Feb 17, 2017
Lopata Hall, Room 101
Bio-behavioral signals and systems: From signal representations to novel health applications
Department of Electrical Engineering
University of Southern California
Bio-behavioral signal processing and systems modeling enable an integrated computational approach to the study of human behavior and human physical and mental well-being through the use of overt behavioral signals information and covert biomarkers. Recent converging advances in sensing and computing, including wearable technologies, allow the unobtrusive long-term tracking of individuals yielding rich multimodal signal measurements from real-life. In this talk, we will present the development of data-scientific and context-rich bio-behavioral approaches for analyzing, quantifying, and interpreting these bio-behavioral signals. The first part of the talk will describe a novel knowledge-driven signal representation framework able to efficiently handle the large volume of acquired data and the noisy signal measurements. Our approach involves the use of sparse approximation techniques and the design of signal-specific dictionaries learned through Bayesian methods, outperforming previously proposed models in terms of signal reconstruction and information retrieval criteria. The second part will focus on translating the derived signal representations into novel intuitive quantitative measures analyzed with probabilistic and statistical models in relation to external factors of observable behavior. This work has found applications in Autism intervention for detecting beneficial regulation mechanisms during child-therapist interactions, as well as in the family studies domain for identifying instances of emotional escalation and interpersonal conflict. The final part of the talk will discuss how the results from the aforementioned analysis can be employed toward designing human-assistive personalized bio-feedback systems able to promote healthy routines, increase emotional wellness and awareness, and revolutionize clinical assessment and intervention for empowering individuals and clinical experts in mental health and well-being applications.
Theodora Chaspari is a PhD candidate at the Department of Electrical Engineering in University of Southern California. She has received the diploma in Electrical and Computer Engineering from the National Technical University of Athens, Greece (2010) and the Master of Science in Electrical Engineering from the University of Southern California (2012). Since 2010 she is working as a Research Assistant at the Signal Analysis and Interpretation Laboratory at USC. She has also been a Lab Associate Intern at Disney Research (summer 2015). Ms. Chaspari's research interests lie in the areas of biomedical signal processing, human-computer interaction, behavioral signal processing, data science, and machine learning. She is a recipient of the USC Annenberg Graduate Fellowship, USC Women in Science and Engineering Merit Fellowship, and the IEEE Signal Processing Society Travel Grant.