Analysis of human hand kinesiology has been used to justify the design characteristics of proposed prosthetic terminal devices.1–3 It seems that a common assumption in these studies is that the entire articular structure of the hand is required to mechanically reproduce its grasp functions. Another assumption seems to be that biomimetic mechanics and control systems will be a reasonable prosthetic solution for people with upper-limb amputations.
The human hand is a complex mechanical structure consisting of 15 articulations controlled by more than 30 muscles for the five fingers. It functions as a communication, sensory, grasp, and manipulation tool. Separating essential hand kinesiology that is necessary for each hand function might allow for the efficient development of terminal devices where there is one primary function, such as grasping. It follows that to develop an effective grasping mechanism for a terminal device there should be a clear understanding of hand motions and contact characteristics because they relate directly to the grasping function. Understanding grasp kinesiology might lead to the simplification of design specifications that in turn may also reduce cost and increase the rate of learning for the amputee.
The conceptual framework used to characterize grasping functions will most likely influence the development of terminal devices. There is a growing body of literature that is refining the way we think of grasping activities, and may impact terminal device design. For example, researchers have described the simplified coordination of hand motions during grasp as “synergies” or “virtual fingers,” where there are predictable dependencies of one finger on another. Using conceptual approaches, such as “virtual fingers,” can reduce the organizational complexity of hand/object/task interaction.4 In turn, these approaches to understanding grasp may result in novel terminal device mechanisms.
Recent studies indicate that only a small part of the hand's movement potential is used during grasp. Mason et al.5 concluded that most grasps use a base posture with small refinements in finger and thumb positions. In another study, Baud-Bovy and Soechting6 found that when two fingers oppose the thumb, there is a predictable movement pattern that produced a balanced lever. Kamper et al.7 studied a variety of grasping tasks using five everyday objects. They found that the thumb, for example, used less than 5% of the available range. If one were to extrapolate these findings, prosthetic grasping mechanisms may only need a portion of the hand's total range of motion to be effective. Grasping patterns of the normal hand should be studied using a wide variety of objects to compare normal grasp patterns with prosthetic devices.
The current study examined finger angular positions during two- and three-fingered grasp of a wide range of geometrical objects. The results from motion analysis of people with normal hands were compared with two terminal devices grasping the same objects.
Seven subjects participated in the study (two male and five female). A six-camera video motion analysis system (Qualisys, Gothenburg, Sweden) was used to collect marker locations in a calibrated volume approximately 1 m3. Fourteen reflective markers were taped to the interphalangeal (IP) joints, MCP joints, and the carpals of the subjects (Figure 1). Seated subjects were asked to grasp objects, one at a time, using either two fingers or three fingers. Subjects placed their hands in a standard starting position with the palm down with the fingers flat on the table. This was considered the subjects anatomical zero position. On the start command, the subjects reached to grasp the objects at a self-selected speed. Elapsed time for initiation and reaching to grasp the object varied from 2 to 4 seconds. The nine objects were chosen based on standard geometric shapes, sizes, and mass distribution properties. The objects included discs, rods, cubes, rectangles, and spheres. The smallest object was a 1.9-cm diameter sphere and the largest object was a 9.5 × 12.5 × 1.7 cm rectangular prism. The objects were painted brown to avoid inferring a particular use. Data were collected at 120 frames/second and the markers' 3-dimensional coordinates were exported for further analysis using mathematical software. The first grasp trial, where the object was novel to the subject, was used for analysis.
The hand biomechanical model was modified from previous studies of grasp.8 There were 11 rigid segments generated for the three-dimensional motions of each of the finger articulations relative to the wrist (Figure 2). From the three-dimensional coordinates of the markers, joint angles and contact angles were calculated. Total angular ranges of motion for the fingers and thumb were defined as the sum of angular displacements along successive articulations.
Pearson product moment correlation coefficients (r) were used to determine the degree of association between distal to proximal joints as well as parallel joints between the fingers at a single frame at the point of grasp. The point of grasp was identified by object motion. Distal to proximal correlations consisted of all possible pairs along a single digit (distal IP joint with proximal IP joint, distal phalangeal joint with metacarpal phalangeal joint, and proximal IP joint with metacarpal phalangeal joint). Parallel joint correlations consisted of homologous joints between digits.
A descriptive comparison was made between grasp position of the normal hand and two terminal devices during grasp of the same objects. A hook terminal device (Figure 3) and a modified hook (Figure 4) were passively closed on the objects by one of the investigators. The two hook variations were oriented along similar contact planes compared with normal hand orientation.
Descriptive statistics of joint angles at the time of grasp are listed in Table 1. For two- and three-fingered grasp using this specific set of objects, thumb IP and MCP angular displacements and variability were similar. The thumb total range of motion was from 20° to 43° for 80% of grasps. In other words, the range of motion requirement for 80% of grasps in this sample was 23°.
The largest variability of any of the finger joints was the index MCP joint (a range of −44° to 49°). There seemed to be a predisposition for the distal segments of the thumb and index finger to be angled rather than being parallel. The average angle between the index and thumb distal phalanges was approximately 30°.
