Robotic 'coaches' could provide personalised support and increase motivation during crucial stroke and brain injury rehabilitation exercises, researchers have found.

During the successful international pilot study, led by researchers from the National Robotarium, the UK's centre for robotics and AI, socially assistive robots communicated with patients through a headset that detects the brain’s neural activity.

The robots were found to offer real-time motivation and feedback to improve adherence.

“Strange, but being used to human physio (physiotherapist) for the last six months it's hard to replace but having this and having nothing... Some people are isolated. Like I can see this sitting on my hospital bed and the physios in the hospital are so busy," said Jake MacDonald, while wearing a brain wave detecting headset.

In March the 59 year old from Stirling, had a Basal Ganglia stroke which affects the motor neurons on the left side of the brain.

"This is a good key thing for interacting with people," he said while being instructed on arm movements by a small robot.

The VITALISE (Virtual Health and Wellbeing Living Lab Infrastructure) project, led by the UK's National Robotarium’s Human Robot Interaction (HRI) team, has developed a system where socially assistive robots communicate with patients using a headset that detects brain neural activity.

"When Jake thinks 'I'm going to raise my left arm' to do the exercise, the headset picks it up, reads it into the computer, and we take that information and tell the robot, okay, now you need to lift your arm as well. So in that way, the robot actually mimics Jake's movements in real time,” EmilyAnn Nault, HRI co-investigator, said.

The robots interpret the patients brain waves to understand the user's intent during exercises, providing real-time motivation, visual mimicry, and feedback.

The primary objective is to get patients doing rehabilitation routines without their physiotherapist present. The robots provide personalized support to patients, taking into account their individual needs, cognitive abilities, and exercise pace.

"What we're hoping is that this movement at the same time to the person undertaking the rehab, will motivate the person to exercise more to undertake their stroke rehabilitation," said Lynne Baillie, Professor of Computer Science at Heriot-Watt University and academic lead at the National Robotarium for human-robot interaction, assistive living and health.

Upper limb impairment affects approximately 80% of acute stroke and brain injury survivors, significantly impacting their ability to perform daily activities and live independently.

Continued, repetitive exercise specific to the task is required for cognitive improvement post-injury. However, this can be difficult for patients to maintain due to issues such as forgetfulness and a lack of motivation.

The project worked with 16 stroke and brain injury survivors along with six therapists, testing the robot-assisted rehabilitation system. The primary focus was to evaluate the robot's ability to understand user intent and provide personalized motivation.

The team also examined the robot's capacity to detect an individual's intended exercise movement based on brain signal data and then mimic that motion in real-time to demonstrate an exercise. - Report from Reuters