With Training, People in Mind-Controlled Wheelchairs Can Navigate Normal, Cluttered Spaces

Summary: Brain Interface Technology – The new engine allows immobile people to control their wheelchairs through mind control. BMI allows users to go through a natural and messy environment after training.

Source: Click a cell

Mind-controlled wheelchairs can help paralyzed people gain new mobility by translating user ideas into mechanical commands.

November 18 in the magazine iScienceResearchers have shown that tetraplegic users can operate a mind-controlled wheelchair in a natural and congested environment after long-term training.

José del R. Millán, co-author of the study at the University of Texas at Austin, says: “The operation of such a wheelchair was successful.” “Our research shows a potential path for improving clinical translation of non-invasive brain-machine interface technology.”

Millán and his colleagues selected three tetraplegic people for the longitudinal study. Each participant received three training sessions per week for 2 to 5 months.

Participants wore a skull cap that detects their brain activity via electroencephalography (EEG), which will be converted into mechanical commands for the wheelchair via a brain-machine interface device.

Participants were asked to control the direction of the wheelchair by thinking about their physical movement. In particular, they need to think about moving both arms to the left and both legs to turn right.

Credit: Luca Tonin

In the first training session, three participants had a similar level of accuracy when the instrument responses were in line with the user’s opinion of about 43% to 55%. During the training, the brain-machine interface team saw a significant improvement in accuracy in the first participant, who reached more than 95% accuracy at the end of his training. .

Also Read :  IonQ Wins $13.4M Quantum Contract with U.S. Air Force Research Lab - High-Performance Computing News Analysis

The team also observed an increase in accuracy in participants from 3 to 98% half of his training before the team updated his equipment with new algorithms.

The improvements seen in the first and third participants were related to the improved feature discrimination, the ability of the algorithm to discriminate brain activity patterns encoded for the “left” concept from there. For “to the right”.

The team found that better discrimination was not only the result of machine learning of the device, but also learning in the participants’ brains. The EEG of the first and third participants showed clear changes in the brain wave pattern as they improved the accuracy of instrument control.

“We see EEG results where the subject unites the skills of modifying different parts of their brain to create patterns for ‘left’ and different patterns for ‘to right’,” says Millán. . “We believe that there is a cortical reorganization that has taken place as a result of the participants’ learning process.”

This shows a person in a wheelchair
The improvements seen in the first and third participants were related to the improved feature discrimination, the ability of the algorithm to discriminate brain activity patterns encoded for the “left” concept from there. For “to the right”. Image is a public domain

Compared with the first and third participants, the second participant did not have a significant change in brain activity pattern throughout the training session. His accuracy increased slightly during the first session, which remained stable for the rest of the training session. Millán says it shows that machine learning alone is not enough for the successful implementation of mind control tools.

Also Read :  Don't let trust issues derail your operational resilience plan

At the end of the training, all participants were asked to drive their wheelchairs through a jammed hospital room. They have to walk around obstacles such as room divisions and hospital beds designed to simulate a real-world environment. Both the first and third participants completed the task, while the second participant did not complete it.

“It seems that for someone to get control of a brain-machine interface that allows them to perform complex daily activities, such as driving a wheelchair in a natural environment, it requires preparation,” says Millán. “The neuroplastic substance in our cortex.”

The study also emphasizes the role of long-term training for consumers. Millán says that although the first participant did particularly well at the end, he struggled in the first few training sessions. Longitudinal study is one of the first evaluations to evaluate the clinical interpretation of non-invasive brain-machine interface technology in tetraplegic humans.

The team then wanted to find out why the second participant did not have a learning experience. They hope to conduct a more detailed analysis of the brain signals of all participants to understand their differences and possible interventions for people struggling with future learning processes.

About this neurotech research information.

Author: Press Office
Source: Click a cell
Contact: Press Office – Cell Press
Image: Image is in public domain

See also

This shows a sleeping man

Early research: Enable access.
“Learn BMI-Driven Wheelchair Management for People with Severe Tetanus” by José del R. Millán et al. iScience


Abstract

Also Read :  State Funding of Morgan Research Centers Supports R1 Aspirations While Addressing Societal Challenges

Learn to control BMI-driven wheelchairs for people with severe tetanus

Highlights

  • Three participants learned to drive a wheelchair using non-invasive BMI
  • Direct transfer of learned BMI skills to wheelchair management
  • Learning the subject and the intelligence of the robot is the key for a robot that works with BMI

Summary

A mind-controlled wheelchair is an interesting mobile assistance solution that can be applied in complete paralysis. Despite advances in Brain-machine interface (BMI) technology, its translation is still difficult to understand.

The main purpose of this study was to investigate the hypothesis that the acquisition of BMI skills by end users is fundamental to the management of non-invasive, non-invasive smart wheelchairs in real-world situations.

We show that three tetraplegic spine injury users can be trained to operate a non-invasive self-controlled wheelchair and perform complex exploration tasks. However, only two users demonstrated increased performance, decoding and discrimination features, significant neuroplasticity changes, and improved BMI command delays achieved higher browsing performance.

In addition, we demonstrate that agile and consistent robot control is possible through low levels of freedom, separate and indistinct control networks, such as BMI imaging, by a combination of human and artificial intelligence. Term Joint management approach.

We assume that subject learning and co-management are key components that pave the way for non-invasive BMI interpretation.

Source

Leave a Reply

Your email address will not be published.