The research of my team is in the area of nonlinear dynamics and complex systems. We study how dynamics impacts life: We are focused on the intrinsic dynamics of living systems such as the continuous shape changes, motions, and membrane potential modulations of epithelial and immune cells, and the continuous changes in ionic signals in neural networks. We also learn from nature's way of harnessing dynamics to discover and validate new AI algorithms and architectures.
Biosensing with multimodal excitability: I led an AFOSR Biophysics program MURI on multimodal excitability from 2015-2022. We discovered that electrical, mechanical and chemical dynamics can often be described as excitable systems with characteristic rhythms and refractory periods. Just like electrical excitability carries information in neurons - biomechanical and chemical excitability can also retain and transmit actionable information. We discovered, for example, that biomechanical waves can act as primary sensors of the physical environment of a cell, in particular electric fields and texture (Yang et al PNAS, 2023).
Computing with multimodal excitability: My team investigates the computing characteristics of living neural networks with a focus on the role of astrocytes (which exhibit mechano-chemical excitability). Since 2025 I lead a MURI team to pursue reasearch on hybrid-AI - taking the multimodal characteristics of living neural networks one step further and combining the best of living systems with hardware based computing. We discovered new algorithms Hybrid-AI MURI
Sensing Multimodal Excitability with Quantum Technology: New quantum Sensor Technologies show great promise for sensing and driving the subtle changes in the charge and metabolic environments associated with cellular and tissue excitability. We explore how Quantum Technologies will allow us to detect and control the multimodal excitable rhythms of cells. Based on our prior work we expect that these measurements will complement existing bioanalytical and -omics tools. My group co-organized an NSF sponsored workshop on Quantum Biotechnology
Excitablility as a foundation for new AI algorithms: My team develops new approaches to AI based on multimodal excitability concepts. We are finding that the addition of multimodal exciability architectures and algorithms can make machine learning and AI algorithms more robust, sensitve, and effective. One key astrocyte inspired algorithm, which we call rhythmic sharing, won the invention of the year award of the University of Maryland in 2025.
AI and Physics: AI is emerging as a powerful driver of innovation across all areas of physics including the physics of living systems. Since I started working on Machine Learning in 2014, I have contributed to groups within the American Physical Society that highlight the role of physicists in shaping the AI frontier. I served on the inaugural leadership team that started a new APS Group on Data Science (2017-2022), and now serve on the leadership team of the Division of Statistical and Nonlinear Physics (2025-2029).
