Melbourne brain-computer interface startup Fluent has raised $2 million to help speech-impaired people communicate.
The University of Melbourne spinout was backed by the uni’s Genesis Pre-Seed Fund, Pacific Channel, Galileo Ventures, Multiple Sclerosis WA, paediatric audiologist Professor David Graydon, and by New York’s Jumpspace Ventures and London’s Founder’s Factory.
The low-risk interface, inserted under the scalp, but outside of the skull, helps people with neurological disorders that lead to a loss of speech, such as motor neurone disease and multiple sclerosis, to communicate using their thoughts.
Fluent cofounder and biomedical engineer, Dr Tim Mahoney, said the device sits above a part of the brain called the motor cortex, which controls speech muscles. Machine learning is used to convert brain activity into text or audio.
Like a QR code
Dr Mahoney compares it to a routine cochlear implant, with the potential to be more accessible to the broader population.
“If you think of electrical signals as QR codes: when a person speaks, every individual mouth and jaw movement produces different ‘QR codes’ in their motor cortex,” he said.
Get the best of Startup Daily straight to your inbox
Want to know the latest in startup news? Subscribe to our daily news and analysis coverage on what’s happening to ANZ startups, investors and the broader ecosystem. And best of all, it’s FREE!
By continuing, you agree to our Terms & Conditions and Privacy Policy.
“These ‘QR codes’ also occur when a person with impaired speech is attempting to speak. Our device can capture these codes in a sequence, which tells us what someone is trying to say.”
Low-risk surgery
During his PhD in accessible brain-computer interfacing, Dr Mahoney demonstrated that the quality of electrical signals captured is comparable whether it’s placed beneath or outside the scalp, thus avoiding the complex surgery normally involving electrodes implanted inside the skull.
Preliminary human testing at St Vincent’s Hospital Melbourne confirmed the results for Fluent’s interface.
“Participants had 144 electrodes placed on their scalp and the electrodes recorded brain activity occurring in their motor cortex while they spoke, mimed and imagined saying different phrases,” Dr Mahoney explained.
“After building the largest English dataset of its kind, we partnered with a Japanese team leveraging an even larger dataset. This collaboration proved that a model can accurately isolate the correct phrase from a pool of 128 options with 96% accuracy.”
University of Melbourne deputy vice-chancellor Professor Mark Cassidy said: “Fluent was born out of the University’s innovation ecosystem with its researchers securing Proof of Concept funding, nurtured by the Melbourne Entrepreneurial Centre’s accelerator program and now securing investment to develop human clinical trials.”
Clinical studies of Fluent’s insertable electrodes are planned for later this year.