The Artifact Subspace Reconstruction (ASR) for EEG Signal Correction. A Comparative Study

Plechawska–Wójcik, Małgorzata; Kaczorowska, Monika; Zapała, Dariusz

doi:10.1007/978-3-319-99996-8_12

Cited by 61 publications

(46 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In both cases, the data are processed in chunks of 500 ms. As a result, ASR has a very short processing delay and relatively low computational complexity. It is well-suited for online applications and it has been shown to perform very well on EEG data in different recording conditions (Arad et al, 2018; Kim and Kim, 2018; Plechawska-Wojcik et al, 2019).…”

Section: Introductionmentioning

confidence: 99%

A Riemannian Modification of Artifact Subspace Reconstruction for EEG Artifact Handling

Blum

Jacobsen

Bleichner

et al. 2019

Front. Hum. Neurosci.

132

View full text Add to dashboard Cite

Artifact Subspace Reconstruction (ASR) is an adaptive method for the online or offline correction of artifacts comprising multichannel electroencephalography (EEG) recordings. It repeatedly computes a principal component analysis (PCA) on covariance matrices to detect artifacts based on their statistical properties in the component subspace. We adapted the existing ASR implementation by using Riemannian geometry for covariance matrix processing. EEG data that were recorded on smartphone in both outdoors and indoors conditions were used for evaluation ( N = 27). A direct comparison between the original ASR and Riemannian ASR (rASR) was conducted for three performance measures: reduction of eye-blinks (sensitivity), improvement of visual-evoked potentials (VEPs) (specificity), and computation time (efficiency). Compared to ASR, our rASR algorithm performed favorably on all three measures. We conclude that rASR is suitable for the offline and online correction of multichannel EEG data acquired in laboratory and in field conditions.

show abstract

Section: Introductionmentioning

confidence: 99%

A Riemannian Modification of Artifact Subspace Reconstruction for EEG Artifact Handling

Blum

Jacobsen

Bleichner

et al. 2019

Front. Hum. Neurosci.

132

View full text Add to dashboard Cite

show abstract

“…For data cleaning, EEGLAB plugin clean_rawdata() was applied including Artifact Subspace Reconstruction was applied to reduce high-amplitude artifacts (Blum et al, 2019;Chang et al, 2018Chang et al, , 2020Gabard-Durnam et al, 2018;Kothe and Makeig, 2013;Mullen et al, 2015;Plechawska-Wojcik M, 2019). The parameters used were: flat line removal, 10 s; electrode correlation, 0.7; ASR, 20; window rejection, 0.5.…”

Section: Eeg Recording and Preprocessingmentioning

confidence: 99%

Unique contributions of sensory discrimination and gamma synchronization deficits to cognitive, clinical, and psychosocial functional impairments in schizophrenia

Koshiyama

Miyakoshi

Michael

et al. 2020

Preprint

View full text Add to dashboard Cite

BackgroundSchizophrenia patients show widespread deficits in neurocognitive, clinical and psychosocial functioning. Mismatch negativity (MMN) and gamma-band auditory steady-state response (ASSR) are robust biomarkers for domains of neuropsychiatric disorders that are impaired in schizophrenia patients and are separately associated with cognitive dysfunction, negative symptom severity and psychosocial disability. Although these measures of early auditory information processing are conceptually linked, it is unclear, whether these measures are redundant or account for unique variance in important outcome measures. In this study, we aimed to determine whether MMN and gamma-band ASSR are associated with cognitive, clinical, and functional variables and, if so, whether they account for shared vs. non-shared variance in those important domains.MethodsMultiple regression analyses with MMN, gamma-band ASSR and clinical measures were performed in large cohorts of schizophrenia outpatients (N=428) and healthy comparison subjects (N=283).ResultsReduced MMN (d = 0.67), gamma-band ASSR (d = –0.40), and lower cognitive function were confirmed in schizophrenia patients compared to healthy comparison subjects. Regression analyses revealed that both MMN and gamma-band ASSR have significant unique associations with tasks measuring of working memory, and daily functioning in schizophrenia patients.ConclusionThese findings suggest that MMN and ASSR measures are non-redundant and complementary measures. Studies are needed to clarify the neural substrates of MMN and gamma-band ASSR in order to improve our understanding of pathophysiology of schizophrenia and accelerate their use in the development of novel therapeutic interventions.

show abstract

“…(4) Removes the bursts via Artifact Subspace Reconstructionapplies PCA decomposition to the channels in sliding window and rejects and reconstructs the components for which the standard deviation differs from the most representative part of the signal. The standard deviation threshold was set at 5 (Mullen et al, 2013;Plechawska-Wojcik et al, 2018). 5Removes the unrepaired windows-a sliding window of 1 s and 66% overlap deletes the windows that contain more than four "bad channels."…”

Section: Data Processingmentioning

confidence: 99%

Predicting Sex From EEG: Validity and Generalizability of Deep-Learning-Based Interpretable Classifier

et al. 2020

View full text Add to dashboard Cite

Explainable artificial intelligence holds a great promise for neuroscience and plays an important role in the hypothesis generation process. We follow-up a recent machine learning-oriented study that constructed a deep convolutional neural network to automatically identify biological sex from EEG recordings in healthy individuals and highlighted the discriminative role of beta-band power. If generalizing, this finding would be relevant not only theoretically by pointing to some specific neurobiological sexual dimorphisms, but potentially also as a relevant confound in quantitative EEG diagnostic practice. To put this finding to test, we assess whether the automatic identification of biological sex generalizes to another dataset, particularly in the presence of a psychiatric disease, by testing the hypothesis of higher beta power in women compared to men on 134 patients suffering from Major Depressive Disorder. Moreover, we construct ROC curves and compare the performance of the classifiers in determining sex both before and after the antidepressant treatment. We replicate the observation of a significant difference in beta-band power between men and women, providing classification accuracy of nearly 77%. The difference was consistent across the majority of electrodes, however multivariate classification models did not generally improve the performance. Similar results were observed also after the antidepressant treatment (classification accuracy above 70%), further supporting the robustness of the initial finding.

show abstract

The Artifact Subspace Reconstruction (ASR) for EEG Signal Correction. A Comparative Study

Cited by 61 publications

References 27 publications

A Riemannian Modification of Artifact Subspace Reconstruction for EEG Artifact Handling

A Riemannian Modification of Artifact Subspace Reconstruction for EEG Artifact Handling

Unique contributions of sensory discrimination and gamma synchronization deficits to cognitive, clinical, and psychosocial functional impairments in schizophrenia

Predicting Sex From EEG: Validity and Generalizability of Deep-Learning-Based Interpretable Classifier

Contact Info

Product

Resources

About