Modelling assistive technology adoption for people with dementia

Area of Science:

• Gerontology
• Human-Computer Interaction
• Health Informatics

Background:

• Assistive technologies offer potential for enhancing elderly care, improving quality of life and independence.
• However, widespread adoption is hindered by elderly individuals' general technophobia and limited experience.
• Evaluating technology acceptance potential before deployment is crucial for successful integration.

Purpose of the Study:

• To develop a refined model for predicting assistive technology adoption among the elderly.
• To identify key features that improve the accuracy of technology adoption prediction.
• To minimize the likelihood of non-adoption by understanding user profiles.

Main Methods:

• Utilized data from the Cache County Study on Memory and Aging (CCSMA) with 31 features.
• Extracted user background features including age, gender, education, and health conditions.
• Applied feature selection and reduction techniques to build and optimize an adoption model, followed by classification model identification.

Main Results:

• A technology adoption model was constructed using a reduced set of labeled features.
• The model achieved an average prediction accuracy of 92.48% when tested on 173 participants.
• The refined feature set significantly improved prediction accuracy for technology adoption.

Conclusions:

• Modeling user adoption based on physical, environmental, and social perspectives is effective.
• Personalized technology recommendations can be made by matching user profiles with suitable assistive technologies.
• This approach facilitates better integration of assistive technologies in elderly care settings.