Best Current Practice for Obtaining High Quality EEG Data During Simultaneous fMRI

Simultaneous electroencephalography (EEG) and functional Magnetic Resonance imaging (fMRI) is a powerful neuroimaging tool. However, the inside of an MRI scanner forms a difficult environment for EEG data recording and safety must be considered whenever operating EEG equipment inside a scanner. Here, we present an optimised EEG-fMRI data acquisition protocol.

The overall goal of this procedure is to set up Mr compatible EEG equipment inside an MRI scanner for simultaneous acquisition of high quality EEG and FMRI data. This is accomplished by first correctly setting up the EEG filters, sampling rate, and synchronization. The next step is to ensure that there are good connections between the EEG electrodes and the subject scalp.

The third step is to correctly position the EEG hardware around the MRI scanner. The final step is to optimally position the subject within the MR Scanner for standard E-E-G-F-M-R-I data acquisition via commercially available hardware. This procedure ensures that artifacts in the EEG data are minimized and optimally sampled, thus allowing maximal artifact removal via post-processing methods.

The main advantage of combining EEG and FMI data acquisition in E-E-G-F-M-I experiments is that this approach allows the simultaneous monitoring of the electrical and hemodynamic signals from the human brain. Glen Spencer, a PhD student from my laboratory, will be helping me demonstrate this experiment Prior to the subject's arrival. Set up the EEG equipment in the control room where the scanner operator will sit.

Connect the laptop computer to the EEG turn on the computer, which will record the EEG data and open brain vision recorder. Make sure the workspace for recording the data is set to the highest available temporal resolution. Next, set the filters.

AC coupling with a filter band ranging from 0.016 to 250 hertz is usually optimal. Set up the stimulus computer. This study utilizes visual stimuli.

Markers are read into the brain vision recorder at the beginning and end of each stimulation period. Now, confirm that the EEG and MRI scanner clocks are synchronized. First turn on the synchronization of the scanner and EEG clocks using the software control panel.

Then check on the synchronization if it is correct. The green dot icon and sync on marker will appear. Next, set up the MR scanner.

In this case, the body transmit RF coil and a 32 channel head Receive RF coil are in use. When possible, it is best to use a head size transmit coil as this minimizes the risk of RF heating of the EEG cap and associated cables. An access port on the head coil is useful.

It allows cables from the EEG cap to run along a straight path from the scanner. Now, check on the sequences. The FMRI sequence must use a slice tr that is a multiple of 200 microseconds, which is the EEG clock period.

Lastly, make one final check. The tall equipment is recording as expected. First, explain to the subject the purpose of the experiment and what will happen.

Next, measure the subject's head circumference for the cap size. Place the cap on the head, starting at the front of the head and pulling backwards. Position the cap correctly.

The CZ electrode needs to be directly between the nasion and nian and also centered on the left right axis. Now connect the electrodes to the head First, move the hair out of the way. Then apply alcohol followed by ABRY light gel to make the electrical connection between the electrode and the head.

Then attach an ECG electrode to the base of the back using a similar method to that used for the cap electrodes. This electrode measures the heartbeat with the electrodes attached. Work their contacts to reduce their impedances to less than 10 kilo ohms.

This excludes the resistance of the internal resistors in each electrode. Lastly, check the data to see that the EEG data quality is satisfactory. Ask the subject to be seated while you set up the EEG equipment in the MR scanner room.

Then take the amplifier into the shielded room and place it on a table at the back of the scanner. Attach the amplifier to a long fiber optic cable. Pass the fiber optic cable through the waveguide and attach it to the brain AMP USB adapter in the control room.

Now, take the subject into the room and ask him or her to sit on the scanner bed. Give the subject earplugs headphones and the call button. Then make absolutely certain that the subject is comfortable.

Now pad the subject's head to minimize head movement. Put the head coil over the subject's head. The EEG cables must leave the head coil along the shortest path available.

Then move the subject into the scanner bore. Ensure the electrodes FP one and FP two are ISO center to the MR scanner in the Z axis. Now attach the EEG cap to the amplifier at the rear of the scanner.

There should be no wire loops in the EEG leads. Maximally isolate the EEG cables from the MR scanner vibrations here. This is done using a cantilevered beam.

The amplifiers can also be placed directly in the scanner board as shown here. In this instance, it is important to use the shortest available ribbon cables. Make sure cables and amplifiers are isolated against vibrations and the system is central in set axis.

Talk to the subject from the console room to confirm that they can hear the scanner operator and that they feel okay. A second experimenter should monitor the EEG checking for noisy channels in the traces, as well as for the green sink on. at the bottom of the screen.

Now, switch off the cryo pumps to stop the deleterious effect of the cryo pumps on the recording quality. Next, ask the subject to move their head by a small amount. The importance of keeping the head still can be seen from the large voltages in the EEG recording that result from small head movements.

Then test the recording of neuronal activity by asking the subject to open and close their eyes. Occipital alpha activity should be measured above the noise floor. The pulse artifact can clearly be seen in the raw data, particularly on electrodes over the temples.

Use the ECG trace to correct this artifact in real time using rec view. Once the data quality has been optimized and the subject is ready, start the MRI preparatory scans and plan the slice positioning for the FMRI. As soon as each MRI scan begins, the gradients will cause large artifacts in the EEG.

When the FMRI experiment is ready to start, begin saving the data from the EEG. Now, start the experiment. Make sure the markers from the stimulus presentation and the MR scanner are seen in the brain vision recorder.

The EEG quality will appear very poor, but can be cleaned up in rec view or during post-processing. First, correct the gradient artifact done here in rec view before removing the pulse artifact. With the gradient artifact removed.

Proceed with making the pulse artifact correction. This is a signal quality to be expected when no artifact correction has been performed. It is clear that any neuronal activity is obscured.

The gradient artifact occurs at distinct frequencies that are harmonics of the frequency of slice acquisition in the FMRI sequence spanning the entire frequency range of the recording. Once the gradient artifact has been removed, the pulse artifact is revealed. There is considerable spatial variation of this artifact and that oz, one of the channels of interest for this visual experiment displays a particularly large pulse artifact.

This artifact has a lower frequency than the gradient artifact and is linked to the cardiac activity. In this data, the pulse artifact was corrected using average artifact subtraction in analyzer two and the R peaks of the cardiac waveform were detected from the ECG trace. The remaining signals are far smaller and reveal neuronal signals.

Now segment the data according to the stimulus presentation for analysis, the simplest being plots of the average evoked response for each channel. These evoked response for channels oh one and oh two are each averages across 300 stimuli. A topographic map for the P one 20 is on the right, examining the evoked responses as averaged over 32nd blocks measured from channel oh one reveals a natural and unpredictable variation of the responses.

This variation may be used to interrogate correlations between simultaneous recordings of bold responses and EEG responses. After watching this video, you should have a good idea of the best current practice for obtaining high quality EG data with simultaneous FMRI, using commercially available hardware Once the measurements been made using the setup described in this video. Further analysis methods can be applied to the EEG and FMI data in order to identify the spatiotemporal characteristics of the electrical and hemodynamic signals from the human brain.