Non-Invasive Von Mises Strain Imaging for Longitudinal Artery Wall

Identification of vulnerable plaques plays an important role in the diagnosis and prognosis of atherosclerosis. In clinic ultrasound imaging, the vessel wall is often imaged in the longitude plane for better resolution. However, there is still lack of efficient parameters. The aim of this study is to construct a reliable elasticity map of carotid arterial wall from longitudinal ultrasonic images. The approximation of Von Mises equivalent strain in the longitudinal plane is developed as an efficient and robust strain index. It synthesizes all three basic train components available in longitudinal imaging plane to be a single strain index, and shows its ability to provide accurate elasticity map of the plaque. A 3D pathological plaque model is constructed and used for finite element analysis. The experiment result shows that Von Mises strain is robust under different SNR level, and Von mises can depict the mechanical property of the soft plaque much better than the basic strain components even with the presence of heavy noise, from both visualization and contrast point of view. In addition, Von Mises strain imaging on carotid artery is also provided, which suggests that the proposed Von Mises strain can be a promising metric for imaging mechanical properties of vessel wall.