Ke Lin1, Yonghua Hu2, Guilan Kong3. 1. National Institute of Health Data Science, Peking University, Beijing, China; Center for Data Science in Health and Medicine, Peking University, Beijing, China; Department of Epidemiology and Biostatistics, School of Public Health, Peking University, Beijing, China. 2. Department of Epidemiology and Biostatistics, School of Public Health, Peking University, Beijing, China; Medical Informatics Center, Peking University, Beijing, China. 3. National Institute of Health Data Science, Peking University, Beijing, China; Center for Data Science in Health and Medicine, Peking University, Beijing, China. Electronic address: guilan.kong@hsc.pku.edu.cn.
Abstract
OBJECTIVES: We aimed to construct a mortality prediction model using the random forest (RF) algorithm for acute kidney injury (AKI) patients in the intensive care unit (ICU), and compared its performance with that of two other machine learning models and the customized simplified acute physiology score (SAPS) II model. METHODS: We used medical information mart for intensive care (MIMIC) III database for model development and performance comparison. The RF model uses the same predictor variable set as that of the SAPS II model. We also developed three other models and compared the RF model with the other three models in prediction performance. Three other models include support vector machine (SVM) model, artificial neural network (ANN) model and customized SAPS II model. In model comparison, the prediction performance of each model was assessed by the Brier score, the area under the receiver operating characteristic curve (AUROC), accuracy and F1 score. RESULTS: The final cohort consisted of 19044 patients with AKI in the ICU. The observed in-hospital mortality of AKI patients is 13.6% in the final cohort. The results of model performance comparison show that the Brier score of the RF model is 0.085 (95%CI: 0.084-0.086) and AUROC of the RF model is 0.866 (95%CI: 0.862-0.870). The accuracy of the RF model is 0.728 (95%CI: 0.715-0.741). The F1 score of the RF model is 0.459 (95%CI: 0.449-0.470). The calibration plots show that the RF model slightly overestimates mortality in patients with low risk of death and underestimates mortality in patients with high risk of death. CONCLUSION: There is great potential for the RF model in mortality prediction for AKI patients in ICU. The RF model may be helpful to aid ICU clinicians to make timely clinical intervention decisions for AKI patients, which is critical to help reduce the in-hospital mortality of AKI patients. A prospective study is necessary to evaluate the clinical utility of the RF model.
OBJECTIVES: We aimed to construct a mortality prediction model using the random forest (RF) algorithm for acute kidney injury (AKI) patients in the intensive care unit (ICU), and compared its performance with that of two other machine learning models and the customized simplified acute physiology score (SAPS) II model. METHODS: We used medical information mart for intensive care (MIMIC) III database for model development and performance comparison. The RF model uses the same predictor variable set as that of the SAPS II model. We also developed three other models and compared the RF model with the other three models in prediction performance. Three other models include support vector machine (SVM) model, artificial neural network (ANN) model and customized SAPS II model. In model comparison, the prediction performance of each model was assessed by the Brier score, the area under the receiver operating characteristic curve (AUROC), accuracy and F1 score. RESULTS: The final cohort consisted of 19044 patients with AKI in the ICU. The observed in-hospital mortality of AKI patients is 13.6% in the final cohort. The results of model performance comparison show that the Brier score of the RF model is 0.085 (95%CI: 0.084-0.086) and AUROC of the RF model is 0.866 (95%CI: 0.862-0.870). The accuracy of the RF model is 0.728 (95%CI: 0.715-0.741). The F1 score of the RF model is 0.459 (95%CI: 0.449-0.470). The calibration plots show that the RF model slightly overestimates mortality in patients with low risk of death and underestimates mortality in patients with high risk of death. CONCLUSION: There is great potential for the RF model in mortality prediction for AKI patients in ICU. The RF model may be helpful to aid ICU clinicians to make timely clinical intervention decisions for AKI patients, which is critical to help reduce the in-hospital mortality of AKI patients. A prospective study is necessary to evaluate the clinical utility of the RF model.
Authors: Xinsong Du; Jae Min; Chintan P Shah; Rohit Bishnoi; William R Hogan; Dominick J Lemas Journal: Int J Med Inform Date: 2020-04-15 Impact factor: 4.046
Authors: Behrooz Mamandipoor; Fernando Frutos-Vivar; Oscar Peñuelas; Richard Rezar; Konstantinos Raymondos; Alfonso Muriel; Bin Du; Arnaud W Thille; Fernando Ríos; Marco González; Lorenzo Del-Sorbo; Maria Del Carmen Marín; Bruno Valle Pinheiro; Marco Antonio Soares; Nicolas Nin; Salvatore M Maggiore; Andrew Bersten; Malte Kelm; Raphael Romano Bruno; Pravin Amin; Nahit Cakar; Gee Young Suh; Fekri Abroug; Manuel Jibaja; Dimitros Matamis; Amine Ali Zeggwagh; Yuda Sutherasan; Antonio Anzueto; Bernhard Wernly; Andrés Esteban; Christian Jung; Venet Osmani Journal: BMC Med Inform Decis Mak Date: 2021-05-07 Impact factor: 2.796