EEG-Based Continuous Hand Movement Decoding Using Improved Center-Out Paradigm

Abstract

The continuous decoding of human movement intention based on electroencephalogram (EEG) signals is valuable for developing a more natural motor augmented or assistive system instead of its discrete classifications. The classic center-out paradigm has been widely used to study discrete and continuous hand movement parameter decoding. However, when applying it in studying continuous movement decoding, the classic paradigm needs to be improved to increase the decoding performance, especially generalization performance. In this paper, we first discuss the limitations of the classic center-out paradigm in exploring the hand movement's continuous decoding. Then, an improved paradigm is proposed to enhance the continuous decoding performance. Besides, an adaptive decoder-ensemble framework is developed for continuous kinematic parameter decoding. Finally, with the improved center-out paradigm and the ensemble decoding framework, the average Pearson's correlation coefficients between the predicted and recorded movement kinematic parameters improve significantly by about 75 percent for the directional parameters and about 10 percent for the nondirectional parameters. Furthermore, its generalization performance improves significantly by about 20 percent for the directional parameters. This study indicates the advantage of the improved paradigm in predicting the hand movement's kinematic information from low-frequency scalp EEG signals. It can advance the applications of the noninvasive motor brain-computer interface (BCI) in rehabilitation, daily assistance, and human augmentation areas.

State-Based Decoding of Continuous Hand Movements Using EEG Signals

“…The pursuit tracking task was also considered in some studies [19], [52]. Although early studies employed linear decoding models, non-linear methods such as unscented Kalman filter (UKF) and neural networks have attracted considerable attention recently [21], [52]. Decoding upper limb movement based on non-invasive recordings has progressed noticeably during the last decade.…”

“…The feasibility of using non-invasive brain recordings for continuous decoding of hand movement parameters was first proved in a paper by Bradbery et al [16] where the velocity trajectories of hand movements were reconstructed from lowdelta EEG potentials in a 3D center-out task. The problem of estimating hand motion parameters from EEG signals has attracted widespread attention recently [17]- [21]. Korik et al [18] demonstrated the role of band-power features from mu, beta, and low-gamma frequency bands in EEG-based motion trajectory prediction.…”

Robust Decoding of Upper-Limb Movement Direction Under Cognitive Distraction With Invariant Patterns in Embedding Manifold

“…Since we used 14 EEG channels, 14 was the highest possible Average overall robustness of our proposed method and the baseline method using different training datasets: 1) Mixed dataset with samples of both attentive and distracted states; 2) samples of attentive state only; 3) samples of distractive state only. The robustness values are calculated on all subjects and direction pairs according to (18). 8, most of the models' accuracies (13/16, about 81.3% of all subjects and direction conditions) showed a cliff-like trend that was either stable or increasing before the peak and dropped significantly thereafter.…”

“…Therefore, ME-BCI systems are more capable of decoding more complex movement fashions, including hand grasp vs relaxing [14], wrist and finger extension [15], hand gestures [16], and combined movement such as grasp and lift [17]. In studies of ME-BCI systems related to upper-limb movement decoding, the center-out task is a paradigm widely used for its fundamentality and generalizability [18]. Zeng et al [19] examined slow oscillation phase representation from low-delta frequency bands, and reconstructed the two-dimensional kinematics parameters with two commonly used linear decoding models under the centerout task.…”

A Review of Myoelectric Control for Prosthetic Hand Manipulation

“…Susannah et al [ 131 ] explored the effects of different socket loads, arm positions, and motion patterns on training paradigms and verified, for the first time, the feasibility of using sonomyography to control prosthetic hands. Jiarong Wang et al [ 132 ] improved the classic center-out paradigm in the field of EEG signal research, enhancing the training paradigm’s movement prediction performance and generalizability, significantly reducing subjects’ physical exertion.…”