Kasper Eskelund
Research Subjects
Kasper Eskelund has a total of 38 co-authors. Here are the top co-authors by number of shared publications:
- Karen-Inge Karstoft
Department of Psychology, University of Copenhagen, Copenhagen, Denmark
5 shared publications - Tobias Søren Andersen
Department of Applied Mathematics and Computer Science, Technical University of Denmark (DTU), Kgs. Lyngby, 2800, Denmark
5 shared publications - Florine L. Bachmann
Eriksholm Research Centre, Snekkersten, Denmark
4 shared publications - Hamish Innes-Brown
Eriksholm Research Centre, Snekkersten, Denmark
4 shared publications - Joshua P. Kulasingham
Automatic Control, Department of Electrical Engineering, Linköping University
4 shared publications - Søren Bo Andersen
Research and Knowledge Centre, the Danish Veterans Centre, Ringsted, Denmark
4 shared publications
Latest Publications
- Open Access
- 26 February 2026
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 21 (2) , e0333716
Study protocol for a randomized clinical pilot trial investigating feasibility and efficacy of augmenting a virtual reality-assisted intervention targeting auditory verbal hallucinations with biofeedback: The Neuro-VR study
Introduction: Auditory Verbal Hallucinations (AVH) are among the most frequent and severe symptoms in schizophrenia and related psychotic disorders. Virtual Reality (VR)-assisted interventions have emerged, demonstrating promising potential in reducing AVH severity. This treatment approach may be challenged with regards to feasibility, particularly when therapeutically managing the anxiety-related reactions associated with AVH. This pilot study evaluates the feasibility and acceptability of augmenting VR-assisted therapy with real-time biofeedback to address these challenges. The integration of biofeedback enables continuous adaptation of therapy based on physiological responses while allowing participants to train self-regulation of these parameters. Methods: Neuro-VR is a randomized clinical pilot trial utilizing a mixed-methods design. Thirty participants with schizophrenia spectrum disorders and AVH will be randomized to either eight sessions of VR-assisted therapy or eight sessions of VR-assisted therapy augmented with real-time biofeedback. Assessments will be conducted at baseline and post-treatment. Outcome measures include both clinical metrics, electroencephalogram recordings, and qualitative interviews to evaluate feasibility, acceptability, and potential treatment effects of the combined approach. Discussion: This study will explore whether integrating biofeedback into VR-assisted therapy enhances personalization, supports emotion regulation, and improves tolerability. The findings will provide preliminary evidence on the utility of physiological markers to guide VR-based interventions for AVH and inform the development of individualized, effective treatments for patients with schizophrenia. Trial registration: ClinicalTrials.gov NCT06628323 .
- 26 February 2026
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 21 (2) , e0333716
Open AccessStudy protocol for a randomized clinical pilot trial investigating feasibility and efficacy of augmenting a virtual reality-assisted intervention targeting auditory verbal hallucinations with biofeedback: The Neuro-VR study
Introduction: Auditory Verbal Hallucinations (AVH) are among the most frequent and severe symptoms in schizophrenia and related psychotic disorders. Virtual Reality (VR)-assisted interventions have emerged, demonstrating promising potential in reducing AVH severity. This treatment approach may be challenged with regards to feasibility, particularly when therapeutically managing the anxiety-related reactions associated with AVH. This pilot study evaluates the feasibility and acceptability of augmenting VR-assisted therapy with real-time biofeedback to address these challenges. The integration of biofeedback enables continuous adaptation of therapy based on physiological responses while allowing participants to train self-regulation of these parameters. Methods: Neuro-VR is a randomized clinical pilot trial utilizing a mixed-methods design. Thirty participants with schizophrenia spectrum disorders and AVH will be randomized to either eight sessions of VR-assisted therapy or eight sessions of VR-assisted therapy augmented with real-time biofeedback. Assessments will be conducted at baseline and post-treatment. Outcome measures include both clinical metrics, electroencephalogram recordings, and qualitative interviews to evaluate feasibility, acceptability, and potential treatment effects of the combined approach. Discussion: This study will explore whether integrating biofeedback into VR-assisted therapy enhances personalization, supports emotion regulation, and improves tolerability. The findings will provide preliminary evidence on the utility of physiological markers to guide VR-based interventions for AVH and inform the development of individualized, effective treatments for patients with schizophrenia. Trial registration: ClinicalTrials.gov NCT06628323 .
