Resumen de Frailty assessment based on the instrumented version on the 30-s chair stand test
Nora Millor Muruzabal
FRAILTY ASSESSMENT BASED ON THE INSTRUMENTED VERSION OF THE 30-S CHAIR STAND TEST. The primary aim of this thesis was to evaluate the use of inertial and magnetic measurement systems to outperform the 30-s CST, motivated by the urge of objective parameters to identify as soon as the complex frailty syndrome. Chapter 2 begins the development of this thesis with a systematic review for the assessment of the SiSt and StSi transitions performance by using IMUs (inertial units). The aim was to assess the role of IUs in physical performance evaluation so that clinicians can have at their disposal new devices and objective measures to improve their diagnostics. Moreover, the most significant inertial sensor-based parameters and the more interesting results for each pathological condition were listed in different tables. Chapter 3 shows the novel ¿PB-algorithm¿ to obtain drift-free position estimation for periodic movements. Z-axis acceleration data from a single IMU located on the lower back was used to estimate the up and down position of the subject when performing the different cycles of the 30-s CST. Drift disturbances were removed based on polynomial data fitting, splines interpolation and the wavelet transform, one after the other. The Vicon optical measurement system was used as gold standard and also another drift-correction algorithm was tested to compare results. High accuracy respect of the Vicon¿s reference was showed and a reduction of the mean-square-error (RMSE) was obtained from the previously proposed algorithm. This study means the first step to analyze the kinematic parameters within the 30-s CST duration since Z-position is actively used in the cycle¿s recognition. Chapter 4 explains, for the first time, the inertial measures of the 30-s CST using a single IMU attached on the lower back in a pre-frail population. Z-position drift-free signal was used to automatically count the number of performed cycles and the three corresponding cycle¿s phases (i.e. ¿impulse¿, ¿stand-up¿ and ¿sit-down¿) were detected thanks to the IMU data. A collection of meaningful data based on kinematic parameters is selected for further analysis. Chapter 5 extends the study done for the time-domain information from the 30-s CST in pre-frail subject. The aim is to identify subjects with different frailty levels (i.e. frail, pre-frail and healthy). Her, other parameters such as the X-orientation ranges and the standing-up and sitting-down ¿modified-impulses¿ are also evaluated. Moreover, a subset of pre-frail and healthy subjects with the similar number of performed cycles was assessed in terms of kinematic parameters. So, the developed procedure seems to be a clear improvement of the current clinical evaluations based on the 30-s CST. Finally, chapter 6 sums up the most important results of the thesis. Here, the conclusions and the recommendations for future research in this field were included.
