EVALUATION OF DIAGNOSTIC ALGORITHM BASED ON COLLAGEN ORGANIZATION PARAMETERS FOR BREAST TUMORS

Area of Science:

• Oncology
• Biomedical Engineering
• Materials Science

Background:

• Collagen structure alterations are crucial in malignant neoplasms, including breast cancer (BCa).
• Quantitative and spatial collagen parameters serve as vital diagnostic and prognostic factors.
• Understanding these changes is essential for developing advanced cancer detection technologies.

Purpose of the Study:

• To develop and validate an algorithm for assessing collagen organization parameters in breast tumor tissues.
• To identify informative attributes of collagen associated with breast cancer for machine learning applications.
• To contribute to the development of an intelligent system for cancer diagnostics.

Main Methods:

• Histochemical identification of collagen using the Mallory method.
• Digital microscopy for capturing high-resolution images of tissue samples.
• Morphometric analysis using specialized software (CurveAlign v. 4.0. beta and ImageJ) to quantify collagen characteristics.

Main Results:

• A novel algorithm was developed and tested for analyzing quantitative and spatial collagen matrix characteristics.
• Collagen fibers in BCa tissue exhibited significantly reduced length and width (p < 0.001).
• BCa collagen fibers showed increased straightness (p < 0.001) and angle (p < 0.05) compared to fibroadenoma tissue, with no significant difference in density.

Conclusions:

• The developed algorithm effectively assesses diverse collagen fiber parameters, including spatial orientation, arrangement, and density.
• This quantitative assessment of the collagen matrix provides valuable data for machine learning models.
• The findings support the potential of collagen analysis as a tool for intelligent breast cancer diagnostics.