A method and device for determining the centerline of an anatomical structure in medical imaging

Abstract

The embodiment of the present invention discloses a method for determining the centerline of the anatomical structure of medical images. A classifier with multi-level resolution is used to estimate the center point of the target. First, a classifier with low resolution is used for estimation. The center point of the target is determined by the shift amount, and then the high-resolution classifier is used for estimation. The center point of the ROI image in this estimation is the target center point obtained after the previous estimation, so that the target center close to the center line is gradually obtained. points through which the centerline is precisely determined. In this method, classifiers with different resolutions are used to gradually determine the center point of the target, taking into account the speed and accuracy of the determination of the center line, which can be applied to the determination of the center line of anatomical structures with different structural characteristics, and does not interfere with image inhomogeneity and noise. The factor is insensitive, and the accuracy and stability of centerline determination are high.

Description

technical field [0001] The invention relates to the technical field of image processing of medical images, in particular to a method and device for determining a centerline of an anatomical structure of a medical image. Background technique [0002] Medical image scanning is a commonly used auxiliary diagnostic method in medicine. In the process of using medical images for diagnosis, it is usually necessary to mark the centerline position of certain specific anatomical structures on the image for further image scanning, analysis or Diagnosis provides positional reference and basis. [0003] At present, the methods for determining the position of the centerline mainly include the thinning method and the key point method. The thinning method is to define a template of a certain size through the thinning algorithm, and make the template move along the scanning direction during the calculation process and delete it according to preset rules. Or keep the pixels of the target ima...

AI Technical Summary

Problems solved by technology

The key point method is to select multiple key points on the target image, identify the key points on the target image, or segment the anatomical structure in the target image, then identify the key points on the segmented anatomical structure, and finally use these key points The connection line of the target image is used as the center line. This method has high requirements for the influence of the target image and anatomical structure. When the target image quality is low, the key points are not clear, or the outline of the anatomical structure is not clear, it is difficult to guarantee the recognition accuracy of the center line.

Images