Praveen Sripad
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Praveen Sripad has a total of 102 co-authors. Here are the top co-authors by number of shared publications:
- Lucia Melloni
Predictive Brain, Research Center “One Health” Ruhr, University Alliance Ruhr, Faculty of Psychology, Ruhr-University Bochum, Bochum, Germany
5 shared publications - Rony Hirschhorn
Sagol School of Neuroscience, Tel Aviv University, Tel Aviv, Israel
5 shared publications - Jürgen Dammers
Institute of Neuroscience and Medicine (INM-4), Forschungszentrum Jülich GmbH, Jülich, Germany
4 shared publications - Niccolò Bonacchi
William James Center for Research, ISPA - Instituto Universitario, Lisbon, 1149-041, Portugal
4 shared publications - Tanya Brown
Neural Circuits, Consciousness and Cognition Research Group, Max Planck Institute for Empirical Aesthetics, Frankfurt am Main, Germany
4 shared publications - Alex Lepauvre
Neural Circuits, Consciousness and Cognition Research Group, Max Planck Institute for Empirical Aesthetics, Frankfurt am Main, Germany
3 shared publications
Latest Publications
- Open Access
- 29 May 2026
- Published by Springer Nature in Scientific Data
- Vol. 13 (1)
An open multi-center MEG-EEG dataset for studying conscious visual perception
Here, we present a large-scale, multi-center dataset of combined magnetoencephalographic (MEG) and electroencephalographic (EEG) recordings, along with eye-tracking data and high-resolution structural MRI (T1); complementing with iEEG and fMRI datasets that are shared in accompanying data papers. The data was obtained through an adversarial collaboration between advocates of two neuroscientific theories of consciousness: the Global Neuronal Workspace Theory and the Integrated Information Theory. The dataset includes recordings from 100 individuals (mean age 22.79 ± 3.59 years, 54 female, all right-handed) across two research centers (UK and China), using a standardized data collection protocol. During the experiment, participants were asked to perform a non-speeded Go/No-Go target detection task, during which they were exposed to visual stimuli from four distinct categories (faces, objects, letters, false fonts) presented at different orientations (front, left, right view), and for varying durations (0.5, 1.0, 1.5 s), under different task conditions. The quality of the data was assessed and organized according to the Brain Imaging Data Structure (BIDS). It is accompanied by extensive metadata to enhance reusability.
- 29 May 2026
- Published by Springer Nature in Scientific Data
- Vol. 13 (1)
Open AccessAn open multi-center MEG-EEG dataset for studying conscious visual perception
Here, we present a large-scale, multi-center dataset of combined magnetoencephalographic (MEG) and electroencephalographic (EEG) recordings, along with eye-tracking data and high-resolution structural MRI (T1); complementing with iEEG and fMRI datasets that are shared in accompanying data papers. The data was obtained through an adversarial collaboration between advocates of two neuroscientific theories of consciousness: the Global Neuronal Workspace Theory and the Integrated Information Theory. The dataset includes recordings from 100 individuals (mean age 22.79 ± 3.59 years, 54 female, all right-handed) across two research centers (UK and China), using a standardized data collection protocol. During the experiment, participants were asked to perform a non-speeded Go/No-Go target detection task, during which they were exposed to visual stimuli from four distinct categories (faces, objects, letters, false fonts) presented at different orientations (front, left, right view), and for varying durations (0.5, 1.0, 1.5 s), under different task conditions. The quality of the data was assessed and organized according to the Brain Imaging Data Structure (BIDS). It is accompanied by extensive metadata to enhance reusability.
- Open Access
- 7 May 2026
- Published by Springer Nature in Scientific Data
- Vol. 13 (1)
An open-access multi-site fMRI dataset for investigating conscious visual perception
We present a functional magnetic resonance (fMRI) dataset collected as part of an adversarial collaboration aimed at arbitrating between the Global Neuronal Workspace theory (GNWT) and the Integrated Information Theory (IIT) of consciousness. Participants (N = 118) were presented with suprathreshold visual stimuli belonging to four different categories (faces, objects, letters, false fonts) with three orientations (front, left, right view), and three durations (0.5, 1.0, 1.5 seconds). Participants were asked to identify infrequent targets that changed in each block, thereby rendering two categories task-relevant and two task-irrelevant. The simplicity of the experimental design and of the task given to the participants ensures that these data are broadly reusable. Besides testing predictions from other theories of consciousness, these data can be used to examine various aspects of visual processing. The anonymized data were converted to Brain Imaging Data Structure (BIDS), and can be easily accessed through a web platform or an API. The dataset contains quality reports, demographics, behavioral performance, and eye-tracking data. We also provide code for preprocessing and analyzing the data.
- 7 May 2026
- Published by Springer Nature in Scientific Data
- Vol. 13 (1)
Open AccessAn open-access multi-site fMRI dataset for investigating conscious visual perception
We present a functional magnetic resonance (fMRI) dataset collected as part of an adversarial collaboration aimed at arbitrating between the Global Neuronal Workspace theory (GNWT) and the Integrated Information Theory (IIT) of consciousness. Participants (N = 118) were presented with suprathreshold visual stimuli belonging to four different categories (faces, objects, letters, false fonts) with three orientations (front, left, right view), and three durations (0.5, 1.0, 1.5 seconds). Participants were asked to identify infrequent targets that changed in each block, thereby rendering two categories task-relevant and two task-irrelevant. The simplicity of the experimental design and of the task given to the participants ensures that these data are broadly reusable. Besides testing predictions from other theories of consciousness, these data can be used to examine various aspects of visual processing. The anonymized data were converted to Brain Imaging Data Structure (BIDS), and can be easily accessed through a web platform or an API. The dataset contains quality reports, demographics, behavioral performance, and eye-tracking data. We also provide code for preprocessing and analyzing the data.
- Open Access
- 23 May 2025
- Published by Springer Nature in Scientific Data
- Vol. 12 (1) , 1-14
Open multi-center intracranial electroencephalography dataset with task probing conscious visual perception
We introduce an intracranial EEG (iEEG) dataset collected as part of an adversarial collaboration between proponents of two theories of consciousness: Global Neuronal Workspace Theory and Integrated Information Theory. The data were recorded from 38 patients undergoing intracranial monitoring of epileptic seizures across three research centers using the same experimental protocol. Participants were presented with suprathreshold visual stimuli belonging to four different categories (faces, objects, letters, false fonts) in three orientations (front, left, right view), and for three durations (0.5, 1.0, 1.5 s). Participants engaged in a non-speeded Go/No-Go target detection task to identify infrequent targets with some stimuli becoming task-relevant and others task-irrelevant. Participants also engaged in a motor localizer task. The data were checked for its quality and converted to Brain Imaging Data Structure (BIDS). The de-identified dataset contains demographics, clinical information, electrode reconstruction, behavioral performance, and eye-tracking data. We also provide code to preprocess and analyze the data. This dataset holds promise for reuse in consciousness science and vision neuroscience to answer questions related to stimulus processing, target detection, and task-relevance, among many others.
- 23 May 2025
- Published by Springer Nature in Scientific Data
- Vol. 12 (1) , 1-14
Open AccessOpen multi-center intracranial electroencephalography dataset with task probing conscious visual perception
We introduce an intracranial EEG (iEEG) dataset collected as part of an adversarial collaboration between proponents of two theories of consciousness: Global Neuronal Workspace Theory and Integrated Information Theory. The data were recorded from 38 patients undergoing intracranial monitoring of epileptic seizures across three research centers using the same experimental protocol. Participants were presented with suprathreshold visual stimuli belonging to four different categories (faces, objects, letters, false fonts) in three orientations (front, left, right view), and for three durations (0.5, 1.0, 1.5 s). Participants engaged in a non-speeded Go/No-Go target detection task to identify infrequent targets with some stimuli becoming task-relevant and others task-irrelevant. Participants also engaged in a motor localizer task. The data were checked for its quality and converted to Brain Imaging Data Structure (BIDS). The de-identified dataset contains demographics, clinical information, electrode reconstruction, behavioral performance, and eye-tracking data. We also provide code to preprocess and analyze the data. This dataset holds promise for reuse in consciousness science and vision neuroscience to answer questions related to stimulus processing, target detection, and task-relevance, among many others.
