Pediatric severe traumatic brain injury mortality prediction determined with machine learning-based modeling

• Published in: Injury 2022-03, Vol.53 (3), p.992-998
• Main Authors: Daley, Mark, Cameron, Saoirse, Ganesan, Saptharishi Lalgudi, Patel, Maitray A., Stewart, Tanya Charyk, Miller, Michael R., Alharfi, Ibrahim, Fraser, Douglas D.
• Format: Article
• Language: English
• Subjects: Adolescent; Brain Injuries, Traumatic - therapy; Child; Children; Glasgow Coma Scale; Humans; Machine Learning; Mortality; Prognosis; Prognostic modeling; Retrospective Studies; Severe traumatic brain injury; Tomography, X-Ray Computed
• ISSN: 0020-1383; 1879-0267
• DOI: 10.1016/j.injury.2022.01.008

•Accurate prognostication of severe TBI, a leading killer of youth, can help guide clinicians for interventions and/or end-of-life discussions.•A novel, highly discriminative clinical pediatric sTBI outcome prediction model was created by applying advanced machine learning techniques.•The final six-variable mortality prediction model included PTT, motor GCS, serum glucose, fixed pupil(s), platelet count and creatinine.•The resulting parsimonious model had an 82% classification accuracy and an AUC of 0.90, signifying high discriminative ability. Severe traumatic brain injury (sTBI) is a leading cause of mortality in children. As clinical prognostication is important in guiding optimal care and decision making, our goal was to create a highly discriminative sTBI outcome prediction model for mortality. Machine learning and advanced analytics were applied to the patient admission variables obtained from a comprehensive pediatric sTBI database. Demographic and clinical data, head CT imaging abnormalities and blood biochemical data from 196 children and adolescents admitted to a tertiary pediatric intensive care unit (PICU) with sTBI were integrated using feature ranking by way of a forest of randomized decision trees, and a model was generated from a reduced number of admission variables with maximal ability to discriminate outcome. In total, 36 admission variables were analyzed using feature ranking with variable weighting to determine their predictive importance for mortality following sTBI. Reduction analysis utilizing Borata feature selection resulted in a parsimonious six-variable model with a mortality classification accuracy of 82%. The final admission variables that predicted mortality were: partial thromboplastin time (22%); motor Glasgow Coma Scale (21%); serum glucose (16%); fixed pupil(s) (16%); platelet count (13%) and creatinine (12%). Using only these six admission variables, a t-distributed stochastic nearest neighbor embedding algorithm plot demonstrated visual separation of sTBI patients that lived or died, with high mortality predictive ability of this model on the validation dataset (AUC = 0.90) which was confirmed with a conventional area-under-the-curve statistical approach on the total dataset (AUC = 0.91; P