- Open Access
- 10 November 2025
- Published by Society for Neuroscience in eneuro
- Vol. 12 (11)
Visual Speech Reduces Cognitive Effort as Measured by EEG Theta Power and Pupil Dilation
Listening effort reflects the cognitive and motivational resources allocated to speech comprehension, particularly under challenging conditions. Visual cues are known to enhance speech perception, potentially by reducing the cognitive demands of the task. However, the neurophysiological mechanisms underlying this facilitation, especially in terms of effort-related changes, remain unclear. In this study, we combined pupillometry and electroencephalography (EEG) to investigate how visual speech cues modulate cognitive effort during speech recognition. Twenty-two participants (seven females) performed a speech-in-noise task under three modalities: (1) auditory-only, (2) audiovisual, and (3) visual-only. Task difficulty was manipulated via signal-to-noise ratio (SNR) in the first two modalities. Firstly, we found an inverted U-shape relationship between pupil dilation and frontal midline theta with SNR for audiovisual and auditory-only speech, consistent with prior models of effort allocation. Secondly, we observed the SNR at which the neurophysiological measures peaked was at a lower SNR for audiovisual speech. Surprisingly, we found pupil dilation to be larger overall in audiovisual speech, while frontal midline theta did not show differences in either modality. These findings highlight the complexity of interpreting physiological markers of effort and suggest that visual cues may alter the temporal dynamics or resource allocation strategies during speech processing. Our results support the extension of auditory-based models of listening effort to audiovisual contexts and underscore the value of integrating multimodal neurophysiological measures to better understand the cognitive and neural mechanisms of effortful listening.
- 10 November 2025
- Published by Society for Neuroscience in eneuro
- Vol. 12 (11)
Open AccessVisual Speech Reduces Cognitive Effort as Measured by EEG Theta Power and Pupil Dilation
Listening effort reflects the cognitive and motivational resources allocated to speech comprehension, particularly under challenging conditions. Visual cues are known to enhance speech perception, potentially by reducing the cognitive demands of the task. However, the neurophysiological mechanisms underlying this facilitation, especially in terms of effort-related changes, remain unclear. In this study, we combined pupillometry and electroencephalography (EEG) to investigate how visual speech cues modulate cognitive effort during speech recognition. Twenty-two participants (seven females) performed a speech-in-noise task under three modalities: (1) auditory-only, (2) audiovisual, and (3) visual-only. Task difficulty was manipulated via signal-to-noise ratio (SNR) in the first two modalities. Firstly, we found an inverted U-shape relationship between pupil dilation and frontal midline theta with SNR for audiovisual and auditory-only speech, consistent with prior models of effort allocation. Secondly, we observed the SNR at which the neurophysiological measures peaked was at a lower SNR for audiovisual speech. Surprisingly, we found pupil dilation to be larger overall in audiovisual speech, while frontal midline theta did not show differences in either modality. These findings highlight the complexity of interpreting physiological markers of effort and suggest that visual cues may alter the temporal dynamics or resource allocation strategies during speech processing. Our results support the extension of auditory-based models of listening effort to audiovisual contexts and underscore the value of integrating multimodal neurophysiological measures to better understand the cognitive and neural mechanisms of effortful listening.
