10.1109@embc.2015.7319114

Prévia do material em texto

 
Abstract—The coordination and combination of motion 
and sensation are critical to realize a natural and precise control 
of prosthetic hands. Transcutaneous electrical stimulation (TES) 
is one of possible methods to develop an intuitive perception 
feedback for limb amputees. However, the perception afferent 
sites would be a critical issue that is still unexplored in depth. 
This paper reports a preliminary study on using somatosensory 
evoked potentials (SEP) to determine the proper afferent sites of 
perceptions on residual arms of transradial amputees. In this 
study, two transradial amputees with phantom finger 
perception (PFP) were recruited and SEP for the stimulation of 
median nerves and ulnar nerves were recorded and analyzed. 
PFP distribution maps on subjects’ stumps were obtained by 
mechanical stimulations performed manually. Electrical 
stimulation was then applied to some selected sites on the 
stumps of their residual arms with surface electrodes to evoke 
SEP. In the experiments, SEP were successfully recorded, which 
means that the proposed method might be a suitable approach 
for localizing the afferent sites of perceptions, and could provide 
technique support for possible intuitive neural feedback for 
limb amputees in future work. 
I. INTRODUCTION 
Now available prostheses is helpful for amputees in 
doing their activities, but a major drawback with current hand 
prostheses is that they do not provide the users with proper 
sensory feedback [1]. For human beings, the interaction 
between sensory and motor functions is essential [2]. 
Transfering the sensor-detected sensation signals into users’ 
nerve systems is very critical to realize intuitive feedback of 
perception for arm amputees [3]. 
Many methods have been proposed to provide sensation 
feedback for amputees, including indirect stimulation 
methods (visual, vibrotactile or auditory feedback) and 
direct neural stimulation methods (electrical stimulation of 
somatosensory cortex or peripheral nerves) [3]. Electrical 
stimulations of both somatosensory cortex and peripheral 
 
This work was supported in part by the National Key Basic Research 
Program of China (#2013CB329505), the National Natural Science 
Foundation of China (#61203209, #91420301, #61135004), the Guangdong 
Province Natural Science Fund for Distinguished Young Scholars 
(#2014A030306029), the Shenzhen Peacock Plan Grant 
(#JCYJ20130402113127532) and the Guangdong Innovation Research 
Team Fund for Low-cost Healthcare Technologies. 
H. Wang, P. Fang, L. Tian, Y. Zheng, H. Zhou and G. Li are with the Key 
Lab of Human-Machine Intelligence-Synergy Systems, Shenzhen Institutes 
of Advanced Technology, Chinese Academy of Sciences, Shenzhen, 518055 
China. H. Wang is also with the Shenzhen College of Advanced Technology, 
University of Chinese Academy of Sciences, Shenzhen, 518055 China. X. 
Zhang is with National Research Center for Rehabilitation Technical Aids, 
Beijing, China. 
 (Corresponding author: Guanglin Li; Tel: +86-755-86392219; Fax: 
+86-755-86392299; E-mail: gl.li@siat.ac.cn). 
nerves are possible approaches to regenerate intuitive and 
accurate perception feedback for limb amputees [4-6]. But the 
direct neural stimulation methods require delicate surgery to 
implant the electrodes. In addition, more research work and 
clinical verification should be performed before actual 
applications. Electrical stimulation of sensory afferents using 
transcutaneous electrical nerve stimulation (TES) has been 
confirmed to generate somatic sensations in an amputee's 
phantom limb directly and invasively [7]. TES can induce 
various types of sensation, including touch, vibration, warmth, 
wetness and so on [8, 9]. Tactile perception was found to be 
induced more easily on the median and ulnar aspects than the 
dorsal and radial aspects of the forearm with TES [8]. 
Phantom limb sensation (PLS) is a remarkable and 
important sensory phenomena [10]. More than 80% of 
amputees report phantom limb sensation, which refers to 
non-painful sensations that are felt at a missing limb or a 
portion of the missing limb [11]. Transradial amputees often 
experience sensation on specific phantom fingers (phantom 
finger perception, PFP) with cutaneous stimulation of 
specific stump areas [12]. Some amputees experienced a 
“hand-map phenomenon” corresponding to somatic 
sensations in specific phantom finger elicited by touch of 
specific localized parts of the amputation stump skin [13]. 
Although the neuronal basis of referred phantom limb 
sensations is unclear, several existing studies have 
demonstrated that deafferented cortex remains responsive 
when provided with artificial phantom input and could 
provide a neuronal substrate for spontaneous phantom limb 
sensations [14]. By stimulating the forearm stump skin, PLS 
may be an important method for sensation input for amputee 
[15, 16]. The users of prosthesis can understand the source 
location of sensation with stimulating the phantom finger area. 
PFP evoked by TES may be a promising technique to 
establish sensory feedback from the prosthetic fingers to the 
amputees [11]. 
One of the critical problems for sensory feedback based 
on TES technology is how to determine the proper 
stimulation sites. Stimulation sites are usually chosen along 
the arm nerves. The location of the sensations is evaluated 
subjectively when the stimulations are applied on the forearm 
skin. And an objective evaluation of sensation site is needed. 
Somatosensory evoked potentials (SEP) is a useful and 
noninvasive approach for assessing somatosensory system. 
By combining SEP recordings at different levels of the 
somatosensory pathways, it is possible to assess the 
transmission of the afferent volley from the periphery up to 
the cortex. In that way, we can monitor the afferent nerve 
pathway of perception, from the residual limbs to the 
Towards determining the afferent sites of perception feedback on 
residual arms of amputees with transcutaneous electrical stimulation 
Hui Wang, Peng Fang, Member, IEEE, Lan Tian, Yue Zheng, Hui Zhou, Guanglin Li, Senior 
Member, IEEE , and Xiufeng Zhang.
 
