Date of Award
Doctor of Philosophy (PhD)
Electrical and Computer Engineering
Older people--Orientation and mobility., Wearable technology., Older people--Health risk assessment., Smartphones., Mobile apps.
Assessment of mobility in older adults is important for early detection and prevention of falls. The Timed Up and Go (TUG) and the 30 Second Chair Stand (30SCS) tests are recom-mended and routinely used for assessing overall mobility, but they provide a single parameter to quantify mobility. These tests are still subjective and prone to errors. Therefore, we need cost effective new diagnostic procedures that provide more detailed assessment parameters related to fall risks. Modern smartphones enable the development of new mobile health (mHealth) appli-cations by integrating inertial and environmental sensors along with the increasing data pro-cessing and communication capabilities. We developed a suite of smartphone applications for assessing mobility to automate and quantify the TUG test, the 30SCS test and the 4-Stage-Balance test (4SBT). We developed a personalized three-segment control model that quantifies torques/forces during sit-to-stand (S2ST) posture transitions, and assesses optimality of each S2ST transition using inputs from smartphone’s inertial sensors. The model assesses energy expenditure using action, defined as an integral of mechanical energy over time during the transition. We demonstrated that the theoretical optimal transition time can be determined for each person by finding the minimum action using a personalized dynamic model. We proposed additional methods of assessment of stability using spectral and harmonic analysis of signals during walking in the TUG test. We tested the model by evaluating optimum action and opti-mum S2ST transition time for a group geriatric patients undergoing a mobility improvement program by comparing their performance with the optimum performance generated by the model.
Madhushri, Priyanka, "A model based framework for mobility assessment of older adults using wearable systems" (2017). Dissertations. 130.