A fully automated method for spinal canal detection in computed tomography images

Area of Science:

• Medical Imaging
• Radiology
• Computer-Aided Diagnosis

Background:

• Accurate spinal canal detection is crucial for diagnosing spinal conditions.
• Manual segmentation of the spinal canal in Computed Tomography (CT) images is time-consuming and subjective.
• Automated methods are needed to improve efficiency and objectivity in spinal canal analysis.

Purpose of the Study:

• To develop and evaluate a novel automated method for spinal canal detection in CT images.
• To assess the method's ability to reduce radiologist workload and provide objective measurements.
• To leverage both 2D and 3D image information for robust spinal canal extraction.

Main Methods:

• The automated method involves thresholding and morphological operations for initial segmentation.
• 3D connectivity analysis is employed to identify and extract spinal canal components.
• Centroid computation, interpolation, and extrapolation are used to define the spinal canal's geometry.

Main Results:

• The method was applied to 25 patients (8704 images) from two different acquisition systems.
• Qualitative evaluation by an experienced radiologist confirmed that all extracted points accurately located within the spinal canal.
• The automated approach demonstrated potential for reducing workload and enabling objective spinal canal detection.

Conclusions:

• The developed automated method shows promising results for spinal canal detection in CT images.
• The technique offers an objective and efficient alternative to manual segmentation.
• Future research will focus on quantitative evaluation to further validate the method's performance.