Miniscope v4 Electrical Noise

[Ashley Choi]

Hello, 

We recently set up the Miniscope v4 system in our lab. We record miniscope imaging along with EEG/EMG of our mice and had not had any issues with the v3 system. However, with the Miniscope v4, we are getting electrical noise that is introduced in our EEG/EMG recording when the miniscope is also hooked up/running. This noise is only present when the miniscope is running and goes away once it is turned off. Does anyone know what might be causing this or has anyone had a similar issue? 

Thanks! 

Ashley 

[Daniel Aharoni]

Hi Ashley,

This is very interesting. Can you tell me a bit more about the noise you are seeing? Frequency? Does it scale with distance to the V4 Miniscope? Can you upload a screenshot of the noise?

[Nat Kinsky]

Hi All,

I also encountered this issue. Currently recording with a v4 miniscope (rat PFC) and a 32 channel NeuroNexus linear probe (rat HPC) with and Intan 1024 channel system and OpenEphys software. The probe and camera are approximately 7-8 mm apart.  The miniscope I am using is v4.41, ordered from LabMaker, which I self assembled.  Like Ashley, I did not see any such noise in a previous experiment with electrodes and a v3 miniscope in the same locations.

For me, this noise does NOT arise when I first connect the miniscope up and see the red light turn on. It emerges immediately upon hitting run. See attached screenshots (note that the channels are not ordered properly in these shots).  It stays on even after I close the software and only goes away when physically disconnect the USB cable from the DAQ box.

The noise is very high frequency and amplitude and completely swamps my signal. However, the good news is that if I low-pass filter around 1000-1250 Hz I can partially recover signal, enough to see oscillations like ripples and theta, and even perhaps some multi-unit activity! Though it is still much noisier than before and I am not optimistic I will be able to discern any single units out. 

At first I thought this might be the LED. However, I did a quick mock up with a 32 channel Intan headstage unconnected to a probe and a different v4 miniscope and the signal remains fine when I move the headstage pins close to the LED. However, when I move the pins into the vicinity of the EWL (~5-10mm away) I see crazy, high frequency noise pop up.  A super quick-and-dirty attempt to build a faraday cage around the EWL with copper tape (see attached) seems like it helped block the noise: with the tape in place I only saw the noise pop up when I was physically touching the pins the to tape.  Sorry, not shots of this yet.

I am hoping to do a more thorough test of this engineering fix soon and will keep folks posted. In the meantime, any thoughts? Could this originate from the same issue where the EWL introduces row noise to the recordings (https://github.com/Aharoni-Lab/Miniscope-v4/wiki/Removing-Horizontal-Noise-from-Recordings)? I have yet to do a frequency analysis on this data but could easily characterize the noise in the next week or two.

Nat

[Daniel Aharoni]

Hi Nat,

Yep this definitely sounds like noise generated by the EWL driver and EWL. We have some ideas to better shield this portion of the Miniscope for future iterations but your solution of adding some copper tape seems to be helping a bit. The EWL gets driven at around 5KHz with a 0 to 80V signal so no surprise it is showing up in your ephys channels... especially since the 2 traces running to the EWL aren't well grounded. Also, the EWL drive powers up in standby mode and isn't enabled until the software connects it... this is why you don't see the noise right away.

As we get around to dealing with this noise I'll post here with updates. 

[Nat Kinsky]

Thanks Daniel,

You anticipated my question about when the EWL boots up - hadn't had time to investigate that myself.

Glad to hear you have some ideas for fixes.  I will also post any progress I make to this thread.

Nat

[Nat Kinsky]

Hi Daniel,

Quick update. I got around to doing a power spectrum analysis on the ephys data and confirmed that the main source of electrical noise seems to be coming from the EWL. See the attached power spectrum showing a huge peak ~4800Hz (x-axis = frequency, y-axis = power).

However, I still see a bunch of noise even when I filter this signal out online. Upon further investigation, there seems to be a bunch of 60Hz noise (and its associated harmonics at 180Hz, 300Hz, 420Hz, etc.) too:

I have verified that this 60Hz noise is not present without the camera connected.

Any idea where this 60Hz noise might come from? I didn't notice it until now because the EWL noise was so much more obvious, but it does seem to be hindering unit detection a bit.  If there's a potential culprit component I could try to shield that as well to mitigate this issue.

Nat