Automated Model Discovery for Mechanics of Artificial and Real Meats
Publication: St Pierre SR, Sibley LS, Tran S, Tran V, Darwin EC, Kuhl E. Biaxial testing and sensory texture evaluation of plant-based and animal deli meat. Curr Res Food Sci. 2025 Jun 4;10:101080.
PMID: 40529651 PMCID: PMC12173621 DOI: 10.1016/j.crfs.2025.101080
Summer research conducted with the Living Matters Lab of the Stanford Mechanical Engineering Department. Mentor: Skyler St. Pierre. Principal investigator: Ellen Kuhl.
Previous methods for discovering mathematical equations to predict material behavior are outdated. We used a neural network architecture ( machine learning ) to predict the elastic behavior of meat and artificial meat.
I collected raw training data in the biaxial tensile tester, processed the data, and adopted neural network code to fit mathematical models.
| Hard Skills |
|---|
| Experimental Testing |
| Machine Learning and Neural Networks |
| Data Collection |
| Data Processing |
| Research and Development |
| Scientific Writing |
| Soft Skills |
|---|
| Science Communication |
| Meticulous Mentor-Mentee Communication |
| Task Management |
| Delegation of Work |
| Positive Influence |
| Presentating at Research Symposium |
| Devices / Tools / Software |
|---|
| Biaxial Machine |
| Python: Matlib Plot, SciPy, Tensor Flow |
The Behind Story
Challenge: Skill Acquisitions In an Interdisciplinary Field
Solution: Dedication, Diligence, Communication, Research
With any challenge comes acknowledging what you know, don’t know, and how to get to a spot where you can be successful with mentorship. My research in biomaterial characterization and testing resides in the interdisciplinary sector of material science and engineering, research, data collection, machine learning, and mechancial engineering. Thus, I was out of my element coming from only a mechanical engineering background. In a fast-paced research setting, a lot of onboarding and learning had to be done; however, as a student, I am used to this. My natural curiosity and efforts to learn made me a quick and valuable researcher. Any time I did not know something, I would ask my research peers questions, and then I would communicate with my mentor. Ultimately, learning how to use a neural network for the first time, a principal concept in machine learning, I felt rewarded with rainbow data plots and mathematical equations to describe my precious meats. This research can be used to inform and improve artificial meat products.
