Institute of Information Science, Academia Sinica

Events

Print

Press Ctrl+P to print from browser

TIGP (BIO)—Development of a Brain-Machine Interface Device for Studying Large-scale Neural Activity in Behaving Animals

:::

TIGP (BIO)—Development of a Brain-Machine Interface Device for Studying Large-scale Neural Activity in Behaving Animals

  • LecturerDr. Yu-Wei Wu (Institute of Molecular Biology, Academia Sinica)
    Host: TIGP (BIO)
  • Time2022-03-03 (Thu.) 14:00 – 16:00
  • LocationAuditorium 101 at IIS New Building
Abstract

Monitoring the activities of neural networks is critical in studying brain function. Chronic recording of neural action potentials from a large-scale neuronal population enables us to understand functional changes in learning and memory formation. However, commonly used recording devices are limited for recording in planar structures, i.e., culture cells and thin brain slices, due to lacking a 3D interface. We developed a strategy to interface a silicon-based multi-electrode array (MEA) with a three-dimensional microwire bundle (Fig. A). This approach provides the link between rapidly developing silicon-based electronics and high-density neural interfaces. The system consists of a large-scale MEA, such as commercial electrophysiology MEAs and camera chips, connecting with a bundle of microwires (Fig. B). The bundles can be chronically implanted into deep cortical layers of a mouse and recorded for more than six months with stable recording yield (> 1000 neurons) and high signal quality (Fig. C). The same strategy potentially can be applied for recordings of organoids or other thick brain tissues. This modular design enables the integration between a variety of microwire types and sizes with different types of MEAs, connecting the rapid progress of commercial multiplexing, digitization, and data acquisition hardware together with a three-dimensional neural interface. We further applied deep-learning models in decoding the brain signal and studying the underlying mechanisms of large-scale neural dynamics.

deep-learning models

BIO

Dr. Yu-Wei Wu is currently an Assistant Research Fellow in the Institute of Molecular Biology, Academia Sinica. He obtained his Ph.D. training in a joint program between University College London (UCL) in the UK and RIKEN Brain Science Institute in Japan. He is specialized in using state-of-the-art imaging and electrophysiological techniques to study the neuron's function across multiple scales – from a single synapse to a large neural network. During his postdoctoral training at Stanford University, he further applied material sciences techniques in developing a new brain-machine interface (BMI) device for recording more than a thousand neurons simultaneously in living mice. His lab will continue to unveil the neural circuit mechanisms underlying the dynamic brain with this new technology.

Lab website: 【Yu-Wei Wu Laboratory