- Top Cited PapersOpen Access
- 30 April 2025
- Published by Springer Nature in Nature
- Vol. 642 (8066) , 133-142
Adversarial testing of global neuronal workspace and integrated information theories of consciousness
Different theories explain how subjective experience arises from brain activity 1,2 . These theories have independently accrued evidence, but have not been directly compared 3 . Here we present an open science adversarial collaboration directly juxtaposing integrated information theory (IIT) 4,5 and global neuronal workspace theory (GNWT) 6–10 via a theory-neutral consortium 11–13 . The theory proponents and the consortium developed and preregistered the experimental design, divergent predictions, expected outcomes and interpretation thereof 12 . Human participants ( n = 256) viewed suprathreshold stimuli for variable durations while neural activity was measured with functional magnetic resonance imaging, magnetoencephalography and intracranial electroencephalography. We found information about conscious content in visual, ventrotemporal and inferior frontal cortex, with sustained responses in occipital and lateral temporal cortex reflecting stimulus duration, and content-specific synchronization between frontal and early visual areas. These results align with some predictions of IIT and GNWT, while substantially challenging key tenets of both theories. For IIT, a lack of sustained synchronization within the posterior cortex contradicts the claim that network connectivity specifies consciousness. GNWT is challenged by the general lack of ignition at stimulus offset and limited representation of certain conscious dimensions in the prefrontal cortex. These challenges extend to other theories of consciousness that share some of the predictions tested here 14–17 . Beyond challenging the theories, we present an alternative approach to advance cognitive neuroscience through principled, theory-driven, collaborative research and highlight the need for a quantitative framework for systematic theory testing and building.
- 30 April 2025
- Published by Springer Nature in Nature
- Vol. 642 (8066) , 133-142
Top Cited PapersOpen AccessAdversarial testing of global neuronal workspace and integrated information theories of consciousness
Different theories explain how subjective experience arises from brain activity 1,2 . These theories have independently accrued evidence, but have not been directly compared 3 . Here we present an open science adversarial collaboration directly juxtaposing integrated information theory (IIT) 4,5 and global neuronal workspace theory (GNWT) 6–10 via a theory-neutral consortium 11–13 . The theory proponents and the consortium developed and preregistered the experimental design, divergent predictions, expected outcomes and interpretation thereof 12 . Human participants ( n = 256) viewed suprathreshold stimuli for variable durations while neural activity was measured with functional magnetic resonance imaging, magnetoencephalography and intracranial electroencephalography. We found information about conscious content in visual, ventrotemporal and inferior frontal cortex, with sustained responses in occipital and lateral temporal cortex reflecting stimulus duration, and content-specific synchronization between frontal and early visual areas. These results align with some predictions of IIT and GNWT, while substantially challenging key tenets of both theories. For IIT, a lack of sustained synchronization within the posterior cortex contradicts the claim that network connectivity specifies consciousness. GNWT is challenged by the general lack of ignition at stimulus offset and limited representation of certain conscious dimensions in the prefrontal cortex. These challenges extend to other theories of consciousness that share some of the predictions tested here 14–17 . Beyond challenging the theories, we present an alternative approach to advance cognitive neuroscience through principled, theory-driven, collaborative research and highlight the need for a quantitative framework for systematic theory testing and building.
- Open Access
An adversarial collaboration to critically evaluate theories of consciousness
Summary: Different theories explain how subjective experience arises from brain activity 1,2 . These theories have independently accrued evidence, yet, confirmation bias and dependence on design choices hamper progress in the field 3 . Here, we present an open science adversarial collaboration which directly juxtaposes Integrated Information Theory (IIT) 4,5 and Global Neuronal Workspace Theory (GNWT) 6–10 , employing a theory-neutral consortium approach 11,12 . We investigate neural correlates of the content and duration of visual experience. The theory proponents and the consortium developed and preregistered the experimental design, divergent predictions, expected outcomes, and their interpretation 12 . 256 human subjects viewed suprathreshold stimuli for variable durations while neural activity was measured with functional magnetic resonance imaging, magnetoencephalography, and electrocorticography. We find information about conscious content in visual, ventro-temporal and inferior frontal cortex, with sustained responses in occipital and lateral temporal cortex reflecting stimulus duration, and content-specific synchronization between frontal and early visual areas. These results confirm some predictions of IIT and GNWT, while substantially challenging both theories: for IIT, a lack of sustained synchronization within posterior cortex contradicts the claim that network connectivity specifies consciousness. GNWT is challenged by the general lack of ignition at stimulus offset and limited representation of certain conscious dimensions in prefrontal cortex. Beyond challenging the theories themselves, we present an alternative approach to advance cognitive neuroscience through a principled, theory-driven, collaborative effort. We highlight the challenges to change people’s mind 13 and the need for a quantitative framework integrating evidence for systematic theory testing and building.