978-1-4244-9270-1/15/$31.00 ©2015 IEEE 3367
 
 
 
somatosensory cortical hand area. SEP may be an objective 
tool to evaluate sensory feedback. 
As far as we know, there are no systematic studies of 
regional distribution of phantom fingers and how to 
determine reliable and proper stimulation site of TES for 
amputee sensory feedback. The goal of this work is to explore 
the method to evaluate the stimulation site for TES using the 
phenomenon of phantom finger perception by SEP in 
amputees for sensory feedback of prosthetic hand. Our 
preliminary results of this study indicated that SEP would be 
a promising technique to determine the stimulation sites for 
establishing sensory feedback from fingers of the prosthetic 
hand to amputee with TES and PFP. 
II. MATERIALS AND METHODS 
A. Principle 
The median nerve innervation territory includes thumb, 
index finger and middle finger. Ulnar nerve dominates the 
little finger and part of ring finger (Fig. 1). We can reasonably 
assume that stump median nerve can be identified from stump 
ulnar nerve by identifying the phantom thumb perception and 
phantom little finger perception. PFP can be aroused by 
touching a specific part of the skin in the stump area of 
amputees. TES can evoke a similar PFP in amputees with 
electrodes placed at the same skin area and we hypothesized 
that the stimulation sites that aroused the PFP were along the 
stump nerves. So in that way, we can evaluate the sensory 
feedbackfunction with SEP. 
In this study, we firstly identified the PFP points in 
stump skin of amputees, so as to prove that PFP area is 
located along the forearm stump nerves. In order to evaluate 
the stimulation site, we then measured the SEP when 
electrocutaneous nerve stimulation was employed with 
electrodes site along the PFP area 
 
