An electrode implanted into the brain of a man who is unable to move or communicate has enabled him to use a speech synthesizer to produce vowel sounds as he thinks them.

The work could one day help similar patients to produce whole sentences using signals from their brains, say the researchers.

A paralysed man has been able to use a voice synthesizer thanks to an electrode implanted in his brain.GETTY

Frank Guenther of Boston University in Massachusetts and his colleagues worked with a patient who has locked-in syndrome, a condition in which patients are almost completely paralysed — often able to move only their eyelids — but still fully conscious.

Guenther and his team first had to determine whether the man's brain could produce the same speech signals as a healthy person's. So they scanned his brain using functional magnetic resonance imaging (fMRI) while he attempted to say certain vowels.

Once the researchers were happy that the signals were the same, they implanted an electrode — designed by neuroscientist Philip Kennedy of the firm Neural Signals in Duluth, Georgia — into the speech-production areas of the man's brain. The electrode will remain there for the forseeable future.

The electrode is different to others used for brain–computer interfaces, most of which are fixed to the skull rather than within a specific part of the brain. This means that the electrodes can move around, making it difficult to record from the same neurons every time or to leave the electrode in place in for more than a few months at a time.

The electrode used by Guenther's team is impregnated with neurotrophic factors, which encourage neurons to grow into and around the electrode, anchoring it in place and allowing it to be recorded from for a much longer time.

Unpicking the code

Once the electrode was implanted, the team used a computer model of speech that Guenther had developed over the past 15 years to decode the signals coming from the man's brain and discern which vowel sounds he was thinking about. Guenther presented the results at the Society for Neuroscience meeting in Washington DC on 19 November¹.

So far, the patient has been able "to produce three vowel sounds with good accuracy", says Guenther. This happens as quickly as normal speech, he says.

"The long-term goal within five years is to have him use the speech brain–computer interface to produce words directly," Guenther says.

Most of the interfaces currently being developed transmit signals from the region of the brain that controls movement to either a prosthetic arm² or even, as shown by a recent study³ in monkeys, the subject's own arm. According to Guenther, this is the first brain–computer interface that has been tailored for speech.

Thinking aloud

Dorina Papageorgiou, a neuroscientist who works on decoding speech from fMRI signals at MD Anderson Cancer Center in Houston, Texas, says that the research is "cutting-edge work in the area of brain–computer interface speech output".

But brain signals for speech can also be decoded by electrodes positioned outside the brain, on the skull, or from fMRI, as in Papageorgiou's work, and she believes that, for many patients, non-invasive methods would be a better bet than a brain electrode.

Guenther and his colleagues say that they feel privileged to be involved in the project. "This was the first application where we see an individual improve his abilities based on something we theorized years ago," he says.

Their efforts are appreciated by the patient too. "When we first arrived to install this system he was obviously very excited — you can tell from his involuntary movements, and he was trying to look at us the whole time," Guenther says. As the man's father told the team, "he really has a new lease on life".

The team's next step is to train their computer decoder to recognize consonants so that patients can form whole words, and even sentences. They also hope that with developments in technology, they can implant more electrodes in their next patient to transmit a more detailed signal.

• References

  1. Guenther, F. H. et al. Abstract 712.1, presented at Society for Neuroscience annual meeting 19 November 2008. Available at http://tinyurl.com/5j8qdk.
  2. Velliste, M., Perel, S., Chance Spalding, M., Whitford, A. S. & Schwartz, A. B. Nature 453, 1098–1101 (2008).
  3. Moritz, C. T., Perlmutter, S. I. & Fetz, E. E. Nature Advance online publication doi:10.1038/nature07418 (2008).