Additional Projects

Graduate Work

Early COVID-19 respiratory risk stratification using machine learning

Collaborators: Dr. Molly Douglas, Brian Bell, Brian Toner, Sarah Pungitore

We develop machine learning models to assist in triage by predicting patients needing intubation within the next 48 hours based on objective clinical parameters.

Citation: [1] Douglas, Molly J., Bell, B. W., Kinney, A. C., Pungitore, S. A., Toner, B. P., "Early COVID-19 respiratory risk stratification using machine learning." Trauma surgery & acute care open 7.1 (2022): e000892.

Figure 1

Visual abstract of [1] showing the variables used in a regularized logistic regression model used to predict intubation within 48 hours. Figure reproduced from [1].

Undergraduate Work

Modeling Aggregation of Proliferating Microglia in Response to Amyloid-beta in Dementia

Advisor: Dr. Ellen Swanson

We develop a model describing the relationship between accumulation of amyloid-beta and abnormalities in microglial cell clusters in Alzheimer's Disease brains. We start with a published model of PDEs predicting changes in spatila concentration of microglial cells in reponse to amyloid-beta. The PDE model assumed constant numbers of microglial cells, but these cells proliferate in response to amyloid-beta. Thus, we couple the PDE model with an ODE model predicting changes in the number of glial cells.

Citation: [2] Kinney, Adrienne C., and Ellen R. Swanson. "Modeling aggregation of proliferating microglia in response to amyloid-beta in dementia." Spora: A Journal of Biomathematics 3.1 (2017): 4.

(a)

(b)

Figure 2

Predicted concentration of reactive microglia cells compared to the pixel intensity profile of microglia clusters in a PD brain. (a) Chemotaxis model and (b) Chemotaxis model with nonuniform proliferation of reactive microglia. Figure reproduced from [2].

Page updated

Report abuse