 
Figure 1.Schematic diagram for brachial plexus nerves innervations and 
the phenomenon of phantom finger perceptions 
B. Subjects and Data Acquisition 
Two male transradial amputees (designated as A1 and 
A2) with ages of 31 and 27, and amputated in 2007 and 2010, 
respectively, were recruited in the study. The two subjects 
reported spontaneous phantom limb sensations but no 
phantom limb pain. A1 was was with left forearm amputation , 
and A2 was with right foream amputaion.Results from 
subject A1 were typically presented. The protocol of this 
study was approved by the Institutional Review Board of 
Shenzhen Institutes of Advanced Technology. The subjects 
gave the written informed consent and provided permission 
for publication of photographs with a scientific and 
educational purpose. 
We used an 8-channel EMG/EP system (NTS-2000-A38, 
Shanghai NCC Electronic co) with one current stimulator and 
eight EMG recording channels. The stimulation parameters 
were: square wave pulses, current intensity 2 mA, pulse width 
0.2 ms, inter-pulse delay 0.2 ms, and frequency 4.7 Hz. 
C. Procedures 
We first identified the PFP area in the remaining arm of 
amputee, so as to prove that PFP area is along the forearm 
stump nerves. The experiment protocol was as follows: 
Subject was asked to be seated on a chair, with his eyes 
always closed. The subject’s residual forearm was properly 
surrounded with a clinical sterile cloth with coordinate scales. 
Pressures were applied directly on the subject’s residual 
forearm with a hard plastic stick (such as a pen), and the 
corresponding perceptions of different fingers or 
imperceptions were subjectively reported by the subject. 
SEP to median nerve stimulation and ulnar nerve 
stimulation were recorded for intact limb and amputated limb. 
For intra individual side-to-side comparisons (amputated vs. 
intact side), a mixed sensory-motor arm nerve (median or 
ulnar nerve) was stimulated electrically 100 times proximal to 
the amputation site. N9 was recorded at the Erb's point 
(reference electrode: contralateral Erb's point), N13 was 
measured over the fifth cervical spine (reference electrode: 
FPz), and the N20 was recorded over the somatosensory 
cortex (reference electrode: FPz). 
The experiment protocol for intact limb SEP followed the 
standard procedure [17]. Surface electrodes were used to 
record SEP. Stimulation electrode and reference electrode 
were positioned carefully so as to evoke sensation projected 
exclusively to the median nerve's territory or ulnar nerve’s 
territory (above the motor threshold for the healthy hand, 
causing a thumb or a little finger to shake with amplitude 
about 1 cm). 
The detailed experiment protocol for stump limb SEP 
was as follows [11]: 1) Marked the PFP regions in the subject 
stump limb; 2) Attached the SEP recording electrodes to the 
assigned position; 3) Attached the stimulation electrode to 
labeled skin area, and the reference electrode was positioned 
carefully so as to evoke sensation projected exclusively to the 
median nerve's territory in the amputated phantom hand 
through phantom thumb perception; 4) Turned on the 
stimulator, increased current intensity from 0 until pricking 
sensation was reached, and recorded this current level as 
upper limit of stimulation; 5) Chose the stimulator parameter 
3368
 
 
 
to arouse phantom thumb perception and the subject reported 
that phantom thumb was shaking and without painful 
sensation; 6) Used the selected stimulation site and parameter 
for SEP recording; 7) Chose stimulation site among the 
phantom little finger points, so as to stimulate the ulnar nerve 
territory, and did steps 5-6 again. 
III. RESULTS 
A. Phantom finger perception 
As Fig. 2 shows, the central part indicated the “hand 
–map” for different phantom finger perception. The areas of 
single finger-PFP representing the thumb (1), index (2), 
middle (3), ring (4) and little finger (5) were marked. Some 
peculiar points with mixed finger-PFP were also labeled. For 
example, (1+2) indicates mixed PFP of thumb and index 
fingers, and (3+4+5) for mixed PFP for middle, ring and little 
fingers. Those areas not eliciting any perception of touch 
were identified with (0). These labeled single finger-PFP 
points are exactly the positions of stimulating electrode. In 
this way, the accurate hand-map for PFP can be drawn and 
digitized. The results showed that most points were for 
phantom thumb perception or phantom little finger 
perception.Few points were for other phantom fingers 
perception. It was in accordance with our hypotheses that 
stump median nerve can be distinguished from stump ulnar 
nerve by identifying the phantom thumb perception and 
phantom little finger perception. 
 
Figure 2.Experiment set up with identifying ‘hand-map’ for phantom finger 
perception. The right part of the figure is about the identification of PFP 
points on the stump skin and marking points of PFP with a plastic rod. The 
central part is a continuous ‘hand-map’ on the stump skin 
B. Somatosensory evoked potentials 
SEP to median nerve and SEP to ulnar nerve were 
recorded. N9, N13 and N20 were detected for SEP to median 
nerve at the corresponding recording site. Only N20 was 
detected on C3/C4 for stimulation on the ulnar nerve. The two 
amputees showed significant SEP responses upon electrical 
stimulation of the truncated nerve on the residual limb 
(phantom thumb points or phantom little finger points) as 
well as of the nerve at the intact side. These preliminary 
results showed the amplitude and latency of corresponding 
SEP. Representative SEP component can be obtained when 
choosing phantom thumb points or phantom little finger 
points as simulation sites (Fig. 3). Results from subject A2 
were similar, the hand-map for PFP and reasonable SEP 
component were obtained. 
 
