A research group led by Professor Toshiyuki Fujiwara of the Department of Rehabilitation Medicine, Juntendo University Graduate School of Medicine, is collaborating with Meltin MMI Co., Ltd. to develop the world's first ``distinguishing a patient's intentions from bioelectrical signals and moving a paralyzed hand as desired.'' We performed stroke rehabilitation treatment using an AI robot and showed improvement in upper limb dysfunction.
It is said that only 15 to 20% of patients with residual hand paralysis after stroke recover from paralysis to a practical level.Remaining paralysis of the hands greatly interferes with activities of daily living, and is also a cause of hindering return to work.
In this research, we will use AI to determine the patient's intention from bioelectrical signals and use AI to help patients with severe paralysis who are unable to move their hands as they wish to move their hands, so that they can be rehabilitated for hand paralysis after stroke. We used an AI robot that can.This AI robot was developed by Meltin MMI, and uses three pairs of electrodes attached to its forearm to record weak bioelectrical signals sent from the brain to the hand when the patient tries to move his paralyzed hand.Based on the signal pattern, AI determines brain activity and determines whether the patient is trying to "stretch their fingers," "bend their fingers," or "relax" them without applying force. The robot reads the patient's intentions and moves the patient's paralyzed hand according to the patient's intention.
Using this AI robot, a patient with a post-stroke paralyzed hand was trained to bend and straighten their fingers, grasp objects, and move, twice a week for 1 minutes, a total of 40 times. However, improvement in upper extremity motor function was observed after training.In addition, it was said that the effect of rehabilitation was maintained 2 weeks after the completion of rehabilitation, and improvement was confirmed in the frequency of use of the paralyzed hand in daily life.
The results of this research are the first to show the effectiveness of stroke rehabilitation treatment using an AI robot, and it will be a new rehabilitation treatment method that makes it possible to recover the paralyzed hand after a stroke, which was considered difficult to recover with conventional methods. There is expected.
