Spring 2019

Thurs, Apr 18th 2019, 12:00PM, Venue: HH A306
Sergio Pequito, RPI

Title: Towards personalized real-time closed-loop brain-machine-brain neurotechnologies

Abstract: In the context of healthcare applications, cyber-physical systems (CPSs) bring the promise of enabling a continuous and automatic monitoring and interaction with subjects, with the objective of ensuring or enhancing their quality of life.

Consequently, CPSs possess the potential to lower healthcare costs for prevention and therapies associated with chronic diseases. Neurotechnologies enable CPS applications like brain-machine interfaces (e.g., interaction with a computer to move a cursor or a speller, or with a prosthetic device), and brain-machine-brain (e.g., neurostimulation devices used to mitigate neurological diseases such as epilepsy). Notwithstanding, such applications are still to fulfill their promise as they currently need to be highly tuned ad hoc and/or under very controlled environments.

In this talk, we provide an overview of some of the ongoing work towards establish a control systems principled analysis of electroencephalographic (EEG) data in the context of neurotechnology applications. Specifically, we discuss how spatiotemporal models described by fractional-order dynamics enables a succinct parametric description of the EEG evolution possibly under unknown unknowns. Equipped with such models, we can propose to use model predictive control schemes that are suitable for personalized neurostimulation schemes per opposition to the current event-trigged open-loop schemes used in common practice. To illustrate the use of the proposed framework, we will consider different public available data where our methods outperform state-of-the-art approaches.

Bio: Dr. Sérgio Pequito is an assistant professor at the Department of Industrial and Systems Engineering at the Rensselaer Polytechnic Institute. He also holds a courtesy appointment at the Electrical, Computer, and Systems Engineering department. From 2014 to 2017, he was a postdoctoral researcher in General Robotics, Automation, Sensing & Perception Laboratory (GRASP lab) at University of Pennsylvania. He obtained his Ph.D. in Electrical and Computer Engineering from Carnegie Mellon University and Instituto Superior Técnico, through the CMU-Portugal program, in 2014. Previously, he received his B.Sc. and M.Sc. in Applied Mathematics from the Instituto Superior Técnico in 2007 and 2009, respectively.

Pequito's research consists of understanding the global qualitative behavior of large-scale systems from their structural or parametric descriptions and provide a rigorous framework for the design, analysis, optimization and control of large scale (real-world) systems. Currently, his interests span to neuroscience and biomedicine, where dynamical systems and control theoretic tools can be leveraged to develop new analysis tools for brain dynamics that, ultimately, will lead to new diagnostics and treatments of neural disorders. In addition, these tools can be used towards effective personalized medicine and improve brain-computer and brain-machine-brain interfaces that will improve people's life quality.

Pequito was awarded the best student paper finalist in the 48th IEEE Conference on Decision and Control (2009). Also, Pequito received the ECE Outstanding Teaching Assistant Award from the Electrical and Computer Engineering Department at Carnegie Mellon University, and the Carnegie Mellon Graduate Teaching Award (University-wide) honorable mention, both in 2012. Also, Pequito was a 2016 EECI European Ph.D. Award on Control for Complex and Heterogeneous Systems finalist and received the 2016 O. Hugo Schuck Award in the Theory Category.

link to video