Discerning true beta spindles on raw EEG

57 views
Skip to first unread message

Bar Lehmann

unread,
Apr 18, 2022, 1:27:57 PM4/18/22
to ISNR_Members_Forum
While looking at the raw EEG, I am having a lot of challenges with separating out beta spindles from EMG in certain cases. My understanding is the following: it seems quite clear if Fz and Cz, (which I understand to be sitting on bone, not on muscle) are involved in the spindles, they are probably real. My brainmaster amp. goes up to 70Hz and I always take filters off through the whole recording to ensure I can see all the way up to 70 and I look out for the biphasic spikes that are often calling cards for EMG. I try to get clients to reduce EMG in recording as much as possible. 

1) Should beta spindles reduce significantly during eyes closed (when I do see a significant reduction of them, I have a feeling often that the beta spindles must be false since why wouldn't they last in the EC)? 

2) Should beta spindles reduce very significantly during alpha trains, or does that suggest its EMG? 

3) If the beta spindles are not fronto-central, how suspicious should I be of them? in other words, how common is it to see beta spindles Non-frontocentrally?  

Any responses to any questions or other tips to indicate which beta spindles are real will be greatly appreciated. 

daniel.lane

unread,
Apr 28, 2022, 9:26:29 AM4/28/22
to ISNR_Members_Forum
Hi Bar, these are GREAT questions - andI have often thought the same. I often think I see beta spindles in temporal sites - but then think "no they can't be I was taught they only occur in frontal regions..." I hope that you get some great replies to these question ! Warm regards, Daniel.

Dave Siever

unread,
Apr 28, 2022, 11:51:02 AM4/28/22
to isnr_memb...@googlegroups.com

Single motor unit firing in the 15-20ish range can sometimes look a lot like beta, even with open filters. To ensure that it’s not in the reference, check to see that it does not build fairly evenly towards the midline. If so, check that the channels are not in phase. If they are in phase, then the corruption is in the reference. You can double-check this by doing a bipolar montage – which removes the references.

 

 

 

 

Dave Siever

- dancing in the dendrites!

 

Mind Alive Inc.

Edmonton, AB, Canada

800-661-6463

 

This is the youngest day of the rest of your life.

Now go do the youngest thing possible! 

--
You received this message because you are subscribed to the Google Groups "ISNR_Members_Forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to isnr_members_fo...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/isnr_members_forum/cb114b73-871d-4062-8632-c9b34e2d8499n%40googlegroups.com.

Robert Thatcher

unread,
Apr 28, 2022, 12:21:02 PM4/28/22
to ISNR_Members_Forum
Daniel,
   Beta spindles are a normal aspect of the human EEG and are present in all electrode locations.  Abnormal beta spindle activity depends on the duration, how often they occur and amplitude of the spindles and one can compare beta spindles to normative databases (both instantaneous and averaged over time) to better understand whether they are likely normal or not.  If highly deviant from a normal reference then what is important is the location and the spatial gradient of the deviant spindles, e.g., frontal and excecutive functioning, occipital and vision, central regions and movement, etc.  

As for muscle artifact, this is determined by the gradient of change in amplitude and location.  For example, all EMG is produced external to the brain by three sets of muscles, e.g., frontalis muscles that are maximum in Fp1/2 and decrease as a function of distance (e.g., Fp1/d>F3/4>C3/4) and temporalis muscles (T3/4 > C3/4) and neck muscles O1/2 > P3/4, etc.  Also, EMG is usually > 15 Hz.  What is important is the location and gradient of change in amplitude with distance from the EMG scalp source and keeping in mind that there are no muscles inside the skull.

This is what is commonly taught to medical students and graduate students using multiple examples, etc.  

Robert W. Thatcher, Ph.D., BCIA, QEEG-D, ECNS


--
You received this message because you are subscribed to the Google Groups "ISNR_Members_Forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to isnr_members_fo...@googlegroups.com.
To view this discussion on the web visit

daniel.lane

unread,
Apr 29, 2022, 8:08:07 AM4/29/22
to ISNR_Members_Forum
Hi Bob, that's a brilliant answer and very helpful, thank you so much :-)

Bar Lehmann

unread,
May 10, 2022, 8:23:32 AM5/10/22
to ISNR_Members_Forum
Dr. Thatcher - apart from the location, gradient, and amplitude, you mention the duration of the spindles which sounds quite important, are there any good guidelines you might point me to that help discern when/if they are more likely to be abnormal per the duration aspect, or is that something you would gauge solely using the database norms? 

Dave, does what you are saying re. the montage also apply to the global average just as well as the bipolar since it does away with the reference(s) too? 

Thanks very very much for your responses and looking forward to hear,
Bar

Robert Thatcher

unread,
May 10, 2022, 10:00:12 AM5/10/22
to ISNR_Members_Forum
Hi Bar,
    ANI has developed FDA registered lifespan norms from birth to 82 years of age for "Burst Metrics".  As seen in the Neuroguide screen capture "Click Report > Report Selections".  This includes the "Number of Bursts"; "Burst Amplitudes", "Burst Durations" and the "Burst Intervals".  These metrics use standard statistics based on a Bell Shaped curve and thus one can evaluate locations of burst metrics and see if a patient's burst metrics are + - 2 or 3 or 4 standard deviations outside of normal etc.   The location of the deviant burst metrics is still the most important factor, e.g., deviant occipital bursts and vision and reading, etc., temporal lobe deviance and memory and verbal functions, frontal lobe deviance and judgment and concentration, etc.

      Inline image


I hope that this is helpful,

Robert W. Thatcher, Ph.D., BCIA, QEEG-D, ECNS



Robert Thatcher

unread,
May 10, 2022, 12:50:07 PM5/10/22
to ISNR_Members_Forum
Bar,
  I forgot to mention that the Laplacian montage is also reference independent.  The Laplacian only evaluates the spatial gradient of EEG voltage change surrounding each scalp electrode for surface EEG independent of the reference. Also with sLORETA and swLORETA the spatial gradient around the center voxel of each Brodmann area or region of interest again independent of the reference electrode.

Robert W. Thatcher, Ph.D., BCIA, QEEG-D, ECNS



Reply all
Reply to author
Forward
0 new messages