Validity and Reliability of a Depth Camera-based Automated Physical Function and Fall Risk Assessments

Falls significantly reduce physical function as well as quality of life which leads to failure of successful aging among elders. There is a great need for automated assessing physical functions that can be address fall risks for those elderly living alone. The study aims to test validity and reliability of a depth camera-based physical function and fall risk assessment among healthy young and elderly adults with and without fall histories. The present study explores the concurrent, convergent, and discriminative validity of Azure Kinect-based physical function assessments, along with the intra-rater and inter-rater reliability of these measurements. The diagnostic accuracy and the optimal cut-off values of the Azure Kinect-based tests in differentiating fall risks will be provided. The enrolled participants will complete the following evaluations during a single session: Tandem Stance Test (TST), Five Times Sit to Stand Test (FTSS), Turn Up and Go Test (TUG), Berg Balance Scale (BBS), Functional Reach Test (FRT), Four Square Step Test (FSST), Grip Strength (GS), and Short Physical Performance Battery (SPPB). Each of the AK-based evaluation will be measured three times with simultaneous measurements of the same physical therapist directed tests: two trials will be measured by one physical therapist, while one trial will be measured by the another physical therapist.