- Open Access
Study protocol for a randomized clinical pilot trial investigating feasibility and efficacy of augmenting a virtual reality-assisted intervention targeting auditory verbal hallucinations with biofeedback: the Neuro-VR study
Introduction Auditory Verbal Hallucinations (AVH) are among the most frequent and severe symptoms in schizophrenia and related psychotic disorders. Virtual Reality (VR)-assisted interventions have emerged, demonstrating promising potential in reducing AVH severity. This treatment approach may be challenged with regards to feasibility, particularly when therapeutically managing the anxiety-related reactions associated with AVH. This pilot study evaluates the feasibility and acceptability of augmenting VR-assisted therapy with real-time biofeedback to address these challenges. The integration of biofeedback enables continuous adaptation of therapy based on physiological responses while allowing participants to train self-regulation of these parameters. Methods Neuro-VR is a randomized clinical pilot trial utilizing a mixed-methods design. Thirty participants with schizophrenia spectrum disorders and AVH will be randomized to either eight sessions of VR-assisted therapy or eight sessions of VR-assisted therapy augmented with real-time biofeedback. Assessments will be conducted at baseline and post-treatment. Outcome measures include both clinical metrics, electroencephalogram recordings, and qualitative interviews to evaluate feasibility, acceptability, and potential treatment effects of the combined approach. Discussion This study will explore whether integrating biofeedback into VR-assisted therapy enhances personalization, supports emotion regulation, and improves tolerability. The findings will provide preliminary evidence on the utility of physiological markers to guide VR-based interventions for AVH and inform the development of individualized, effective treatments for patients with schizophrenia. Trial registration ClinicalTrials.gov, NCT06628323 (https://clinicaltrials.gov/study/NCT06628323). Registered August 19, 2024.
Open AccessStudy protocol for a randomized clinical pilot trial investigating feasibility and efficacy of augmenting a virtual reality-assisted intervention targeting auditory verbal hallucinations with biofeedback: the Neuro-VR study
Introduction Auditory Verbal Hallucinations (AVH) are among the most frequent and severe symptoms in schizophrenia and related psychotic disorders. Virtual Reality (VR)-assisted interventions have emerged, demonstrating promising potential in reducing AVH severity. This treatment approach may be challenged with regards to feasibility, particularly when therapeutically managing the anxiety-related reactions associated with AVH. This pilot study evaluates the feasibility and acceptability of augmenting VR-assisted therapy with real-time biofeedback to address these challenges. The integration of biofeedback enables continuous adaptation of therapy based on physiological responses while allowing participants to train self-regulation of these parameters. Methods Neuro-VR is a randomized clinical pilot trial utilizing a mixed-methods design. Thirty participants with schizophrenia spectrum disorders and AVH will be randomized to either eight sessions of VR-assisted therapy or eight sessions of VR-assisted therapy augmented with real-time biofeedback. Assessments will be conducted at baseline and post-treatment. Outcome measures include both clinical metrics, electroencephalogram recordings, and qualitative interviews to evaluate feasibility, acceptability, and potential treatment effects of the combined approach. Discussion This study will explore whether integrating biofeedback into VR-assisted therapy enhances personalization, supports emotion regulation, and improves tolerability. The findings will provide preliminary evidence on the utility of physiological markers to guide VR-based interventions for AVH and inform the development of individualized, effective treatments for patients with schizophrenia. Trial registration ClinicalTrials.gov, NCT06628323 (https://clinicaltrials.gov/study/NCT06628323). Registered August 19, 2024.
