First insights exploring Activities of Daily Living performance captured through activity sensors in a Smart Home Environment as an indicator of cognitive decline: A cross‐sectional Analysis

Abstract Background Smart homes offer a unique potential not only for supporting various users’ needs but also for monitoring their activities. Herein the findings of the RADAR‐AD Smart Home study are presented. Using data from activity sensors installed in a smart home environment simulating real‐life conditions, we aim to detect deficits while performing Activities of Daily Living (ADLs) in three known groups of people in different stages of cognitive decline related to Alzheimer’s Disease (AD). Method In a fully equipped Smart Home (https://smarthome.iti.gr/), Fibaro smart plugs (to monitor power consumption), motion‐, door‐, flood‐sensors and panic buttons were installed. During their 24 hour visit, participants [13 Healthy Control (HC), 14 Preclinical AD (PreAD), 13 Prodromal AD (ProAD)] (Table 1) followed a protocol listing a number of ADLs (e.g., meal/beverage preparation) (Table 1). Data collection, feature extraction (i.e., activity duration, number of repetitions) and visualization of ADLs were performed through the CARL data collection and analysis platform for assisted living ( https://carl.iti.gr/ ). (Figure 1). Two‐Way ANOVA and Mann‐Whitney were used to test the statistical significance between the groups. Result Comparing Activity Duration for different ADLs and groups (Figure 2), HC executed the ADLs in less time compared to PreAD and ProAD, while PreAD showed no difference in the majority of the tasks compared to ProAD. Three activities were completed by 50% of the HC, 36% of the PreAD and only 17% of the ProAD. Two‐way ANOVA revealed a statistically significant interaction between the effects of cognitive decline and Activity Duration F(3,106) = 3.504, p = 0.034. Mann‐Whitney analysis for the complex task “Meal Preparation” showed decreased duration for HC compared to PreAD (U = 33.00, p = 0.042) and ProAD (U = 17.00, p = 0.052). Conclusion The RADAR‐AD Smart Home study provides a proof‐of‐concept for the use of home‐based sensors for investigating ADLs in patients with cognitive decline. This work has received support from the EU/EFPIA Innovative Medicines Initiative Joint Undertaking (grant No 806999). www.imi.europa.eu. This communication reflects the views of the RADAR‐AD consortium and neither IMI nor the European Union and EFPIA are liable for any use that may be made of the information contained herein .