The following is a summary of “Noninvasive and reliable automated detection of spreading depolarization in severe traumatic brain injury using scalp EEG,” published in the August 2023 issue of Neurology by Chamanzar et al.
Spreading depolarizations (SDs), a biomarker of worsening brain injury after traumatic brain injury (TBI), could be treated if noninvasively detected.
Researchers performed a retrospective study to demonstrate the viability of using noninvasive scalp electroencephalography (EEG) to automate SD detection in patients with severe TBI.
They built upon the recent WAVEFRONT algorithm to develop an automated SD detection approach. This method tracks power depressions in EEG using adjustable parameters and improved velocity estimation. The evaluation dataset encompasses 700 total spreading depolarizations in 12 patients with severe TBI who underwent decompressive hemicraniectomy (DHC). These instances were annotated using authentic intracranial EEG recordings. Simultaneously recorded, continuous, low-density (19 electrodes) scalp EEG signals are employed to assess the accuracy of WAVEFRONT in terms of valid positive rate (TPR), false positive rate (FPR), and the precision of estimating SD frequency.
The results demonstrated that WAVEFRONT attains the highest average validation accuracy by employing Delta band EEG, achieving a 74% TPR and an FPR under 1.5%. There are initial indications that WAVEFRONT can estimate the occurrence frequency of SDs.
They concluded the WAVEFRONT algorithm enables noninvasive SD detection after severe TBI using an automated algorithm for diagnosis, monitoring, and treatment tailored to brain injury.
Source: nature.com/articles/s43856-023-00344-3#Tab1
