Principal Component Regression on Motor Evoked Potential in Single-Pulse Transcranial Magnetic Stimulation

Nguyen, Dao T A; Rissanen, Saara M.; Julkunen, Petro; Kallioniemi, Elisa; Karjalainen, Pasi A.

doi:10.1109/tnsre.2019.2923724

Cited by 14 publications

(17 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One might be tempted to use simpler metrics than ApEn to quantify the abnormality of the morphology. A different metric could be counting the number of peaks (or, equivalently, the number of zero-line crossings) to count the number of phases (Nguyen et al, 2019 ). This works less well: the number of peaks feature has AUC = 0.76 on average on the 1-votes of the training set, compared to AUC = 0.92 for ApEn.…”

Section: Resultsmentioning

confidence: 99%

“…To significantly lower their dimensionality and to capture their salient information, evoked potential time series (EPTS) are often condensed into a single EP score (Schlaeger et al, 2016 ). Recent work has also investigated in reducing the dimensionality of an EP by using principal component regression (Nguyen et al, 2019 ). The EP score is a composite score, for which three variables are commonly extracted from the EPTS: latency, amplitude, and presence of morphological abnormality (Schlaeger et al, 2016 ).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deciphering the Morphology of Motor Evoked Potentials

Yperman

Becker

Valkenborg

et al. 2020

Front. Neuroinform.

View full text Add to dashboard Cite

Motor Evoked Potentials (MEPs) are used to monitor disability progression in multiple sclerosis (MS). Their morphology plays an important role in this process. Currently, however, there is no clear definition of what constitutes a normal or abnormal morphology. To address this, five experts independently labeled the morphology (normal or abnormal) of the same set of 1,000 MEPs. The intra-and inter-rater agreement between the experts indicates they agree on the concept of morphology, but differ in their choice of threshold between normal and abnormal morphology. We subsequently performed an automated extraction of 5,943 time series features from the MEPs to identify a valid proxy for morphology, based on the provided labels. To do this, we compared the cross-validation performances of one-dimensional logistic regression models fitted to each of the features individually. We find that the approximate entropy (ApEn) feature can accurately reproduce the majority-vote labels. The performance of this feature is evaluated on an independent test set by comparing to the majority vote of the neurologists, obtaining an AUC score of 0.92. The model slightly outperforms the average neurologist at reproducing the neurologists consensus-vote labels. We can conclude that MEP morphology can be consistently defined by pooling the interpretations from multiple neurologists and that ApEn is a valid continuous score for this. Having an objective and reproducible MEP morphological abnormality score will allow researchers to include this feature in their models, without manual annotation becoming a bottleneck. This is crucial for large-scale, multi-center datasets. An exploratory analysis on a large single-center dataset shows that ApEn is potentially clinically useful. Introducing an automated, objective, and reproducible definition of morphology could help overcome some of the barriers that are currently obstructing broad adoption of evoked potentials in daily care and patient follow-up, such as standardization of measurements between different centers, and formulating guidelines for clinical use.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Deciphering the Morphology of Motor Evoked Potentials

Yperman

Becker

Valkenborg

et al. 2020

Front. Neuroinform.

View full text Add to dashboard Cite

show abstract

“…Hence, all mentioned algorithms (AHTE and SM) give a relatively accurate estimation of the MEP signal latency with higher PTP amplitudes (exceeding 100 µV). However, a lower estimation (precision) rate is documented for the latency of the MEP signals with lower PTP amplitudes (i.e., lower than 100 µV) [18][19][20][21][22]. Therefore, there is a need for a better algorithm that will improve the latency estimation of MEP signals with lower PTP amplitudes.…”

Section: B Algorithms Based On Statistical Measuresmentioning

confidence: 99%

“…Firstly, MEPs are interpreted concerning the performance of actions (resting-state vs. execution); secondly, to probe the physiology of the motor cortex (i.e., pharmacological manipulations or investigation of cortical excitability in the context of psychiatric or neurological illnesses) and thirdly, to probe the physiology that occurs outside of the M1. The amplitude and the latency of the MEP signal are the most used features for quantifying MEP signals [17] - [19]. Although the MEP amplitude varies significantly in patients with pathologies, as well as in healthy subjects [18], [20] - [22], the peak-to-peak (PTP) amplitude defined as the difference between the maximum and minimum value of the MEP response, represents an accurate indicator for estimating the MEP amplitude oscillation [23]- [26].…”

