The Kitakanto Medical Journal
Online ISSN : 1881-1191
Print ISSN : 1343-2826
ISSN-L : 1343-2826
ORIGINALS
Effect and Task of the Dementia Care-assist AI System for Prediction and Prevention of Behavioral and Psychological Symptoms of Dementia
Naoto OkonogiMasaharu HatanoTaiga FujuTetsuya YamagamiHaruyasu Yamaguchi
Author information
JOURNAL FREE ACCESS

2024 Volume 74 Issue 1 Pages 11-20

Details
Abstract

Background & Aims: In the treatment of people with dementia (PwD), it is necessary to consider behavioral and psychological symptoms of dementia (BPSD). We developed a Dementia Care-assist AI system (DeCaAI) and examined its effectiveness and tasks.

Methods: DeCaAI was introduced in 10 dementia group homes. We enrolled 172 PwD and 206 caregivers. The control group received conventional care, and an intervention group received a 19-week intervention. In the intervention group, we fitted vital sensors to PwD and installed sensors in their rooms. Caregivers recorded cases of BPSD. DeCaAI predicted the occurrence of BPSD from the sensor data and care records and notified caregivers of preventive care measures. We examined severity of BPSD, quality of life, and activities of daily living for PwD in both groups. Moreover, we administered a questionnaire to assess the workload of caregivers.

Results: Four of 10 facilities were excluded because of insufficient care record data and no final evaluation. There were no significant interactions in any of the evaluations of the PwD and caregivers. Moreover, the questionnaire revealed that improvement of the predictive accuracy and usability of DeCaAI is necessary.

Conclusion: We found no significant improvement on BPSD of PwD and workload of caregivers.

Content from these authors
© © 2024, The Kitakanto Medical Society
Previous article Next article
feedback
Top