- Top Cited PapersOpen Access
- 8 February 2024
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 19 (2) , e0297826
Predictors for estimating subcortical EEG responses to continuous speech
Perception of sounds and speech involves structures in the auditory brainstem that rapidly process ongoing auditory stimuli. The role of these structures in speech processing can be investigated by measuring their electrical activity using scalp-mounted electrodes. However, typical analysis methods involve averaging neural responses to many short repetitive stimuli that bear little relevance to daily listening environments. Recently, subcortical responses to more ecologically relevant continuous speech were detected using linear encoding models. These methods estimate the temporal response function (TRF), which is a regression model that minimises the error between the measured neural signal and a predictor derived from the stimulus. Using predictors that model the highly non-linear peripheral auditory system may improve linear TRF estimation accuracy and peak detection. Here, we compare predictors from both simple and complex peripheral auditory models for estimating brainstem TRFs on electroencephalography (EEG) data from 24 participants listening to continuous speech. We also investigate the data length required for estimating subcortical TRFs, and find that around 12 minutes of data is sufficient for clear wave V peaks (>3 dB SNR) to be seen in nearly all participants. Interestingly, predictors derived from simple filterbank-based models of the peripheral auditory system yield TRF wave V peak SNRs that are not significantly different from those estimated using a complex model of the auditory nerve, provided that the nonlinear effects of adaptation in the auditory system are appropriately modelled. Crucially, computing predictors from these simpler models is more than 50 times faster compared to the complex model. This work paves the way for efficient modelling and detection of subcortical processing of continuous speech, which may lead to improved diagnosis metrics for hearing impairment and assistive hearing technology.
- 8 February 2024
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 19 (2) , e0297826
Top Cited PapersOpen AccessPredictors for estimating subcortical EEG responses to continuous speech
Perception of sounds and speech involves structures in the auditory brainstem that rapidly process ongoing auditory stimuli. The role of these structures in speech processing can be investigated by measuring their electrical activity using scalp-mounted electrodes. However, typical analysis methods involve averaging neural responses to many short repetitive stimuli that bear little relevance to daily listening environments. Recently, subcortical responses to more ecologically relevant continuous speech were detected using linear encoding models. These methods estimate the temporal response function (TRF), which is a regression model that minimises the error between the measured neural signal and a predictor derived from the stimulus. Using predictors that model the highly non-linear peripheral auditory system may improve linear TRF estimation accuracy and peak detection. Here, we compare predictors from both simple and complex peripheral auditory models for estimating brainstem TRFs on electroencephalography (EEG) data from 24 participants listening to continuous speech. We also investigate the data length required for estimating subcortical TRFs, and find that around 12 minutes of data is sufficient for clear wave V peaks (>3 dB SNR) to be seen in nearly all participants. Interestingly, predictors derived from simple filterbank-based models of the peripheral auditory system yield TRF wave V peak SNRs that are not significantly different from those estimated using a complex model of the auditory nerve, provided that the nonlinear effects of adaptation in the auditory system are appropriately modelled. Crucially, computing predictors from these simpler models is more than 50 times faster compared to the complex model. This work paves the way for efficient modelling and detection of subcortical processing of continuous speech, which may lead to improved diagnosis metrics for hearing impairment and assistive hearing technology.
- Open Access
- 1 January 2024
- Published by SAGE Publications in Trends in Hearing
- Vol. 28
Extending Subcortical EEG Responses to Continuous Speech to the Sound-Field
The auditory brainstem response (ABR) is a valuable clinical tool for objective hearing assessment, which is conventionally detected by averaging neural responses to thousands of short stimuli. Progressing beyond these unnatural stimuli, brainstem responses to continuous speech presented via earphones have been recently detected using linear temporal response functions (TRFs). Here, we extend earlier studies by measuring subcortical responses to continuous speech presented in the sound-field, and assess the amount of data needed to estimate brainstem TRFs. Electroencephalography (EEG) was recorded from 24 normal hearing participants while they listened to clicks and stories presented via earphones and loudspeakers. Subcortical TRFs were computed after accounting for non-linear processing in the auditory periphery by either stimulus rectification or an auditory nerve model. Our results demonstrated that subcortical responses to continuous speech could be reliably measured in the sound-field. TRFs estimated using auditory nerve models outperformed simple rectification, and 16 minutes of data was sufficient for the TRFs of all participants to show clear wave V peaks for both earphones and sound-field stimuli. Subcortical TRFs to continuous speech were highly consistent in both earphone and sound-field conditions, and with click ABRs. However, sound-field TRFs required slightly more data (16 minutes) to achieve clear wave V peaks compared to earphone TRFs (12 minutes), possibly due to effects of room acoustics. By investigating subcortical responses to sound-field speech stimuli, this study lays the groundwork for bringing objective hearing assessment closer to real-life conditions, which may lead to improved hearing evaluations and smart hearing technologies.
