Chi Peng1, Fan Yang2, Lulu Li3, Liwei Peng4, Jian Yu1, Peng Wang4, Zhichao Jin5. 1. Department of Health Statistics, Second Military Medical University, Shanghai, China. 2. Institute of Pathology and Southwest Cancer Center, Southwest Hospital, Third Military Medical University (Army Medical University) and Key Laboratory of Tumor Immunopathology, Ministry of Education of China, Chongqing, China. 3. Department of Orthopedics, Changhai Hospital, Second Military Medical University, Shanghai, China. 4. Department of Neurosurgery, Tangdu Hospital, Fourth Military Medical University, Xi'an, China. 5. Department of Health Statistics, Second Military Medical University, Shanghai, China. jinzhichao@smmu.edu.cn.
Abstract
BACKGROUND: Acute kidney injury (AKI), a prevalent non-neurological complication following traumatic brain injury (TBI), is a major clinical issue with an unfavorable prognosis. This study aimed to develop and validate machine learning models to predict severe AKI (stage 3 or greater) incidence in patients with TBI. METHODS: A retrospective cohort study was conducted by using two public databases: the Medical Information Mart for Intensive Care IV (MIMIC)-IV and the eICU Collaborative Research Database (eICU-CRD). Recursive feature elimination was used to select candidate predictors obtained within 24 h of intensive care unit admission. The area under the curve and decision curve analysis curves were used to determine the discriminatory ability. On the other hand, the calibration curve was employed to evaluate the calibrated performance of the newly developed machine learning models. RESULTS: In the MIMIC-IV database, there were 808 patients diagnosed with moderate and severe TBI (msTBI) (msTBI is defined as Glasgow Coma Score < 12). Of these, 60 (7.43%) patients experienced severe AKI. External validation in the eICU-CRD indicated that the random forest (RF) model had the highest area under the curve of 0.819 (95% confidence interval 0.783-0.851). Furthermore, in the calibration curve, the RF model was well calibrated (P = 0.795). CONCLUSIONS: In this study, the RF model demonstrated better discrimination in predicting severe AKI than other models. An online calculator could facilitate its application, potentially improving the early detection of severe AKI and subsequently improving the clinical outcomes among patients with msTBI.
BACKGROUND: Acute kidney injury (AKI), a prevalent non-neurological complication following traumatic brain injury (TBI), is a major clinical issue with an unfavorable prognosis. This study aimed to develop and validate machine learning models to predict severe AKI (stage 3 or greater) incidence in patients with TBI. METHODS: A retrospective cohort study was conducted by using two public databases: the Medical Information Mart for Intensive Care IV (MIMIC)-IV and the eICU Collaborative Research Database (eICU-CRD). Recursive feature elimination was used to select candidate predictors obtained within 24 h of intensive care unit admission. The area under the curve and decision curve analysis curves were used to determine the discriminatory ability. On the other hand, the calibration curve was employed to evaluate the calibrated performance of the newly developed machine learning models. RESULTS: In the MIMIC-IV database, there were 808 patients diagnosed with moderate and severe TBI (msTBI) (msTBI is defined as Glasgow Coma Score < 12). Of these, 60 (7.43%) patients experienced severe AKI. External validation in the eICU-CRD indicated that the random forest (RF) model had the highest area under the curve of 0.819 (95% confidence interval 0.783-0.851). Furthermore, in the calibration curve, the RF model was well calibrated (P = 0.795). CONCLUSIONS: In this study, the RF model demonstrated better discrimination in predicting severe AKI than other models. An online calculator could facilitate its application, potentially improving the early detection of severe AKI and subsequently improving the clinical outcomes among patients with msTBI.
Authors: Angelos G Kolias; Andres M Rubiano; Anthony Figaji; Franco Servadei; Peter J Hutchinson Journal: Lancet Neurol Date: 2019-02 Impact factor: 44.182
Authors: Elizabeth M Moore; Rinaldo Bellomo; Alistair Nichol; Nerina Harley; Christopher Macisaac; D James Cooper Journal: Ren Fail Date: 2010 Impact factor: 2.606
Authors: Feras Akbik; Cederic Pimentel-Farias; Di'Jonai A Press; Niara E Foster; Kevin Luu; Merin G Williams; Sena G Andea; Regina K Kyei; Grace M Wetsel; Jonathan A Grossberg; Brian M Howard; Frank Tong; C Michael Cawley; Owen B Samuels; Ofer Sadan Journal: Neurocrit Care Date: 2021-12-21 Impact factor: 3.210
Authors: Stefan Büttner; Andrea Stadler; Christoph Mayer; Sammy Patyna; Christoph Betz; Christian Senft; Helmut Geiger; Oliver Jung; Fabian Finkelmeier Journal: J Intensive Care Med Date: 2018-01-29 Impact factor: 3.510
Authors: Ofer Sadan; Kai Singbartl; Jacqueline Kraft; Joao McONeil Plancher; Alexander C M Greven; Prem Kandiah; Cederic Pimentel; C L Hall; Alexander Papangelou; William H Asbury; John J Hanfelt; Owen Samuels Journal: J Intensive Care Date: 2020-05-04