Stability of mental motor-imagery classification in EEG depends on the choice of classifier model and experiment design, but not on signal preprocessing
IntroductionModern consciousness research has developed diagnostic tests to improve the diagnostic accuracy of different states of consciousness via electroencephalography (EEG)-based mental motor imagery (MI), which is still challenging and lacks a consensus on how to best analyse MI EEG-data. An o...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-04-01
|
| Series: | Frontiers in Computational Neuroscience |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fncom.2023.1142948/full |
| _version_ | 1797839153082138624 |
|---|---|
| author | Martin Justinus Rosenfelder Martin Justinus Rosenfelder Myra Spiliopoulou Burkhard Hoppenstedt Rüdiger Pryss Rüdiger Pryss Patrick Fissler Patrick Fissler Patrick Fissler Mario della Piedra Walter Mario della Piedra Walter Iris-Tatjana Kolassa Andreas Bender Andreas Bender |
| author_facet | Martin Justinus Rosenfelder Martin Justinus Rosenfelder Myra Spiliopoulou Burkhard Hoppenstedt Rüdiger Pryss Rüdiger Pryss Patrick Fissler Patrick Fissler Patrick Fissler Mario della Piedra Walter Mario della Piedra Walter Iris-Tatjana Kolassa Andreas Bender Andreas Bender |
| author_sort | Martin Justinus Rosenfelder |
| collection | DOAJ |
| description | IntroductionModern consciousness research has developed diagnostic tests to improve the diagnostic accuracy of different states of consciousness via electroencephalography (EEG)-based mental motor imagery (MI), which is still challenging and lacks a consensus on how to best analyse MI EEG-data. An optimally designed and analyzed paradigm must detect command-following in all healthy individuals, before it can be applied in patients, e.g., for the diagnosis of disorders of consciousness (DOC).MethodsWe investigated the effects of two important steps in the raw signal preprocessing on predicting participant performance (F1) and machine-learning classifier performance (area-under-curve, AUC) in eight healthy individuals, that are based solely on MI using high-density EEG (HD-EEG): artifact correction (manual correction with vs. without Independent Component Analysis [ICA]), region of interest (ROI; motor area vs. whole brain), and machine-learning algorithm (support-vector machine [SVM] vs. k-nearest neighbor [KNN]).ResultsResults revealed no significant effects of artifact correction and ROI on predicting participant performance (F1) and classifier performance (AUC) scores (all ps > 0.05) in the SVM classification model. In the KNN model, ROI had a significant influence on the classifier performance [F(1,8.939) = 7.585, p = 0.023]. There was no evidence for artifact correction and ROI selection changing the prediction of participants performance and classifier performance in EEG-based mental MI if using SVM-based classification (71–100% correct classifications across different signal preprocessing methods). The variance in the prediction of participant performance was significantly higher when the experiment started with a resting-state compared to a mental MI task block [X2(1) = 5.849, p = 0.016].DiscussionOverall, we could show that classification is stable across different modes of EEG signal preprocessing when using SVM models. Exploratory analysis gave a hint toward potential effects of the sequence of task execution on the prediction of participant performance, which should be taken into account in future studies. |
| first_indexed | 2024-04-09T15:53:34Z |
| format | Article |
| id | doaj.art-efc1d6d7d7b1435c9c6e7fbdaa5ca6f4 |
| institution | Directory Open Access Journal |
| issn | 1662-5188 |
| language | English |
| last_indexed | 2024-04-09T15:53:34Z |
| publishDate | 2023-04-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Computational Neuroscience |
| spelling | doaj.art-efc1d6d7d7b1435c9c6e7fbdaa5ca6f42023-04-26T04:42:59ZengFrontiers Media S.A.Frontiers in Computational Neuroscience1662-51882023-04-011710.3389/fncom.2023.11429481142948Stability of mental motor-imagery classification in EEG depends on the choice of classifier model and experiment design, but not on signal preprocessingMartin Justinus Rosenfelder0Martin Justinus Rosenfelder1Myra Spiliopoulou2Burkhard Hoppenstedt3Rüdiger Pryss4Rüdiger Pryss5Patrick Fissler6Patrick Fissler7Patrick Fissler8Mario della Piedra Walter9Mario della Piedra Walter10Iris-Tatjana Kolassa11Andreas Bender12Andreas Bender13Institute of Psychology and Education, Clinical and Biological Psychology, Ulm University, Ulm, GermanyTherapiezentrum Burgau, Burgau, GermanyKnowledge Management and Discovery Lab, Faculty of Computer Science, Otto-von-Guericke-University Magdeburg, Magdeburg, GermanyInstitute of Databases and Information Systems, Ulm University, Ulm, GermanyInstitute of Databases and Information Systems, Ulm University, Ulm, GermanyInstitute of Clinical Epidemiology and Biometry, University of Würzburg, Würzburg, GermanyInstitute of Psychology and Education, Clinical and Biological Psychology, Ulm University, Ulm, GermanyPsychiatric Services Thurgau, Münsterlingen, SwitzerlandUniversity Hospital for Psychiatry and Psychotherapy, Paracelsus Medical University, Salzburg, AustriaTherapiezentrum Burgau, Burgau, GermanyFaculty 2: Biology/Chemistry, University of Bremen, Bremen, GermanyInstitute of Psychology and Education, Clinical and Biological Psychology, Ulm University, Ulm, GermanyTherapiezentrum Burgau, Burgau, GermanyDepartment of Neurology, University of Munich, Munich, GermanyIntroductionModern consciousness research has developed diagnostic tests to improve the diagnostic accuracy of different states of consciousness via electroencephalography (EEG)-based mental motor imagery (MI), which is still challenging and lacks a consensus on how to best analyse MI EEG-data. An optimally designed and analyzed paradigm must detect