Date on Master's Thesis/Doctoral Dissertation
5-2022
Document Type
Master's Thesis
Degree Name
M.S.
Department
Industrial Engineering
Degree Program
Industrial Engineering, MS
Committee Chair
Bai, Lihui
Committee Co-Chair (if applicable)
Chen, Xiaoyu
Committee Member
Chen, Xiaoyu
Committee Member
Huang, Jiapeng
Committee Member
Zhang, Hui
Author's Keywords
Hypotension; neural network; mixed response; multivariate; electronic healthcare record
Abstract
Blood Pressure is the main determinant of blood flow to organs. Hypotension is defined as a systolic blood pressure less than 90 mmHg or a diastolic blood pressure less than 50 mmHg. The severity and duration of hypotension is associated with low blood flow to organs often result in organ damage and a high mortality rate. Predicting hypotension prior to surgery and during the surgery can reduce the incidence and duration resulting in better patient outcomes. This thesis uses preoperative bloodwork and vital signs as well as perioperative vital signs in 5-minute increments as inputs to forecast hypotension. Hypotension can be represented by multivariate mixed responses which follows both continuous and binary distributions. The main focus of this thesis is to apply a new method known as an “Interpretable Neural Network” (INN) to this clinical predictive application by simultaneously modeling mixed hypotension responses considering human domain knowledge. The customized INN method was developed and tested with a dataset containing 588 hysterectomy surgeries. It was benchmarked against other models including an Artificial Neural Network (ANN), logistic regression, k-nearest neighbors, support vector classifier, stochastic gradient descent, decision tree, random forest and extra trees. The results suggest while the ANN classification model had the best test accuracy overall, the customized INN model produced better test accuracy with the mixed response.
Recommended Citation
Ritter, Jodie, "Forecasting hypotension by learning from multivariate mixed responses.." (2022). Electronic Theses and Dissertations. Paper 3894.
https://doi.org/10.18297/etd/3894
Included in
Industrial Engineering Commons, Operational Research Commons, Systems Engineering Commons