Quantifying balance control during stance: a multivariate system identification approach

Balance control involves the contribution of neural, muscular and sensory systems, which work together via complex feedback pathways in a closed loop. With age or disease, the underlying systems in balance control can deteriorate; e.g. muscle strength decreases, the sensory systems become less accurate, the processing time of sensory information increases and the neural conduction time increases. To maintain balance in various situations and to prevent falling, the underlying systems can compensate for each other’s deterioration; i.e. there exists some redundancy within the closed loop system of balance control. However, when the deterioration in the underlying system is too severe, or when the compensation mechanism is also affected, impaired balance tends to become symptomatic and the risk of falling increases. To prescribe targeted therapy on an individual level to reduce the consequences of falls, it is important to detect the primarily deteriorated underlying system and the compensation strategies that are at work. Available clinical balance tests have little influence on clinical decision making. In this thesis, a novel experimental set-up and data-analysis method was introduced to assess the contribution of the underlying mechanisms in standing balance control; a multivariate system identification approach. With this experimental approach we quantified age-related changes in standing balance control and we validated the applicability for in clinical practice. The application of system identification techniques give more insight in the underlying physiology of balance control and the changes with age. Using the techniques as a diagnostic tool, can help to detect impaired balance and ultimately reduce the consequences of falls in the elderly