Expert Knowledge for Modeling Functional Health from Sensor Data.

Area of Science:

• Gerontology
• Biomedical Engineering
• Health Informatics

Background:

• Information and Communication Technology (ICT) solutions are increasingly used for active and healthy ageing.
• Continuous monitoring of older adults using domestic sensor systems offers valuable insights into functional health.
• A robust model for interpreting sensor data in this context is currently lacking.

Purpose of the Study:

• To define functional health predictors measurable by domestic sensors.
• To establish thresholds for identifying significant changes in these predictors.
• To develop a standardized model for assessing functional health via sensor data.

Main Methods:

• A literature review was conducted to develop a model linking functional health predictors to sensor-derived features.
• Model parameters were determined through a study involving 38 health experts.
• The model's application was demonstrated using three case studies.

Main Results:

• Twenty-five functional health predictors and their attributes were identified.
• Parameters, including attribute and urgency thresholds, were determined for 12 predictors measurable by passive infrared motion sensors.

Conclusions:

• The developed model, with its parametrized predictors, allows for standardized deployment of domestic sensors to assess functional health.
• Case examples illustrated the model's utility as a screening instrument for detecting functional decline in older adults.