Classification algorithms for the identification of structural injury in TBI using brain electrical activity

•The derivation of algorithms for the identification of structural TBI are described.•EEG recordings were obtained using only forehead electrodes on a handheld device.•Classifiers methodologies used included genetic algorithms and LASSO logistic regression.•The classifiers achieved an average sensitivity/specificity of 97.5%/59.5%.•Performance well surpasses that of standard clinical practice for CT scan referrals in TBI.