CenBRAIN members design and implement state-of-the-art integrated circuits, System-on-a-chip and microsystems to tackle various challenges in the area of wearable and implantable medical devices and brain-machine interfaces, Figure 1.
Figure 1. Layout of our first intelligent biosignal processing 3.5x3.5 mm2 CMOS chip
Electrophysiology is the essential signal-based method to help understand the brain functions. However, the amplitude of these signals is weak and prone to be interfered by different types of noise. Consequently, to obtain high-quality electrophysiology signals, challenges such as signal transmission, bandwidth and power considerations come one after another when taking into account the fact that the electrophysiology needs to be acquired invasively or with a wearable device. We, in CenBRAIN, design and implement analog, mixed-signal and RF integrated circuits to tackle these challenges and bridge the brain to the electronic world [1].
Specifically, we work on low-power analog front-end design including amplifier and analog-to-digital converters for neural signal recording, wireless links for data and power transmission for neural signal recording. We are devoted to propose various circuit metrics including signal-to-noise ratio, energy-efficiency circuits, etc, and system architecture to the state-of-the-art level.
Analog, mixed-signal and RF integrated circuits mentioned in Section 1.1 and AI accelerators can be integrated on a System-on-a-chip (SoC) platform or packaged together to construct smart closed-loop neuromodulation systems [2]. With advanced silicon and packaging technology, the SoC and packaged system are given low-power, miniaturization features and rich functionality. These in-house designed and assembled systems can be deployed to various applications mentioned in section 3 and serve as effective tools to help us better understand the brain functions.
[1] M. Sawan, J. Yang, M. Tarkhan, J. Chen, M. Wang, C. Wang, F. Xia, and Y.-H. Chen, “Emerging Trends of Biomedical Circuits and Systems,” Foundations and Trends® in Integrated Circuits and Systems, vol. 1, no. 4, pp. 217-411, 2021.
[2] J. Yang and M. Sawan, "From Seizure Detection to Smart and Fully Embedded Seizure Prediction Engine: A Review," IEEE Transactions on Biomedical Circuits and Systems, vol. 14, no. 5, pp. 1008-1023, 2020, doi: 10.1109/TBCAS.2020.3018465.