I am a postdoctoral fellow in the Windreich Department of Artificial Intelligence and Human Health at Icahn School of Medicine at Mount Sinai, specializing in AI-driven modeling and inference. I have obtained my PhD in the Biomedical Engineering program at New York University (NYU) Tandon School of Engineering, where I was advised by Dr. Rose T. Faghih at the Computational Medicine Lab (CML).

My research lies at the nexus of theory, computing, and data, where I developed state-space frameworks to infer latent physiological conditions from multimodal data. Due to the complex human anatomy and physiology, inferring underlying physiological conditions poses challenges such as inter-individual and intra-individual variability and sparsity of data. My research aims to address these challenges by employing personalized, automated, and system-theoretic toolboxes that can infer the underlying physiological condition and facilitate the diagnosis and treatment.

Selected Lines of Research

Identifying the cognitive arousal-performance link during a working memory experiment
The cognitive arousal and performance are two hidden brain states that are linked closely, and the well-known Yerkes-Dodson law proposed an inverted-U link between the arousal and performance states. This line of research decodes these hidden states and identifies a mathematical model that can express the link between the decoded arousal and performance aligned with the Yerkes-Dodson law.

Modeling and decoding the hidden performance state in an adaptive framework
The cognitive performance state is often modeled by assuming a linear state model and time-invariant model parameters, such as process noise variance. Such assumptions may not resemble a real-world scenario in which the environmental stimuli can impact the hidden cognitive state and lead to a non-linear and time-varying process noise dynamic. This line of research explores the models and decoders that can account for the aforementioned impacts in a non-linear and time-varying paradigm.