Applying Universal Semiconductor Processes Lowers Production Cost to One-Hundredth, Enabling Commercialization
Simultaneous Analysis of Neural Activity Timing and Cell Types Raises Hopes for Precision Treatment of Alzheimer’s and Parkinson’s Disease

A domestically developed semiconductor chip has been created that could accelerate the commercialization of next-generation brain-computer interface (BCI) technologies, such as moving robots with just thoughts or treating Alzheimer’s and Parkinson’s disease.


Previously, it was only possible to determine ‘when’ a neuron was active or to identify ‘which type of cell’ was involved, but not both at once. Now, by simultaneously reading both electrical and optical signals from a single chip, it is possible to analyze these two types of information together. In particular, by employing a universal semiconductor process used in smartphones and similar devices, the production cost has been reduced to about one-hundredth of its previous level, greatly increasing the potential for industrialization.

Schematic of an optoelectronic BCI chip. Electrodes and optical sensors are integrated into a single chip, allowing simultaneous analysis of the timing of neural cell activity and cell types. By applying a general-purpose CMOS process, the production cost was significantly reduced, enhancing the potential for commercialization of next-generation BCI. Provided by the research team

Schematic of an optoelectronic BCI chip. Electrodes and optical sensors are integrated into a single chip, allowing simultaneous analysis of the timing of neural cell activity and cell types. By applying a general-purpose CMOS process, the production cost was significantly reduced, enhancing the potential for commercialization of next-generation BCI. Provided by the research team

View original image

The Korea Institute of Science and Technology (KIST) announced on August 4 that the research team led by Principal Researcher Lee Changhyuk at the Brain Science Institute has, for the first time in the world, developed a next-generation optoelectronic BCI chip that can simultaneously measure both electrical and optical signals on a single semiconductor chip. The results were published in the latest issue of the international journal Advanced Science and selected as the Front Cover Article.


In March of this year, the government designated BCI as one of the ‘Seven National Priority Tasks with Tangible Benefits for the People’ and is fostering it as a core technology for next-generation healthcare and artificial intelligence (AI). BCI refers to technology that reads brain signals to control computers or machines, or directly stimulates the brain for therapeutic purposes.


Traditional BCI technologies could accurately measure the timing of neural activity, but it was difficult to distinguish which types of cells were responding. Conversely, using optical signals such as fluorescence made it possible to identify cell types, but it was hard to measure these signals simultaneously with electrical signals, limiting the precision of analysis.


Simultaneous Measurement of Electrical and Optical Signals... Laying the Foundation for BCI Commercialization


The research team integrated electrodes for measuring neural activity timing and photonic sensors for identifying cell types onto a single semiconductor chip.


They arranged 416 electrodes and 832 photonic sensor pixels on 13 silicon probes, each as thick as a strand of hair, enabling simultaneous analysis of when neural cells responded and which types of cells were involved.


In addition, the chip was manufactured by combining a universal complementary metal–oxide–semiconductor (CMOS) process, commonly used in smartphones, with a three-dimensional (3D) printing post-process. Instead of using research-specific dedicated processes, existing semiconductor production infrastructure can be utilized, reducing production cost to about one-hundredth of the previous level. This also secures the potential for mass production using domestic foundries and fabless companies in the future.

KIST-developed optoelectronic BCI chip (center) and the cover paper of the international journal Advanced Science (bottom left). Inside the circle at the top right are co-first author student researchers Joohee Moon and Miji Kim, and corresponding author principal researcher Changhyuk Lee. This achievement was developed with domestic technology by integrating system semiconductor design, semiconductor processes, and neuroscience. Provided by KIST

KIST-developed optoelectronic BCI chip (center) and the cover paper of the international journal Advanced Science (bottom left). Inside the circle at the top right are co-first author student researchers Joohee Moon and Miji Kim, and corresponding author principal researcher Changhyuk Lee. This achievement was developed with domestic technology by integrating system semiconductor design, semiconductor processes, and neuroscience. Provided by KIST

View original image

The research team validated the chip's performance by implanting it into three brain regions of living mice. They simultaneously measured changes in brain waves according to anesthesia depth and neural responses to visual stimuli, and observed that even after one week following implantation, there were no signs of inflammation or motor function abnormalities.


The researchers expect that this technology will be used for precise diagnosis of brain diseases such as Parkinson’s disease, Alzheimer’s disease, and depression, for developing customized electronic therapeutics, and for implementing advanced brain-AI interface systems.



Lee Changhyuk, Principal Researcher at KIST, stated, “To achieve precision BCI, it is necessary not only to determine when neural cells are active, but also to distinguish the types of cells involved. This research presents a new technological direction with optoelectronic BCI and will serve as a foundation linking the domestic semiconductor industry and BCI sector.”


This content was produced with the assistance of AI translation services.

© The Asia Business Daily. All rights reserved. Unauthorized AI training and use prohibited.

Today’s Briefing