Magnetoencephalographic Virtual Recording: a novel diagnostic tool for concussionKeywords: traumatic brain injury, technique, magnetoencephalography, diagnosis, neurophysiological monitoringInteractive Manuscript
Ask Questions of this Manuscript:
What is the background behind your study?
There is no formal, objective neuroimaging modality for the diagnosis of concussion.
What is the purpose of your study?
We present a novel functional neuroimaging strategy using magnetoencephalography (MEG) Virtual Recording to provide objective, non-invasive, diagnostic data in concussion.
Describe your patient group.
With IRB approval, subjects were recruited from our university hospital neurotrauma service and from the general University population (controls). All subjects underwent MEG evaluation during performance of a language/spatial task. Ten subjects, 5 control subjects and 5 individuals with a history of concussion were enrolled in this pilot study.
Describe what you did.
Task performance by each subject produced 960 responses to 320 sentence stimuli. 0.3 seconds of MEG data preceding the response to each word and test figure presentation were extracted. The data from each subject was then classified using a rule constructed from the data from the other nine participants.
Describe your main findings.
Analysis of response times using multiple ANOVA showed significant differences (p < 10-4) between concussed and normal groups, demonstrating the sensitivity of the task. Leave 1 out linear classification of MEG measures enabled the correct classification of eight of 10 individuals as concussed vs non-concussed (p = 0.055) and 70% of the single trials (p < 10-10) as originating from concussed vs non-concussed individuals. Concussed patients showed increased activation in occipito-parietal and temporal regions.
Describe the main limitation of this study.
This is a retrospective study.
Describe your main conclusion.
These pilot findings are evidence of the utility of MEG Virtual Recordings coupled with a language/spatial task in diagnosing concussion.
Describe the importance of your findings and how they can be used by others.
With further refinements, this approach may represent a noninvasive test to diagnose concussion objectively and also monitor its resolution.
There is no formal, objective neuroimaging modality for the diagnosis of concussion.
We present a novel functional neuroimaging strategy using magnetoencephalography (MEG) Virtual Recording to provide objective, non-invasive, diagnostic data in concussion.
With IRB approval, subjects were recruited from our university hospital neurotrauma service and from the general University population (controls). All subjects underwent MEG evaluation during performance of a language/spatial task. Ten subjects, 5 control subjects and 5 individuals with a history of concussion were enrolled in this pilot study.
Task performance by each subject produced 960 responses to 320 sentence stimuli. 0.3 seconds of MEG data preceding the response to each word and test figure presentation were extracted. The data from each subject was then classified using a rule constructed from the data from the other nine participants.
Analysis of response times using multiple ANOVA showed significant differences (p < 10-4) between concussed and normal groups, demonstrating the sensitivity of the task. Leave 1 out linear classification of MEG measures enabled the correct classification of eight of 10 individuals as concussed vs non-concussed (p = 0.055) and 70% of the single trials (p < 10-10) as originating from concussed vs non-concussed individuals. Concussed patients showed increased activation in occipito-parietal and temporal regions.
This is a retrospective study.
These pilot findings are evidence of the utility of MEG Virtual Recordings coupled with a language/spatial task in diagnosing concussion.
With further refinements, this approach may represent a noninvasive test to diagnose concussion objectively and also monitor its resolution.
Project Roles: