Abstract
INTRODUCTION: Although sensor technologies hold potential for addressing challenges for the detection and management of behavioural and psychological symptoms of dementia (BPSD), their application in this context remains in its early stages. This review evaluated the accuracy of sensor technologies to detect and/or monitor neuropsychiatric symptoms in individuals with dementia, particularly in real-world home settings. METHODS: Systematic searches were conducted in five databases on February 20, 2025. Two independent reviewers performed data extraction, with a third reviewer resolving any disagreements. RESULTS: Of 109 records meeting the inclusion criteria for full-text review, 17 studies were included. Sensor-based detection of agitation (number of studies, n = 6) and other common BPSD, specifically sleep (n = 5) and Activities of Daily Living (ADLs) (n = 3), shows early promise, particularly when multimodal data and machine learning techniques are employed. However, the current evidence base is limited by small, predominantly observational studies and inconsistent accuracy across studies. Predictive capabilities remain underdeveloped (n = 3), and the generalizability of findings across different settings is unproven. CONCLUSIONS: This review highlights that the application of sensor technology for detecting and monitoring BPSD remains at an early stage, with existing evidence largely confined to a small subset of symptoms and providing limited clinimetric validation. Additionally, it emphasizes the need for consensus on standard definitions, outcome measures, and validation methodologies to guide and inform future research, which can then leverage these technologies to improve the care and quality of life of persons with dementia and their caregivers.