Researchers at Duke University’s School of Medicine, led by neurology professor Gregory Cogan, PhD, are making strides in brain decoding technology, offering hope for individuals dealing with conditions like ALS or locked-in syndrome.

Cogan emphasized the significance of this technology in addressing the limitations of current communication tools, which are often slow and cumbersome. Present speech decoding technology operates at a rate roughly half the speed of an audiobook, around 78 words per minute, while the average human speech rate is approximately 150 words per minute.

To overcome this limitation, Cogan collaborated with Jonathan Viventi, PhD, a faculty member at the Duke Institute for Brain Sciences specializing in high-density, flexible brain sensors. Viventi’s team successfully incorporated 256 minuscule brain sensors onto a postage stamp-sized, medical-grade plastic substrate. This increased sensor density allows for more comprehensive information about brain activity, crucial for accurate speech predictions.

Working with neurosurgeons from Duke University Hospital, including Derek Southwell, MD, PhD; Nandan Lad, MD, PhD; and Allan Friedman, MD, the researchers conducted tests on four patients. The device, temporarily implanted in individuals undergoing brain surgery for other conditions, recorded activity from the patients’ speech motor cortex during a listen-and-repeat activity.

Biomedical engineering graduate student Suseendrakumar Duraivel, the study’s lead author, processed the neural and speech data using a machine learning algorithm. The algorithm aimed to predict sounds based solely on brain activity recordings. Results demonstrated an 84% accuracy for certain sounds at the beginning of a sequence of three forming a given nonsense word. Overall, the decoder achieved a 40% accuracy rate, an impressive feat considering it operated with only 90 seconds of spoken data from a 15-minute test.

The researchers are now working on a cordless version of the device, supported by a $2.4 million grant from the National Institutes of Health. While acknowledging the technology’s potential, it is recognized that further progress is needed before it becomes widely accessible. Viventi envisions ongoing improvements in speech decoding speed, paving the way for enhanced communication capabilities in the future. The study’s findings were published in the journal Nature Communications.

By Impact Lab