Ambient Assisted Living: a scoping review of artificial intelligence models, domains, technology and concerns

Background: Ambient Assisted Living (AAL) is a common name for various Artificial Intelligence (AI)-infused applications and platforms that support their users in need in multiple activities, from health to daily living. These systems use different approaches to learn about their users and make automated decisions, known as AI models, for personalizing their services and increasing outcomes. Given the numerous systems developed and deployed for people with different needs, health conditions, and dispositions towards the technology, it is critical to obtain clear and comprehensive insights concerning AI models employed, along with their domains, technology, and concerns, to identify promising directions for future work. Objective: This study provides a scoping review of the literature on AI models in AAL. In particular, we analyze: 1) specific AI models employed in A?L systems, 2) the target domains of the models, 3) the technology using the models, and 4) the major concerns from the end-user perspective. Our goal is to consolidate research on the topic and inform end-users, healthcare professionals and providers, researchers, and practitioners in developing, deploying, and evaluating future intelligent AAL systems. Methods: The study was conducted as a scoping review to identify, analyze and extract the relevant literature. It used a natural language processing (NLP) toolkit to retrieve the article corpus for an efficient and comprehensive automated literature search. The relevant articles were then extracted from the corpus and analyzed manually. The review included five digital libraries: the Institute of Electrical and Electronics Engineers (IEEE), PubMed, Springer, Elsevier, and the Multidisciplinary Digital Publishing Institute (MDPI). Results: The annual distribution of relevant articles shows a growing trend for all categories from January 2010 to November 2021. The AI models started with unsupervised approaches as the leader, followed by deep learning (dominant from 2020), instance-based learning, and supervised techniques. Activity recognition and assistance were the most common target domains of the models. Ambient sensing, wearable, and mobile technologies mainly implemented the models. Older adults were primary beneficiaries, followed by patients and frail persons of various ages. Availability was a top beneficiary concern, and to less extent, reliability, safety, privacy, and security. Conclusions: The study presents the analytical evidence of AI models in AAL and their domains, technologies, beneficiaries, and concerns. Future research on intelligent AAL should: involve healthcare professionals and caregivers as designers and users, comply with health-related regulation, improve transparency and privacy, integrate with healthcare technological infrastructure, explain their decisions to the users, and establish evaluation metrics and design guidelines.