In vitro experiments on human blood samples are crucial to assess correlation between thrombus disorders and genetic variance. A blood flow is drained through collagen fibers resulting in a progressive aggregation of platelets. This paper proposes a method of segmentation of thrombi during such acquisitions. Three complementary gradient-based features are introduced and included in a regularized machine learning framework. A novel tracking method of thrombi as deformable growing objects under split and merge conditions is also introduced.     «

In vitro experiments on human blood samples are crucial to assess correlation between thrombus disorders and genetic variance. A blood flow is drained through collagen fibers resulting in a progressive aggregation of platelets. This paper proposes a method of segmentation of thrombi during such acquisitions. Three complementary gradient-based features are introduced and included in a regularized machine learning framework. A novel tracking method of thrombi as deformable growing objects under sp...     »