Artificial Intelligence–Powered Support for Mild Cognitive Impairment and Dementia: A Systematic Review (Preprint)

Abstract

<sec> <title>BACKGROUND</title> AI has emerged as a promising technology in healthcare, offering potential benefits in assessment, diagnosis, drug discovery, and clinical trial for various diseases. Dementia, an incurable condition characterized by progressive cognitive decline, remains a major neurodegenerative disorder with no effective treatment to slow its progression. This systematic review evaluates AI-based interventions in enhancing the quality of life for individuals with Mild Cognitive Impairment (MCI), which is a condition characterized by cognitive decline that is more pronounced than expected for an individual’s age but does not yet meet the criteria for dementia, or dementia. </sec> <sec> <title>OBJECTIVE</title> The primary objective of this review is to evaluate the existing evidence and suggestions on the effectiveness and usefulness of AI-powered applications in healthcare for supporting individuals with MCI or dementia, as well as their formal and informal caregivers. While the cognitive deficits observed in MCI may not be as marked as those observed in dementia, they can nonetheless have a notable impact on the lives of individuals and their carer. In this review, we consider outcomes in the domains of physical functioning, cognitive support, and psychological/behavioural well-being as indicators of quality of life. Based on our findings, our aim is to identify gaps in the current literature and outline future research directions for developers, healthcare providers, and decision makers. </sec> <sec> <title>METHODS</title> This systematic review was conducted following the PRISMA guidelines. We searched multiple databases, including PubMed, Scopus, Web of Science, Engineering Village, ACM Digital Library, and Medline, without restrictions on language or time. Eligibility criteria were defined using the PICOS framework and data extraction was performed using a standardized form. </sec> <sec> <title>RESULTS</title> From 10,514 records, 122 studies were included following inclusion criteria (112 primary studies and 10 secondary studies). Findings indicate that AI-powered technologies have been implemented to provide cognitive, physical, and psychological well-being support specifically for individuals with MCI or dementia and their caregivers. While AI-driven interventions have demonstrated potential benefits in improving the quality of life, challenges remain regarding accessibility, data privacy, bias in AI training datasets, and the need for inclusive design. Additionally, generative AI applications in dementia care are still in early development stages, with limited research on their long-term impact and real-world effectiveness. </sec> <sec> <title>CONCLUSIONS</title> This review demonstrates that AI-based technologies hold measurable promises for improving quality of life in people with MCI and dementia by supporting cognitive, physical, and emotional needs. AI can assist patients and their formal and informal caregivers through monitoring, memory aids, and social interaction tools. A key emerging direction is caregiver-focused AI, including systems that support their well-being, education, and decision-making. Further validation is required to ensure reliability and long-term effectiveness, and future research should prioritize agentic, context-aware, and co-designed systems that adapt to the progressive nature of cognitive decline. </sec>

Ähnliche Arbeiten