solidarity networks and/or human services exchange networks but is limited to local trust results. In this paper, we propose a new aggregator that extends the Bi-lattice model to enable a global reputation calculation. This new aggregator discovers the trust links from the member whose score is to be evaluated to every other members he is connected to on the trust network. It then computes the global reputation of this member based on these trust links. Furthermore, it enables a lightweight approach, as it is able to compute a global reputation based only on a partial knowledge of the trust network. Throughout the paper, the proposed aggregator is presented, evaluated and compared to Eigentrust to show its effectiveness.

Older adults with cognitive deficits face difficulties recalling daily obstacles and lack self-awareness, amplifying the challenges for homecare clinicians to obtain reliable information on functional decline and homecare needs. The result may be suboptimal service delivery. Telemonitoring of ADL has emerged as a tool to optimize ADL homecare needs evaluation. Utilizing ambient sensors, telemonitoring of ADL gathers information about an individual's ADL behaviors within the home, such as preparing meals and sleeping. However, there is a significant gap in the comprehension of how ADL telemonitoring data can be integrated into clinical reasoning to better target homecare services. The current paper aimed to describe 1) how ADL telemonitoring data is used by clinicians in the process of maintaining care recipients with cognitive deficits at home as well as 2) the impact of ADL telemonitoring on homecare service delivery. We used an embedded mixed-methods multiple-case study design in which our cases of interest were three health institutions located in the greater Montreal region and offering public homecare services. An ADL telemonitoring system, named NEARS-SAPA, was deployed within those three health institutions for 4 years. Within each case were embedded sub-cases (care recipient, informal caregiver, clinician(s)). For the objectives of the present paper, we used the data collected during 45-60 min interviews with clinicians only. Quantitative metadata were also collected on each service provided to care recipients before and after the implementation of NEARS-SAPA to triangulate the qualitative data. We analyzed 27 sub-cases, comprising 23 clinicians, that completed a total of 57 post-implementation interviews concerning 147 telemonitoring reports. Data analysis showed a 4-step decision-making process used by clinicians 1) Extraction of relevant telemonitoring data, 2) Comparison of telemonitoring data with other sources of information, 3) Risk assessment of the care recipient’s ADL performance and ability to remain at home, and 4) Maintenance or modification of the intervention plan. Quantitative data reporting the number of services received allowed to triangulate qualitative data pertaining to step 4. Overall, the results suggest a stabilization in monthly services following the introduction of the ADL telemonitoring system, particularly in cases where services were increasing prior to its implementation. This is consistent with qualitative data indicating that, in light of the telemonitoring data, most HSCP decided to maintain the current intervention plan rather than increasing or reducing services. Results suggest that ADL telemonitoring contributed to service optimization on a case-to-case basis. ADL telemonitoring may have an important role in reassuring clinicians about their risk management and the appropriateness of services delivery, especially when questions remain as to the relevance of services. Future studies may further explore the benefits of ADL telemonitoring for public healthcare systems with larger-scale implementation studies. RR2-10.2196/52284

Abstract Context Assessing older adults’ abilities to carry out their activities of daily living (ADLs) is a key determinant in the provision of homecare services for aging in place. Amidst a growing aging population and lack of human resources, continuous remote monitoring technology appears promising to support health and social care professionals (HSCPs) in identifying service needs. However, implementation studies conducted in real-life settings are lacking. This study is part of an on-going action design research project aimed at developing an ambient telemonitoring system monitoring ADLs to support clinical decision making. It focused on the initial step of implementation and aimed to understand 1) which HSCPs would want to use the system, 2) for which care recipient they requested it, and 3) for which reasons. Methods A multiple embedded case study utilizing mixed methods was conducted across 3 healthcare establishments in Quebec, Canada. Descriptive statistics from surveys and medical records was conducted to describe the profile of HSCPs and their care recipients. An inductive qualitative analysis was carried out through interviews with 23 HSCPs, in charge of 31 care recipients, to deepen our understanding of the reasons why they requested the system. Results HSCPs were primarily women (89%) occupational therapists (43%). Home care recipients were also primarily women (74%), with documented refusal of homecare services (65%), diagnosed with cognitive decline (94%), and living in a single-family home or apartment (68%). Overall, interviews revealed HSCPs challenges in getting the necessary information to assess their care recipients needs, despite the presence of in-home services and other strategies in place (e.g. informal carer support). Moreover, the telemonitoring system was perceived as promising for risk management. Conclusions There is an interest for the use of ADL telemonitoring technology in the delivery of home healthcare services for aging in place. Key messages • To facilitate its integration in practice, we explored the need for, and value of, ADLs telemonitoring technology by health and social care professionals (HSCPs) in real-life contexts. • By studying the integration of innovative technologies in home healthcare practices, such as ADLs telemonitoring, we aim to support HSCPs practice in fulfilling older adults desire to age in place.

Towns and cities are currently equipping themselves with a host of connected devices, with a view to transforming themselves into ''smart cities''. To manage this mass of connected objects, autonomous software entities, known as agents, can be attached to them to cooperate and use these devices to offer personalized services. However, this object infrastructure needs to be semantically structured in order to be exploited. This is why the proposal of this article is an ontology, formatted in OWL, describing the object infrastructures, their links with the organization of the multi-agent system and the services to be delivered according to the users of the system. The ontology is applied to smart mobility for people with reduced mobility, and could be adapted to other smart city axes.