Behavioral Assessment within Smart Homes using Activity Curves

Traditional individual behavioral health studies require a controlled laboratory or clinical environment distinct from an individual's usual living space.  This can be costly, induce artificial stress on vulnerable subjects, and impact observed behavior.

This demonstrated method and implementation provides an alternative or additional tool for unobtrusive activity-based behavior and wellbeing monitoring, using data from pervasive computing and sensing devices.  Automated analysis and recognition of data patterns creates an activity curve representing an individual's normal daily routine.  Comparison of activity curves allows for detection of changes in behavioral routines and cognitive or physical health.  Model change detection approach was evaluated using a longitudinal smart home sensor data set collected from smart homes with older adult residents.  Also demonstrated is how big data-based analytics such as activity curve-based change detection can be used to perform functional health assessment.  The Clinical Assessment using Activity Behavior (CAAB) approach models a smart home resident’s daily behavior.  CAAB uses statistical features that describe characteristics of a resident’s daily activity performance to train machine learning algorithms that predict clinical assessment scores.

•       Home health monitoring services.

•       Diagnostic tool for medical providers.

•       Behavioral evaluation within home or less intrusive assisted living facilities.

•       Increased responsiveness to adverse activities detected in home environment.

•       Reduced exposure to clinical environments.

•       Increased patient quality of life.

•       Potential medical cost savings.