The US Army has co-developed a new machine learning (ML) algorithm that can analyse patterns in brain signals, potentially providing soldiers with behavioural feedback in factors like fatigue and stress.
The new algorithm can isolate patterns in brain signals related to specific behaviours before decoding them; for example, it could establish whether a particular pattern suggested a soldier was growing tired or thirsty. The algorithm was developed by researchers at the University of Southern California as part of a wider effort by American and British universities.
This was part of a joint Multidisciplinary University Research Initiative (MURI) grant awarded by the US Army Research Office (ARO) and the UK Defence Science and Technology Laboratory (DSTL). The research was first published in Nature Neuroscience . ARO is part of the US Army Combat Capabilities Development Command’s Army Research Laboratory (ARL).
Dr Hamid Krim, an ARO programme manager, told Janes that the algorithm was tested by analysing a monkey’s brain signals. He said the overarching goal of the research “is to be able to actually read brain signals, just like you would read a textbook, and subsequently exploit that for feedback to the individual for possible corrective action”.
The brain is effectively the central processor for all primates, including humans, and interpreting its many signals is a complex process, Krim explained, with “signals that pertain to your heartbeat, signals that pertain to your breathing, signals that represent what you’re thinking about right now”. Other signals can be more mechanical, he said, for example, those related to moving a hand or performing some other deliberate function, rather than the types of underlying states that continue subconsciously.
Looking to read the full article?
Gain unlimited access to Janes news and more...