Open AccessAn adversarial collaboration to critically evaluate theories of consciousness
Summary: Different theories explain how subjective experience arises from brain activity 1,2 . These theories have independently accrued evidence, yet, confirmation bias and dependence on design choices hamper progress in the field 3 . Here, we present an open science adversarial collaboration which directly juxtaposes Integrated Information Theory (IIT) 4,5 and Global Neuronal Workspace Theory (GNWT) 6–10 , employing a theory-neutral consortium approach 11,12 . We investigate neural correlates of the content and duration of visual experience. The theory proponents and the consortium developed and preregistered the experimental design, divergent predictions, expected outcomes, and their interpretation 12 . 256 human subjects viewed suprathreshold stimuli for variable durations while neural activity was measured with functional magnetic resonance imaging, magnetoencephalography, and electrocorticography. We find information about conscious content in visual, ventro-temporal and inferior frontal cortex, with sustained responses in occipital and lateral temporal cortex reflecting stimulus duration, and content-specific synchronization between frontal and early visual areas. These results confirm some predictions of IIT and GNWT, while substantially challenging both theories: for IIT, a lack of sustained synchronization within posterior cortex contradicts the claim that network connectivity specifies consciousness. GNWT is challenged by the general lack of ignition at stimulus offset and limited representation of certain conscious dimensions in prefrontal cortex. Beyond challenging the theories themselves, we present an alternative approach to advance cognitive neuroscience through a principled, theory-driven, collaborative effort. We highlight the challenges to change people’s mind 13 and the need for a quantitative framework integrating evidence for systematic theory testing and building.
- Open Access
- 25 February 2021
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 16 (2) , e0247408
Conflict processing networks: A directional analysis of stimulus-response compatibilities using MEG
The suppression of distracting information in order to focus on an actual cognitive goal is a key feature of executive functions. The use of brain imaging methods to investigate the underlying neurobiological brain activations that occur during conflict processing have demonstrated a strong involvement of the fronto-parietal attention network (FPAN). Surprisingly, the directional interconnections, their time courses and activations at different frequency bands remain to be elucidated, and thus, this constitutes the focus of this study. The shared information flow between brain areas of the FPAN is provided for frequency bands ranging from the theta to the lower gamma band (4–40 Hz). We employed an adaptation of the Simon task utilizing Magnetoencephalography (MEG). Granger causality was applied to investigate interconnections between the active brain regions, as well as their directionality. Following stimulus onset, the middle frontal precentral cortex and superior parietal cortex were significantly activated during conflict processing in a time window of between 300 to 600ms. Important differences in causality were found across frequency bands between processing of conflicting stimuli in the left as compared to the right visual hemifield. The exchange of information from and to the FPAN was most prominent in the beta band. Moreover, the anterior cingulate cortex and the anterior insula represented key areas for conflict monitoring, either by receiving input from other areas of the FPAN or by generating output themselves. This indicates that the salience network is at least partly involved in processing conflict information. The present study provides detailed insights into the underlying neural mechanisms of the FPAN, especially regarding its temporal characteristics and directional interconnections.
- 25 February 2021
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 16 (2) , e0247408
Open AccessConflict processing networks: A directional analysis of stimulus-response compatibilities using MEG
The suppression of distracting information in order to focus on an actual cognitive goal is a key feature of executive functions. The use of brain imaging methods to investigate the underlying neurobiological brain activations that occur during conflict processing have demonstrated a strong involvement of the fronto-parietal attention network (FPAN). Surprisingly, the directional interconnections, their time courses and activations at different frequency bands remain to be elucidated, and thus, this constitutes the focus of this study. The shared information flow between brain areas of the FPAN is provided for frequency bands ranging from the theta to the lower gamma band (4–40 Hz). We employed an adaptation of the Simon task utilizing Magnetoencephalography (MEG). Granger causality was applied to investigate interconnections between the active brain regions, as well as their directionality. Following stimulus onset, the middle frontal precentral cortex and superior parietal cortex were significantly activated during conflict processing in a time window of between 300 to 600ms. Important differences in causality were found across frequency bands between processing of conflicting stimuli in the left as compared to the right visual hemifield. The exchange of information from and to the FPAN was most prominent in the beta band. Moreover, the anterior cingulate cortex and the anterior insula represented key areas for conflict monitoring, either by receiving input from other areas of the FPAN or by generating output themselves. This indicates that the salience network is at least partly involved in processing conflict information. The present study provides detailed insights into the underlying neural mechanisms of the FPAN, especially regarding its temporal characteristics and directional interconnections.
