Identifying Intraoperative Spinal Cord Injury Location from Somatosensory Evoked Potentials’ Time-Frequency Components

This study explores using somatosensory evoked potentials (SEPs) to pinpoint the location of spinal cord injuries (SCI) during corrective scoliosis surgery. Identifying the injury location quickly can help surgeons take immediate action to minimize damage. The researchers used a rat model to simulate distraction spinal cord injuries at various levels. They then analyzed the time-frequency components of the SEPs to see if they could accurately classify the injury location. The findings suggest that analyzing the time-frequency components of SEPs, combined with machine learning techniques, holds promise for diagnosing the location of spinal cord injuries during surgery.

• 1
  There was a significant delay in the latency of the time-frequency components distributed between 15 and 30 ms and 50 and 150 Hz in all spinal cord injury groups.
• 2
  The overall classification accuracy using k-medoid clustering and naive Bayes methods was 88.28% for cervical SCI and 84.87% for cervical, thoracic, and lumbar SCI.
• 3
  The k-medoid clustering and naive Bayes methods are capable of extracting the time-frequency component information depending on the spinal cord injury location.