Organ Of The Ultrasound Image Boundary Extraction And Auxiliary Explained

Ultrasound image diagnosis is an important medical image diagnosis means like x-ray CT, isotope scanning and MRI. It is an important precondition and a base of ultrasound image diagnosis to effectively extract organic boundary from an ultrasound image. The diseases can be diagnosed by analyzing changes of both the boundary and inside contents of the organic boundary. However, compared with other medical images (e.g. CT and MRI), ultrasound images are particularly difficult to extract an organic boundary as the quality of the images is relatively low. So it is necessary to study the approaches of boundary extraction and set up computer assistant interpretative systems of ultrasound images based on the organic boundary extraction.In this paper, both boundary extraction and image interpretation aspects are mainly studied, and an experimental prototype system is set up for ultrasound image assistant interpretation. In the research, the organs are separated into two kinds, which are cavum organ kind and substantial organ kind. The class models of organs in the ultrasound images are subsequently constructed. There are two different methods for the boundary extraction. The cavum organic boundary is extracted by minimum error thresholding based on Rayleigh-distribution. The substantial organic boundary is extracted by a modified energy-based active contour model, which is developed in this thesis. In image interpretation, the rules of describing different organs in ultrasound images are collected and formed as relative models. Shape parameters and inner features of an organ are extracted based on the boundary extraction, which are invariant to different imaging conditions, and formed to object description table according to the organic class. Based on decision-making methods of a production system, the assistant interpretation of an organic state in the ultrasound images is obtained. The proposed method is applied to real-world ultrasound images, and expectant experimental results are obtained.