Segmentation of elongated structures processed on breast MRI for the detection of vessels

Abstract The multi-seed region growing (MSRG) algorithm from previous work is extended to extract elongated segments breast Magnetic Resonance Imaging (MRI) stacks. A model created adjust the MSRG parameters such that may reveal vessels can support clinicians in their diagnosis of diseases or provide them with useful information before surgery during e. g. a neoadjuvant therapy. pipeline tasks and contains user-defined influence segmentation result. crucial task relies on skeletonization-like collects about segments’ thickness, length, etc. Length, gradient pixel intensity along segment helps determine whether extracted have tubular structure, which assumed be case for vessels. In this work, we show how results are derived capable extracting vessel-like even noisy MR images.