- 1 January 2024
- Published by SAGE Publications in Trends in Hearing
- Vol. 28
Open AccessExtending Subcortical EEG Responses to Continuous Speech to the Sound-Field
The auditory brainstem response (ABR) is a valuable clinical tool for objective hearing assessment, which is conventionally detected by averaging neural responses to thousands of short stimuli. Progressing beyond these unnatural stimuli, brainstem responses to continuous speech presented via earphones have been recently detected using linear temporal response functions (TRFs). Here, we extend earlier studies by measuring subcortical responses to continuous speech presented in the sound-field, and assess the amount of data needed to estimate brainstem TRFs. Electroencephalography (EEG) was recorded from 24 normal hearing participants while they listened to clicks and stories presented via earphones and loudspeakers. Subcortical TRFs were computed after accounting for non-linear processing in the auditory periphery by either stimulus rectification or an auditory nerve model. Our results demonstrated that subcortical responses to continuous speech could be reliably measured in the sound-field. TRFs estimated using auditory nerve models outperformed simple rectification, and 16 minutes of data was sufficient for the TRFs of all participants to show clear wave V peaks for both earphones and sound-field stimuli. Subcortical TRFs to continuous speech were highly consistent in both earphone and sound-field conditions, and with click ABRs. However, sound-field TRFs required slightly more data (16 minutes) to achieve clear wave V peaks compared to earphone TRFs (12 minutes), possibly due to effects of room acoustics. By investigating subcortical responses to sound-field speech stimuli, this study lays the groundwork for bringing objective hearing assessment closer to real-life conditions, which may lead to improved hearing evaluations and smart hearing technologies.
- Open Access
Extending Subcortical EEG Responses to Continuous Speech to the Sound-Field
The auditory brainstem response (ABR) is a valuable clinical tool for objective hearing assessment, which is conventionally detected by averaging neural responses to thousands of short stimuli. Progressing beyond these unnatural stimuli, brainstem responses to continuous speech presented via earphones have been recently detected using linear temporal response functions (TRFs). Here, we extend earlier studies by measuring subcortical responses to continuous speech presented in the sound-field, and assess the amount of data needed to estimate brainstem TRFs. Electroencephalography (EEG) was recorded from 24 normal hearing participants while they listened to clicks and stories presented via earphones and loudspeakers. Subcortical TRFs were computed after accounting for non-linear processing in the auditory periphery by either stimulus rectification or an auditory nerve model. Our results demonstrated that subcortical responses to continuous speech could be reliably measured in the sound-field. TRFs estimated using auditory nerve models outperformed simple rectification, and 16 minutes of data was sufficient for the TRFs of all participants to show clear wave V peaks for both earphones and sound-field stimuli. Subcortical TRFs to continuous speech were highly consistent in both earphone and sound-field conditions, and with click ABRs. However, sound-field TRFs required slightly more data (16 minutes) to achieve clear wave V peaks compared to earphone TRFs (12 minutes), possibly due to effects of room acoustics. By investigating subcortical responses to sound-field speech stimuli, this study lays the groundwork for bringing objective hearing assessment closer to real-life conditions, which may lead to improved hearing evaluations and smart hearing technologies.
