LAUSR

Key Points Question Can machine learning algorithms be used to triage patients with head trauma according to their severity before transportation? Findings In this cohort study of 2123 patients with head trauma, a machine learning–based prediction model detected traumatic intracranial hemorrhage with a sensitivity of 74% and a specificity of 75% by using only the patient’s prehospital information. Meaning This study suggests that a machine learning algorithm can accurately stratify patients with head trauma according to severity in prehospital settings and may improve the prognosis of patients with severe traumatic head injury.