A Texture-Based Multi Scale Analysis on Rat’s Arteries Relating to Location and Aging

The bridge between macro and micro scale has been arousing people’s attention for years.As for the vessel wall,the link between material property and microstructural network remains unknown,leaving potential possibility to discover the intrinsic mechanism of biological compound material.The objective of the study is to perform a novel analysis method to investigate how microstructure unit contributes to its mechanical characteristics and what kind of factors relating to macro properties of vessel wall may affect its micro characteristics.In this study,we chose to employ a texture analysis to describe and measure spatial network-like structure and collagen fiber alignment patterns in abdominal aorta,femoral artery and carotid artery of rats,respectively.Several first order texture statistics and second order texture statistics have been selected to be embedded into a feature matrix to characterize significance structural distinction(P<0.01)of the aforementioned types of arteries.Also,aging would also be considered as a chronic factor to affect microstructural network.The featuring matrix was then used for training a SVM classifier to predict the artery’s types,age and mechanical properties based on mechanical tests data.(Accuracy=0.86)This analysis methodreveals the link between micro and macro scale of arterial mechanics and more findings will be uncovered based on the framework in the future.