Digital morphometric characterization of mucosal surface lesion patterns under magnification colonoscopy

Area of Science:

• Gastroenterology
• Endoscopy
• Digital Pathology

Background:

• Colonoscopic examination is crucial for detecting colorectal lesions.
• Histological confirmation remains the gold standard for lesion diagnosis.
• Objective, quantitative analysis of mucosal patterns could enhance diagnostic accuracy.

Purpose of the Study:

• To analyze quantified mucosal surface patterns in digitized colonoscopic images.
• To differentiate hyperplastic lesions, adenomas, and adenocarcinomas using morphometric analysis.
• To assess the feasibility of real-time morphometric evaluation during colonoscopy.

Main Methods:

• Morphometric measures were performed on colonoscopic images of 37 hyperplastic lesions, 42 adenomas, and 26 adenocarcinomas.
• Evaluated pit counts per area, pit diameter, and pit area compared to normal mucosa.
• Images were obtained with 17-40x magnification and chromoscopy.

Main Results:

• Statistically significant differences (p < 0.001) were observed in pit counts per area and pit area between lesions and normal mucosa.
• Morphometric characteristics effectively differentiated the three lesion types.
• These findings were consistent across hyperplastic lesions, adenomas, and adenocarcinomas.

Conclusions:

• Real-time morphometric evaluation during videocolonoscopy is feasible.
• Automated analysis of morphometric characteristics can support colonoscopists' diagnostic decisions.
• This quantitative approach offers a valuable adjunct to traditional colonoscopy.