Mapping brain injury with symmetrical-channels' EEG signal analysis--a pilot study.

A technique for detecting brain injury at the bedside has great clinical value, but conventional imaging techniques (such as computed tomography [CT] and magnetic resonance imaging) are impractical. In this study, a novel method-the symmetrical channel electroencephalogram (EEG) signal analysis-was developed for this purpose. The study population consisted of 45 traumatic brain injury patients and 10 healthy controls. EEG signals in resting and stimulus states were acquired, and approximate entropy (ApEn) and slow-wave coefficient were extracted to calculate the ratio values of ApEn and SWC for injured and uninjured areas. Statistical analyses showed that the ratio values for both ApEn and SWC between injured and uninjured brain areas differed significantly (P<0.05) for both resting and name call stimulus states. A set of criteria (range of ratio values) to determine whether a brain area is injured or uninjured was proposed and its reliability was verified by statistical analyses and CT images.