A fuzzy logic model for hand posture control using human cortical activity recorded by micro-ECog electrodes.
A fuzzy logic model for hand posture control using human cortical activity recorded by micro-ECog electrodes.
复制标题
使用微 ECog 电极记录的人类皮层活动进行手部姿势控制的模糊逻辑模型。
DOI：
10.1109/iembs.2009.5332746
复制
发表时间：
2009
期刊：
影响因子：
--
通讯作者：
Wang,W
中科院分区：
文献类型：
--
作者：
Vinjamuri,R;Weber,DJ;Degenhart,AD;Collinger,JL;Sudre,GP;Adelson,PD;Holder,DL;Boninger,ML;Schwartz,AB;Crammond,DJ;Tyler-Kabara,EC;Wang,W
本文提出了一种模糊逻辑模型，用于从运动皮质区的皮质电图（ECoG）活动中解码手部姿势。其中一名受试者在运动皮层表面植入了微型ecog电极阵列。在受试者参与3次伸手抓握过程中，从该阵列上的14个电极记录神经信号。在每一个环节中，被试手拿一个木制玩具锤子五次。选择在任务过程中活跃的最佳通道/电极。在运动开始前1/2秒和运动开始后1秒的时间内，将最佳通道的功率谱密度平均值输入模糊逻辑模型。该模型解码手的姿势是打开还是关闭，准确率为80%。利用模糊逻辑模型的输出，采用基于速度的译码和基于加速度的译码两种方法对任务时间方向上的手势进行译码。当将模型预测的手部姿势与实验中数据手套记录的姿势进行比较时，后者的表现更好。将该模糊逻辑模型导入到MATLABregSIMULINK中，实现对虚拟手的控制。
This paper presents a fuzzy logic model to decode the hand posture from electro-corticographic (ECoG) activity of the motor cortical areas. One subject was implanted with a micro-ECoG electrode array on the surface of the motor cortex. Neural signals were recorded from 14 electrodes on this array while subject participated in three reach and grasp sessions. In each session, subject reached and grasped a wooden toy hammer for five times. Optimal channels/electrodes which were active during the task were selected. Power spectral densities of optimal channels averaged over a time period of 1/2 second before the onset of the movement and 1 second after the onset of the movement were fed into a fuzzy logic model. This model decoded whether the posture of the hand is open or closed with 80% accuracy. Hand postures along the task time were decoded by using the output from the fuzzy logic model by two methods (i) velocity based decoding (ii) acceleration based decoding. The latter performed better when hand postures predicted by the model were compared to postures recorded by a data glove during the experiment. This fuzzy logic model was imported to MATLABregSIMULINK to control a virtual hand.