Use of Kalman Filtering in State and Parameter Estimation of Diabetes Models

Author

Cassidy LeFollow

Graduation Year

2020

Document Type

Open Access Senior Thesis

Degree Name

Bachelor of Science

Department

Mathematics

Reader 1

Lisette de Pillis

Reader 2

Blerta Shtylla

Terms of Use & License Information

Terms of Use for work posted in Scholarship@Claremont.

Rights Information

2020 Cassidy My Huong Le

Abstract

Diabetes continues to affect many lives every year, putting those affected by it at higher risk of serious health issues. Despite many efforts, there currently is no cure for diabetes. Nevertheless, researchers continue to study diabetes in hopes of understanding the disease and how it affects people, creating mathematical models to simulate the onset and progression of diabetes. Recent research by David J. Albers, Matthew E. Levine, Andrew Stuart, Lena Mamykina, Bruce Gluckman, and George Hripcsak¹ has suggested that these models can be furthered through the use of Data Assimilation, a regression method that synchronizes a model with a particular set of data by estimating the system's states and parameters. In my thesis, I explore how Data Assimilation, specifically different types of Kalman filters, can be applied to various models, including a diabetes model.

¹Albers, David J, Matthew E Levine, Andrew Stuart, Lena Mamykina, Bruce Gluckman, and George Hripcsak. 2018. Mechanistic machine learning: how data assimilation leverages physiologic knowledge using bayesian inference to forecast the future, infer the present, and phenotype. JAMIA 25(10):1392–1401. doi:10.1093/jamia/ocy106. https: //doi.org/10.1371/journal.pone.0048058.

Comments

All code affiliated with this research can be found in the following GitHub repository: https://github.com/CassidyLe98/Thesis_KalmanFilters.

Recommended Citation

Le, Cassidy, "Use of Kalman Filtering in State and Parameter Estimation of Diabetes Models" (2020). HMC Senior Theses. 232.
https://scholarship.claremont.edu/hmc_theses/232