Learning Models for Traumatic Brain Injury Mortality Prediction on Pediatric Electronic Health Records

Abstract

BackgroundTraumatic Brain Injury (TBI) is one of the leading causes of injury related mortality in the world, with severe cases reaching mortality rates of 30-40%. It is highly heterogeneous both in causes and consequences, complicating medical interpretation and prognosis. Gathering clinical, demographic, and laboratory data to perform a prognosis requires time and skill in several clinical specialties. Machine learning (ML) methods can take advantage of the data and guide physicians toward a better prognosis and, consequently, better healthcare. The objective of this study was to develop and test a wide range of machine learning models and evaluate their capability of predicting mortality of TBI, at hospital discharge, while assessing the similarity between the predictive value of the data and clinical significance.MethodsThe used dataset is the Hackathon Pediatric Traumatic Brain Injury (HPTBI) dataset, composed of electronic health records containing clinical annotations and demographic data of 300 patients. Four different classification models were tested, either with or without feature selection. For each combination of the classification model and feature selection method, the area under the receiver operator curve (ROC-AUC), balanced accuracy, precision, and recall were calculated.ResultsMethods based on decision trees perform better when using all features (Random Forest, AUC = 0.86 and XGBoost, AUC = 0.91) but other models require prior feature selection to obtain the best results (k-Nearest Neighbors, AUC = 0.90 and Artificial Neural Networks, AUC = 0.84). Additionally, Random Forest and XGBoost allow assessing the feature's importance, which could give insights for future strategies on the clinical routine.ConclusionPredictive capability depends greatly on the combination of model and feature selection methods used but, overall, ML models showed a very good performance in mortality prediction for TBI. The feature importance results indicate that predictive value is not directly related to clinical significance.

Open source and reproducible and inexpensive infrastructure for data challenges and education

“…An additional example of reuse: one research group that did not participate in the data challenge used only the publicly available training data to conduct a mortality analysis very similar to one of the data challenge tasks 19 .…”

Traumatic Brain Injury Rehabilitation Outcome Prediction Using Machine Learning Methods

“…Compared with LM, advanced ML methods allow greater flexibility for modeling non-linear recovery pattern, interactions between treatments, diminishing returns, ceiling/floor effect, which better reflects real-world settings. 6 A few studies have applied ML methods to rehabilitation data and predict outcomes in different patient populations affected by mild TBI, 7 , 8 stroke, 9 , 10 and predict FIM scores at discharge, 11 survival or mortality probability after TBI, 6 , 12 , 13 , 14 , 15 , 16 , 17 , 18 suicidal ideation after TBI. 19 In contrast, Bruschetta et al 20 did not find ML methods to have superiority over LM in predicting outcome after TBI and was limited by quantity of predictor variables.…”