Research idea on using BCI to write text
I had an idea the other day while I was thinking about ways we can use our brainwaves to type/communicate faster than we can with a keyboard. I feel like an efficient system will require calibration with each individual user. The solution I want to try involves reading a dictionary and recording the brainwaves produced for each word. Here are the steps that I would take to make this work.
1. Read a word twice and see if there are any differences between the two brainwave readings.
2. If the brainwaves are similar, write a program to link each word with the brainwave produced when you think about the word.
3. When a brainwave matches with one in the program's database, have it paste the word associated with the brainwave into a text file.
4. Try making a sentence.
I'm very interested in hearing everyone's thoughts on this idea, and even more interested in the results if someone tests this method.
Additional thoughts:
Test if the brainwaves produced by reading a word are similar to the ones produced by thinking about the word.
Comments
As well, i dont think that we will get any differences using brain waves (FFT theta band or beta...etc).
I think that it would be interesting if you analyze the eeg data like event related potentials. Or using nonlinear mathematics like the Entropy to extract some kind of EEG algorithms that you can relate aftewards with any letter or word.
Well let us know about your progress.
For the moment i dont have my own equipment. But i work with EEG.
Greetings and good luck!
P.
https://www.google.com/search?q="High-speed+spelling+with+a+noninvasive+brain–computer+interface"
http://english.cas.cn/newsroom/research_news/201511/t20151102_155042.shtml
The OpenViBE P300 is discussed on this thread,
http://openbci.com/forum/index.php?p=/discussion/206/openvibe-p300-speller-tutorial-questions
A commercial system, Intendix, here,
https://www.google.com/search?q=intendix+gtec
https://www.google.com/search?q=eeg+word+recognition
Here's two full text papers from 2015, 2014 with many references. Unfortunately, the "state of the art" is still pretty dismal as far as accuracy. The first paper is ECoG, cortical sensor EEG, not scalp sensor EEG.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4464168/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4034498/
https://www.google.com/search?q=darpa+synthetic+telepathy+irvine
https://www.google.com/search?q=michael+d'zmura+uc+irvine
http://cnslab.ss.uci.edu/speechattention/index.html
DARPA has been much in the news the last few months for their new ECoG based project,
https://www.google.com/search?q=darpa+brain+implant
William