A sufferer of tetraplegia – caused by a lesion on the spinal cord that prevents the nervous system from controlling all four limbs – was able to move again and control both arms thanks to AI technology. Using a neuroprosthetic, which records, transmits and decodes brain signals in real time, the patient was able to control an exoskeleton.

The major innovation in this device is its ability to provide chronic high-resolution recording of the brain’s electrical activity related to the intention to move. This ‘intention’ is transmitted wirelessly to a computer for decoding, so the movements of the exoskeleton’s four limbs can be controlled.

The implantable device, named WIMAGINE, collects brain signals in the sensorimotor cortex, emitted when the person imagines moving, which means the exoskeleton moves through mental control, with no external controls unnecessary.

The recorded electrocorticograms are then decoded to predict the deliberate movement imagined by the patient and then, for example, to control the corresponding limb of an exoskeleton. Decoding electrocorticograms required the development of highly sophisticated algorithms based on artificial intelligence (AI) methods (machine learning) and software.

The design of WIMAGINE also involved research engineers from CEA-List, the institute specialising in smart digital systems, who developed the four-limb exoskeleton based on their reversible actuation and control-command bricks.

The long-term goal of the project is to identify fields in which the brain-machine interface could be used to create compensatory systems for various types of motor disabilities and give patients more independence in their everyday lives; for example, by driving a wheelchair or controlling an articulated arm.

Photo credit:  ©FDD Clinatec