[home] [Personal Program] [Help]

---

Session: Prostheses
Session starts: Thursday 24 January, 13:30
Presentation starts: 14:30
Room: Lecture room 558

---

Heidi Witteveen (University of Twente)
Frauke Luft (University of Twente)
Hans Rietman (Roessingh Research and Development)
Peter Veltink (University of Twente)

Abstract:
The percentage of myoelectric forearm prostheses that is being used on a daily basis, is quite low, which is among others caused by the lack of sensory feedback to the users. Earlier studies, e.g. [1], showed the possibilities of vibrotactile stimulation to provide hand opening or grasping force feedback. For optimal object handling, the ability to determine the stiffness of an object is extremely important and it is shown that both proprioceptive and tactile information from the human hand are required for stiffness discrimination. In this study, grasping force and hand opening feedback were combined to provide stiffness feedback. Hand opening feedback was provided through an array of eight coin motors placed along the forearm, while six levels of grasping force were fed back via six amplitude levels of a single C2 tactor. 10 healthy subjects were asked to control a virtual hand by mouse scrolling, grasping virtual objects with varying sizes and stiffness. When closing the hand, by scrolling, hand opening feedback was provided via activation of one of the coin motors. After the virtual object was being touched, grasping force feedback was provided and when scrolling further the grasping force increased. For a stiff object, hand opening was kept constant when increasing the grasping force, but for the most compliant object, each step in increasing force resulted in a decrease in hand opening. Two levels of stiffness, requiring 2 or 4 levels of increase in force to change the hand opening, were simulated in between. Subjects were asked to determine the stiffness of 50 objects using either combined hand opening and grasping force feedback, only one of both feedback methods or no feedback. The outcome parameters were the percentages correctly identified stiffness and the task durations. The stiffness of an object could be correctly identified by the subjects in ±60% of the cases when using combined hand opening and grasping force feedback and when using hand opening feedback alone. These percentages were significantly (p<0.001) higher than those reached with force feedback only or without feedback (±25%). The task duration with combined feedback was significantly longer (5.22±1.5s) than with hand opening feedback alone (4.05±0.9s), which was also significantly longer than without feedback (0.96±0.3s). It is shown that it is possible to provide information about object stiffness by combined hand opening and grasping force feedback or by hand opening feedback alone. Despite the longer task duration with the combined feedback, subjects preferred this combination over hand opening feedback alone. Whether this stiffness feedback is useful for myoelectric prosthesis users in daily life should be further investigated.