A Graph-theoretic Algorithm for Small Bowel Path Tracking in CT Scans.

Area of Science:

• Medical Imaging
• Computational Anatomy
• Graph Theory

Background:

• Accurate small bowel path tracking is crucial for medical diagnosis and surgical planning.
• Existing methods often struggle with indistinct bowel walls and may produce inaccurate paths with shortcuts.
• A need exists for robust methods that can track the entire small bowel course without extensive training data.

Purpose of the Study:

• To develop and evaluate a novel graph-theoretic method for accurate small bowel path tracking.
• To improve upon existing methods by addressing challenges like indistinct bowel walls and the need for ground-truth training data.
• To enable comprehensive evaluation of small bowel path tracking across the entire organ.

Main Methods:

• Formulated small bowel path tracking as a minimum cost path problem on a graph derived from bowel wall detection.
• Incorporated 'must-pass' nodes to ensure complete coverage of the small bowel's course.
• Evaluated the method on 10 abdominal CT scans with acquired ground-truth paths, comparing against a baseline method.

Main Results:

• The proposed graph-theoretic method demonstrated significant improvements over the baseline approach.
• The method successfully avoids short-cuts through indistinct bowel walls by utilizing must-pass nodes.
• On average, the method tracked paths with errors over 800 mm in length per scan, indicating high accuracy for the entire small bowel course.

Conclusions:

• The novel graph-theoretic method offers a robust and accurate solution for small bowel path tracking.
• The approach eliminates the need for ground-truth path training, making it more practical for clinical applications.
• This method enhances the ability to visualize and analyze the complete small bowel anatomy from CT scans.