Heinz-Bodo Schmiedmayer

The GRASP Taxonomy of Human Grasp Types

In this paper, we analyze and compare existing human grasp taxonomies and synthesize them into a single new taxonomy (dubbed “The GRASP Taxonomy” after the GRASP project funded by the European Commission). We consider only static and stable grasps performed by one hand. The goal is to extract the largest set of different grasps that were referenced in the literature and arrange them in a systematic way. The taxonomy provides a common terminology to define human hand configurations and is important in many domains such as human-computer interaction and tangible user interfaces where an understanding of the human is basis for a proper interface. Overall, 33 different grasp types are found and arranged into the GRASP taxonomy. Within the taxonomy, grasps are arranged according to 1) opposition type, 2) the virtual finger assignments, 3) type in terms of power, precision, or intermediate grasp, and 4) the position of the thumb. The resulting taxonomy incorporates all grasps found in the reviewed taxonomies that complied with the grasp definition. We also show that due to the nature of the classification, the 33 grasp types might be reduced to a set of 17 more general grasps if only the hand configuration is considered without the object shape/size.

A Metric for Comparing the Anthropomorphic Motion Capability of Artificial Hands

We propose a metric for comparing the anthropomorphic motion capability of robotic and prosthetic hands. The metric is based on the evaluation of how many different postures or configurations a hand can perform by studying the reachable set of fingertip poses. To define a benchmark for comparison, we first generate data with human subjects based on an extensive grasp taxonomy. We then develop a methodology for comparison using generative, nonlinear dimensionality reduction techniques. We assess the performance of different hands with respect to the human hand and with respect to each other. The method can be used to compare other types of kinematic structures.

Parameters influencing the accuracy of the point of force application determined with piezoelectric force plates

Errors up to +/- 30 mm in determining the point of force application with piezoelectric force plates have been reported in the literature (Kistler, 1984. Multicomponent Measuring Force Plate for Biomechanics and Industry. Kistler, Switzerland; Bobbert and Schamhardt, 1990. Journal of Biomechanics 23, 705-710; Sommer et al., 1997. Proceedings of the XVI th I.S.B. Congress). To explain the main factors influencing the systematic errors the force plate system is modeled as a two-dimensional beam structure. By this model it is strongly indicated that the cause for the errors in determining the point of force application are bending moments in the measurement posts. The main parameters influencing the shape and magnitude of the error function are the ratios between the bending stiffness of the plate and the bending and compressive stiffnesses of the measurement posts. In the current design it is therefore not possible to eliminate the cause for the errors by changing the constructive parameters. By comparing the error functions derived with the beam model to the correction formulas given in the literature an improved algorithm is proposed. This paper shall help biomechanists in understanding the basic concepts of determining the point of force application with force plates and in constructing custom-made force plates for specific applications.

Enhancements in the accuracy of the center of pressure (COP) determined with piezoelectric force plates are dependent on the load distribution.

Errors up to +/- 30 mm in determining the COP with piezoelectric force plates have been reported in the literature. To compensate for these errors, correction formulas were proposed, based on measurements with single point loads. In this paper, it will be shown that the errors in the COP depend on the load distribution. Two examples are presented: (1) simulated balance study, and (2) different pressure patterns during walking. Accurate corrections can only be made for forces distributed over a small area. Errors are expected to be overcompensated if there are only a few pressure peaks separated by large distances. These errors can be as large as the statistical errors (5.8 +/- 3.7 mm) after compensation. For certain situations, it is probably better not to use correction formulas.

Mechanical factors responsible for the obstruction of the gliding mechanism of a dynamic hip screw for stabilizing pertrochanteric femoral fractures.

BACKGROUND In treatment of pertrochanteric femoral fractures with dynamic hip screws (DHSs) (135-degree, Synthes, Bettlach, Switzerland), damage was observed in removed lag screws, leading to the conclusion that the gliding mechanism must have been obstructed as a result of either inappropriate position of the implant or insufficient medial support in the fracture zone. METHODS The forces and moments transmitted in the screw socket are calculated using a mathematical model to find the optimal position of the implant. RESULTS The forces and moments depend on the position and orientation of the lag screw as well as on the position of the contact point between the two main fragments. By changing the point of contact, a better decrease of the load to the DHS can be achieved than by changing the position and orientation of the screw. For a low contact point, the model shows the lowest values for the forces in the socket. CONCLUSION Complete agreement was found between the results of the presented calculations and our own clinical experience in removed DHSs.

