Artificial Intelligence

Transparent Brain-Computer Interface Uses AI and Nanotech

Novel brain-computer interface combines AI machine learning and graphene.

Posted January 24, 2024 | Reviewed by Kaja Perina

David_SMC/Pixabay

Innovative technology such as artificial intelligence (AI), brain-computer interfaces and nanotechnology are accelerating neuroscience research in the quest for improving human health and daily lives. Researchers at the University of California San Diego (UCSD) have created a novel transparent brain-computer interface (BCI) capable of providing high-resolution neural recordings from the brain’s surface utilizing AI machine learning and a nanomaterial called graphene.

Every one out of six people, approximately 16% of the global population, experience significant disability according to the World Health Organization (WHO). Brain-computer interfaces, also called brain-machine interfaces (BMIs), are enabling technologies that offer hope to those who have lost the ability to speak or move.

With the help of a brain-computer interface, a person can manage and operate external electronic devices with just thoughts to communicate via synthesized speech, move prosthetic limbs, operate a computer, and more important functions that improve the quality of life for those with disabilities.

The brain-computer interface market, a USD 2 billion industry in 2023, is expected to reach USD 6.2 billion by 2030 with a compound annual growth rate (CAGR) of 17.5% during 2020-2030 according to the Brain Computer Interface Market Size & Share Report 2030 by Grand View Research.

Per the report, North America had the largest revenue share globally at 39.5 % in 2022. A growing aging population is expected to contribute to the BCI market growth as the prevalence of Alzheimer’s disease, Parkinson’s disease, Huntington’s disease, and other neurodegenerative disorders increases.

“Recordings of neural activity at depth without implanting invasive neural probes could extend the lifetime of neural implants and improve the longevity of BCI technologies and pave the way for their medical translation,” wrote UCSD researchers Duygu Kuzum, Mehrdad Ramezani, Jeong-Hoon Kim, Xin Liu, Chi Ren, Abdullah Alothman, Chawina De-Eknamkul, Madison N. Wilson, Ertugrul Cubukcu, Vikash Gilja, and Takaki Komiyama.

What sets this brain-computer interface apart is the ability to record brain activity via both optical imaging and electrical signals simultaneously. Unlike conventional BCI implants which are opaque, this new BCI is transparent, providing neuroscientists with a window for observation via microscopy. As the transparent graphene electrode array records electrical signals from the neurons located in the brain’s outer layers, at the same time, the calcium spikes from neurons up to 250 micrometers deep are imaged using a two-photon microscope shining laser lights through the array. In this manner, the researchers were able to correlate the electrical signals at the brain’s outer layers with calcium spike activity in the deeper parts of the brain.

The correlation data was used as training data for an AI artificial neural network. The UCSD researchers created an AI model with a linear hidden layer, a single-layer bidirectional LSTM (Long Short-Term Memory), or BiLSTM, and a linear readout layer. The AI model learned from the correlation data in order to predict the calcium activity in the deeper parts of the brain based on the electrical signals on the outer layer. This enables neuroscientists to observe brain activity for longer periods as the organism is moving around freely versus being locked under a microscope for a short duration.

The researchers demonstrated on laboratory mice that the electrical signals in the outer layers recorded by their high-density transparent graphene array could be correlated with calcium activity at deeper parts of the brain. According to the study authors, their nanotechnology array is able to predict average and single-cell calcium activities from surface potential recordings. With this pioneering innovation, the next steps are to expand the research beyond laboratory mouse models.

“This could potentially improve brain computer interfaces and enable less invasive treatments for neurological disorders,” the UCSD researchers concluded.