Active Contour Model And Its Intravascular Ultrasound Image Edge Extraction Study

Coronary artery diseases are characterized by accumulation of plaque on the vessel wall. Atherosclerosis causes partial or total obstruction of human arteries. Early diagnosis and accurate assessment of plaque position and volume are essential for the selection of the appropriate treatment. Intravascular ultrasound (IVUS) is a commonly used diagnostic method. Edge detection and further processing for IVUS images can not only measure accurately the area of vessel cavity but also provides real-time visualization of plaque morphology, detection of typical plaque components, such as calcium, and quantification of plaque eccentricity and wall thickness. So edge detection for IVUS images plays an important role in diagnosing coronary artery disease clinically.Inner and outer boundaries of the vessel wall belong to weak edges of elastomer in serious noise. Classical edge detection operators are sensitive to noise, and besides, they more suit to process rigid images. Since snake model may be used to extract a variety of contours, it can overcome the above shortcomings. The active contour model, or snake, is defined as an energy-minimizing spline. Edge is detected when the energy function is minimum. To overcome the shortage that conventional model is sensitive to noise, contrast instead of gradient is used as external energy. Contrast integrates gray level gradient information and background intensity information. In the process of energy optimization , The energy function is appropriately modified and minimized using a Hopfield neural network. A simulated annealing scheme is integrated to ensure convergence at a global minimum, so accurate edges can be detected. Experiments show that, compared with the former methods of edge detection, our method is algorithmically simple, statistically accurate, reproducible and robust in sequential IVUS frames, and is a kind of global optimal algorithm.Ultrasound imagery technology cause serious Speckle noise in IVUS images. Speckle noise decreases spatial resolution and contrast resolution of images, influencing image postprocessing such as edge detection. So effectively suppressing Speckle noise is a key to detect edge accurately. Some methods, such as adaptive median filter, Weiner filter, wavelet filter, anisotropic diffusion and so on, can not reduce noise effectively or...