Polyps: linear and volumetric measurement at CT colonography

Area of Science:

• Medical Imaging
• Gastroenterology
• Radiology

Background:

• Accurate polyp size measurement is crucial for colorectal cancer screening and surveillance.
• Computed tomographic (CT) colonography offers a less invasive alternative to optical colonoscopy.
• Standardizing polyp size measurements in CT colonography is essential for reliable assessment of change over time.

Purpose of the Study:

• To identify the most accurate CT colonography-based polyp measurement for comparison with optical colonoscopy.
• To determine which CT colonography measurement method is most reliable for assessing changes in polyp size during serial examinations.

Main Methods:

• Retrospective analysis of 216 patients undergoing prone and supine CT colonography with same-day optical colonoscopy.
• Polyp size was measured using three linear and two volume methods, with both manual and automated segmentation.
• Measurements were assessed for compatibility with optical colonoscopy size and reproducibility (inter-scan, inter-observer, intra-observer variability).

Main Results:

• Manual 3D linear measurements (L(M3D)) showed high confidence in identifying polyps <1.0 cm at optical colonoscopy.
• The smallest inter-scan (prone vs. supine) size difference was observed for L(M3D) and linear diameter from automated volume measurements.
• Linear diameter measurements on a 2D display exhibited the lowest inter-observer and intra-observer variability.

Conclusions:

• Manual 3D linear measurement (L(M3D)) best approximates polyp size as measured by optical colonoscopy.
• Linear diameter derived from automated volume measurements demonstrates minimal variation between scans and avoids observer bias, making it optimal for monitoring polyp size changes.