Relation between object properties and EMG during reaching to grasp

Abstract

a b s t r a c tIn order to stably grasp an object with an artificial hand, a priori knowledge of the object's properties is a major advantage, especially to ensure subsequent manipulation of the object held by the hand. This is also true for hand prostheses: pre-shaping of the hand while approaching the object, similar to able-bodied, allows the wearer for a much faster and more intuitive way of handling and grasping an object. For hand prostheses, it would be advantageous to obtain this information about object properties from a surface electromyography (sEMG) signal, which is already present and used to control the active prosthetic hand.We describe experiments in which human subjects grasp different objects at different positions while their muscular activity is recorded through eight sEMG electrodes placed on the forearm. Results show that sEMG data, gathered before the hand is in contact with the object, can be used to obtain relevant information on object properties such as size and weight.

Spatiotemporal distribution of location and object effects in the electromyographic activity of upper extremity muscles during reach-to-grasp

“…Our results are consistent with recent studies, however, in showing that EMG activity in proximal muscles also can vary depending on the object about to be grasped (Martelloni et al 2009) and that such object effects are observable quite early in reaching movements (Fligge et al 2013). Stark and colleagues (2007) found little, if any, object-related variation in shoulder and elbow muscle activity when objects were arranged in a horizontal plane, an orientation that they found elicited considerably less variation in proximal joints than when the same objects were arranged vertically.…”

“…The problems of relating the activity of various muscles to the mechanics of the grasping hand are complex (Schaffelhofer et al 2015;Towles et al 2013;ValeroCuevas 2005;Valero-Cuevas et al 2003). Nevertheless, when 6 -8 objects, which differ in size and shape, are all presented at the same location, the object that is being grasped can be decoded from the EMG activity of 8 -12 forearm and intrinsic hand muscles (Brochier et al 2004;Fligge et al 2013). The decoding accuracy increases progressively during the movement, approaching 100% correct object classification midway through the reach, indicating that patterns of activation across the set of distal muscles become more object specific as the movement evolves.…”

EMG-based decoding of grasp gestures in reaching-to-grasping motions

“…Moreover, it has been reported [27] that it possible to relate the object's characteristics with the muscular activity. Toward this direction, the decode of the preshape by the EMG signals, as presented in our work, could provide valuable information for object before the contact of the fingers with the object.…”

“…As result, these devices would increase their effectiveness and usability, and consequently increase the natural transition between the reaching and grasping phase on the prostheses increasing their acceptance by patients. However, at the moment only a limited number of studies focused on the detection of different grasp movements during reaching and grasping motions [25][26][27], and no measurement were performed to assess when a good classification was achieved respect to the hand's preshape .…”

New design of home-based dynamic hand splint for hemiplegic hands: a preliminary study

“…The rehabilitation program improved grip and finger strength. It was previously reported that functional grasping by the hand assisted by a dynamic hand splint activated the recruitment of motor units 11 , 12 ) . However, in the present study, no significant improvements were seen in the Fugl-Meyer assessments after 1 and 3 months of the intervention.…”