Home > Seminars > Eva Dyer - Neuroscientific Discovery Via Knowledge Extraction Pipelines

Eva Dyer - Neuroscientific Discovery Via Knowledge Extraction Pipelines


4/28/2016 at 1:30PM


4/28/2016 at 2:30PM


356 Fitzpatrick


College of Engineering close button

Walter Scheirer

Walter Scheirer

VIEW FULL PROFILE Email: walter.scheirer@nd.edu
Phone: 574-631-2436
Website: http://www.nd.edu/~wscheire
Office: 321C Stinson-Remick Hall


College of Engineering Assistant Professor
Primary interests in Computer Vision, Machine Learning, Biometrics,and Digital Humanities. Specific areas of research include Open Set Recognition, Extreme Value Theory Models for Visual Recognition, Biologically-inspired Learning Algorithms, and Stylometry.
Click for more information about Walter
Add to calendar:
iCal vCal

Neuroscientific discovery is becoming increasingly powered by high-throughput imaging methods. Extracting knowledge from these data is still challenging, due to the complexity of data acquisition, size of the raw datasets, and scarcity of “ground truth’’ or training data. In this talk, I will discuss my current efforts to build knowledge extraction pipelines, which span from initial image formation and data collection, all the way to building abstractions or models that encapsulate the structure or function of interest. In particular, I will describe an image analysis pipeline that I recently developed, which uses X-ray microtomography to build large-scale brain maps with cellular (micron) resolution. Following this, I will describe our team’s efforts to scale this system up to create whole brain maps. This requires optimizing each stage of the pipeline (i.e., sample preparation, data acquisition, image reconstruction, and analysis) in addition to building effective strategies for feeding back information extracted at later stages of analysis to improve data acquisition and reconstruction.

Seminar Speaker:

Eva Dyer

Northwestern University

Eva Dyer is currently a Postdoctoral Fellow at the Rehabilitation Institute of Chicago and Northwestern University. Eva holds degrees in Electrical & Computer Engineering from Rice University (PhD, MS) and the University of Miami (BS). Her current research interests lie at the intersection of neuroscience, signal processing, and machine learning. While at Rice, she co-designed and served as the Matlab content guru for 301x, an edX course on Discrete-Time Signals and Systems. At the University of Miami, she worked as a multimedia designer and an assistant sound designer for the documentary One Water: A Collaborative Effort for a Sustainable Future. She is the recipient of a NSF Graduate Research Fellowship and a National Library of Medicine Fellowship in Computational Biology.