Automated Segmentation and Diagnostic Measurement for the Evaluation of Cervical Spine Injuries Using X‑Rays

The study focuses on using machine learning to automatically analyze cervical spine X-ray images. This is done by identifying and outlining (segmenting) different parts of the spine and skull on the X-rays. The automated system then measures key distances and angles on the segmented images, which are important for diagnosing injuries. These measurements are compared to manual measurements to see how accurate the automated system is. The goal is to create a tool that can help doctors quickly and accurately assess cervical spine injuries using widely available X-ray technology.

• 1
  The study achieved high average dice coefficient values for cervical spine segmentation (C1-C7), indicating accurate automated identification of these regions on X-rays.
• 2
  Comparison of automatically measured metrics and manually measured metrics showed high Pearson’s correlation coefficients in McGregor’s line, space available cord, cervical sagittal vertical axis, and cervical lordosis.
• 3
  No metric showed adjusted significant differences at P < 0.05 between manual and automatic metric measuring methods.