MORPHOLOGICAL CHARACTERIZATION OF COLORECTAL PITS USING AUTOFLUORESCENCE MICROSCOPY IMAGES

Area of Science:

• Biomedical imaging
• Cancer research
• Microscopy

Background:

• Colorectal cancer diagnosis relies on histological analysis of biopsies.
• Kudo's pit pattern classification is standard for differentiating neoplastic lesions.
• Autofluorescence imaging offers a staining-free alternative for pit characterization.

Purpose of the Study:

• To process and analyze autofluorescence microscopy images of colon tissue.
• To quantify and characterize pit morphology in a murine colorectal cancer model.
• To measure morphology parameters and shape descriptors for pit characterization.

Main Methods:

• Utilized widefield autofluorescence microscopy on fresh colon tissue samples.
• Employed two-dimensional (2D) segmentation and quantification via FIJI macro programming.
• Analyzed samples from azoxymethane-treated and control mice over 20 weeks.

Main Results:

• Type I pit morphology predominated in control groups (53-81%).
• Treated groups showed increased Type I and III-L pit morphologies.
• Specific pit pattern types (IV, III-S, II) were identified in treated samples.

Conclusions:

• Autofluorescence image processing provides rapid 2D pit characterization.
• This method can quantify non-pathological pits and estimate early colorectal cancer stages.
• Offers an advancement over conventional staining techniques for colorectal lesion analysis.