Method and system for reconstructing internal region-of-interest image

Abstract

The invention provides a method and a system for reconstructing an internal region-of-interest image. The method includes: (a), scanning an internal region-of-interest of an object by CT (computed tomography) equipment; (b), acquiring a projection image formed by projection of all rays of the internal region-of-interest, and acquiring a noise weight matrix of the projection image; (c), constructing a reconstruction model related to the internal region-of-interest image according to the projection image, the noise weight matrix and a system matrix expressing CT equipment system parameters; (d), performing optimization solution on the reconstruction model to acquire the reconstructed internal region-of-interest image. According to the method for reconstructing the internal region-of-interest image, existence of a known region in the internal ROI (region-of-interest) is not needed, robustness of data with noise is high, and application range is wider.

Description

technical field [0001] The present invention relates generally to the field of imaging technology. More specifically, it relates to a method and system for reconstructing an image of an internal region of interest. Background technique [0002] In practical engineering applications, due to technology and cost constraints, the size of the detector is limited, and the size of the object to be detected is often too large, so that the scanning field of view (FOV) cannot completely cover the object, and the area to be imaged is generally of interest to us. Both are located inside the object. For example, in dental CT imaging, dentists are often interested in teeth rather than the entire head, so there may be a Region-Of-Interest (ROI) imaging problem. ROI imaging is considered to be an effective method to reduce CT radiation dose. Different from general ROI imaging, the problem of internal ROI imaging is more complicated. FOV only covers ROI, and ROI is completely inside the ob...

AI Technical Summary

Problems solved by technology

The most significant shortcoming of the ART algorithm used in its reconstruction is that it does not statistically model the data noise

If the TV minimization is performed on an image containing noise, it cannot satisfy the condition that the ROI required by this method is a piecewise constant or a piecewise polynomial, and thus cannot fully realize the accurate reconstruction of the ROI.

[0006] Therefore, the existing methods for reconstructing images of internal ROIs have certain limitations. For example, a known region is required in the internal ROI, and clinical applications are limited; or ROIs are required to be piecewise constants or piecewise polynomials. Sensitive to noisy data

Images