Automated Labeling of Movement- Related Cortical Potentials Using Segmented Regression

Rashid, Usman; Niazi, Imran Khan; Jochumsen, Mads; Krol, Laurens R.; Signal, Nada; Taylor, Denise

doi:10.1109/tnsre.2019.2913880

Cited by 13 publications

(10 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To model these two separate motor processes in the sl LRP waveform we used a piecewise regression method (Rashid et al, 2019). Two linear regression lines were fitted through the stimulus locked LRP averaged within each condition in each individual (Fig.…”

Section: Resultsmentioning

confidence: 99%

Reward and expectancy effects on neural signals of motor preparation and execution

Chen

Berg

Kwak

2021

Preprint

View full text Add to dashboard Cite

The prospect of rewards can have strong modulatory effects on response preparation. Importantly, selection and execution of movements in real life happens under an environment characterized by uncertainty and dynamic changes. The current study investigated how the brain's motor system adapts to the dynamic changes in the environment in pursuit of rewards. In addition, we studied how the prefrontal cognitive control system contributes in this adaptive control of motor behavior. To this end, we tested the effect of rewards and expectancy on the hallmark neural signals that reflect activity in motor and prefrontal systems, the lateralized readiness potential (LRP) and the mediofrontal (mPFC) theta oscillations, while participants performed an expected and unexpected action to retrieve rewards. To better capture the dynamic changes in neural processes represented in the LRP waveform, we decomposed the LRP into the preparation (LRPprep) and execution (LRPexec) components. The overall pattern of LRPprep and LRPexec confirmed that they each reflect motor preparation based on the expectancy and motor execution when making a response that is either or not in line with the expectations. In the comparison of LRP magnitude across task conditions, we found a greater LRPprep when large rewards were more likely, reflecting a greater motor preparation to obtain larger rewards. We also found a greater LRPexec when large rewards were presented unexpectedly, suggesting a greater motor effort placed for executing a correct movement when presented with large rewards. In the analysis of mPFC theta, we found a greater theta power prior to performing an unexpected than expected response, indicating its contribution in response conflict resolution. Collectively, these results demonstrate an optimized motor control to maximize rewards under the dynamic changes of real-life environment.

show abstract

Section: Resultsmentioning

confidence: 99%

Reward and expectancy effects on neural signals of motor preparation and execution

Chen

Berg

Kwak

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…A detailed tutorial explaining the particle swarm algorithm along with practical examples can be found in [ 65 ]. The advantage of using the particle swarm algorithm over traditional gradient-based methods is that it does not require a good initial guess for the parameters that can be initialised to random values within their bounds, and its results are less sensitive to the initial guess [ 66 , 67 , 68 , 69 , 70 ]. However, optimisation is computationally expensive and it can take a longer time to converge to a solution when compared to a traditional gradient-based algorithm.…”

Section: Materials and Methodsmentioning

confidence: 99%

On Nonlinear Regression for Trends in Split-Belt Treadmill Training

et al. 2020

Self Cite

View full text Add to dashboard Cite

Single and double exponential models fitted to step length symmetry series are used to evaluate the timecourse of adaptation and de-adaptation in instrumented split-belt treadmill tasks. Whilst the nonlinear regression literature has developed substantially over time, the split-belt treadmill training literature has not been fully utilising the fruits of these developments. In this research area, the current methods of model fitting and evaluation have three significant limitations: (i) optimisation algorithms that are used for model fitting require a good initial guess for regression parameters; (ii) the coefficient of determination (R2) is used for comparing and evaluating models, yet it is considered to be an inadequate measure of fit for nonlinear regression; and, (iii) inference is based on comparison of the confidence intervals for the regression parameters that are obtained under the untested assumption that the nonlinear model has a good linear approximation. In this research, we propose a transformed set of parameters with a common language interpretation that is relevant to split-belt treadmill training for both the single and double exponential models. We propose parameter bounds for the exponential models which allow the use of particle swarm optimisation for model fitting without an initial guess for the regression parameters. For model evaluation and comparison, we propose the use of residual plots and Akaike’s information criterion (AIC). A method for obtaining confidence intervals that does not require the assumption of a good linear approximation is also suggested. A set of MATLAB (MathWorks, Inc., Natick, MA, USA) functions developed in order to apply these methods are also presented. Single and double exponential models are fitted to both the group-averaged and participant step length symmetry series in an experimental dataset generating new insights into split-belt treadmill training. The proposed methods may be useful for research involving analysis of gait symmetry with instrumented split-belt treadmills. Moreover, the demonstration of the suggested statistical methods on an experimental dataset may help the uptake of these methods by a wider community of researchers that are interested in timecourse of motor training.

show abstract

“…These results may also inform research into the development of automated labelling methods that can extract key features of ERPs. Rashid and colleagues [42] recently developed an automated method to identify MRCP features and highlighted the need for a reliable gold standard to which automated algorithms can be compared [63,64]. The present study is the first step towards providing reliable data for the comparison of manual labelling methods by EEG experts, with automated algorithms.…”

Section: Future Recommendationsmentioning

confidence: 96%

“…The morphology of the epochs was quantified using the cosine similarity index. This was defined as the similarity of a single epoch from a participant compared to the average of all 50 epochs from the same participant, which was considered a representation of the expected MRCP characteristics [42],…”

Section: Secondary Analysis: Epoch Selectionmentioning

confidence: 99%

Intra- and Inter-Rater Reliability of Manual Feature Extraction Methods in Movement Related Cortical Potential Analysis

Alder

Signal

Rashid

et al. 2020

Sensors

Self Cite

View full text Add to dashboard Cite

Event related potentials (ERPs) provide insight into the neural activity generated in response to motor, sensory and cognitive processes. Despite the increasing use of ERP data in clinical research little is known about the reliability of human manual ERP labelling methods. Intra-rater and inter-rater reliability were evaluated in five electroencephalography (EEG) experts who labelled the peak negativity of averaged movement related cortical potentials (MRCPs) derived from thirty datasets. Each dataset contained 50 MRCP epochs from healthy people performing cued voluntary or imagined movement, or people with stroke performing cued voluntary movement. Reliability was assessed using the intraclass correlation coefficient and standard error of measurement. Excellent intra- and inter-rater reliability was demonstrated in the voluntary movement conditions in healthy people and people with stroke. In comparison reliability in the imagined condition was low to moderate. Post-hoc secondary epoch analysis revealed that the morphology of the signal contributed to the consistency of epoch inclusion; potentially explaining the differences in reliability seen across conditions. Findings from this study may inform future research focused on developing automated labelling methods for ERP feature extraction and call to the wider community of researchers interested in utilizing ERPs as a measure of neurophysiological change or in the delivery of EEG-driven interventions.

show abstract

Automated Labeling of Movement- Related Cortical Potentials Using Segmented Regression

Cited by 13 publications

References 37 publications

Reward and expectancy effects on neural signals of motor preparation and execution

Reward and expectancy effects on neural signals of motor preparation and execution

On Nonlinear Regression for Trends in Split-Belt Treadmill Training

Intra- and Inter-Rater Reliability of Manual Feature Extraction Methods in Movement Related Cortical Potential Analysis

Contact Info

Product

Resources

About