Section: Introductionmentioning

confidence: 99%

A Novel Latency Estimation Algorithm of Motor Evoked Potential Signals

et al. 2020

View full text Add to dashboard Cite

Evaluation of motor evoked potential (MEP) signals elicited by transcranial magnetic stimulation (TMS) over the motor cortex provide a measure of cortico-motor excitability at the time of stimulation. In the research and clinical medical practice, the MEP latency is a relevant neurophysiological parameter to determine conduction time for neural impulses from the cortex to peripheral muscles. State changes at different levels of signal propagation through the neural tissue can significantly influence MEP latency, based on which different medical diagnoses can be issued. This study aims to present the Squared Hard Threshold Estimator (SHTE), which is a novel and improved algorithm for MEP latency estimation. Analyses presented in the paper were based on the SHTE algorithm, which was efficiently applied to a large number of MEP signals recorded from hand muscles. The SHTE algorithm was compared with other prominent methods such as the absolute hard threshold estimation (AHTE) algorithm, the statistical measures (SM) algorithm, and manual assessment. Results obtained in terms of robustness test and statistical analysis show that the proposed SHTE algorithm is reliable in estimating MEP latency, especially for the MEP signals having peak-to-peak (PTP) amplitudes lower than one hundred microvolts. Compared with the AHTE and SM algorithms, the SHTE shows a lower percentage deviation index in MEP latency estimation of the MEP signals with the PTP amplitudes lower than one hundred microvolts. Hence, the proposed SHTE algorithm represents an improved armamentarium in automatic MEP latency estimation.

show abstract

“…However, the correlation between main and quality variables has not been considered. Then, an external analysis-based regression model [33] was constructed for robust soft sensing, in which principal component regression (PCR) [34], [35] and PLS were employed to construct the regression model. Remarkably, in reference [33], the corresponding nonlinear tools were introduced to resolve the nonlinear relationships although they were not described in detail.…”

Section: Introductionmentioning

confidence: 99%

Fault Detection in Multimode Processes Through Improved Nonlinear External Analysis Regression

Sun

2020

IEEE Access

View full text Add to dashboard Cite

External analysis serve as a cogent approach for multimode process detection in recent years. However, external analysis approach may not detect faults well because of the imprecise extraction of relations between variables. This paper proposes a fault detection approach for multimode processes, called improved nonlinear external analysis regression. External regression models between external variables and main/quality variables are established to remove mode-change-related information in the main/quality variables, ensuring that the following work is performed under a single mode. The remaining information in the main and quality variables is employed to develop an internal regression model for fault detection. Compared with existing approaches, the proposed approach has the following advantages: (1) In external regression models, applying kernel orthogonal projections to latent structures resulted in a relatively smaller number of loadings, reduced model complexity and, not least, efficient extraction of mode-change-related information in main and quality variables. (2) Internal regression model has the capacity to improve separation performance of output-related information and output-unrelated information. (3) Two comprehensive and perspicuous detection statistics are designed to accurately detect process faults. To experimentally verify the stability and superiority of the method, it is applied to a penicillin fermentation process for fault detection. INDEX TERMS Fault detection, kernel orthogonal projections to latent structures, improved nonlinear external analysis regression, penicillin fermentation process, multimode processes.

show abstract

Principal Component Regression on Motor Evoked Potential in Single-Pulse Transcranial Magnetic Stimulation

Cited by 14 publications

References 36 publications

Deciphering the Morphology of Motor Evoked Potentials

Deciphering the Morphology of Motor Evoked Potentials

A Novel Latency Estimation Algorithm of Motor Evoked Potential Signals

Fault Detection in Multimode Processes Through Improved Nonlinear External Analysis Regression

Contact Info

Product

Resources

About