Cheng Chihwen, Huo Xueliang, Ghovanloo Maysam
GT-Bionics Lab School of Electrical and Computer Engineering at Georgia Institute of Technology, Atlanta, GA, 30308, USA.
Annu Int Conf IEEE Eng Med Biol Soc. 2009;2009:563-6. doi: 10.1109/IEMBS.2009.5334058.
This paper presents a new magnetic localization system based on a compact triangular sensor setup and three different optimization algorithms, intended for tracking tongue motion in the 3-D oral space. A small permanent magnet, secured on the tongue by tissue adhesives, will be used as a tracer. The magnetic field variations due to tongue motion are detected by a 3-D magneto-inductive sensor array outside the mouth and wirelessly transmitted to a computer. The position and rotation angles of the tracer are reconstructed based on sensor outputs and magnetic dipole equation using DIRECT, Powell, and Nelder-Mead optimization algorithms. Localization accuracy and processing time of the three algorithms are compared using one data set collected in which source-sensor distance was changed from 40 to 150 mm. Powell algorithm showed the best performance with 0.92 mm accuracy in position and 0.7(o) in orientation. The average processing time was 43.9 ms/sample, which can satisfy real time tracking up to approximately 20 Hz.
本文提出了一种基于紧凑三角形传感器设置和三种不同优化算法的新型磁定位系统，旨在跟踪三维口腔空间中的舌头运动。通过组织粘合剂固定在舌头上的小型永磁体将用作示踪剂。口腔外部的三维磁感应传感器阵列检测由于舌头运动引起的磁场变化，并将其无线传输到计算机。使用DIRECT、Powell和Nelder-Mead优化算法，基于传感器输出和磁偶极方程重建示踪剂的位置和旋转角度。使用收集的一个数据集比较三种算法的定位精度和处理时间，其中源传感器距离从40毫米变化到150毫米。Powell算法表现最佳，位置精度为0.92毫米，方向精度为0.7°。平均处理时间为43.9毫秒/样本，这可以满足高达约20赫兹的实时跟踪。
