The Yousefi Laboratory

 Ali Yousefi

Assistant Professor

Department of Computer Science &

Department of Neuroscience

Worcester Polytechnic Institute (WPI)

100 Institute Rd, Worcester, MA 01609

ayousefi@wpi.edu

(508) 831-5493

My research focuses on 1) developing mathematical and statistical methods to analyze neuroscience data, 2) developing methodologies and tools for brain-machine interfaces with basic and clinical neuroscience applications, and 3) application of these theories and models to neural prostheses (to restore cognitive function after brain damage or degenerative disease) and to common data analysis problems in other research fields (e.g., analysis of noise-contaminated data). In my research, I have worked to integrate methodologies related to signal processing, statistical inference, model identification, stochastic modeling, optimization, and control theory to develop more appropriate tools and techniques for analysis of diverse forms of neural and other physiological signals.

Publications:

Patents:

https://scholar.google.com/citations?user=jieyeRUAAAAJ

Advising:

Ph.D. research advisor: Yalda Amidi

M.Sc. research advisor: Reza Saadati Fard, Mohammad R. Rezaei

Teaching:

(Fall 2019) Machine Learning

(Spring 2020) Computational Neuroscience

Information for prospective students:

I'm always looking for good students (both graduate students and committed, well-prepared undergraduates) to collaborate with. In my current research topics, I listed some projects we're thinking about recently. If you're interested or have a related project in mind that I might be able to contribute to, please contact me by ayousefi@wpi.edu

Links:

Current Research:

Global coherence measure (GCoh) measure for 48 channels. 64-lead EEG recordings are from a subject undergoing anesthesia induction (48 channels after referencing). For 10 Hz (Alpha band), we can see a significant change in synchronized activity (GCoh) staring from time 50 to 100 minutes. For 25 Hz (Beta band), there is no significant change in GCoh over the course of the experiment.

Decoding real data in 2D using the marked point process state-space model. We use the marked point process state-space to decode replay events. The sequential reactivation of place cells corresponds to previously experienced trajectories (memories) or future planned trajectories (mental exploration, decision-making)

Closed-loop BCI elements include brain, behavior, and stimulus. The stimulus modulates brain dynamics and will influence brain state and associated behavior. The stimulus can be invasive and non-invasive; in this research, architectural elements are the stimulus that changes brain dynamics to promote human mental health.