- Open Access
- 20 March 2020
- Published by Wiley in Case Reports in Neurological Medicine
- Vol. 2020, 1-8
Effect of Zolpidem in the Aftermath of Traumatic Brain Injury: An MEG Study
In the past two decades, many studies have shown the paradoxical efficacy of zolpidem, a hypnotic used to induce sleep, in transiently alleviating various disorders of consciousness such as traumatic brain injury (TBI), dystonia, and Parkinson’s disease. The mechanism of action of this effect of zolpidem is of great research interest. In this case study, we use magnetoencephalography (MEG) to investigate a fully conscious, ex-coma patient who suffered from neurological difficulties for a few years due to traumatic brain injury. For a few years after injury, the patient was under medication with zolpidem that drastically improved his symptoms. MEG recordings taken before and after zolpidem showed a reduction in power in the theta-alpha (4–12 Hz) and lower beta (15–20 Hz) frequency bands. An increase in power after zolpidem intake was found in the higher beta/lower gamma (20–43 Hz) frequency band. Source level functional connectivity measured using weighted-phase lag index showed changes after zolpidem intake. Stronger connectivity between left frontal and temporal brain regions was observed. We report that zolpidem induces a change in MEG resting power and functional connectivity in the patient. MEG is an informative and sensitive tool to detect changes in brain activity for TBI.
- 20 March 2020
- Published by Wiley in Case Reports in Neurological Medicine
- Vol. 2020, 1-8
Open AccessEffect of Zolpidem in the Aftermath of Traumatic Brain Injury: An MEG Study
In the past two decades, many studies have shown the paradoxical efficacy of zolpidem, a hypnotic used to induce sleep, in transiently alleviating various disorders of consciousness such as traumatic brain injury (TBI), dystonia, and Parkinson’s disease. The mechanism of action of this effect of zolpidem is of great research interest. In this case study, we use magnetoencephalography (MEG) to investigate a fully conscious, ex-coma patient who suffered from neurological difficulties for a few years due to traumatic brain injury. For a few years after injury, the patient was under medication with zolpidem that drastically improved his symptoms. MEG recordings taken before and after zolpidem showed a reduction in power in the theta-alpha (4–12 Hz) and lower beta (15–20 Hz) frequency bands. An increase in power after zolpidem intake was found in the higher beta/lower gamma (20–43 Hz) frequency band. Source level functional connectivity measured using weighted-phase lag index showed changes after zolpidem intake. Stronger connectivity between left frontal and temporal brain regions was observed. We report that zolpidem induces a change in MEG resting power and functional connectivity in the patient. MEG is an informative and sensitive tool to detect changes in brain activity for TBI.
- 27 October 2018
- Published by Wiley in Human Brain Mapping
- Vol. 42 (13) , 4122-4133
Comparison of EEG microstates with resting state fMRI and FDG‐PET measures in the default mode network via simultaneously recorded trimodal (PET/MR/EEG) data
Simultaneous trimodal positron emission tomography/magnetic resonance imaging/electroencephalography (PET/MRI/EEG) resting state (rs) brain data were acquired from 10 healthy male volunteers. The rs‐functional MRI (fMRI) metrics, such as regional homogeneity (ReHo), degree centrality (DC) and fractional amplitude of low‐frequency fluctuations (fALFFs), as well as 2‐[18F]fluoro‐2‐desoxy‐d‐glucose (FDG)‐PET standardised uptake value (SUV), were calculated and the measures were extracted from the default mode network (DMN) regions of the brain. Similarly, four microstates for each subject, showing the diverse functional states of the whole brain via topographical variations due to global field power (GFP), were estimated from artefact‐corrected EEG signals. In this exploratory analysis, the GFP of microstates was nonparametrically compared to rs‐fMRI metrics and FDG‐PET SUV measured in the DMN of the brain. The rs‐fMRI metrics (ReHO, fALFF) and FDG‐PET SUV did not show any significant correlations with any of the microstates. The DC metric showed a significant positive correlation with microstate C (rs = 0.73, p = .01). FDG‐PET SUVs indicate a trend for a negative correlation with microstates A, B and C. The positive correlation of microstate C with DC metrics suggests a functional relationship between cortical hubs in the frontal and occipital lobes. The results of this study suggest further exploration of this method in a larger sample and in patients with neuropsychiatric disorders. The aim of this exploratory pilot study is to lay the foundation for the development of such multimodal measures to be applied as biomarkers for diagnosis, disease staging, treatment response and monitoring of neuropsychiatric disorders.