Modeling of the Ski-Snow Contact for a Carved Turn

Carved turns with alpine skis are investigated. During the movement of a ski, snow is loaded and unloaded. Compacted snow is not elastic, i.e. deformations remain. Such effects are modeled by a hypoplastic constitutive equation. During a turn the shovel digs into the snow and the tail maintains nearly the same penetration depth as the part under maximum load. This results in a higher resistance against shearing for the afterbody of the ski. In the present work we investigated the benefits of the hypoplastic against the elastic forcepenetration relationship. Simulation results for a sledge on two skis are compared to experimental track data.

Reconstructing the knee joint mechanism from kinematic data

The interpretation of joint kinematics data in terms of displacements is a product of the type of movement, the measurement technique and the underlying model of the joint implemented in optimization procedures. Kinematic constraints reducing the number of degrees of freedom (DOFs) are expected to compensate for measurement errors and noise, thus, increasing the reproducibility of joint angles. One approach already successfully applied by several groups approximates the healthy human knee joint as a compound hinge joint with minimal varus/valgus rotation. Most of these optimizations involve an orthogonality constraint. This contribution compares the effect of a model with and without orthogonality constraint on the obtained joint rotation angles. For this purpose, knee joint motion is simulated to generate kinematic data without noise and with normally distributed noise of varying size. For small noise the unconstrained model provides more accurate results, whereas for larger noise this is the case for the constrained model. This can be attributed to the shape of the objective function of the unconstrained model near its minimum.

A study of constrained models for the kinematic analysis of the human knee joint

In a number of clinical applications such as ligament repair, joint replacement or prosthesis’ design, the understanding of knee kinematics is of fundamental importance. Here, the determination of the joint rotation axes is crucial for the interpretation of joint kinematics. However, concurring definitions of knee joint axes based on anatomic landmarks are in use. Regarding any of these definitions, there is a large variability between observers and different sessions or trials. In order to reduce observer dependence, mathematical procedures based on movement analysis, so-called functional methods, have been developed. Presently, a number of alternative concepts exist allowing for the determination of joint rotation axes [1]. The finite helical axis (FHA) as an invariant description of joint displacements is a powerful tool; however, the FHA is not easy to interpret clinically [2]. Further approaches aim to bridge the gap between clinicians and engineers. The clinical protocol may be conserved while, subsequently, mathematical optimization is employed to reorient the clinically determined rotational axes, aiming at an optimal fit between the data and the underlying kinematic model [3,4]. Therein, the intact human tibio femoral joint (no ligament damage or osteoarthritis) is modeled as a compound hinge joint exhibiting only two rotational axes, flexion/extension (FE) and tibial rotation (TR). It has been shown that, in an angular range of about 40° to 80° of knee flexion during weight bearing flexion exercises, the knee performs an almost plane movement. Furthermore, the flexion range from 0° to 40° is dominated by tibial rotation and flexion [3]. Optimization techniques determine the FE axis in the former angular regime, and the TR axis is found by minimizing the varus-valgus rotation in the latter regime. Here, it is usually assumed that the FE and TR axes are orthogonal and intersect. However, [5] state that the TR axis is anterior to the FE axis and not perpendicular to it. A kinematic model that does not match the natural geometry of the joint is expected to yield unphysical displacements. It is the aim of this contribution to study one particular aspect in detail, namely the effect of an orthogonality constraint when applied to data from a knee exhibiting an arbitrary angle between the rotational axes.

A Mechanical Study to Investigate the Free Play of 3 Different Nail Designs Used for Fixation of Subtrochanteric Femoral Fractures

Patients with subtrochanteric fractures treated by intramedullary fixation frequently report marked post-treatment pain associated with movement. The free distance, clearly visible in the lateral view X-rays, between the inner cortex of the medullary cavity and outer implant surface shows that anterior-posterior movement is possible when using a single bolt for stabilisation of the nail position. The movement behaviour of three standard nail types (Proximal Femur Nail / Synthes, Gamma Nail / Howmedica, Trochanteric Nail / Howmedica) was investigated by mechanical testing and a mathematical model. The results of mechanical testing closely matched the mathematical model. The nail tip length from the locking bolt axis to its distal end was crucial in determining nail movement. The Proximal femur nail (60 mm) shows the least movement (best performance), the Trochanteric nail (20 mm) shows the largest movements, the movement associated with the Gamma nail (40 mm) was between those of the two other nails. We believe that the Trochanteric nail results in too large a movement and is the least suitable implant for stabilisation of subtrochanteric femoral fractures.