Recording Human Electrocorticographic (ECoG) Signals for Neuroscientific Research and Real-time Functional Cortical Mapping

The aim of this procedure is to obtain research quality electro cortico graphic data from patients who are undergoing clinical epilepsy monitoring and use it to perform real-time functional mapping. This is accomplished by first setting up a research quality signal acquisition system that can be connected to the clinical epilepsy monitoring system without compromising or interrupting clinical monitoring. The second step in functional mapping is to record baseline signals while the patient is awake but resting quietly.

Next, the signals are processed to extract features representing brain activity in the high gamma frequency range, and these are used to build a model of the patient's baseline brain state. The final step is to have the patient perform simple cognitive or motor tasks while the novelty detection algorithms of the Siegfried system highlight parts of the cortex where high gamma activity is significantly different from baseline. Ultimately, the coefficient determination at each electrode location shows which cortical areas show significant activation during performance of the task.

Their involvement is then confirmed by electro cortical stimulation mapping the sick food results make the stimulation mapping procedure easier and less time consuming by providing complimentary information. ELECTROCORTICOGRAPHY or ECOG is defined by the use of subdural grid and strip electrodes to record patient's habitual seizures as well. To electrically stimulate these electrodes, either one or two at a time to try to interfere with certain functions like a motoric function or linguistic function to make a map of eloquent cortex that should be spared in the resection.

ECOG has a combination of high temporal and high spatial resolution that you simply don't find with other measurement techniques such as EEG or functional MRI. In particular, it's the only way to get good recordings of electrical brain activity in the high gamma frequency range, which is a very informative correlate of a wide range of brain functions. ECOG is an invasive recording technique where grids are placed directly on the brain surface, so this does require an operation and it's only done in clinical situations that require surgery.

Patients are typically in the epilepsy monitoring unit for about a week until clinicians have enough information to localize the seizure focus during this time. If the patient consents, they can take part in research and provided we don't disrupt clinical monitoring, this provides a unique window of opportunity for researchers to record human brain in action. In this video, I will be demonstrating how to localize the ECOG electrodes by coregistering, a pre-op MRI and a post-op cd, And I will be demonstrating a new form of functional mapping based on gamma features in the ECOG signals Prior to surgery, obtain a preoperative T one weighted structural MRI of the patient's head with 256 by 256 pixels per slice, a full field of view, no interpolation, a one millimeter slice, width, and preferably sagittal slices.

Then on the day of surgery, observe the surgical implantation of the grid and strip electrodes and collect digital photographs of the electrodes in situ, as well as the neurosurgeon's notes on the implanted locations. After surgery, obtain postoperative skull x-ray images, as well as brain CT scans, which should be acquired skin to skin at high resolution with one millimeter slice, width and no angle. Next, use the curry software package to create a three-dimensional cortical model of the patient's brain from the preoperative MRI then coregister this to the post grid implantation CT images.

Then right click on each electrode position in order and select export cursor to localize. To register the electrodes 3D coordinates, export the subject's 3D cortical structure and electrode coordinates in MATLAB format. Then create an export, a movie that shows the electrodes mapped onto the brain surface.

Also, map the electrode coordinates to standard broadman areas using an automated rack atlas. Review the information from the 3D model, x-ray images, photographs, and notes. Finalize a numbering scheme for the electrodes and work with the hospital technicians to ensure that the electrodes are patched into the splitter boxes.

Following this numbering exactly. Also, create a schematic layout for flossing the electrodes such that all the electrode positions can be clearly distinguished without overlapping. Select two electrode locations that are likely to be electro cortico graphically silent away from the presumed eloquent cortex to use as an initial ground and reference.

Next, prepare a linked stack of GUSB amps. Synchronize them by choosing one. As the master connect, its sink out to the other S sink in and enter the master's serial number as BCI two thousand's.

Device ID master parameter. Link the blue reference sockets to each other and link the yellow ground sockets to each other at the extreme right of the stack. This is where you will patch your chosen reference to blue and ground to yellow before the session starts.