 
 
Figure 3.Median SEP and ulnar SEP (amputated side vs. intact side). a) 
Shows the amplitudes of N9, N13, N20 for median SEP, and N20 for 
ulnar SEP. b) Shows the latency. 
3369
 
 
 
IV. DISCUSSION 
To our best knowledge, this is the first study for 
evaluating stimulation site with SEP for limb amputee sensory 
feedback based on TES and PFP. 
According to our results，a continuous “hand-map” for 
phantom finger perception was determined through 
mechanical stimulation of a manual touch applied to skin 
surface of the stump in forearm amputees (Fig. 2). The 
continuous map of the hand on the stump skin for PFP was 
digitized and the coordinates for PFP points were obtained. 
We generally believed that there was a correlation 
relationship between the PFP distribution and brachial plexus 
nerves innervations. These preliminary results demonstrated 
that the distribution of PFP points was along the stump nerves. 
Phantom limb phenomena are very complex, and individual 
differences for PFP are great. The above way may be a useful 
approach for understanding PFP distribution [10]. 
The coordination and combination of motion and 
sensation is very critical to achieve a natural and precise grasp 
for prosthetic hands. However, prosthetic hands still neither 
provide an intuitive sensation neural feedback for prosthesis 
users, nor accurately adjust the grasp motion and strengthaccording to the sensation information. It is possible to relay 
the sensation information from the prosthetic hand to the 
brain of amputees with finger-to-finger specificity via TES 
evoked PFP [11]. It is hard to choose a very efficient and 
robust stimulation site for amputated limb. PFP points are 
along stump nerves innervations, which make them suitable 
stimulation site candidates for TES. 
The mechanisms for phantom limb sensations is 
unknown [16]. It is interesting to note that most points of PFP 
are about the phantom thumb or little finger; the points of 
phantom ring finger are usually congregate with other 
phantom fingers. The phenomenon for PFP distribution can 
be explained by brachial plexus nerves innervations. The 
accurate hand-map for PFP can be used for further study. 
SEP recordings after electrocutaneous nerve stimulation 
were employed to monitor afferent sensory pathway. We 
firstly use SEP for stimulation site selection. Stump median 
nerve can be distinguished from stump ulnar nerve by 
identifying the phantom thumb perception and phantom little 
finger perception. The N20 amplitudes and lantency in 
response to median nerver stimulation were similar with N20 
to ulnar nerver stimulation. Although statistical anlaysis is 
needed, SEP may be a reliable and objective character to 
evaluate the PFP points.In future work, more subjects would 
be recruited to investigate the rules of perception input and 
verify the reliability of the proposed method. 
V. CONCLUSION 
The results obtained in this study demonstrated that 
stump median nerve or stump ulnar nerve can be 
distinguished by identifying the phantom thumb perception 
and phantom little finger perception. SEP may be a promising 
tool for evaluating stimulation site for PFP induced by TES in 
amputees to build an indirectly and invasively sensory 
feedback system. 
ACKNOWLEDGMENT 
The authors would like to thank the two amputees who 
participated in this study. We would also thank the technical 
support provided by the members of Research Center for 
Neural Engineering, at Shenzhen Institutes of Advanced 
Technology, Chinese Academy of Sciences. 
REFERENCES 
[1] C. Antfolk, C. Balkenius, B. Rosen, G. Lundborg, and F. Sebelius, 
