U.S. Department of Energy

Pacific Northwest National Laboratory

Utilizing Topological Visual Analytics for Persistent Human Performance Monitoring

Monday, May 23, 2016
Dr. Ryan Kramer
Air Force Officer and Civilian
Air Force Research Laboratory
The application of machine learning and deep learning approaches to multiple data types is providing increased insights into multivariate and multimodal data. Although inclusion of these approaches has dramatically enhanced the speed of data to decision processes, there are multiple drawbacks that include “black box” and “hidden layers” that obfuscate interrelationships between multivariate data and their conclusions. In addition, as the world changes, these analytic methods are often brittle to the inclusion of emergent or unannotated data. One potential alternative is the extension of topological data analysis (TDA) into a real-time, deep learning, autonomous ensemble network for data exploitation. In this application, black-boxes and hidden layers are replaced by a continuous framework of topological solutions that are each individually addressable, are informatically registered to disseminate annotation across the solution network, provide a rich contextual visualization for data exploration, and can easily identify novel emergent data in near real-time. By creating a deep learning analytical approach that implements TDA as the analytic backbone, underlying methodologies can be created to autonomously formulate hypotheses across the network. Within this talk, we illustrate how TDA can be used as an integrated visual analytic backbone for persistent physiological monitoring whereby the machine learning network can seamlessly interact with an operator to annotate, quantify, and classify multimodal data streams over their lifetime.
Speaker Bio

Dr. Ryan Kramer has been with the Air Force Research Laboratory for 13 years as an Air Force Officer and Civilian. During his time within AFRL, he has worked across multiple disciplines and topics that include biomaterial characterization for next-generation camouflage, development highly multiplex sensor arrays for breath diagnostics, high-throughput image analytics for the discovery of host-directed therapeutics, and genetic and molecular signature profiling for diagnostic detection of bioterror-Ryan related infectious diseases. More recently, Dr. Kramer has focused on new approaches that aim to understand how best to connect the ever increasing amount of data streaming from a new generation of experimental and research paradigms.

| Pacific Northwest National Laboratory