Open AccessExtending Subcortical EEG Responses to Continuous Speech to the Sound-Field
The auditory brainstem response (ABR) is a valuable clinical tool for objective hearing assessment, which is conventionally detected by averaging neural responses to thousands of short stimuli. Progressing beyond these unnatural stimuli, brainstem responses to continuous speech presented via earphones have been recently detected using linear temporal response functions (TRFs). Here, we extend earlier studies by measuring subcortical responses to continuous speech presented in the sound-field, and assess the amount of data needed to estimate brainstem TRFs. Electroencephalography (EEG) was recorded from 24 normal hearing participants while they listened to clicks and stories presented via earphones and loudspeakers. Subcortical TRFs were computed after accounting for non-linear processing in the auditory periphery by either stimulus rectification or an auditory nerve model. Our results demonstrated that subcortical responses to continuous speech could be reliably measured in the sound-field. TRFs estimated using auditory nerve models outperformed simple rectification, and 16 minutes of data was sufficient for the TRFs of all participants to show clear wave V peaks for both earphones and sound-field stimuli. Subcortical TRFs to continuous speech were highly consistent in both earphone and sound-field conditions, and with click ABRs. However, sound-field TRFs required slightly more data (16 minutes) to achieve clear wave V peaks compared to earphone TRFs (12 minutes), possibly due to effects of room acoustics. By investigating subcortical responses to sound-field speech stimuli, this study lays the groundwork for bringing objective hearing assessment closer to real-life conditions, which may lead to improved hearing evaluations and smart hearing technologies.
- Open Access
Predictors for estimating subcortical EEG responses to continuous speech
Perception of sounds and speech involves structures in the auditory brainstem that rapidly process ongoing auditory stimuli. The role of these structures in speech processing can be investigated by measuring their electrical activity using scalp-mounted electrodes. However, typical analysis methods involve averaging neural responses to many short repetitive stimuli that bear little relevance to daily listening environments. Recently, subcortical responses to more ecologically relevant continuous speech were detected using linear encoding models. These methods estimate the temporal response function (TRF), which is a regression model that minimises the error between the measured neural signal and a predictor derived from the stimulus. Using predictors that model the highly non-linear peripheral auditory system may improve linear TRF estimation accuracy and peak detection. Here, we compare predictors from both simple and complex peripheral auditory models for estimating brainstem TRFs on electroencephalography (EEG) data from 24 participants listening to continuous speech. We also discuss the data length required for estimating subcortical TRFs with clear peaks. Interestingly, predictors from simple models resulted in TRFs that were similar to those estimated using complex models, and were much faster to compute. This work paves the way for efficient modelling and detection of subcortical processing of continuous speech, which may lead to improved diagnosis metrics for hearing impairment and assistive hearing technology.
Open AccessPredictors for estimating subcortical EEG responses to continuous speech
Perception of sounds and speech involves structures in the auditory brainstem that rapidly process ongoing auditory stimuli. The role of these structures in speech processing can be investigated by measuring their electrical activity using scalp-mounted electrodes. However, typical analysis methods involve averaging neural responses to many short repetitive stimuli that bear little relevance to daily listening environments. Recently, subcortical responses to more ecologically relevant continuous speech were detected using linear encoding models. These methods estimate the temporal response function (TRF), which is a regression model that minimises the error between the measured neural signal and a predictor derived from the stimulus. Using predictors that model the highly non-linear peripheral auditory system may improve linear TRF estimation accuracy and peak detection. Here, we compare predictors from both simple and complex peripheral auditory models for estimating brainstem TRFs on electroencephalography (EEG) data from 24 participants listening to continuous speech. We also discuss the data length required for estimating subcortical TRFs with clear peaks. Interestingly, predictors from simple models resulted in TRFs that were similar to those estimated using complex models, and were much faster to compute. This work paves the way for efficient modelling and detection of subcortical processing of continuous speech, which may lead to improved diagnosis metrics for hearing impairment and assistive hearing technology.