- 27 October 2018
- Published by Wiley in Human Brain Mapping
- Vol. 42 (13) , 4122-4133
Comparison of EEG microstates with resting state fMRI and FDG‐PET measures in the default mode network via simultaneously recorded trimodal (PET/MR/EEG) data
Simultaneous trimodal positron emission tomography/magnetic resonance imaging/electroencephalography (PET/MRI/EEG) resting state (rs) brain data were acquired from 10 healthy male volunteers. The rs‐functional MRI (fMRI) metrics, such as regional homogeneity (ReHo), degree centrality (DC) and fractional amplitude of low‐frequency fluctuations (fALFFs), as well as 2‐[18F]fluoro‐2‐desoxy‐d‐glucose (FDG)‐PET standardised uptake value (SUV), were calculated and the measures were extracted from the default mode network (DMN) regions of the brain. Similarly, four microstates for each subject, showing the diverse functional states of the whole brain via topographical variations due to global field power (GFP), were estimated from artefact‐corrected EEG signals. In this exploratory analysis, the GFP of microstates was nonparametrically compared to rs‐fMRI metrics and FDG‐PET SUV measured in the DMN of the brain. The rs‐fMRI metrics (ReHO, fALFF) and FDG‐PET SUV did not show any significant correlations with any of the microstates. The DC metric showed a significant positive correlation with microstate C (rs = 0.73, p = .01). FDG‐PET SUVs indicate a trend for a negative correlation with microstates A, B and C. The positive correlation of microstate C with DC metrics suggests a functional relationship between cortical hubs in the frontal and occipital lobes. The results of this study suggest further exploration of this method in a larger sample and in patients with neuropsychiatric disorders. The aim of this exploratory pilot study is to lay the foundation for the development of such multimodal measures to be applied as biomarkers for diagnosis, disease staging, treatment response and monitoring of neuropsychiatric disorders.
- Open Access
- 2 May 2018
- Published by Wiley in Journal of Engineering
- Vol. 2018, 1-10
Deep Learning Approach for Automatic Classification of Ocular and Cardiac Artifacts in MEG Data
We propose an artifact classification scheme based on a combined deep and convolutional neural network (DCNN) model, to automatically identify cardiac and ocular artifacts from neuromagnetic data, without the need for additional electrocardiogram (ECG) and electrooculogram (EOG) recordings. From independent components, the model uses both the spatial and temporal information of the decomposed magnetoencephalography (MEG) data. In total, 7122 samples were used after data augmentation, in which task and nontask related MEG recordings from 48 subjects served as the database for this study. Artifact rejection was applied using the combined model, which achieved a sensitivity and specificity of 91.8% and 97.4%, respectively. The overall accuracy of the model was validated using a cross-validation test and revealed a median accuracy of 94.4%, indicating high reliability of the DCNN-based artifact removal in task and nontask related MEG experiments. The major advantages of the proposed method are as follows: (1) it is a fully automated and user independent workflow of artifact classification in MEG data; (2) once the model is trained there is no need for auxiliary signal recordings; (3) the flexibility in the model design and training allows for various modalities (MEG/EEG) and various sensor types.
- 2 May 2018
- Published by Wiley in Journal of Engineering
- Vol. 2018, 1-10
Open AccessDeep Learning Approach for Automatic Classification of Ocular and Cardiac Artifacts in MEG Data
We propose an artifact classification scheme based on a combined deep and convolutional neural network (DCNN) model, to automatically identify cardiac and ocular artifacts from neuromagnetic data, without the need for additional electrocardiogram (ECG) and electrooculogram (EOG) recordings. From independent components, the model uses both the spatial and temporal information of the decomposed magnetoencephalography (MEG) data. In total, 7122 samples were used after data augmentation, in which task and nontask related MEG recordings from 48 subjects served as the database for this study. Artifact rejection was applied using the combined model, which achieved a sensitivity and specificity of 91.8% and 97.4%, respectively. The overall accuracy of the model was validated using a cross-validation test and revealed a median accuracy of 94.4%, indicating high reliability of the DCNN-based artifact removal in task and nontask related MEG experiments. The major advantages of the proposed method are as follows: (1) it is a fully automated and user independent workflow of artifact classification in MEG data; (2) once the model is trained there is no need for auxiliary signal recordings; (3) the flexibility in the model design and training allows for various modalities (MEG/EEG) and various sensor types.