“We have developed an adaptive algorithm that personalizes a robotic harness to enable and train locomotion in people with spinal cord injury or stroke,” Jean-Baptiste Mignardot, a researcher at the Center for Neuroprosthetics and Brain Mind Institute at the Swiss Federal Institute of Technology Lausanne, told Digital Trends. “We then demonstrated its efficacy to enable natural walking in non-ambulatory individuals, and to enhance skilled locomotor control in the less-impaired subjects.”
Rehab programs have long since involved asking patients to walk on treadmills, while they are held upright by a harness. What this new technology improves on is the use of AI and robotics to help simulate the forces people will encounter in various real-world situations — whether this be walking along a straight path, on a wavy path, or across irregularly positioned rungs on a downward-facing ladder.
“We implemented an algorithm that can predict the optimal upward and forward support given by the robotic harness,” Mignardot continued. “The optimal upward is determined by an artificial neural network which integers kinematic and kinetic data from the patient. Then the algorithm combines the optimal upward support with the preferred speed of the patient — or the speed chosen by the physical therapist — to determine the optimal forward support.”
It’s even possible to set the system to imitate different gravitational pulls, so that users can see what it would be like to walk on other planets!
In trials, the technology was shown to work impressively well. For example, after using the AI-assisted robot harness, three patients with spinal cord injuries were able to walk independently — despite not previously being able to stand. Much the same proved true for stroke patients. A paper describing the research was published in the journal Science Translational Medicine.
“Until now, the general guidelines on the use of bodyweight support system in rehabilitation field advises to stay below 30 to 40 percent of body weight,” Mignardot said. “Here we showed that, with an appropriate forward component, the full range of body weight support can be applied without restriction.”
Going forward, the plan is for the technology to be commercialized for use in rehabilitation centers as part of the clinical routine.