Application of support vector machine in predicting in-hospital mortality risk of patients with acute kidney injury in ICU.
- Author:
Ke LIN
1
,
2
;
Jun Qing XIE
1
,
2
;
Yong Hua HU
1
,
2
;
Gui Lan KONG
3
Author Information
1. Department of Epidemiology and Biostatistics, Peking University School of Public Health, Beijing 100191, China
2. Medical Informatics Center, Peking University, Beijing 100191, China.
3. Medical Informatics Center, Peking University, Beijing 100191, China.
- Publication Type:Journal Article
- MeSH:
Acute Kidney Injury/mortality*;
Critical Care;
Hospital Mortality;
Humans;
Intensive Care Units;
Prognosis;
ROC Curve;
Sensitivity and Specificity;
Support Vector Machine
- From:
Journal of Peking University(Health Sciences)
2018;50(2):239-244
- CountryChina
- Language:Chinese
-
Abstract:
OBJECTIVE:To construct an in-hospital mortality prediction model for patients with acute kidney injury (AKI) in intensive care unit (ICU) by using support vector machine (SVM), and compare it with the simplified acute physiology score II (SAPS-II) which is commonly used in the ICU.
METHODS:We used Medical Information Mart for Intensive Care III (MIMIC-III) database as data source. The AKI patients in the MIMIC-III database were selected according to the 2012 Kidney Disease: Improving Global Outcomes (KDIGO) definition of AKI. We employed the same predictor variable set as used in SAPS-II to construct an SVM model. Meanwhile, we also developed a customized SAPS-II model using MIMIC-III database, and compared performances between the SVM model and the customized SAPS-II model. The performance of each model was evaluated via area under the receiver operation characteristic curve (AUROC), root mean squared error (RMSE), sensitivity, specificity, Youden's index and accuracy based on 5-fold cross-validation. The agreement of the results between the SVM model and the customized SAPS-II model was illustrated using Bland-Altman plots.
RESULTS:A total number of 19 044 patients with AKI were included. The observed in-hospital mortality of the AKI patients was 13.58% in MIMIC-III. The results based on the 5-fold cross validation showed that the average AUROC of the SVM model and the customized SAPS-II model was 0.86 and 0.81, respectively (The difference between the two models was statistically significant with t=13.0, P<0.001). The average RMSE of the SVM model and the customized SAPS-II model was 0.29 and 0.31, respectively (The difference was statistically significant with t=-9.6, P<0.001). The SVM model also outperformed the customized SAPS-II model in terms of sensitivity and Youden's index with significant statistical differences (P=0.002 and <0.001, respectively).The Bland-Altman plot showed that the SVM model and the customized SAPS-II model had similar mortality prediction results when the mortality of a patient was certain, but the consistency between the mortality prediction results of the two models was poor when the mortality of a patient was with high uncertainty.
CONCLUSION:Compared with the SAPS-II model, the SVM model has a better performance, especially when the mortality of a patient is with high uncertainty. The SVM model is more suitable for predicting the mortality of patients with AKI in ICU and early intervention in patients with AKI in ICU. The SVM model can effectively help ICU clinicians improve the quality of medical treatment, which has high clinical value.
