Inspiration

Our inspiration to create this product came from our experiences dealing with mental illness and from the treatment and advice we received during that process. One of the major points that a therapist tries to teach a patient with depression is how to combat depressive thought patterns. These sort of thoughts usually stem from some event or stimuli and while some people are more likely to suffer these kinds of thoughts because of biological reasons, these thoughts can and should be acknowledged as being neurotic. This process is one of the first steps and one that many people have trouble with. These depressive thought patterns are self-reinforcing and a person suffering with depression will almost always interpret events in the context of these neurotic thoughts, interpreting things to be hopeless or somehow their fault causing them to feel self loathing or become even more disconnected from reality.

What it does

What our discrete device would hope to do is to give therapists a tool to help their patients dealing with depression by giving them more information about their patients thought patterns using EEG (electroencephalogram) technology which would track their brain wave activity during their daily life. The device would also seek to help patients to more consistently identify depressive thought patterns and then work to change these neurotic thoughts to healthy, normal and positive thoughts. Using an initial calibration performed by the therapist the usual mental state of the patient while experiencing depressive thought patterns will be recorded and the machine programmed to recognize the same or similar conditions later. The device would record the times that these thought patterns typically arise and keep a running record of the patient's mental state allowing the therapist to specifically focus their questions to try to address the roots of a patient's conditions and what factors might be causing them to worsen. Not only this but the device could alert the patient that the thoughts they are having are neurotic, something that many patients cannot normally do when they start treatment. This is important because it would give the patient the ability and opportunity to correct the neurotic thought patterns, the way that the device alerts the patient could also be tailored to the patient based on questions asked by the therapist. For example, things like calming or happy music could be played by the device, texts from loved ones could be pulled up or in dire circumstances, the therapist could be called directly and alerted to the patient's distress.

How we built it

Because of the hardware dependent nature of this particular project we were not able to make progress towards the creation of our own discrete EEG earpiece. To compensate for this we attempted to pull data from a Muse headset offered here and then use this data to control a wireless microcontroller to select different audio files from an audio codec chip using the data from the EEG waveforms collected.

Challenges we ran into

One of the first problems we encountered while trying to design this system was the terrible interface provided by the Muse headset which is strongly geared towards mobile devices with almost no support to non-mobile applications. Muse Lab, the provided program, was incredibly inconsistent and would stop collecting data or fail to start collecting data almost constantly. To add to this frustration, the waveforms that were collected from the device were incredibly difficult to determine despite the experience we have in reading EEG signals. Their methods used to display the brain wave frequency information was terrible and did not give the actual frequency of the waves in these different brain wave bands (information which is important to fully characterize the activity of the brain wave) but only gave the averaged magnitude for all the frequencies in the band which made interpretation of the bands more difficult and the data available less useful. Lastly, the headset was quite uncomfortable while trying to get all sensors to activate.

Accomplishments that we're proud of

We did manage to successfully implement most of what we set out to achieve. We successfully programmed the microcontroller to receive OSC (Open Sound Control) inputs in UDP (User Datagram Protocol) data from a server over WiFi, but failed to get the Muse Lab software to output the data correctly despite there being a built-in interface to do so. Because the Muse Lab program was unable to successfully pass data to the microcontroller, the whole system was unresponsive.

What we learned

The greatest lessons that our team learned from this weekend was the current state of EEG technology and how it's normally used in many applications. Most EEG headsets like the Muse are marketed as devices to help you improve your brain's functioning or to control certain devices using your mind. There are currently none of these instruments available that seek to help people with mental illness as well as only a handful of devices that attempt to be discrete in their design. None of these devices are really designed to be used almost continuously either as the Muse in particular is designed to be used for only a few hours each day. In addition many brain-training technologies like the Muse and similar brain entrainment are often looked down upon as being pseudo-homeopathic and ridiculous because of the verbiage and ludicrous promises made by those selling them, which we see as a potential market hurdle to be overcome if our device is to receive widespread use.

What's next for Mental Harmony

Moving forward we plan to continue interviewing more therapists and refining the value proposition of our system. We need to define exactly what therapists and patients would want from this type of product and then implement these features. We also are looking to design our own electrode system to improve the reliability of the captured data and create a system whose form-factor is small enough to be encapsulated in a Bluetooth-style earpiece to be as nondescript as possible. We would also want to develop an app that would be able to interface with the device and give the patient feedback about their current mental state and potentially store some of the data recorded by the EEG.

Built With

  • eeg
  • microprocessor
Share this project:
×

Updates