PubMed Health⌕ Search
Home/Archive/PubMed

PubMed · 11776230

Myoelectric signal classification using neural networks.

M Ungureanu·R Strungaru·V Lazarescu
View original source ↗Export metadata ↓

Abstract

A feed-forward neural network is used for diagnosis of spastic paralysis. It is a two-layer perceptron and it is able to classify two kinds of myoelectric signal recorded in surface electromyography: the normal EMG and the EMG in the case of spastic paralysis. The myoelectric signal was recorded with a surface electrode pair and sampled at 10 kHz. The EMG activity is stochastic and the instantaneous amplitude distribution for a fixed level of contraction is Gaussian. The signal variance is considered a measure of muscle force. We can describe any kind of this process by the AR model. For a precisely modeling of EMG there are necessary many AR model parameters. In the classification problem we have it is not necessary to use a high order AR model. We find a 4-th order AR model is good enough for this study. The Hopfield algorithm is used to calculate the parameters of the autoregressive model.

Explore related subjects

ElectrodesElectromyographyHemiplegiaHumansNeural Networks, ComputerReference ValuesSignal Processing, Computer-AssistedJournal Article

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

M Ungureanu, R Strungaru, V Lazarescu. 1998. Myoelectric signal classification using neural networks.. https://doi.org/10.1515/bmte.1998.43.s3.87

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2007

Probing spin accumulation in Ni/Au/Ni single-electron transistors with efficient spin injection and detection electrodes.

We have investigated spin accumulation in Ni/Au/Ni single-electron transistors assembled by atomic force microscopy. The fabrication technique is unique in that unconventional hybrid devices can be realized with unprecedented control, including real-time tunable tunnel resistances. A grid of Au disks, 30 nm in diameter and 30 nm thick, is prepared on a SiO2 surface by conventional e-beam writing. Subsequently, 30 nm thick ferromagnetic Ni source, drain, and side-gate electrodes are formed in similar process steps. The width and length of the source and drain electrodes were different to exhibit different coercive switching fields. Tunnel barriers of NiO are realized by sequential Ar and O2 plasma treatment. By use of an atomic force microscope with specially designed software, a single nonmagnetic Au nanodisk is positioned into the 25 nm gap between the source and drain electrodes. The resistance of the device is monitored in real time while the Au disk is manipulated step-by-step with angstrom-level precision. Transport measurements in magnetic field at 1.7 K reveal no clear spin accumulation in the device, which can be attributed to fast spin relaxation in the Au disk. From numerical simulations using the rate-equation approach of orthodox Coulomb blockade theory, we can put an upper bound of a few nanoseconds on the spin-relaxation time for electrons in the Au disk. To confirm the magnetic switching characteristics and spin injection efficiency of the Ni electrodes, we fabricated a test structure consisting of a Ni/NiO/Ni magnetic tunnel junction with asymmetric dimensions of the electrodes similar to those of the single-electron transistors. Magnetoresistance measurements on the test device exhibited clear signs of magnetic reversal and a maximum tunneling magnetoresistance of 10%, from which we deduced a spin polarization of about 22% in the Ni electrodes.

Electrodes↗
PubMed2006

Organized assemblies of single wall carbon nanotubes and porphyrin for photochemical solar cells: charge injection from excited porphyrin into single-walled carbon nanotubes.

Photochemical solar cells have been constructed from organized assemblies of single-walled carbon nanotubes (SWCNT) and protonated porphyrin on nanostructured SnO2 electrodes. The protonated form of porphyrin (H4P2+) and SWCNT composites form 0.5-3.0 microm-sized rodlike structures and they can be assembled onto nanostructured SnO2 films [optically transparent electrode OTE/SnO2] by an electrophoretic deposition method. These organized assemblies are photoactive and absorb strongly in the entire visible region. The incident photon to photocurrent efficiency (IPCE) of OTE/SnO2/SWCNT-H4P2+ is approximately 13% at an applied potential of 0.2 V versus saturated calomel electrode. Femtosecond pump-probe spectroscopy experiments confirm the decay of the excited porphyrin in the SWCNT-H4P2+ assembly as it injects electrons into SWCNT. The dual role of SWCNT in promoting photoinduced charge separation and facilitating charge transport is presented.

Electrodes↗
PubMed2006

Free flow isotachophoresis in an injection moulded miniaturised separation chamber with integrated electrodes.

An injection moulded free flow isotachophoresis (FFITP) microdevice with integrated carbon fibre loaded electrodes with a separation chamber of 36.4mm wide, 28.7 mm long and 100 microm deep is presented. The microdevice was completely fabricated by injection moulding in carbon fibre loaded polystyrene for the electrodes and crystal polystyrene for the remainder of the chip and was bonded together using ultrasonic welding. Two injection moulded electrode designs were compared, one with the electrode surface level with the separation chamber and one with a recessed electrode. Separations of two anionic dyes, 0.2mM each of amaranth and acid green and separations of 0.2mM each of amaranth, bromophenol blue and glutamate were performed on the microdevice. Flow rates of 1.25 ml min(-1) for the leading and terminating electrolytes were used and a flow rate of 0.63 ml min(-1) for the sample. Electric fields of up to 370 V cm(-1) were applied across the separation chamber. Joule heating was not found to be significant although out-gassing was observed at drive currents greater than 3 mA.

Electrodes↗