The goal of this project was to design and build an electroencephalogram (EEG) that records brainwaves non-invasively through the scalp. Specifically, we were interested in distinguishing between relaxed and focused states of the user, which are frequently found to characterize various control modes in brain-computer interfaces. Our functional, relatively inexpensive EEG is a step towards designing viable, low-cost, yet functional solutions for use across the globe.
A major challenge for recording neuroelectrical activity is decoding a noisy, small magnitude signal via filtering and amplification. We constructed a circuit with notch filtering to remove the surrounding powerline frequency (60 Hz), and used both high and low pass filters to accentuate alpha and beta wave ranges (8-32 Hz). These have been found to correspond to relaxed and focused mental states, respectively.
Three electrodes were placed on the user via conductive paste: a grounding electrode on the earlobe, a reference on the mastoid, and a primary signal electrode on the occipitoparietal region of the scalp.
In preliminary testing, the user was instructed to relax with her eyes closed, blink rapidly, clench her jaw, and focus while performing mental math.
A 10 Hertz wave was observable via WaveForms in real-time during relaxed states. There was also a visible difference between relaxed states and muscle movements. This visual clarity gave a preliminary confidence in the data.
The recorded trials that followed consisted of six thirty-second blocks where the user was instructed to either relax with her eyes closed or perform a mental math task with her eyes opened.
A Fast Fourier Transform was applied to the recorded trials in MATLAB, translating a time series response into the frequency domain. This allowed for a clearer analysis of alpha wave ranges.
The magnitude of alpha waves in the relaxed state are comparatively higher than those in the focused state. The amplitude peak around 10 Hz is a commonly observed phenomenon corresponding to the user’s closed eyes.
These results suggest an effective EEG can be built for distinguishing between relaxed and focused mental states. Further, with its net cost of approximately $25, this circuit is an extremely cost-effective alternative to commercial EEGs, commonly priced between $80 to $3,000.