Princeton University

School of Engineering & Applied Science

Tackling the size and complexity of large-scale neural datasets

Speaker: 
Eva Dyer, Rehabilitation Institute of Chicago, Northwestern University
Location: 
E-Quad, B205
Date/Time: 
Thursday, February 9, 2017 - 4:30pm

Abstract:
There is a recent renaissance in the design of new tools for interrogating the brain's structure and function at finer spatial resolutions, with faster readouts of the activity of more neurons, and over larger brain volumes. As a result, we are now faced with a neural data deluge: raw data is generated at rates that most labs are not equipped to process, let alone extract biological insights from. In this talk, I will describe my efforts in developing algorithms and frameworks to tackle this deluge of data. During the first half of my talk, I will describe methods I have developed to cope with the size of such datasets. I will specifically discuss my work in developing distributed and scalable data analysis pipelines for quantifying mesoscale neuroanatomy with X-ray microtomography. In the latter half of my talk, I will discuss ways I am tackling the complexity of large-scale neural datasets by reducing their dimensionality with unsupervised learning techniques. Throughout, I will provide examples from a variety of cellular-level imaging methods, including: X-ray microtomography, serial two-photon tomography, and electron microscopy. I will conclude with my plans to develop computational approaches for integrating X-rays, light, and electrons to interrogate the structure and function of large brain volumes at the nano and mesoscale.
 
Bio:
Eva Dyer is currently a Research Scientist at the Rehabilitation Institute of Chicago and in the Department of Physical Medicine and Rehabilitation at Northwestern University. Eva received her Ph.D. in Electrical & Computer Engineering from Rice University in 2014. Before that, she received her B.S. in Electrical & Computer Engineering from the University of Miami, where she completed a double major in Audio Engineering and Physics. While at Rice, she served as a Teaching Fellow for the Department of Electrical & Computer Engineering and was a co-developer of the edX course “Discrete-Time Signals and Systems”. She is the recipient of numerous awards including a NSF Graduate Research Fellowship, a National Library of Medicine Fellowship in Computational Biology and Medicine, a Presidential Fellowship from Rice University’s George R. Brown School of Engineering, an award for Best Ph.D. Presenter from the Department of Electrical & Computer Engineering at Rice University, and the Eliahu Jury Award for Scholarship in Electrical Engineering.