Recently, the research teams of Professor Liu Hong and Associate Professor Zhao Chao from the School of Biological Science and Medical Engineering / State Key Laboratory of Digital Medical Engineering at Southeast University were invited to the IEEE BioCAS 2025 (IEEE Biomedical Circuits and Systems Conference), a flagship conference in the field of biomedical circuits and systems, held in Abu Dhabi, UAE. They delivered an oral presentation titled “A Dual-Mode CMOS Microelectrode Array for Chemical and Electrical Neural Recording.”
Currently, high-end brain-computer interface (BCI) chips are predominantly supplied by companies from the United States, Europe, Japan, and South Korea. Domestic research institutions and enterprises in China have only limited access to low-end chips. The 15th Five-Year Plan explicitly identifies BCI-related technologies as a priority. To address this “bottleneck” issue, the teams of Liu Hong and Zhao Chao have spent five years independently developing a CMOS-based microelectrode array (MEA) BCI chip. Achieving a throughput of over ten thousand channels and dual-mode sensing, this chip serves as a domestic substitute for high-end BCI chips and represents a critical tool for accelerating China’s foundational research in brain science and the diagnosis and treatment of brain disorders.
Revealing the information interaction mechanisms of neural networks via BCI technology not only helps understand brain functions and treat brain diseases but also provides key insights for neuromorphic computing and next-generation artificial intelligence. CMOS MEAs, owing to their high throughput, high spatiotemporal resolution, and efficient analog/digital signal processing capabilities, are becoming essential tools in BCI research. However, dual-mode BCI sensing circuits must simultaneously meet stringent specifications such as wide dynamic range, high bandwidth, and low input-referred noise, making their design extremely challenging. Consequently, very few CMOS MEAs capable of high-throughput dual-mode sensing have been reported internationally.
To address these challenges, the Liu Hong and Zhao Chao teams proposed and implemented on a single CMOS chip a dual-mode architecture based on continuous-time/discrete-time current sensing modules, targeting high-fidelity acquisition and readout of both chemical and electrical neural signals. The chip integrates 24,576 microelectrodes (each 10 μm in diameter) within a sensing area of 2.3 × 3.5 mm², achieving high-density, large-scale integration. The current detection dynamic range reaches 123 dB, covering wide variations from weak electrophysiological signals to electrochemical signals. The electrochemical and electrophysiological sensing bandwidths reach 70 kHz and 5 kHz, respectively, accommodating both rapid chemical event capture and neural spike recording. The noise density is as low as 2 fA/√Hz, providing a solid foundation for recording weak neural electrophysiological currents. Experimental results demonstrate that the chip exhibits excellent performance and stability in both in vitro neurotransmitter detection and neuronal spike recording.
The first author of the paper is Xu Jingyi, a Ph.D. student at the School of Biological Science and Medical Engineering, Southeast University. Professor Liu Hong and Associate Professor Zhao Chao are co-corresponding authors. Dr. Wang Yingfei, a Zhishan postdoctoral fellow, contributed to this work. This research was supported by the National Key Research and Development Program of China, the National Natural Science Foundation of China, the Key Program of Basic Research of Jiangsu Province, the Young Elite Program of the State Key Laboratory of Digital Medical Engineering, the Open Project of the State Key Laboratory of Analytical Chemistry for Life Sciences, and the Project of Jiangsu Provincial Scientific Research Center of Applied Mathematics.
Paper link: https://ieeexplore.ieee.org/document/11327961
