Quantitative analysis for lumen and media-adventitia border detection in intravascular ultrasound images

摘要

Therosclerosis causes partial or total obstruction of human arteries. Early quantitative analysis and accurate assessment of plaque position and volume are essential for the selection of the appropriate treatment. Several imaging techniques can be used for the estimation of the severity of the disease in vivo. Intravascular ultrasound (IVUS) is a commonly used diagnostic tool, which provides real-time visualization of plaque morphology and detection of typical plaque components, such as calcium, quantification of plaque eccentricity and wall thickness. In this paper, we firstly used a spatio-temporal filter to reduce the effect of speckles and enhance the image. Then we translated the problem of image segmentation to the problem of the minimum resolution of energy function, using an improved deformable models, we detected the border of lumen and media-adventitia in sequential intravascular ultrasound frames, and optimized it by dynamic programming. Finally, through the identification of the internal and external elastic lamina and the plaque-lumen interface, we figured out the parameter of plaque load, maximal and minimal diameters of the internal and external elastic lamina and so on. The obtained results demonstrate that our method is statistically accurate, reproducible, and capable to identify the regions of interest in sequences of IVUS frames.