The short message from 2nd rotation IMLP Alana Edumunds, "GE IMLP," that appeared on Twitter is not an ordinary message.

The unusual thing about this message is how she wrote it:  with her brain waves.

On July 13th, a group of 5 Schenectady Infra Energy IMLP interns accompanied by their buddy Alana Edmunds and IMLP champion visited the Wadsworth Research Center in Albany, NY to learn about the impressive research going on in the area of developing a brain-computer interface (BCI).

The BCI interface is primarily being built for patients suffering from amyotrophic lateral sclerosis (ALS), a disabling condition created by brain-stem stroke or spinal-cord injury. People with these disorders gradually lose their ability to use their muscles and therefore lose their capability to communicate with the outside world by speaking, nodding, or even with eye movement.

For almost 20 years, researchers at the Wadsworth Center have been developing a BCI system to help such paralyzed people communicate and hence lead productive and fulfilling lives.

Gerwin Schalk, chief software engineer of the Wadsworth BCI project was our tour guide. He started the tour by explaining how the BCI system records the brain's electrical activity. First, he showed us a swimming cap with a series of electrodes fitted to record Electrophysiological signals (EEG signals) from the scalp, and then explained the process of recording these signals.

To actually type via brainwaves, a user watches a computer screen that displays a flashing matrix of letters that correspond to the keys of a standard computer keyboard. To type, the user must focus on the next letter of the word he or she wishes to spell and when that particular letter flashes, the user must convey this selection message to the brain in some way - which could be thinking"that's it," or tapping a finger.

In doing so, there is a momentary change in brain activity which appears as a spike in the EEG signals. Analyzing the brain's electrical activity during this process helps the software map out the user's intentions, and types the correct letter.

Editor's warning: the following paragraph is for the "technically minded" reader:

With the team mesmerized, Schalk explained the mathematical concepts involved in signal processing of the EEG. The signal processing has two stages. The first stage is feature extraction, or the calculation of the values of specific features of the signals. These features may be relatively simple measures such as amplitudes or latencies of specific potential amplitudes, frequencies of specific rhythms (e.g., sensorimotor rhythms), or firing rates of individual cortical neurons, or they may be more complex measures such as spectral coherences. The second stage is a translation algorithm that translates these features into device commands. Features such as rhythm amplitudes or neuronal firing rates are translated into commands that specify outputs such as cursor movements, icon selection, prosthesis operation or invoking a word processor. I was amazed with the idea that how using this simple 3 stage BCI system a paralyzed person can communicate to the outside world just like a normal person.

The highlight of the tour was undoubtedly the live demonstration. Alana volunteered to try the device, a bit shocked after finding that a gel would need to be injected into her hair for the signals to be carried from her scalp.

"Don't worry," our demonstrators reassured a suddenly hesitant Alana, "it's water soluble. Plus, we have a place you can rinse your hair out after." They weren't joking.

Our jaws dropped when she posted on the BCI Twitter account (without using a keyboard, of course) the message she had in her mind, fittingly: "GE_IMLP." The Wadsworth BCI Twitter account (Twitter ID: UWBCI) has over 360,000 users following the tweets visitors post via brainwaves - definitely some good exposure for the IMLPs!

While the Twittering application was for fun and to test the capabilities of the system, the BCI base system is continually being tested and currently used in several places around the world. The cost of the system is about $4,000 and the team is working on decreasing that cost significantly in the future. The Wadsworth Research Center hopes to establish a self-sustaining nonprofit organization that would distribute its BCI systems to those in need.

We greatly appreciate the work led by the researchers of the Wadsworth Research Center and wish them the best to help free patients currently restricted from communicating due to their disabilities.

You can check out more photos from the IMLP Blog Brain Twitter Tour flickr album here.

Sugandha Arora is a first-year IMLP Summer 2009 intern based in Schenectady, NY.