English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Agitation is one of the most prevalent symptoms in people with dementia (PwD)that can place themselves and the caregiver's safety at risk. Developingobjective agitation detection approaches is important to support health andsafety of PwD living in a residential setting. In a previous study, wecollected multimodal wearable sensor data from 17 participants for 600 days anddeveloped machine learning models for detecting agitation in one-minutewindows. However, there are significant limitations in the dataset, such asimbalance problem and potential imprecise labelsas the occurrence of agitationis much rarer in comparison to the normal behaviours. In this paper, we firstimplemented different undersampling methods to eliminate the imbalance problem,and came to the conclusion that only 20% of normal behaviour data were adequateto train a competitive agitation detection model. Then, we designed a weightedundersampling method to evaluate the manual labeling mechanism given theambiguous time interval assumption. After that, the postprocessing method ofcumulative class re-decision (CCR) was proposed based on the historicalsequential information and continuity characteristic of agitation, improvingthe decision-making performance for the potential application of agitationdetection system. The results showed that a combination of undersampling andCCR improved F1-score and other metrics to varying degrees with less trainingtime and data.

Undersampling and Cumulative Class Re-decision Methods to Improve Detection of Agitation in People with Dementia