Thanks to a new, smart personal rehabilitation algorithm that is used for a robot-assisted rehabilitation approach, patients who have suffered a stroke and other neurological injuries may be able to walk again.
Rehabilitation programs for spinal cord injuries or strokes normally require many hours of supported walking on treadmills at pre-defined paces to work. The problem is, everyday life activities are very different than steady, pre-defined movements – often, they require moving around in all directions and with differing gaits.
Now, thanks to a new algorithm, an alternative rehabilitation approach can be used. With it, Jean-Baptiste Mignardot and colleagues created a robotic harness capable of helping patients resist the downward force of gravity while at the same time allowing them to walk forwards, backward and side-to-side. Since it’s personalized, the algorithm supports each patient in a specific manner, helping them address their specific motor needs.
The system is controlled by an artificial neural network (which is why it’s smart) that varies the amount of upward and forward force through a cable harness. This is based on information about 120 different variables which are related to body movement.
Currently, under investigation in a clinical trial, the new system with the harness and algorithm was able to lead to significant improvements in unsupported walking for five patients with spinal cord injury with only one hour of overground training.
Source:
American Association for the Advancement of Science via ScienceDaily (https://www.sciencedaily.com/releases/2017/07/170719140942.htm)
The Wonderful World of Algorithms 