Abstract
Background and Purpose: Dementia significantly impacts families, particularly caregivers, who bear the responsibility for daily management and emotional support. This review explores the potential of AI tools, such as decision support systems, chatbots, and monitoring technologies, in enhancing the decision-making capacity, emotional resilience, and care management efficiency of family caregivers for dementia patients. Methods: A scoping review was conducted to identify empirical research on AI technologies designed to support family caregivers of dementia patients. A systematic search of scientific databases including PubMed, Scopus, and Web of Science was conducted in August- September 2024. The review focused on studies involving AI-based interventions such as virtual assistants, machine learning algorithms, and behaviour tracking systems. Data was extracted on the types of AI technologies, caregiver roles, intervention purposes, caregiver outcomes, usability, and methodological designs. Results: A search identified 77 studies on AI technologies supporting dementia caregivers. After title and abstract screening, 36 were included, with 10 selected for full-text screening. AI technologies demonstrated considerable promise in supporting caregivers through enhanced decision-making, real-time behavioural monitoring, and emotional support. AI-driven decision support systems facilitated better care management by providing timely, data-driven insights, while chatbots and virtual companions helped reduce caregiver stress and provided emotional support. However, barriers to adoption, such as trust in AI and usability concerns, were noted. Facilitators of successful AI implementation included co-development with caregivers and intuitive design. Conclusions and Implications: AI technologies hold significant potential to strengthen the role of family caregivers in dementia care, offering solutions to improve decision-making, emotional well-being, and daily care management. However, overcoming barriers related to trust and usability is crucial for successful integration. Future research should focus on refining AI tools through co-development with caregivers to ensure broader adoption and efficacy.