command-following in all healthy individuals, before it can be applied in patients, e.g., for the diagnosis of disorders of consciousness (DOC).MethodsWe investigated the effects of two important steps in the raw signal preprocessing on predicting participant performance (F1) and machine-learning classifier performance (area-under-curve, AUC) in eight healthy individuals, that are based solely on MI using high-density EEG (HD-EEG): artifact correction (manual correction with vs. without Independent Component Analysis [ICA]), region of interest (ROI; motor area vs. whole brain), and machine-learning algorithm (support-vector machine [SVM] vs. k-nearest neighbor [KNN]).ResultsResults revealed no significant effects of artifact correction and ROI on predicting participant performance (F1) and classifier performance (AUC) scores (all ps > 0.05) in the SVM classification model. In the KNN model, ROI had a significant influence on the classifier performance [F(1,8.939) = 7.585, p = 0.023]. There was no evidence for artifact correction and ROI selection changing the prediction of participants performance and classifier performance in EEG-based mental MI if using SVM-based classification (71–100% correct classifications across different signal preprocessing methods). The variance in the prediction of participant performance was significantly higher when the experiment started with a resting-state compared to a mental MI task block [X2(1) = 5.849, p = 0.016].DiscussionOverall, we could show that classification is stable across different modes of EEG signal preprocessing when using SVM models. Exploratory analysis gave a hint toward potential effects of the sequence of task execution on the prediction of participant performance, which should be taken into account in future studies.https://www.frontiersin.org/articles/10.3389/fncom.2023.1142948/fullmotor-imageryencephalographymachine learningsupport vector machinek-nearest neighborsclassification |
| spellingShingle | Martin Justinus Rosenfelder Martin Justinus Rosenfelder Myra Spiliopoulou Burkhard Hoppenstedt Rüdiger Pryss Rüdiger Pryss Patrick Fissler Patrick Fissler Patrick Fissler Mario della Piedra Walter Mario della Piedra Walter Iris-Tatjana Kolassa Andreas Bender Andreas Bender Stability of mental motor-imagery classification in EEG depends on the choice of classifier model and experiment design, but not on signal preprocessing Frontiers in Computational Neuroscience motor-imagery encephalography machine learning support vector machine k-nearest neighbors classification |
| title | Stability of mental motor-imagery classification in EEG depends on the choice of classifier model and experiment design, but not on signal preprocessing |
| title_full | Stability of mental motor-imagery classification in EEG depends on the choice of classifier model and experiment design, but not on signal preprocessing |
| title_fullStr | Stability of mental motor-imagery classification in EEG depends on the choice of classifier model and experiment design, but not on signal preprocessing |
| title_full_unstemmed | Stability of mental motor-imagery classification in EEG depends on the choice of classifier model and experiment design, but not on signal preprocessing |
| title_short | Stability of mental motor-imagery classification in EEG depends on the choice of classifier model and experiment design, but not on signal preprocessing |
| title_sort | stability of mental motor imagery classification in eeg depends on the choice of classifier model and experiment design but not on signal preprocessing |
| topic | motor-imagery encephalography machine learning support vector machine k-nearest neighbors classification |
| url | https://www.frontiersin.org/articles/10.3389/fncom.2023.1142948/full |
| work_keys_str_mv | AT martinjustinusrosenfelder stabilityofmentalmotorimageryclassificationineegdependsonthechoiceofclassifiermodelandexperimentdesignbutnotonsignalpreprocessing AT martinjustinusrosenfelder stabilityofmentalmotorimageryclassificationineegdependsonthechoiceofclassifiermodelandexperimentdesignbutnotonsignalpreprocessing AT myraspiliopoulou stabilityofmentalmotorimageryclassificationineegdependsonthechoiceofclassifiermodelandexperimentdesignbutnotonsignalpreprocessing AT burkhardhoppenstedt stabilityofmentalmotorimageryclassificationineegdependsonthechoiceofclassifiermodelandexperimentdesignbutnotonsignalpreprocessing AT rudigerpryss stabilityofmentalmotorimageryclassificationineegdependsonthechoiceofclassifiermodelandexperimentdesignbutnotonsignalpreprocessing AT rudigerpryss stabilityofmentalmotorimageryclassificationineegdependsonthechoiceofclassifiermodelandexperimentdesignbutnotonsignalpreprocessing AT patrickfissler stabilityofmentalmotorimageryclassificationineegdependsonthechoiceofclassifiermodelandexperimentdesignbutnotonsignalpreprocessing AT patrickfissler stabilityofmentalmotorimageryclassificationineegdependsonthechoiceofclassifiermodelandexperimentdesignbutnotonsignalpreprocessing AT patrickfissler stabilityofmentalmotorimageryclassificationineegdependsonthechoiceofclassifiermodelandexperimentdesignbutnotonsignalpreprocessing AT mariodellapiedrawalter stabilityofmentalmotorimageryclassificationineegdependsonthechoiceofclassifiermodelandexperimentdesignbutnotonsignalpreprocessing AT mariodellapiedrawalter stabilityofmentalmotorimageryclassificationineegdependsonthechoiceofclassifiermodelandexperimentdesignbutnotonsignalpreprocessing AT iristatjanakolassa stabilityofmentalmotorimageryclassificationineegdependsonthechoiceofclassifiermodelandexperimentdesignbutnotonsignalpreprocessing AT andreasbender stabilityofmentalmotorimageryclassificationineegdependsonthechoiceofclassifiermodelandexperimentdesignbutnotonsignalpreprocessing AT andreasbender stabilityofmentalmotorimageryclassificationineegdependsonthechoiceofclassifiermodelandexperimentdesignbutnotonsignalpreprocessing |