Prediction of Length of Stay and Hospital Readmission for Diabetic Patients

Jain, Silky (2021) Prediction of Length of Stay and Hospital Readmission for Diabetic Patients. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration manual) Download (2MB) | Preview

Abstract

Unplanned readmission and unexpected long stay of diabetic patients is one of the biggest problems of the hospitals. This impacts the management and patient care services provided at hospitals. For efficient management and optimal utilization of hospital resources such as bed allocation, diagnoses test labs etc., it becomes necessary to predict the risk of readmission and length of stay of patients. This research proposed a stacked ensemble learning model with a comprehensive methodological approach including data cleaning and transformation techniques like feature encoding, selection, normalization, feature importance etc., to predict readmission and length of stay of patients. The motive is to ensure the quality of data to produce a highly effective and efficient predictive model that helps in saving high healthcare expenses. The models like Random Forest (RF), Decision Tree (DT), Support Vector Machines (SVM), Logistic Regression (LR), k Nearest Neighbors (kNN), Gradient Boosting (GB) and Extreme Gradient Boosting (XGB) are used as base models to build the stacked model based on the performance and parameter tuning of the models. The final stacked ensemble model not only outperforms all the base classifiers but also the existing work in the same field. For hospital readmission prediction stacked model gives the accuracy, precision, specificity, F1Score, and AUC of 94%, 90%, 95%, 90% and 98% respectively and for the prediction of length of stay gives 90% for each evaluation metric. However, the research could include Deep learning for better models and results which is addressed in future works.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Healthcare; Diabetes; Predictive Analytics; Hospital Readmission; Length of Stay; Stacked Ensemble Learning Model; Data Pre-processing
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > QA Mathematics > Computer software T Technology > T Technology (General) > Information Technology > Computer software H Social Sciences > HM Sociology > Information Science > Communication > Medical Informatics
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Clara Chan
Date Deposited:	03 Dec 2021 14:47
Last Modified:	03 Dec 2021 14:47
URI:	https://norma.ncirl.ie/id/eprint/5168

Actions (login required)

View Item