In the Netherlands approximately 3750 persons have an arm defect: they miss (part of) their hand, forearm or even their entire arm. The majority of these people are in the possession of a prosthesis. This prosthesis can be purely cosmetic, or offer the user some grasping function. The latter can either be a body-powered or a myo-electric prosthesis. A myo-electric prosthesis is controlled by electrical signals originating from the contraction of muscles of the user and is powered by electric motors. Body-powered prostheses are operated by body movements, which are captured by a harness and transmitted through a Bowden cable to the prehensor. Unfortunately, 23-45% of the users are so dissatisfied with their chosen prosthesis that they decide not to wear it. Thus, prostheses need to be improved.
This thesis focuses on the improvement of body-powered prostheses, which offer several advantages compared to myo-electric prostheses: they are much lighter, cheaper and more reliable and – perhaps most importantly – offer the user extended proprioceptive feedback about the prehensor’s movements and exerted grip force. On the down side, body-powered prostheses currently require high operation forces, causing pain and fatigue during or after use, and potentially limiting the inherent advantages in perception and control. Additionally, users complain about the comfort and outer appearance of the harness, the design of which still looks like that of the Count of Beaufort in 1860.
Lowering the operation forces will most likely increase the pinch force control accuracy and reduce fatigue and pain during or after operation and therefore improve the prosthesis’ functionality. To which level cable forces need to be lowered is up till now unknown; it is assumed that lowering operation forces is effective, but only up the point where the control forces are still clearly distinguishable from noise (like inefficiencies in prehensor or cable friction).
The goal of this thesis is to quantify the perception and control capabilities of prosthesis users as a function of body-powered prosthesis design elements, such as mechanical properties of the prehensor, or an alternative harness. The obtained quantified understanding is intended to guide improvements in body-powered prosthesis design, to enhance the quality of life of upper-limb prosthesis users and to prevent (repetitive strain) injuries.
First, a range of maximum cable operation forces between 87 N and 538 N was established for a representative group of prosthesis users (Chapter 2). When the corrected values for fatigue-free operation (20% of the individually measured maximum force) were compared to the required operation forces of ten commercially available body-powered prostheses, it was concluded that only one of these could be operated fatigue-free. Based on the available results, cable forces should not exceed 38 N for the average female, and 66 N for the average male for most activities in daily life, to enable users to operate their prosthesis fatigue-free.
A second study investigated the effect of cable operation forces (15 N versus 51 N) on the ability to transport a test object (Chapter 3). The object was a mechanical egg: too high cable forces would ‘break’ the object; too low cable forces would cause the operator to drop it. The results indicated that the egg was transferred successfully more often at the low cable operation force settings than at the high force setting.
A third study investigated users’ perception and control abilities by utilizing a force reproduction task (Chapter 4). For successful object manipulation we desire a small difference between the intended and actually applied force on an object, as well as only minor fluctuations in the applied force level. In a force reproduction task the force reproduction error resembles the difference between the intended and actually applied force, whereas the force variability indicates the force fluctuations. The results showed a decreasing force reproduction error with increasing cable excursions for force levels of 10 and 20 N, and a decreasing force variability for decreasing operation force levels varying between 10 and 40 N. Thus, low force levels and large cable excursions contribute to improved force perception and control.
In the fourth and final study an alternative harness design, the Ipsilateral Scapular Cutaneous Anchor System, was compared with the traditional figure-of-nine harness, as comfort of the harness was identified as being an issue in body-powered prosthesis (Chapter 5). In terms of perception and control capacities of users no differences between the two systems were found for operation forces ranging from 10 to 40 N. It could thus be concluded that the Anchor system appears to be a valid alternative to the traditional harness at low operation force levels as performance is comparable while comfort is reportedly better.
In conclusion, this thesis shows that the operation forces which prosthesis users are required to exert are an important factor in body-powered prosthesis design. For most commercially available body-powered prostheses, the control cable forces are too high to be used on a daily basis. To enable users to operate a body-powered prosthesis fatigue-free during the day ‒ every day – with the provision of high quality feedback and adequate prehensor control, operation forces should not exceed 38 N for the average female and 66 N for the average male user. A long operation movement stroke and thus a large cable excursion does contribute to increased prehensor control. For the suggested low operation force levels the Ipsilateral Scapular Cutaneous Anchor System provides a good alternative for the traditional harness.