Single Trial Extraction of Somatosensory Evoked Potentials for Monitoring Spinal Cord Injury: An Animal Study

摘要

Blind Source Separation is a promising method for extracting somatosensory-evoked potentials. Constrained second order blind identification has been reported as the most appropriate Blind Source Separation method for fast somatosensory-evoked potential extraction with few channels. In this study, 15 rats were evaluated to determine the effectiveness of a proposed constrained second order blind identification method. This study performed single trial extraction of somatosensory-evoked potentials during contusion injury to the spinal cord. Single trial somatosensory-evoked potential was successfully extracted in 84% of trials before injury, but in only 40-60% in the early stage of spinal cord injury. Amplitude and latency could be detected by single trial somatosensory-evoked potential and defined as an abnormality in terms of a 50% decrease in amplitude or a 10% delay in latency. Somatosensory-evoked potential measured by constrained second order blind identification displayed levels that met the warning criterion for spinal cord deficit immediately after the injury, whereas that from ensemble averaging appeared at approximately 22.5 minutes after injury. Compared with the traditional ensemble averaging method, results from constrained second order blind identification showed that single trial somatosensory-evoked potential can predict spinal cord injury in the early stage.