"SmartHand tactile display: a new concept for providing sensory 
feedback in hand prostheses," Scand J Plast Reconstr Surg Hand Surg, 
vol. 44, pp. 50-3, Feb 2010. 
[2] D. M. Rager, D. Alvares, I. Birznieks, S. J. Redmond, J. W. Morley, N. 
H. Lovell, et al., "Generating tactile afferent stimulation patterns for slip 
and touch feedback in neural prosthetics," Conf Proc IEEE Eng Med 
Biol Soc, vol. 2013, pp. 5922-5, Jul 2013. 
[3] C. Antfolk, M. D'Alonzo, B. Rosen, G. Lundborg, F. Sebelius, and C. 
Cipriani, "Sensory feedback in upper limb prosthetics," Expert Rev Med 
Devices, vol. 10, pp. 45-54, Jan 2013. 
[4] J. E. O'Doherty, M. A. Lebedev, P. J. Ifft, K. Z. Zhuang, S. Shokur, H. 
Bleuler, et al., "Active tactile exploration using a brain-machine-brain 
interface," Nature, vol. 479, pp. 228-31, Nov 10 2011. 
[5] M. Ortiz-Catalan, B. Håkansson, and R. Brånemark, "An 
osseointegrated human-machine gateway for long-term sensory 
feedback and motor control of artificial limbs," Science Translational 
Medicine, vol. 6, p. 257re6, October 8, 2014. 
[6] D. W. Tan, M. A. Schiefer, M. W. Keith, J. R. Anderson, J. Tyler, and 
D. J. Tyler, "A neural interface provides long-term stable natural touch 
perception," Science Translational Medicine, vol. 6, p. 257ra138, 
October 8, 2014. 
[7] M. R. Mulvey, H. J. Fawkner, H. E. Radford, and M. I. Johnson, 
"Perceptual embodiment of prosthetic limbs by transcutaneous 
electrical nerve stimulation," Neuromodulation, vol. 15, pp. 42-6; 
discussion 47, Jan-Feb 2012. 
[8] B. Geng, K. Yoshida, L. Petrini, and W. Jensen, "Evaluation of 
sensation evoked by electrocutaneous stimulation on forearm in 
nondisabled subjects," J Rehabil Res Dev, vol. 49, pp. 297-308, 2012. 
[9] B. Geng and W. Jensen, "Human ability in identification of location and 
pulse number for electrocutaneous stimulation applied on the forearm," 
J Neuroeng Rehabil, vol. 11, p. 97, 2014. 
[10] A. Woodhouse, "Phantom limb sensation," Clin Exp Pharmacol 
Physiol, vol. 32, pp. 132-4, Jan-Feb 2005. 
[11] G. H. Chai, S. Li, X. H. Sui, Z. Mei, L. W. He, C. L. Zhong, et al., 
"Phantom finger perception evoked with transcutaneous electrical 
stimulation for sensory feedback of prosthetic hand," in Neural 
Engineering (NER), 2013 6th International IEEE/EMBS Conference on, 
2013, pp. 271-274. 
[12] A. Björkman, A. Weibull, J. Olsrud, H. Henrik Ehrsson, B. Rosén, and I. 
M. Björkman-Burtscher, "Phantom digit somatotopy: a functional 
magnetic resonance imaging study in forearm amputees," European 
Journal of Neuroscience, vol. 36, pp. 2098-2106, 2012. 
[13] H. H. Ehrsson, B. Rosén, A. Stockselius, C. Ragnö, P. Köhler, and G. 
Lundborg, Upper limb amputees can be induced to experience a rubber 
hand as their own vol. 131, 2008. 
[14] B. M. Mackert, T. Sappok, S. Grusser, H. Flor, and G. Curio, "The 
eloquence of silent cortex: analysis of afferent input to deafferented 
cortex in arm amputees," Neuroreport, vol. 14, pp. 409-12, Mar 3 2003. 
[15] RAMACHANDRAN, #160, V. S., HIRSTEIN, #160, and W., The 
perception of phantom limbs : The D.O. Hebb lecture vol. 121. Oxford, 
ROYAUME-UNI: Oxford University Press, 1998. 
[16] M. J. Giummarra, S. J. Gibson, N. Georgiou-Karistianis, and J. L. 
Bradshaw, "Central mechanisms in phantom limb perception: The past, 
present and future," Brain Research Reviews, vol. 54, pp. 219-232, 4// 
2007. 
[17] M. Tinazzi, G. Zanette, A. Polo, D. Volpato, P. Manganotti, C. Bonato, 
et al., "Transient deafferentation in humans induces rapid modulation of 
primary sensory cortex not associated with subcortical changes: a 
somatosensory evoked potential study," Neuroscience Letters, vol. 223, 
pp. 21-24, 2/14/ 1997. 
 
3370