Gradient-Based Algorithms for Characterizing the Structure of Fibrin Clots

The human body’s ability to close wounds through clotting is vital to everyday function—but irregular clotting can cause diseases like deep vein thrombosis, Von Willebrand disease, or hemophilia, which lead to hundreds of thousands of deaths each year. Understanding how various clotting mechanisms affect the mechanical and structural properties of a blood clot’s fibrin fiber network is integral in working to prevent and treat these clot-related diseases. Two structural characteristics of the network, the average fiber diameter and branch point density, lend themselves to discovery by applying various computational image analysis techniques to images acquired using scanning electron microscopy or fluorescence microscopy. Algorithms using gradient-based thresholding were implemented in Python to minimize data loss from classic image analysis techniques and to quantify the network’s structural characteristics.