Brain Data Created by AI to Assist the Disabled

data brain

Researchers are employing generative adversarial networks (GANs), a technology best known for making deepfake films and photorealistic human faces, to develop brain-computer interfaces for persons with disabilities. The researchers described the construction and application of a generative model—a model that generates an almost infinite number of new data distributions from a taught data distribution—in a study published in Nature Biomedical Engineering. The researchers were able to train an AI to create synthetic brain activity data. To increase the usefulness of Brain-Computer Interfaces, the data, specifically brain impulses termed spike trains, can be fed into machine-learning algorithms (BCI).


BCI systems analyze a person’s brain signals and translate them into commands, allowing the user to control digital devices such as computer cursors with only their thoughts. People with motor dysfunction or paralysis, as well as those suffering from locked-in syndrome (when a person is completely cognizant but unable to move or talk), may benefit from these devices.


BCI is presently accessible in a variety of forms, ranging from caps that measure brain impulses to devices implanted in brain tissue. New applications are being discovered all the time, ranging from neurorehabilitation to depression treatment. Despite all of this promise, making these systems quick and reliable enough for real-world use has been difficult. BCIs, in particular, require a large volume of brain data as well as extensive training, calibration, and learning to make sense of their inputs.


The researchers used one session of data recorded from a monkey reaching for an object to train a deep-learning spike synthesizer in an experiment detailed in the study. The synthesizer was then used to generate vast amounts of similar—albeit fictitious—neural data.


The team then used the synthesized data to train a BCI by combining it with small quantities of new real data—either from the same animal on a different day or from a different monkey. This procedure was far faster than the current standard approaches in getting the system up and operating. The researchers discovered that using GAN-synthesised neural data increased the overall training pace of a BCI by up to 20 times. It’s the first time we’ve seen AI construct a thought or movement recipe through the use of synthetic spike trains. This study is an important step toward making BCIs more practical in the real world.


Furthermore, utilizing minimum additional brain data, the system quickly responded to new sessions, or subjects, following training on one experimental session. The main breakthrough is creating artificial spike trains that appear just like they come from this individual as they imagine executing certain motions, then using this data to aid in learning on the next person.


Beyond BCIs, GAN-generated synthetic data has the potential to speed up training and improve performance in other data-hungry fields of artificial intelligence. When a corporation is ready to commercialize a robotic skeleton, robotic arm, or voice synthesis system, they should consider this strategy since it could speed up the training and retraining process.


According to a study funded by the National Institutes of Health’s Brain Research Through Advancing Innovative Neurotechnologies, BCI can help patients who have lost their capacity to move or speak communicate again (NIH BRAIN). The BCI was created with spinal cord injuries and neurological illnesses like amyotrophic lateral sclerosis in mind (ALS).


According to the NIH BRAIN Initiative, this discovery is a significant step forward in the development of BCIs and machine learning technologies for understanding how the human brain governs communication processes. This understanding is laying the groundwork for improving the lives of individuals who have suffered from neurological injuries or diseases. As a result, the findings provide a novel method for BCIs and show that it is possible to accurately decode quick, dexterous movements even years after paralysis.