TY  - JOUR
PY  - 2022//
TI  - Machine learning-based prognostic model for the prediction of early death after traumatic brain injury: Comparison with the Corticosteroid Randomization after Significant Head Injury (CRASH) model
JO  - World neurosurgery
A1  - Lee, Sang Hyub
A1  - Lee, Chul Hee
A1  - Hwang, Soo Hyun
A1  - Kang, Dong Ho
SP  - ePub
EP  - ePub
VL  - ePub
IS  - ePub
N2  - BACKGROUND: Machine learning (ML) has been used to predict the outcomes of traumatic brain injury (TBI). However, few studies have reported the use of ML models to predict early death. This study aimed to develop ML models for early death prediction and to compare performance with the corticosteroid randomization after significant head injury (CRASH) model. <br><br>METHODS: We retrospectively reviewed TBI patients between February 2017 and August 2021. The patients were randomly assigned to a training set and a test set. Predictive variables included clinical findings, laboratory values, and computed tomography findings. The ML models (random forest, support vector machine [SVM], logistic regression) were developed with the training set. The CRASH model is a prognostic model that was developed based on 10,008 patients included in the CRASH trial. The ML and CRASH models were applied to the test set to evaluate the performance. <br><br>RESULTS: A total of 423 patients were included; 317 and 106 patients were randomly assigned to the training and test sets, respectively. The area under the curve (AUC) was highest in the SVM (0.952, 95% CI 0.906-0.990) and lowest in the CRASH model (0.942, 95% CI 0.886-0.999). There were no significant differences between the AUCs of the ML and CRASH models (P =.899 for random forest versus the CRASH model, P =.760 for SVM versus the CRASH model, P =.806 for logistic regression versus the CRASH model). <br><br>CONCLUSIONS: The ML models may have comparable performances compared to the CRASH model despite being developed with a smaller sample size.<p />  <p>Language: en</p>
LA  - en
SN  - 1878-8750
UR  - http://dx.doi.org/10.1016/j.wneu.2022.06.130
ID  - ref1
ER  -