Bilaterally Mirrored Movements Improve the Accuracy and Precision of Training Data for Supervised Learning of Neural or Myoelectric Prosthetic Control
Bilaterally Mirrored Movements Improve the Accuracy and Precision of Training Data for Supervised Learning of Neural or Myoelectric Prosthetic Control
复制标题
双边镜像运动提高了神经或肌电假肢控制监督学习训练数据的准确性和精确度
DOI：
10.1109/embc44109.2020.9175388
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
Clark, G. A.
中科院分区：
文献类型：
--
作者：
George, J. A.;Tully, T. N.;Colgan, P. C.;Clark, G. A.
假肢的直觉控制依赖于训练算法来将生物记录与运动意图联系起来。训练数据集的质量对运行时性能至关重要，但在手部截肢后，很难准确地标记手部运动学。我们量化了两种不同训练方法的手部运动学标记的准确性和精度：1)假设参与者完美地模仿假体的预定运动（模仿训练），2)假设参与者在相同的双侧运动中完美地镜像他们的对侧手（镜像训练）。我们在非截肢者中比较了这些方法，使用红外摄像机实时跟踪手部八个不同的关节角度。综合数据显示，模拟训练不能解释生物力学耦合或手部姿势的时间变化。镜像训练在标记手部运动学方面显着更准确和精确。然而，当训练一个改进的卡尔曼滤波器来估计运动意图时，模仿和镜像训练方法没有显著差异。结果表明，镜像训练方法创建了一个更忠实但更复杂的数据集。更先进的算法，更有能力学习复杂的镜像训练数据集，可能会产生更好的运行时假肢控制。
Intuitive control of prostheses relies on training algorithms to correlate biological recordings to motor intent. The quality of the training dataset is critical to run-time performance, but it is difficult to label hand kinematics accurately after the hand has been amputated. We quantified the accuracy and precision of labeling hand kinematics for two different training approaches: 1) assuming a participant is perfectly mimicking predetermined motions of a prosthesis (mimicked training), and 2) assuming a participant is perfectly mirroring their contralateral hand during identical bilateral movements (mirrored training). We compared these approaches in non-amputee individuals, using an infrared camera to track eight different joint angles of the hands in real-time. Aggregate data revealed that mimicked training does not account for biomechanical coupling or temporal changes in hand posture. Mirrored training was significantly more accurate and precise at labeling hand kinematics. However, when training a modified Kalman filter to estimate motor intent, the mimicked and mirrored training approaches were not significantly different. The results suggest that the mirrored training approach creates a more faithful but more complex dataset. Advanced algorithms, more capable of learning the complex mirrored training dataset, may yield better run-time prosthetic control.
登录
查看更多内容
影响因子：
3
作者：
George, Jacob A.;Davis, Tyler S.;Clark, Gregory A.
通讯作者：
Clark, Gregory A.
关于镜子的运动
影响因子：
11.2
作者：
M. E. Blaw
通讯作者：
M. E. Blaw
影响因子：
25
作者：
George, J. A.;Kluger, D. T.;Clark, G. A.
通讯作者：
Clark, G. A.
影响因子：
2.9
作者：
Page DM;George JA;Kluger DT;Duncan C;Wendelken S;Davis T;Hutchinson DT;Clark GA
通讯作者：
Clark GA