Extraction of breast cancer related biomarkers in T1 weighted MR images of a rodent model

Abstract

In studying the relationship between risk factors and breast cancer, growth patterns of the fat pads and glandular tissues are important features. In other words, these features can be regarded as breast cancer related biomarkers. Accordingly, the goal of this rodent model study is to measure the size of mammary pads over the time and to quantify the development of glandular structures. To achieve this goal, we propose a hierarchical approach to segmenting out rat body, mammary fat pads and glandular tissues in Tl weighted magnetic resonance (T1W-MR) images. Particularly, we have developed an approach combining watershed transform and region competition for improved fat pad segmentation, and a method for glandular tissue segmentation through scale-space analysis. For segmenting fat pads, an efficient strategy, termed as competition propagation, is developed to propagate the region competition result from one slice to next slice, resulting in a fast convergence in region competition algorithm otherwise computationally costly. The glandular tissues, as the fine structures within the fat pads, are then located using scale-space analysis and segmented out by analyzing their local iso-shape patterns. These methods have been applied to 18 volumetric sets of T1W-MR images acquired from this study. The experimental results showed the great utility of the proposed approaches as they can provide accurate measurements to assess novel risk factors for breast cancer.