Muse Projects and questions


So I’ve been having fun with my Muse in the first few days. I’m really not much of a developer but I can follow howtos ok.

First off, here are a few links I think people might be interested in: This uses node.js and should let you view some data in a browser, I haven’t finished getting it setup yet. This looks interesting but I had some problems with my python setup… I’ll come back to it later.

So, I’ve looked at EEG output, and alpha/beta/theta/delta/gamma waves, and I was wondering if anyone might be able to answer a question. When people do things like write a control for a sphero robot, or the mind-controlled RC quads/helis, they say they “think left/right” or “think of making a fist” with one or both hands:

What are they measuring? Can anyone give a summary answer about that?

Also if anyone feels like explaining something like the difference between absolute/relative/“score” elements, or could explain quantifier/FFT/etc stuff is, I’ve love to read it. I’m still plowing through info about brainwaves so if I’ve missed something or haven’t gotten to it yet I apologize. I thought maybe this could stimulate some discussion.


Edit: I got MuseJS working on my VPS. I finally realized I had to edit the .js files to replace with the server ip instead.

I’m not 100% sure it made a difference but I also reinstalled node using these instructions instead of just with the ubuntu package manager

as well as a: sudo npm install


Hey mdorre,

Those are great links! Jasna is actually our web developer here, that’s a personal project of hers :slight_smile:

CloudBrain was recently used in a really amazing exhibit featuring Muse at the SF Exploratorium:

All the data available from Muse is described here:

…including the relative/absolute/session scores. Take a look and let me know if you have any questions. Their definition and application have been discussed quite a bit on the forums too, so feel free to take a look around here.

For your question regarding interactive/control applications, typically what people do is construct a classifier using some fairly well-established machine learning techniques. To learn more about Brain Computer Interfaces, you should check out our Intro to BCI and EEG page (formerly called Intro to Brains). There is a bunch of helpful information listed there, especially if you are interested in classifiers and signal processing techniques (FFTs, for instance), like you’ve mentioned.


Hey mdorre,

Let me know if you have questions about setting up or using the web library. I have plans to extend it much further (the current version was just a weekend project) so feedback is always welcome!


When you say classifier, would something like an independent component analyser be included?


I was really just referring to machine learning techniques in general. ICA is often used to extract features in EEG analysis, though, so in that sense it’s an example of such a technique.