- Open Access
- 9 November 2022
- Published by IOP Publishing in Journal of Neural Engineering
- Vol. 19 (6) , 066005
Resting-state EEG functional connectivity predicts post-traumatic stress disorder subtypes in veterans
Objective. Post-traumatic stress disorder (PTSD) is highly heterogeneous, and identification of quantifiable biomarkers that could pave the way for targeted treatment remains a challenge. Most previous electroencephalography (EEG) studies on PTSD have been limited to specific handpicked features, and their findings have been highly variable and inconsistent. Therefore, to disentangle the role of promising EEG biomarkers, we developed a machine learning framework to investigate a wide range of commonly used EEG biomarkers in order to identify which features or combinations of features are capable of characterizing PTSD and potential subtypes. Approach. We recorded 5 min of eyes-closed and 5 min of eyes-open resting-state EEG from 202 combat-exposed veterans (53% with probable PTSD and 47% combat-exposed controls). Multiple spectral, temporal, and connectivity features were computed and logistic regression, random forest, and support vector machines with feature selection methods were employed to classify PTSD. To obtain robust results, we performed repeated two-layer cross-validation to test on an entirely unseen test set. Main results. Our classifiers obtained a balanced test accuracy of up to 62.9% for predicting PTSD patients. In addition, we identified two subtypes within PTSD: one where EEG patterns were similar to those of the combat-exposed controls, and another that were characterized by increased global functional connectivity. Our classifier obtained a balanced test accuracy of 79.4% when classifying this PTSD subtype from controls, a clear improvement compared to predicting the whole PTSD group. Interestingly, alpha connectivity in the dorsal and ventral attention network was particularly important for the prediction, and these connections were positively correlated with arousal symptom scores, a central symptom cluster of PTSD. Significance. Taken together, the novel framework presented here demonstrates how unsupervised subtyping can delineate heterogeneity and improve machine learning prediction of PTSD, and may pave the way for better identification of quantifiable biomarkers.
- 9 November 2022
- Published by IOP Publishing in Journal of Neural Engineering
- Vol. 19 (6) , 066005
Open AccessResting-state EEG functional connectivity predicts post-traumatic stress disorder subtypes in veterans
Objective. Post-traumatic stress disorder (PTSD) is highly heterogeneous, and identification of quantifiable biomarkers that could pave the way for targeted treatment remains a challenge. Most previous electroencephalography (EEG) studies on PTSD have been limited to specific handpicked features, and their findings have been highly variable and inconsistent. Therefore, to disentangle the role of promising EEG biomarkers, we developed a machine learning framework to investigate a wide range of commonly used EEG biomarkers in order to identify which features or combinations of features are capable of characterizing PTSD and potential subtypes. Approach. We recorded 5 min of eyes-closed and 5 min of eyes-open resting-state EEG from 202 combat-exposed veterans (53% with probable PTSD and 47% combat-exposed controls). Multiple spectral, temporal, and connectivity features were computed and logistic regression, random forest, and support vector machines with feature selection methods were employed to classify PTSD. To obtain robust results, we performed repeated two-layer cross-validation to test on an entirely unseen test set. Main results. Our classifiers obtained a balanced test accuracy of up to 62.9% for predicting PTSD patients. In addition, we identified two subtypes within PTSD: one where EEG patterns were similar to those of the combat-exposed controls, and another that were characterized by increased global functional connectivity. Our classifier obtained a balanced test accuracy of 79.4% when classifying this PTSD subtype from controls, a clear improvement compared to predicting the whole PTSD group. Interestingly, alpha connectivity in the dorsal and ventral attention network was particularly important for the prediction, and these connections were positively correlated with arousal symptom scores, a central symptom cluster of PTSD. Significance. Taken together, the novel framework presented here demonstrates how unsupervised subtyping can delineate heterogeneity and improve machine learning prediction of PTSD, and may pave the way for better identification of quantifiable biomarkers.
- 1 December 2020
- Published by American Psychological Association (APA) in Traumatology
- Vol. 26 (4) , 361-369
An explorative clinical pilot study into the effect of service dogs on chronic posttraumatic stress disorder.
- 1 December 2020
- Published by American Psychological Association (APA) in Traumatology
- Vol. 26 (4) , 361-369
An explorative clinical pilot study into the effect of service dogs on chronic posttraumatic stress disorder.