Rationalizing polyp matching criteria in colon capsule endoscopy: an international expert consensus through RAND...

Area of Science:

• Gastroenterology
• Medical Imaging
• Artificial Intelligence in Medicine

Background:

• Colon capsule endoscopy (CCE) is an emerging lower gastrointestinal tract investigation tool.
• Accurate polyp matching in CCE videos is crucial to prevent double reporting and false positives, especially with AI integration.
• Lack of standardized spatial orientation information complicates polyp comparison in CCE.

Purpose of the Study:

• To identify key factors for matching polyps within colon capsule endoscopy videos.
• To develop comprehensive polyp-matching criteria using an international expert consensus.
• To address challenges in CCE polyp comparison and reporting accuracy.

Main Methods:

• A modified Delphi process involving a panel of 11 international CCE experts.
• Systematic qualitative study utilizing surveys with 60 statements rated on a 1-9 scale.
• Two rounds of surveys and virtual group discussions to achieve expert consensus.

Main Results:

• Eight essential factors for polyp matching were identified: timestamp, localization, vascular pattern, size, time interval between camera colors, surrounding tissue, morphology, and surface/contour.
• Agreement was reached that matching five or more factors indicates the same polyp.
• Established the first complete criteria for polyp matching in CCE.

Conclusions:

• The developed criteria provide a framework to guide and facilitate polyp matching in CCE.
• These criteria aim to improve accuracy and reduce double reporting in CCE procedures.
• While not a definitive solution for all polyp types, the criteria enhance CCE interpretation.