Yue Ruan1,2,3, Alexis Bellot4,5, Zuzana Moysova6, Garry D Tan1,2, Alistair Lumb1,2, Jim Davies6, Mihaela van der Schaar4,5, Rustam Rea7,2. 1. Oxford Centre for Diabetes, Endocrinology and Metabolism, Oxford University Hospitals National Health Service Foundation Trust, Oxford, U.K. 2. Oxford National Institute for Health Research Biomedical Research Centre, Oxford, U.K. 3. Oxford Centre for Diabetes, Endocrinology and Metabolism, University of Oxford, Oxford, U.K. 4. Department of Mathematics, University of Cambridge, Cambridge, U.K. 5. Alan Turing Institute, London, U.K. 6. Big Data Institute, University of Oxford Li Ka Shing Centre for Health Information and Discovery, University of Oxford, Oxford, U.K. 7. Oxford Centre for Diabetes, Endocrinology and Metabolism, Oxford University Hospitals National Health Service Foundation Trust, Oxford, U.K. rustam.rea@nhs.net.
Abstract
OBJECTIVE: We analyzed data from inpatients with diabetes admitted to a large university hospital to predict the risk of hypoglycemia through the use of machine learning algorithms. RESEARCH DESIGN AND METHODS: Four years of data were extracted from a hospital electronic health record system. This included laboratory and point-of-care blood glucose (BG) values to identify biochemical and clinically significant hypoglycemic episodes (BG ≤3.9 and ≤2.9 mmol/L, respectively). We used patient demographics, administered medications, vital signs, laboratory results, and procedures performed during the hospital stays to inform the model. Two iterations of the data set included the doses of insulin administered and the past history of inpatient hypoglycemia. Eighteen different prediction models were compared using the area under the receiver operating characteristic curve (AUROC) through a 10-fold cross validation. RESULTS: We analyzed data obtained from 17,658 inpatients with diabetes who underwent 32,758 admissions between July 2014 and August 2018. The predictive factors from the logistic regression model included people undergoing procedures, weight, type of diabetes, oxygen saturation level, use of medications (insulin, sulfonylurea, and metformin), and albumin levels. The machine learning model with the best performance was the XGBoost model (AUROC 0.96). This outperformed the logistic regression model, which had an AUROC of 0.75 for the estimation of the risk of clinically significant hypoglycemia. CONCLUSIONS: Advanced machine learning models are superior to logistic regression models in predicting the risk of hypoglycemia in inpatients with diabetes. Trials of such models should be conducted in real time to evaluate their utility to reduce inpatient hypoglycemia.
OBJECTIVE: We analyzed data from inpatients with diabetes admitted to a large university hospital to predict the risk of hypoglycemia through the use of machine learning algorithms. RESEARCH DESIGN AND METHODS: Four years of data were extracted from a hospital electronic health record system. This included laboratory and point-of-care blood glucose (BG) values to identify biochemical and clinically significant hypoglycemic episodes (BG ≤3.9 and ≤2.9 mmol/L, respectively). We used patient demographics, administered medications, vital signs, laboratory results, and procedures performed during the hospital stays to inform the model. Two iterations of the data set included the doses of insulin administered and the past history of inpatient hypoglycemia. Eighteen different prediction models were compared using the area under the receiver operating characteristic curve (AUROC) through a 10-fold cross validation. RESULTS: We analyzed data obtained from 17,658 inpatients with diabetes who underwent 32,758 admissions between July 2014 and August 2018. The predictive factors from the logistic regression model included people undergoing procedures, weight, type of diabetes, oxygen saturation level, use of medications (insulin, sulfonylurea, and metformin), and albumin levels. The machine learning model with the best performance was the XGBoost model (AUROC 0.96). This outperformed the logistic regression model, which had an AUROC of 0.75 for the estimation of the risk of clinically significant hypoglycemia. CONCLUSIONS: Advanced machine learning models are superior to logistic regression models in predicting the risk of hypoglycemia in inpatients with diabetes. Trials of such models should be conducted in real time to evaluate their utility to reduce inpatient hypoglycemia.
Authors: Jingtong Huang; Andrea M Yeung; Kevin T Nguyen; Nicole Y Xu; Jean-Charles Preiser; Robert J Rushakoff; Jane Jeffrie Seley; Guillermo E Umpierrez; Amisha Wallia; Andjela T Drincic; Roma Gianchandani; M Cecilia Lansang; Umesh Masharani; Nestoras Mathioudakis; Francisco J Pasquel; Signe Schmidt; Viral N Shah; Elias K Spanakis; Andreas Stuhr; Gerlies M Treiber; David C Klonoff Journal: J Diabetes Sci Technol Date: 2022-07-29
Authors: Sreekar Mantena; Aldo Robles Arévalo; Jason H Maley; Susana M da Silva Vieira; Roselyn Mateo-Collado; João M da Costa Sousa; Leo Anthony Celi Journal: J Clin Monit Comput Date: 2021-10-04 Impact factor: 1.977
Authors: Nestoras N Mathioudakis; Mohammed S Abusamaan; Ahmed F Shakarchi; Sam Sokolinsky; Shamil Fayzullin; John McGready; Mihail Zilbermint; Suchi Saria; Sherita Hill Golden Journal: JAMA Netw Open Date: 2021-01-04
Authors: Kristin E Criner; Han Na Kim; Hira Ali; Smita J Kumar; Justin E Kanter; Li Wang; Mary T Korytkowski Journal: J Diabetes Complications Date: 2021-06-19 Impact factor: 3.219