![Lu Mi](/sites/default/files/styles/utexas_image_style_500w_500h/public/2022-09/Lu_Mi%20-%20Lu%20Mi.jpg?itok=JlkJr7q9)
(she/her/hers)
Allen Institute & University of Washington, Seattle
Deep Learning, Computational Neuroscience, Connectomics
Lu Mi is currently a Shanahan Foundation Fellow at the Allen Institute and postdoctoral researcher at the University of Washington, Seattle. Prior to that, she received her Ph.D. and M.S. in the MIT CSAIL advised by Dr. Nir Shavit. She received a B.S. from Tsinghua University in 2013. Her research interests are deep learning and computational neuroscience. In particular, she is interested in developing a fast automatic connectomics pipeline to discover the brain; bridging the gap between the anatomical structure and function of the brain; building brain-inspired AI frameworks. She received the Shanahan Foundation Fellowship at the Interface of Data and Neuroscience and MathWorks Fellowship. She also has held internships in Google Research and Waymo Research Team, and was selected for Rising Stars in EECS in 2022.
Deep Learning Tools for Next-Generation Connectomics
The field of connectomics has witnessed exciting developments. Efficient algorithms are being developed to reconstruct nanoscale maps of large-scale images, allowing us a better understanding of how neural tissue computes. However, our ability to build powerful tools for the next generation of connectomics is dependent on navigating an inherent accuracy v.s. speed v.s. scalability trade-off. My research goal is to address this tradeoff by introducing four deep learning tools and techniques applied to the acquisition, reconstruction and modeling stages of connectomics pipelines. First, we propose a way to speed up the acquisition of images using learning-guided electron microscope (EM). Second, we proposed a faster and more scalable 3D reconstruction algorithm -- cross-classification clustering (3C), for large-scale connectomics datasets. Third, we introduce a cross-modality image translation technique mapping fast X-ray images to EM images with enhanced segmentation quality. Finally, we introduced a technique to bridge the gaps between structural and functional data with connectome-constrained latent variable models (CC-LVMs) of the unobserved voltage dynamics for the whole-brain nervous system. We hope these advanced applications of deep learning techniques will help address the performance and accuracy trade-offs of next-generation connectomics studies.