summary: Researchers leveraged 3D X-ray videography and machine learning to study the complex tongue movements during feeding in nonhuman primates. In this study, we recorded neural activity in the sensorimotor cortex and found that it accurately decoded the 3D shape of the tongue.
This groundbreaking finding suggests potential advances in brain-computer interface-based prostheses to restore eating and language functions. The technology could also help address dysphagia, a common problem among older people.
important facts:
- Researchers were able to accurately decipher the 3D shape of the tongue from brain signals during feeding.
- This new study of basic behavioral mechanisms such as eating, drinking and communication opens the door to the advancement of brain-computer interface-based prosthetics.
- This technique could potentially be used to stimulate the proper muscle set and improve quality of life in patients with dysphagia.
sauce; University of Chicago
Neuroscientists have learned a great deal about how the brain interprets and controls the actions that make up everyday actions such as walking, reaching, and grabbing.
But measuring basic behavioral mechanics such as eating, drinking, and communication is even more difficult, largely because the critical component, the tongue, is largely hidden from view.
A new study from the University of Chicago addresses this challenge by using 3D X-ray videography and machine learning to record the complex movements of the tongue during feeding in nonhuman primates.
Combined with recordings of neuronal activity taken simultaneously from sensorimotor areas of the brain, this study Nature Communicationshave shown that the 3D shape of the tongue can be accurately decoded from the brain, opening the possibility of brain-computer interface-based prostheses to restore lost feeding and speech functions.
infinite degrees of freedom
Not only is the tongue hidden in the mouth, but it also poses another biomechanical challenge. The movements of the arms and legs are constrained by skeletal bones and joints, giving them a degree of predictability.
Since the tongue is made entirely of muscle and other soft tissue, its freedom of movement is almost limitless (except for the unlucky few who can’t curl their tongue into a U-shape).
“When we think of the brain controlling muscles, we almost invariably think of it like moving an arm or leg that has rigid bones moving around its joints,” said the study’s lead author and former postdoc. Dr. JD Lawrence Chasen said. An academic at the University of Chicago, he currently works as a researcher at the National Renewable Energy Laboratory in Golden, Colorado.
“The tongue has a completely different anatomy.
As a postdoc and PhD student, Lawrence Chaisen uses data analysis and machine learning tools to study how the brain controls dynamic tongue and jaw movements critical to eating and speaking. Did. In his latest study, he collaborated with biobiology and anatomy professors Dr. Niko Hatsopoulos and Dr. Callum Roth to capture the tongue movements of two male rhesus monkeys while eating grapes. rice field.
Each monkey had seven markers on its tongue. These markers can be detected by his two X-ray video cameras that record the movement and shape of his tongue in his mouth, similar to motion capture technology used for special effects in movies and video games. increase.
Because monkeys are fast eaters, chewing two to three times per second, the researchers used a new 3D imaging technique called X-ray moving morphological reconstruction (XROMM) to visualize different movements and shapes of the tongue. of high-speed data was acquired and processed. Change, deformation.
At the same time, microelectrode arrays implanted in the motor cortex recorded neural activity while the monkeys were feeding. Lawrence Chaisen and team used deep neural networks, a type of machine learning software, to analyze brain activity and learn from that information.
When compared with real movements recorded by X-ray cameras, they found that information about the 3D shape and movement of the tongue resides in the motor cortex. That data was then used to accurately decipher and predict tongue shape based solely on neuronal activity.
“Previous studies have shown that the cortex is involved in basic tongue movements, but we were surprised at the range and resolution of information about tongue shape that could be extracted so easily.” Lawrence Chaisen said.
The future of soft prostheses
Interestingly, this data is represented in the same way arm movements and hand 3D positions are represented in the brain.
Dr. Hatsopoulos and Sulimane Benzmire, professors of biobiology and anatomy at the University of California, Chicago, and James Frank and Karen Frank Family Professors have already used a body of research to interpret signals from the brain. It translates it into a software algorithm that allows amputees and quadriplegics to drive the movements of robotic prostheses. Move according to your heart and receive a natural touch in return.
Techniques applied to the tongue are less advanced, but a similar approach could help patients who have lost their eating and speech functions.
“Dysphagia and dysphagia is a big problem, especially for older people,” says Hatsopoulos.
“If we can use this information about the tongue and its shape to decipher when swallowing occurs, we could connect it to a device that stimulates the appropriate muscles to aid swallowing.”
“It’s new that JD was able to decipher the shape of the soft tissue instead of the skeletal system here,” he says. “I think it’s very exciting.”
The study, “Robust Cortical Encoding of 3D Tongue Geometry During Feeding in Male Macaques,” was co-authored by Fritzie Arce-McShane of the University of Washington.
Funding: This research was supported by the National Science Foundation and the National Institutes of Health.
About this neuroscience research news
author: Matt Wood
sauce: University of Chicago
contact: Matt Wood – University of Chicago
image: Image credited to Neuroscience News
Original research: open access.
“Robust cortical encoding of 3D tongue geometry during macaque monkey feeding” Nicho Hatsopoulos et al. Nature Communications
overview
Robust cortical encoding of 3D tongue geometry during macaque feeding
Dexterous tongue deformation is the basis for eating, drinking and speaking. Although the orofacial sensorimotor cortex is thought to be involved in controlling the coordinated kinematics of the tongue, how the brain encodes and ultimately drives the 3D soft body deformation of the tongue remains to be explored. is little known.
Here we combine biplanar X-ray video techniques, multi-electrode cortical recordings, and machine learning-based decoding to investigate cortical representations of tongue deformation. We trained a long short-term memory (LSTM) neural network to decipher different aspects of intraoral tongue deformation from cortical activity during feeding in male rhesus monkeys.
We found that both tongue movements and complex tongue shapes across different feeding behaviors could be deciphered with high accuracy, and the distribution of deformation-related information across cortical regions was consistent with previous studies of the arm and hand. show.
