Date of Award
Level of Access Assigned by Author
Master of Science (MS)
Second Committee Member
Third Committee Member
Breast cancer is the second most occurring cancer type and is ranked fifth in terms of mortality. X-ray mammography is the most common methodology of breast imaging and can show radiographic signs of cancer, such as masses and calcifcations. From these mammograms, radiologists can also assess breast density, which is a known cancer risk factor. However, since not all dense tissue is cancer-prone, we hypothesize that dense tissue can be segregated into healthy vs. risky subtypes. We propose that risky dense tissue is associated with tissue microenvironment disorganization, which can be quantified via a computational characterization of the whole breast to provide an image-based risk assessment. The two-dimensional wavelet transform modulus maxima (2D WTMM) method is a strategy previously utilized on mammographic images to characterize the loss of tissue homeostasis and tissue disorganization. A sliding window protocol is used within the 2D WTMM method to analyze thousands of overlapping subregions of size 256 × 256 pixels from the original mammogram. This approach starts in the top left corner and ends in the bottom right corner in a step size of 32-pixel increments. The subregions of mammographic breast tissue are categorized according to Hurst exponent (H) values and colors based upon these values: fatty (H ≤ 0.45, blue), healthy dense (H ≥ 0.55, red), and risky dense tissue (0.45 < H < 0.55, yellow) [24, 25]. To decrease computational time and cost, an investigation into the efficiency of the sliding window approach was conducted by considering different pixel step size increments. Increments of 32 pixels, 64 pixels, 128 pixels, and 256 pixels were compared using the percent composition of each tissue type and a statistical Wilcoxon Rank Sum test. Optimized iterations of color representations can be created and compared to accompany the statistical analysis of tissue composition. The creation and comparison of multi-layer intensity, single-layer maxima intensity, and single-layer raw intensity heatmaps provide the conclusion that the multi-layer intensity heatmaps show the most accurate visual representation of the proposed tissue types. Through this investigation, we conclude that setting the increment of the sliding window protocol to 128 pixels provides the best comparison of mammograms using multi-layer heatmaps as a visual tool. The optimization of these images will allow the multi-layer intensity heatmaps created at an increment of 128 pixels to aid medical professionals in their identification of patients at a higher risk of developing invasive cancer.
McCarthy, Margaret R., "A Quantitative Visualization Tool for the Assessment of Mammographic Risky Dense Tissue Types" (2023). Electronic Theses and Dissertations. 3862.
Files over 10MB may be slow to open. For best results, right-click and select "save as..."