Army-funded algorithm decodes brain signals responsible for behaviors like walking and breathing

A machine learning algorithm has shown the ability to link specific behaviors, such as walking and breathing, to their related brain signals – a first step to developing brain-machine interfaces.

The algorithm, funded by the US Army, was tested on two monkeys that made various arm and eye movements.

The technology successfully isolated the neural patterns in each of the animal’s brain signals and determined which control these specific movements.

Researchers foresee the technology being used to provide soldiers with feedback when they are stressed or fatigued before the individual themselves realize.

The brain decoding algorithm could also be designed to restore lost functions in those suffering with neurological and mental disorders.

Scroll down for video 

A machine learning algorithm has shown the ability to link specific behaviors, such as walking and breathing, to their related brain signals. Researchers foresee the technology being used to provide soldiers with feedback when they are stressed or fatigued before the individual themselves realize

Inside our brain signals are dynamic neural patterns that allow us to do a number of tasks at once.

However, because they reflect a combination of activities simultaneously, scientists have struggled to narrow down which signals are related to a specific behavior – until now.

Krim told Nextgov that the algorithm is capable of separating ‘what they call behaviorally relevant brain signals from behaviorally irrelevant brain signals.’  

Maryam Shanechi, assistant professor at the University Of Southern California Viterbi School Of Engineering and lead of the study, said: ‘Our algorithm can, for the first time, dissociate the dynamic patterns in brain signals that relate to specific behaviors and is much better at decoding these behaviors.’

Along with linking signals to various movements, the algorithm was able to decode the movements from brain signals. This allowed the algorithm to predict what the movement kinematics are just by analyzing brain signals that control a specific behavior

Along with linking signals to various movements, the algorithm was able to decode the movements from brain signals. This allowed the algorithm to predict what the movement kinematics are just by analyzing brain signals that control a specific behavior

Researchers said the algorithm is capable of separating 'what they call behaviorally relevant brain signals from behaviorally irrelevant brain signals'

Researchers said the algorithm is capable of separating ‘what they call behaviorally relevant brain signals from behaviorally irrelevant brain signals’

Along with linking signals to various movements, the algorithm was able to decode the movements from brain signals.

This allowed the algorithm to predict what the movement kinematics are just by analyzing brain signals that control a specific behavior.

Dr. Hamid Krim, program manager, Army Research Office, an element of the U.S. Army Combat Capabilities Develop Command, now known as DEVCOM, Army Research Laboratory, said: ‘The algorithm has significant implications for basic science discoveries.’

‘The algorithm can discover shared dynamic patterns between any signals beyond brain signals, which is widely applicable for the military and many other medical and commercial applications.’

Shanechi said that the success of the algorithm is its ability to identify both brain signals and behavioral signals such as movement kinematics together, and then find the dynamic patterns that were common to these signals.

The algorithm, funded by the US Army, was tested on two monkeys that made various arm and eye movements. The technology successfully isolated the neural patterns in each of the animal's brain signals and determined which control these specific movements

The algorithm, funded by the US Army, was tested on two monkeys that made various arm and eye movements. The technology successfully isolated the neural patterns in each of the animal’s brain signals and determined which control these specific movements

‘This decoding also depends on our ability to isolate neural patterns related to the specific behavior,’ the team shared in statement.

‘These neural patterns can be masked by patterns related to other activities and can be missed by standard algorithms.’

Although the algorithm is still in the development phase, the team sees it being used in brain-machine interfaces.

This could allow people to control prosthetics simply by thinking about the movement they want to achieve.

The US Army is just one of many groups working on hacking the brain to improve human lives. 

California-based Synchron, a neurovascular bioelectronics medicine company, recently announced its Stentrode brain computer interface (BCI) that has allowed patients with limited mobility carry out tasks on a computer just by using their mind.

Using the implant, patients achieved an average click accuracy of 92 percent and 93 percent and typing speeds of 14 and 20 characters per minute – without lifting a finger.

The team is using blood vessels as a natural highway to the brain, which are laced with sensors that record activity.

These signals are then sent through a telemetry unit to a small computer taped to the patient’s chest, which interprets what actions the individual wants to perform on the nearby PC, such as texting, emailing and shopping online.