Week 12 (01/25/2021)

This week, the team met in-person for the first time. The goals for our lab meetings are to assemble a fully functional EEG and work to develop an algorithm to successfully detect emotion in EEG data.

Two electrodes were connected to the forehead, and one to the mastoid (behind the ear for ground). The output waveform and frequency spectrum of one electrode connected to the head after amplification can be seen in the image below. There is clearly a spike in the frequency spectrum at around 60 Hz, which will hopefully be reduced by filtering.

The amplifier is working as intended, however the signal appeared to fluctuate in amplitude dramatically as the electrode chords moved around. It is the team's thought that the signal being displayed is not due to actual brainwaves, but rather, slight movements in the electrode chords.

The amplifier and high and low pass filters for the helmet were built and can be seen in the image below. When the team left the lab, we were working on verifying the working of the filter.

This Week

Tyler

- Set up lab station with necessary equipment.

- Connected electrodes to amplifier and verified amplifier working.

- Built the amplifying and filtering stages, but did not verify their working.

Mark

-Set up lab station with necessary equipment.

- Performed k-means clustering on active electrodes and observed event-related potential.

Next Week

Tyler

-Verify working of filter

-Observe the signal after the filter and determine if the brainwaves are being measured. If not, try different combinations of electrode placement.

Mark

- Write MATLAB code for machine learning algorithm

Memo: Tyler and Mark spoke with Dr. Hossein Asghari regarding the current progress they have made. Dr. Asghari suggested that mark plot the Alpha, Beta, and Gamma waves of a dataset and work on the K-means clustering machine learning algorithm from there. Dr. Asghari said he will assist Mark if he need helps at any time. Dr. Asghari mentioned to Tyler to make sure to limit the movement of the wires on the electrode since the slightest movement can affect the data being collected.