Location of innervation zone determined with multichannel surface electromyography using an optical flow technique

Östlund, Nils,1977-Umeå universitet,Linköpings universitet,Hälsouniversitetet,Rehabiliteringsmedicin,Institutionen för strålningsvetenskaper (författare)

Multichannel surface electromyography has developed towards more channels and higher spatial resolution. This allows the study of multichannel electromyograms as images of the potential distribution on the skin. In this paper, a method that estimates the motion of the potential distribution using an optical-flow-based technique is introduced. The optical flow is a vector field that describes how images change with time. The aim of this study was to introduce a new method for innervation zone (IZ) localization and to evaluate its performance. The new method was compared with a method that uses the position of the lowest root-mean-square (RMS) value in an electrode array as an estimate of the IZ localization. Comparisons were made with both simulated signals and with recorded multichannel electromyogram signals. Simulations showed that the methods performed similarly for high signal-to-noise ratio (SNR) and that the optical-flow-based method was superior for lower SNR. When the experimental signals were used, localization with the optical-flow-based method gave a mean absolute deviation of 2.4 mm from the location given by an expert group. The lowest RMS method gave a significantly higher deviation (13.6 mm). Due to the low computational complexity of the optical flow algorithm it is possible to get the estimations of the IZ localization in real time. © 2006 Elsevier Ltd. All rights reserved.

Gerdle, Björn,1953-Östergötlands Läns Landsting,Linköpings universitet,Hälsouniversitetet,Rehabiliteringsmedicin,Smärt- och rehabiliteringscentrum(Swepub:liu)bjoge88 (författare)
Karlsson, StefanUmeå Universitetssjukhus (författare)
Linköpings universitetHälsouniversitetet (creator_code:org_t)

Ingår i:Journal of Electromyography & Kinesiology: Elsevier BV17:5, s. 549-5551050-64111873-5711

Av författaren/redakt...: Östlund, Nils, 1 ...; Gerdle, Björn, 1 ...; Karlsson, Stefan

Kungliga biblioteket hanterar dina personuppgifter i enlighet med EU:s dataskyddsförordning (2018), GDPR. Läs mer om hur det funkar här.
Så här hanterar KB dina uppgifter vid användning av denna tjänst.