A multiple segment model of human balance for objective assessment of clinical scales

An increasingly aging population has led to a growth in subjects with balance disorders. For an elderly person, falling is a risk that has to be challenged on a day to day basis and must be monitored efficiently. For a stroke patient, the inability to maintain upright posture is a crippling limitation that affects the performance of abilities of daily living and often requires a long process of rehabilitation. A review of the state of the art in balance recovery and rehabilitation has revealed gaps in the process of assessing and monitoring balance performance in patients at risk. The currently accepted practice is to use Clinical Assessment Scales to evaluate the performance of the subjects. Although simple and easy to use, these methods rely on visual analysis, and the results are given on a qualitative numerical scale. The outcomes are highly subjective and can often lead to confusion for both the patient and the therapist in interpreting the results. This can lead to loss of patient motivation and lack of confidence in the tests themselves. More tools are required to provide accurate and reliable information regarding balance assessment. Up to now, research in the assessment of human balance has almost exclusively been done using simple two dimensional kinematic models, which has been shown to not fully incorporate significant parameters needed for advanced analyses. The main objective of this research is to develop a new assessment method that provides quantitative information of a subject’s performance. To develop this system, the currently used clinical scales have been analyzed, to identify the key parameters that are evaluated and propose an objective assessment tool that focuses on measurable data instead of visual observations. For this purpose, each individual test on the most commonly used scales has been classified into three categories according to the type of motion. Based on the analysis of the clinical scales we have selected two parameters, the center of pressure and the center of mass as the main focus of study. We propose a hybrid posturography/motion capture system to extract both kinematic and ground reaction force information from test subjects. This data is used with a 12-segment, 31- degree-of-freedom articulated linkage model that is implemented in the OpenSim dynamic simulation software. 19 tests were performed to a cohort of 20 healthy subjects, and the relationship of the CoM and CoP was analyzed in the context of static and dynamic stability. The results show a notable increase in the CoP-CoM distance as the intensity of motion increases. This variation is more significant at the transition phases between different stable poses. These results are consistent with current theories regarding the behavior of the CoM and CoP under static and dynamic conditions. The data reveals that the velocity of the CoM is also an important factor to consider when analyzing stability in dynamic conditions. The proposed system has shown to be an accurate method to determine the CoP and CoM of a subject while performing clinical evaluation tests. We have determined that our model is significantly more accurate than existing models in the literature. This alternative method has the potential to overcome the limitations of using only posturography as a method for determining stability limits. In the future it can be used to evaluate the performance and evolution of the patient and subsequently evaluate the effectiveness of the rehabilitation process.Doctor of Philosophy (MAE