Paper Info
Title
A Surface Electromyography-Driven Magnetic Resonance Sequence Controller for Real-Time Myoelectric Triggered Imaging.
Abstract
Combination of surface electromyography (EMG) with diffusion-weighted magnetic resonance imaging (DW-MRI) enables improved studies of spontaneous mechanical contractions in resting human musculature (SMAM). Mechanical muscular activities follow characteristic electrical neuromuscular activities after a delay of several tens of milliseconds. A low-cost standalone system for simultaneous surface EMG measurements during DW-MRI with a real-time model-based surface EMG activity detection is demonstrated which controls the MR sequence. Therefore, a multilayer perceptron (MLP) with sequential forward selection (SFS) was investigated. MLP achieved an area under curve (AUC) of 0.933 in the detection of small surface EMG activities based on five time-domain features. Integration on a microcontroller system enabled fast real-time surface EMG activity detection with highly flexible trigger time settings. Small deviations with only 7.2±1.7 ms time delay between decision of MLP activity detection and onset of MR sequence were measured.
Year
DOI
Venue
2018
10.1109/EMBC.2018.8512570
EMBC
Field
DocType
Volume
Biomedical engineering,Computer vision,Signal processing,Control theory,Computer science,Electromyography,Multilayer perceptron,Activity detection,Millisecond,Artificial intelligence,Forward selection,Magnetic resonance imaging
Conference
2018
Citations 
PageRank 
References 
0
0.34
0
Authors
4
Name
Order
Citations
PageRank
Martin Schwartz151.49
Petros Martirosian282.97
Bin Yang320149.22
Fritz Schick4103.71