Distal to proximal articulations in separate fingers (thumb, index, and long fingers) had relatively low correlations (r = −0.20 to 0.40). Thumb total range of motion had low correlation to index and long finger motion. Higher correlations were found with index finger and long finger parallel joints (r = 0.41 to 0.65) and total ranges of motion (r = 0.74).
The standard hook is a single hinge system (Figure 3). The hinge allowed for 70° of motion for the distal grasp surface. Grasp characteristics of the standard hook terminal device depend entirely on the single degree of freedom articulation relative to the size and surface geometry of the object. As a consequence, contact areas are relatively small. As the size of the object increased and the opening aperture increased, the single proximal hinge created an applied force that pushed the object away compared with the angle of the finger tips.
The modified hook terminal device included the proximal hinge of the standard hook and additional rotation through 4-bar linkages that allow contact surface motion of 90° at each of the prosthetic “fingers” (Figure 4). The adapted hook device has a second degree of freedom at each contact surface mechanism that rotates in response to applied forces. The passive mechanisms become tangent to curved or flat object surfaces and can approximate the contact positions of the finger tips while grasping some objects. Forces acting on the object tended to be perpendicular to the contact surface.
The hand has a series of hinge-like articulations to achieve a wide variety of everyday functions such as communication and grasping. The available sagittal plane range of motion of index and long finger distal phalanges relative to the metacarpals is approximately 320 degrees, and the distal phalanx of the thumb can move 180 degrees relative to the first metacarpal, but grasping patterns use only a small proportion of the total range. To develop an effective grasping terminal device, it might be possible to reproduce the most frequently used ranges of motion more easily than the total potential range of motion.
The subjects had very stereotypical positioning at the thumb MCP and IP joints, as well as the finger distal IP and proximal IP joints. In this study, 95% of grasps were accomplished with only 12 to 28 degrees of motion depending on the joint. The index and long finger MCP joints were more variable in positioning to create the grasp aperture and to grasp.
Published proposed finger designs with series articulation and fewer control degrees of freedom seem to link motions sequential finger joints.1,9 Findings of the current study would suggest that a more normal movement pattern would be to link motions of parallel joints between the fingers.
One goal of a terminal device might be to obtain a similar contact position for the distal phalanx of the thumb, index, and long fingers in relation to the objects. For example, the range of motion of a terminal device contact surface should be at least 90 degrees relative to the metacarpals while accommodating to the required aperture.
The single hinge of the hook device can create an aperture for grasp but there is no mechanism to bring the contact surfaces into a more normal finger-like configuration relative to the object surface. The standard hook surface is unresponsive to forces at object interface. Friction coefficients are low, and there are small surface contact areas. The location of the hinge causes resultant forces on the object that push it away from the terminal device, grasping is less effective compared with subjects in the current study who maintained a small angle between the thumb and fingers in the opposite direction.
The adapted hook device has a second level of articulations that allow high-friction material to contact the object. Passive rotation of the 4-bar linkages allows improved contact geometry compared with the hook. Despite the potential for multiple contact points, the grasp patterns are not similar to the two- or three-finger grasp patterns of the normal hand. However, adaptations of this mechanism may permit a more normal contact distribution.
Although it is possible that only the biomimetic approach can accomplish all possible functions of the hand including communication, sensory experiences, manipulation, and grasp, other approaches with simpler controls or mechanisms have not been explored thoroughly. In creating an artificial hand for grasping, the optimal configuration may differ from the anatomical configuration. Articular designs and control optimization may lead to contact interfaces that are similar to the normal hand or in some way, improved.
Thumb position during two- and three-finger grasp of standard geometric objects was relatively consistent using less than half of the available range of motion. Most grasps were accomplished within a 20° range for the thumb. The index finger and long finger also used less than 50% of their available range of motion. Design criteria for prosthetic terminal devices for grasp should be derived from normal functional ranges of motion and finger positions used during the grasping process or from understanding the contact mechanics of the fingers relative to the object.
1. Carrozza MC, Massa B, Dario P, et al. A two DoF finger for a biomechatronic artificial hand. Technol Health Care
2. Schulz S, Pylatiuk C, Reischl M, et al. A hydraulically driven multifunctional prosthetic hand. Robotica
3. Zollo L, Roccella S, Guglielmelli E, et al. BioMechatronic design and control of an anthropomorphic artificial hand for prosthetics and robotic applications. IEEE ASME Trans Mechatron
4. Iberall T. The nature of human prehension: three dextrous hands in one. In: Proceedings of the 1987 IEEE International Conference on Robotics and Automation Raleigh. NC, 1987;396–401.
5. Mason CR, Gomez JE, Ebner TJ. Hand synergies during reach-to-grasp
. J Neurophysiol
6. Baud-Bovy G, Soechting JF. Two virtual fingers in the control of the tripod grasp
. J Neurophysiol
7. Kamper DG, Cruz EG, Siegel MP. Stereotypical fingertip trajectories during grasp
. J Neurophysiol
8. Lee S-W, Zhang X. Development and evaluation of an optimization-based model for power-grip posture prediction. J Biomech
9. Ceccarelli M, Rodriguez NEN, Carbone G. Design and tests of a three finger hand with 1-DOF articulated fingers. Robotica