Integrated Measuring Systerm For Humen Balance Function And Its Clinical Study

Balance ability evaluation has been widely applied in clinical diagnose and treatment,rehabilitation training.Biomechanical multi-dimensional force platform is considered to be the basic equipment of evaluating balance ability and cross-axis coupling is a major factor which affects the measurement accuracy of force platform.This thesis studies on decoupling design of multi-dimensional force platform,generalized force loading calibration and nonlinear decoupling.This thesis designed a synthesized force measurement system which synchronously combines ground reaction force and plantar pressure distribution which provide more effective and comprehensive information for equilibrium evaluation.The decoupling design of multi-dimensional force platform is discussed in detail.The inflences on the decoupling of orientation and position errors of strain gages and the deflection of platform top board are analyzed in depth.The geometric and structure parameters of the elastomer and the platform are optimized via FEM analysis tool.A generalized force calibration stratege is put forward which features applicability for field implementation,satisfied decoupling capabilty and high calibration efficiency.The influence of the setting of loading points on the decoupling performance has been investigated.Experiment results have demonstrated the validity of the optimized loading point set.This research introduced artificial intelligence method into calibration decoupling.The influence of noise on LSM is studied,and the application of least square support vector machine(LSSVM)combining with KNN method in decoupling of multi-dimensional force platform is presented.A decoupling approach of KNN-LSSVM local support vector method based on the weighted neighborhood measurement is proposed.Compared with LSM,BP neural network and LSSVM,experiment results demonstrated that the decoupling effect of KNN-LSSVM is better than LSM and LSSVM by 50% and 30%,respectively.A FPGA micro-controller based synchronous data acquisition system of six-component force and plantar pressure distribution is constructed.Cross-talk elimitation of pressure-sensitive array of the pressure distribution measurement and the noise supression are studied thoroughly.Filtering parameters is optimized taking both the sampling rate and noise reduction into account.And an improved median filtering algorithm with the self-adaptive threshold is proposed.The synthesized force measurement system is applied in the clinical diagnose of chronical ankle instability.With the synchronous acquisition of six-component force and plantar pressure distribution,the measurement precision of Time-To-Boundary(TTB)and plantar grip is improved.Based on the comprehensive plantar force and pressure information,the statistic differences of center of pressure(COP),indexes of pressure distribution,TTB and plantar grip in evaluating balance ability of chronic ankle instability(CAI)patients have analyzed thoroughly.Experiment results showed that multi-parameter combined evaliuation diagnose has higher diagnose rate than that of sole six-component force plateform.Statistic analysis comparisons showed that TTB parameters have better performance than traditional COP parameters in the clinical diagnose of CAI patients.