Join the Alabama HIMSS Chapter for a free webinar, entitled:  Mitigating ED Overcrowding Problem Using Machine Learning

When:  June 23, 2022 11:30am

Description:
Emergency department (ED) overcrowding is a major issue affecting hospitals globally. Overcrowding can result in many adverse effects such as increased in-hospital mortality rates, long waiting and treatment times, ambulance diversions, etc. Early prediction of a patient admission status helps to manage the ED’s downstream recourses and mitigate overcrowding. We propose a machine learning framework is to predict the admission status of incoming patients at EDs. The initial triage information such as vital signs, demographic data, and chief complaints are utilized to train and test the proposed models. A retrospective large dataset is obtained from a large hospital located in the Midwest for patients visits records between 2017 and 2019. The framework is composed of two main phases. In phase I, Topic Modeling and Latent Dirichlet Allocation (LDA) are utilized to handle patients’ chief complaints text data. In phase II, patient information and the topics extracted from phase I are used to train and test three classification models: neural network, extreme gradient boosting (XGB), and Adaboost. The models are evaluated based on five performance measures: Accuracy, sensitivity, specificity, F1 measure, and area under the curve (AUC).


 

Speaker Biography:

Ahmed Ph.D. is an Assistant Professor in the Health Informatics Graduate Programs in the Department of Health Services Administration, School of Health Professions (SHP) at the University of Alabama at Birmingham (UAB). Before joining UAB, Dr. Ahmed worked as Assistant Professor at the Business Department at the University of Minnesota Crookston. Dr. Ahmed received his B.Sc. and M.Sc. degrees in Industrial Engineering from Jordan University of Science and Technology (JUST) in 2009 and 2012, respectively. Dr. Ahmed received his Ph.D. in Industrial and Systems Engineering from the State University of New York at Binghamton. He is also a Lean Six Sigma Green and Black Belt certified. Dr. Ahmed teaches in the Graduate Programs in Health Informatics focusing on machine learning courses in the Data Analytics track. His research focuses on the applications of novel optimization and machine learning techniques in improving complex systems in healthcare. Dr. Ahmed has published several conference proceedings and journal articles in prestigious journals such as Expert Systems with Applications, Operations Research for Health Care, and Healthcare Management Science.

Register Here