Prepare a model do I nni file containing the signal processing settings that the Siegfried software will use to build a model. Also prepare a dot PRM file or separate dot PRM fragments containing the parameters that the BCI 2000 system will use for acquisition and real time visualization. Two key parameters for Siegfried are electrode locations, specifying the 2D layout chosen for the patient's electrodes and electrode condition, which specifies which tasks will be mapped under which conditions.

Be aware that the time with the patient is limited. Hence, all procedures need to be robust, optimized, and tried out in advance. Choose an appropriate moment for suggesting experimental recordings to the patient giving notice early in the day as well as 15 minutes before the session.

Will the equipment into place at the patient's bedside, then plug in and power on the patient's video monitor and bring the computer out of hibernation iation. Also, connect the data acquisition system to the ECOG grids via the splitter connectors. Launch the BCI 2000 software.

Then with the visualized source parameter enabled select set config right click on the viewer and set the high pass filter to a five hertz cutoff. Note, this setting will only affect visualization and not data collection. Check for interference from how a line noise by noting if activating a notch filter in the viewer makes a large difference to the signal.

If it does, try to reduce this by removing any sources of power interference such as unused cross-talking cables. Change the electrodes used for reference and ground if necessary. If you are using an eye tracker, calibrate it Using the calibration software provided by the manufacturer to avoid distractions and interruptions and to minimize possible signal interference.

Ensure that any TVs, radios, and mobile phones in the room are turned off. Begin the experimental session by first recording baseline activity. Instruct the subject to relax and remain motionless with eyes open.

Then start the GUSB amp source dummy signal processing and stimulus presentation. Task modules. Record six minutes of baseline activity with comfortable illumination in a quiet environment.

Next, start the data to model GUI tool configured to extract features in five hertz bins from 70 to 110 hertz, using the maximum entropy method for every 500 milliseconds of data. Press build model to build a probabilistic model of the selected spectral features using galium mixtures. Now give precise instructions to the patient for the experimental task that will be run in this task.

The subject will perform each task for 10 seconds at a time on each of five repetitions. Start the GUSB amp source Siegfried Sig Pro lava and stimulus presentation task modules, and configure the operator to load the probabilistic model, the cortical model, and the two and three dimensional electrode coordinates. Press start on the operator to start the experiment and initiate the mapping process.

Raw data is streamed automatically along with event markers until suspend is pressed or the experimental run finishes, each time start or resume is pressed, a new file will be created to avoid overwriting previous data during each task. The Siegfried software detects task related ECOG activity that is presented in continuously updated two and three dimensional maps of the cortex. The size and redness of each circle represents its importance in the running task.

The scaling of the circles to maximum R squared values can be controlled using the sliders throughout the session. Be sure to monitor the patient's behavior and the ECOG signals for suspected seizures and be ready to immediately alert and respond to instructions from medical staff. Here we see an example lateral x-ray for a representative patient.

The yellow circles mark the electrodes implicated in receptive language as subsequently identified by electro cortical stimulation mapping. Shown here is a photograph taken during implantation for this patient. These schematic show the Siegfried mapping results in a schematic two-dimensional layout.

The size and redness of each disc represents the significance of the involvement of each electrode in the task relative to baseline, and here the same statistic is mapped to color on a three dimensional brain model rendered from the patient's MRI. This video has demonstrated how human ECOG data can be recorded for research purposes and processed in real time using the BCI 2000 software suite. One example of A BCI 2000 real time application is the functional mapping tool.

Siegfried Siegfried mapping can be performed within 10 minutes depending on the number of tasks that are performed. Preliminary results can be seen within seconds and stable results can be obtained within five to 10 repetitions, which each take 10 seconds. The advantages of a Siegfried mapping is the functional relevance across the whole grid at once, whereas electrical stimulation requires current stimulation at each electrode site, which is a much more time consuming and can can potentially induce seizures.

We still do electrical mapping, but using Siegfried initially we get a a rapid preliminary map that greatly attenuates the process. In the future. Improvement of grid design will hopefully allow for applications apart from epilepsy monitoring, for example, in stroke patients, for restoring function, for movement of prosthetic limbs, and also enhancing communication in people with a paralysis of the vocal musculature or an a LS type diseases.

Thank you for watching.