(Subscribe to our Today's Cache newsletter for a quick snapshot of top 5 tech stories. Click here to subscribe for free.)
A machine learning algorithm has been designed to detect patterns in brain signals that relate to specific actions like walking and typing.
These signals can be used to develop brain-machine interfaces that help restore lost functions for people with neurological and mental disorders, the team of researchers said in a study titled 'Modeling behaviorally relevant neural dynamics enabled by preferential subspace identification.'
The team tested the algorithm on standard brain datasets during the performance of arm and eye movements. The algorithm could predict the movement kinematics by simply looking at the neural patterns in brain signals that generate the movement, the team stated.
Also read | A portable brain-monitoring device to tackle mental health issues
The algorithm can also identify multiple actions by assessing brain signals. For example, it can recognise when the brain is typing a message on a keyboard and acknowledge if a person is thirsty at that same time.
Funded by the U.S. Army, the system could be used to foresee and notify soldiers if they are stressed or fatigued, before they realise themselves.