Research Article

A Study on Decoding Models for the Reconstruction of Hand Trajectories from the Human Magnetoencephalography

Table 1

Movement prediction performance by different algorithms.

AlgorithmMovementsODCMDCMEMV

KalmanReach + rest19.728 ± 0.22723.262 ± 0.2420.148 ± 0.0030.190 ± 0.003
Hybrid Reach + rest8.015 ± 0.1759.321 ± 0.1800.191 ± 0.0040.251 ± 0.005
LinearReach + rest20.010 ± 0.27721.585 ± 0.3330.215 ± 0.0040.300 ± 0.006
KalmanReach12.995 ± 0.16716.992 ± 0.1810.125 ± 0.0010.169 ± 0.002
Hybrid Reach7.956 ± 0.1139.841 ± 0.1530.188 ± 0.0020.258 ± 0.002
LinearReach15.410 ± 0.19317.441 ± 0.2440.159 ± 0.0020.222 ± 0.003

All values are the mean ± standard error of the mean. ODC: orthogonal direction changes; MDC: movement direction changes; ME: movement error; MV: movement variability.