TECHNIQUE FOR THE ESTIMATION OF THE KINEMATIC PARAMETERS OF A SPATIAL ANATOMICAL JOINT MODEL.

This paper describes a general technique for fitting a kinematic model to an in-vivo anatomical joint under typical physiological loading conditions. The method employs a nonlinear least squares algorithm to minimize the aggregate deviation over multiple joint positions between postulated model performance and experimentally measured anatomical joint performance. Formulation of a universal joint with skew oblique revolutes to best reproduce wrist motion was selected as an example. The uniqueness and power of the joint modeling technique described herein are three-fold. First, the technique permits the in-vivo fitting of a model for any anatomical joint under physiological loading conditions simulating a given activity under study (e. g. , an implement striking task). Second, the technique directly evaluates all joint parameters for a postulated kinematic model via least squares optimization and may be applied to a broad range of spatial joint model types. Third, the key factor in this technique is an inner sub-optimization to effectively close the postulated model mechansim at selected measured positions thus allowing formulation of joint models which more exactly match anatomical performance.