Method and apparatus for segmenting structure in CT angiography

Abstract

The present invention provides a technique for automatically generating bone shielding (82) in CTA angiography. This technique can preprocess the image data set (70) to perform various functions, such as removing (90) the image data associated with the table top, cutting (92) the volume into regionally consistent sub-volumes, calculating according to the gradient ( 110) Structure edges, and / or computing (112) seed points for subsequent region growing. The preprocessed data can then be automatically segmented for bone and vascular structures (78). This automatic vessel segmentation (118) can be accomplished using suppression region growing, where the suppression is dynamically updated according to the local statistics of the image data. The vasculature (78) may be subtracted (80) from the bone structure (76) to generate a bone shield (82). The bone mask (82) may in turn be subtracted (84) from the image dataset (70) to generate a boneless CTA volume (86) for volume-rendered reconstruction.

Description

Technical field [0001] The present invention generally relates to the field of medical imaging, and particularly relates to the field of volumetric medical imaging. In particular, the present invention relates to a technique for segmenting bone and vasculature data in computed tomography. Background technique [0002] Volumetric medical imaging technology uses a variety of technologies to gather three-dimensional information of the human body. For example, computed tomography (CT) imaging systems measure the attenuation of X-ray beams passing through a patient from multiple angles. Based on these measurements, the computer can reconstruct the image of the patient's body that caused the radiation attenuation. As those skilled in the art will understand, these images are individually detected based on a series of angularly displaced cross sections. It should be pointed out that the CT system generates data representing the distribution of the linear attenuation coefficient of the s...

AI Technical Summary

Problems solved by technology

As a result, the operator may have to repeatedly re-draw or re-contour the entire procedure, possibly taking more than 1 hour to process the imaged head and neck volume

Furthermore, operator intervention may lead to variability in the segmentation of structures between users and within the same user

[0008] Since this procedure is time consuming, the operator may attempt to reduce processing time by restricting processing to regions or volumes of interest corresponding to the location of the pathology

However, limiting the processing in this way does not take full advantage of all available volume data sets

Furthermore, restricting processing to regions of known or suspected pathology may have hampered the discovery of otherwise obscure pathologies, regardless of data availability

Images