DocumentCode
3078744
Title
A quadratic particle swarm optimization method for magnetic tracking of tongue motion in speech disorders
Author
Wang, Jia ; Huo, Xueliang ; Ghovanloo, Maysam
Author_Institution
NC-Bionics Lab, Department of Electrical and Computer Engineering, NCSU, Raleigh, NC, USA
fYear
2008
fDate
20-25 Aug. 2008
Firstpage
4222
Lastpage
4225
Abstract
We have devised a new magnetic localization technique to accurately track 3-D tongue movements during speech and ingestion. A small permanent magnet secured on the tongue by tissue adhesives, is utilized as a tracer. The magnetic field variations due to tongue motion are detected by three 3-axis magneto-inductive sensor modules outside the mouth, and wirelessly transmitted to a computer for further processing. The tracer is modeled as a magnetic dipole, which position (x, y, z) and orientation (θ, γ) are estimated using a quadratic particle swarm optimization (PSO) algorithm. This algorithm provides faster and more reliable computation compared to the linear PSO. Statistical analysis based on hundreds of simulation and experimental results show that using this method, calculations are accelerated by a factor of three and the tracer can be localized in real-time with ∼1 mm resolution in position and 2° in orientation.
Keywords
Computational modeling; Magnetic sensors; Motion detection; Mouth; Particle swarm optimization; Particle tracking; Permanent magnets; Speech; Statistical analysis; Tongue; Algorithms; Computer Peripherals; Electronics; Equipment Design; Humans; Imaging, Three-Dimensional; Magnetics; Reproducibility of Results; Software; Task Performance and Analysis; Telemetry; Tongue; Touch; Transducers; User-Computer Interface;
fLanguage
English
Publisher
ieee
Conference_Titel
Engineering in Medicine and Biology Society, 2008. EMBS 2008. 30th Annual International Conference of the IEEE
Conference_Location
Vancouver, BC
ISSN
1557-170X
Print_ISBN
978-1-4244-1814-5
Electronic_ISBN
1557-170X
Type
conf
DOI
10.1109/IEMBS.2008.4650141
Filename
4650141
Link To Document