Catherine Potts

Understanding how neurons communicate with each other in the brain and any effects that nanoparticles may have on neuronal communication is a subject area of great interest for biomedicine. Surprisingly, most analysis methods focus on individual neuronal calcium behavior, not the behavior of the interactions between neurons. To capture these interactions, we apply the unsupervised machine learning method Archetypal Analysis, whose computational elements can be easily interpreted in the context of a neuronal cellular signaling.

In this talk, Catherine will discuss her Ph.D. research, in collaboration with MSU's Kunze Neuroengineering Lab and the MT PEAKS fellowship, to apply Archetypal Analysis to neuronal calcium images generated from nanoparticle experiments. She will also discuss Archetypal Analysis in terms of a machine learning model, the geometric interpretation of its computational elements, and compare it to Principal Component Analysis using an image processing database. 

Potts is the recipient of a 2020 Kopriva Graduate Fellowship.

About the Kopriva Science Seminar Series

Potts's lecture is presented by the Kopriva Science Seminar Series, which is funded through an endowment created by Phil Kopriva, a 1957 microbiology graduate from MSU. Kopriva, who died in 2002, also created an endowment to fund the Kopriva Graduate Fellowship Program, which provides support and opportunities for graduate students in the College of Letters and Science, particularly in the biomedical sciences. The series features seminars by MSU graduate students, faculty members and guest speakers.

• Press release