Classification of Hand Grasp Kinetics and Types Using Movement-Related Cortical Potentials and EEG Rhythms

Abstract

Detection of single-trial movement intentions from EEG is paramount for brain-computer interfacing in neurorehabilitation. These movement intentions contain task-related information and if this is decoded, the neurorehabilitation could potentially be optimized. The aim of this study was to classify single-trial movement intentions associated with two levels of force and speed and three different grasp types using EEG rhythms and components of the movement-related cortical potential (MRCP) as features. The feature importance was used to estimate encoding of discriminative information. Two data sets were used. 29 healthy subjects executed and imagined different hand movements, while EEG was recorded over the contralateral sensorimotor cortex. The following features were extracted: delta, theta, mu/alpha, beta, and gamma rhythms, readiness potential, negative slope, and motor potential of the MRCP. Sequential forward selection was performed, and classification was performed using linear discriminant analysis and support vector machines. Limited classification accuracies were obtained from the EEG rhythms and MRCP-components: 0.48 ± 0.05 (grasp types), 0.41 ± 0.07 (kinetic profiles, motor execution), and 0.39 ± 0.08 (kinetic profiles, motor imagination). Delta activity contributed the most but all features provided discriminative information. These findings suggest that information from the entire EEG spectrum is needed to discriminate between task-related parameters from single-trial movement intentions.

Decoding kinetic features of hand motor preparation from single‐trial EEG using convolutional neural networks

“…As such, only one channel was used as input, and the signals were band filtered using low cut‐off frequencies values. However, it was recently suggested that information from the entire EEG spectrum is needed to discriminate between task‐related parameters from single‐trial movement intention (Jochumsen et al., 2017).…”

“…It is straightforward to hypothesize that better commands can be achieved if movement kinematics and kinetics are considered in the decoding process (Jerbi et al., 2011). In this regard, recent attempts to predict speed and force from a hand grasping task in a single‐trial, single electrode strategy resulted in a classification accuracy at or slightly above chance level (Jochumsen, Khan Niazi, Taylor, Farina, & Dremstrup, 2015; Jochumsen et al., 2017; Morash, Bai, Furlani, Lin, & Hallett, 2008). These results contrast with recent studies showing that it is indeed possible to decode hand movement velocities (Bradberry, Gentili, & Contreras‐Vidal, 2010; Lv, Li, & Gu, 2010) and 3D trajectories (Kim, Biessmann, & Lee, 2015) as well as predicting speed and force of a specific movement from EEG (Jochumsen et al., 2015; Jochumsen, Niazi, Mrachacz‐Kersting, Farina, & Dremstrup, 2013), albeit with limited accuracy.…”

Prediction of Hand Movement Speed and Force from Single-trial EEG with Convolutional Neural Networks

“…As such, only one channel was used as input, and the signals were band filtered using low cut-off frequencies values. However, it was recently suggested that information from the entire EEG spectrum is needed to discriminate between task-related parameters from single-trial movement intention [17]. Based on this idea, in this study it was possible to significantly improve the movement prediction accuracy using twenty available channels without additional pre-processing, such as artifact removal or epoch selection.…”

“…Indeed, the decoding of hand movement velocities [12,13] and 3D trajectories [14] as well as the prediction of force and speed from a specific movement [15,10] showed promising results. However, recent attempts to predict speed and force from a hand grasping tasks resulted in a classification accuracy not better than chance level [16,17,18], stressing the need for more accurate prediction schemes.…”

Decoding Hand Movement Types and Kinematic Information From Electroencephalogram

“…Typically, MRCPs manifest as a negative shift in amplitude during the process of movement preparation and reach the peak at the onset of movement, followed by a positive rebound in amplitude. It has been shown that features extracted from MRCPs contain sufficient information to decode hand actions [12,21], upper limb movement [17][18][19][20], and even movement-related parameters, such as speed and force [22][23][24].…”