Motion correction system for dual-energy subtraction chest X-ray image

Abstract

The invention discloses a correction system aiming at motion artifact of a dual-energy subtraction chest X-ray image, and belongs to the technical field of digital image analysis. Through a dual-energy subtraction chest X-ray imaging technology, the detection rate of small intrapulmonary nodules can be effectively increased, the detection of pathological changes existing in tracheas, blood vessels, bones and pleura is facilitated, and the reliability of diagnostic differentiation is increased. But due to continuous double exposure, positional difference existing between a high-energy image and a low-energy image is caused by motions such as respiration, heartbeat and the like of patients, the motion artifact is produced, and the image quality is seriously influenced. By using the system, the characteristics of chest motion are analyzed, motions of the bones and soft tissues are effectively differentiated, then a local elastic registration module based on a bone edge and B spline transformation is adopted and is combined with a rigid registration module, and a multi-scale registration strategy is applied to eliminate artifact existing in the dual-energy image. The system is used for registration of a digital image, particularly registration and motion correction of a medical image, and contributes to improving the image quality.

Description

technical field [0001] The invention belongs to the technical field of digital image analysis and processing, in particular to a motion correction system for dual-energy subtraction chest X-ray images. Background technique [0002] Chest X-ray imaging (Computer digital radiography, CDR) is the basic imaging method for early detection of pulmonary nodules. Statistics show that the false negative rate of conventional chest X-ray films for single pulmonary nodules is as high as 18% to 32%. Compared with conventional chest X-ray images, dual energy subtraction (DES) chest X-ray imaging has unique advantages because it removes the interference of the bony thorax, and can effectively improve the sensitivity of detecting calcification and accuracy, and the detection of calcification is one of the reliable imaging signs for the diagnosis of pulmonary nodules. Nodules with calcification appear as shadows on bone images of dual-energy subtraction, but completely or partially disappea...

AI Technical Summary

Problems solved by technology

Motion artifacts are mainly due to heart beat and respiratory movement. There are a large number of black or white stripes in the lungs and heart, which cannot accurately and clearly display the anatomical structure and local details of the lungs and heart. Accurate positioning, qualitative and quantitative analysis, etc. will have a greater impact

[0007] At present, the registration between digital images based on mutual information measurement is widely used, and it is suitable for registration between images of the same modality or different modalities. Its advantage is that it is not sensitive to the range of image density and has strong anti-noise performance, but they do not take advantage of global spatial information and are prone to fall into local minima (J.Pluim, J.Maintz, and M.Viergever, "Mutual-Information-Based Registration of Medical Images: A Survey," IEEE Trans.Medical Imaging, 22(8), 986-1004(2003); A. El-Baz, G Gimel'farb, "A novel approach for global registration of medical images based on learning the prior appearance model," IEEE 5th International Symposium on Biomedical Imaging: From Nano to Macro, 784-787(2008)

To this end, multi-resolution analysis and multi-scale registration strategies are applied to the registration process (P.Thévenaz, M.Unser, "Optimization of mutual information for multiresolution image registration", IEEE Transactions Image Processing, 9(12), 2083 -2099(2000); L.H.Peng, Y.H.Li, D.W.Xu, "A New Multiresolution Medical Image Registration Algorithm Based on Intensity and Edge Information," IEEE 4th International Conference on Natural Computation, ICNC′08, 5, 13-17(2008 )), although the spatial information is considered, the global optimal value of the objective function can be found, and the accuracy and speed of registration can be improved to a certain extent, but because they do not consider the chest motion characteristics, they still cannot accurately and effectively correct the dual-energy Positional differences generated during subtraction imaging

Images