Jung In Park1, Doyub Kim2, Jung-Ah Lee3, Kai Zheng4, Alpesh Amin5. 1. Assistant Professor, Sue & Bill Gross School of Nursing, University of California, Irvine, CA. 2. Principle Software Engineer, NVIDIA, Santa Clara, CA. 3. Associate Professor, Sue & Bill Gross School of Nursing, University of California, Irvine, CA. 4. Professor, Donald Bren School of Information & Computer Sciences, University of California, Irvine, CA. 5. Professor, School of Medicine, University of California, Irvine, CA.
Abstract
PURPOSE: The aim of the study was to develop and validate machine learning models to predict the personalized risk for 30-day readmission with venous thromboembolism (VTE). DESIGN: This study was a retrospective, observational study. METHODS: We extracted and preprocessed the structured electronic health records (EHRs) from a single academic hospital. Then we developed and evaluated three prediction models using logistic regression, the balanced random forest model, and the multilayer perceptron. RESULTS: The study sample included 158,804 total admissions; VTE-positive cases accounted for 2,080 admissions from among 1,695 patients (1.31%). Based on the evaluation results, the balanced random forest model outperformed the other two risk prediction models. CONCLUSIONS: This study delivered a high-performing, validated risk prediction tool using machine learning and EHRs to identify patients at high risk for VTE after discharge. CLINICAL RELEVANCE: The risk prediction model developed in this study can potentially guide treatment decisions for discharged patients for better patient outcomes.
PURPOSE: The aim of the study was to develop and validate machine learning models to predict the personalized risk for 30-day readmission with venous thromboembolism (VTE). DESIGN: This study was a retrospective, observational study. METHODS: We extracted and preprocessed the structured electronic health records (EHRs) from a single academic hospital. Then we developed and evaluated three prediction models using logistic regression, the balanced random forest model, and the multilayer perceptron. RESULTS: The study sample included 158,804 total admissions; VTE-positive cases accounted for 2,080 admissions from among 1,695 patients (1.31%). Based on the evaluation results, the balanced random forest model outperformed the other two risk prediction models. CONCLUSIONS: This study delivered a high-performing, validated risk prediction tool using machine learning and EHRs to identify patients at high risk for VTE after discharge. CLINICAL RELEVANCE: The risk prediction model developed in this study can potentially guide treatment decisions for discharged patients for better patient outcomes.
Authors: Krista J Hachey; Helene Sterbling; Daniel S Choi; Emma Pinjic; Philip D Hewes; Juan Munoz; David McAneny; Yorghos Tripodis; Hiran C Fernando; Virginia R Litle Journal: J Am Coll Surg Date: 2015-12-18 Impact factor: 6.113
Authors: Eric A Secemsky; Kenneth Rosenfield; Kevin F Kennedy; Michael Jaff; Robert W Yeh Journal: J Am Heart Assoc Date: 2018-06-26 Impact factor: 5.501
Authors: Michael B Streiff; Jeffrey P Brady; Althea M Grant; Scott D Grosse; Betty Wong; Tanja Popovic Journal: MMWR Morb Mortal Wkly Rep Date: 2014-03-07 Impact factor: 17.586