Quantitative Biosciences Thesis Proposal

Shu Gong

School of Biological Sciences

Data-Driven Modeling of Muscle Sensorimotor Physiological States 

Date: Tuesday, September 16th

Time: 12:00 PM - 2:00PM ET

Location: (in person) Room 210, J. Erskine Love Jr. Manufacturing Building.

Open to the Community

Advisor:

Dr. Gregory Sawicki (School of Biological Sciences, School of Mechanical Engineering)

Committee Members:

Dr. Simon Sponberg (School of Biological Sciences, School of Physics)

Dr. Hannah Choi (School of Mathematics)

Dr. Aaron Young (School of Mechanical Engineering)

Dr. Monica Daley (School of Biological Sciences, University of California, Irvine)

Abstract:

Muscles enable versatile, dynamic behaviors essential for survival, yet conventional models based on simplified assumptions fail to capture sensorimotor dynamics under unsteady or perturbed conditions. This proposal aims to develop generalizable, interpretable, data-driven muscle models that accurately estimate sensorimotor physiological states in real-world contexts. To achieve this, I will (1) develop a recurrent neural network–based model of muscle force from muscle fiber data under systematically varied perturbations, (2) develop a human muscle length estimation model using a novel joint-level dataset that manipulates both internal motor commands and external environmental factors, and (3) develop a data-driven muscle spindle model trained on in vivo animal datasets, enhanced for interpretability and robustness through integration with biophysical models. Together, these aims will establish a new paradigm for modeling muscle sensorimotor physiological states, providing a foundation for next-generation human–machine interfaces.