12 results about "Filtered backprojection" patented technology

The invention discloses a DDPM-based CT image metal artifact elimination method, and aims to solve the artifact problem caused by a metal object in an existing CT image and improve image quality and diagnosis reliability. A diffusion model of unconditional training is adopted, step-by-step back diffusion repair of an artifact area is carried out in a sinogram domain, an unrepaired area is dynamically adjusted by combining with a metal mask, accurate repair of the artifact area is achieved, and original data of the area which is not affected by artifacts are kept. In the training stage of the system, artifact-free data are gradually converted into standard Gaussian noise through forward diffusion; in the inference stage, data are gradually recovered by utilizing back diffusion, block repair is carried out on an artifact region by combining with a metal mask, and a complete sinogram is generated through region merging. And finally, reconstructing a CT image by using a filtered back projection algorithm, and optimizing boundary transition through a smoothing algorithm to ensure seamless connection between the metal object and surrounding tissues.

According to the analysis and reconstruction method and system of the static CT based on the arc-shaped detector and the arc-shaped ray source, the analysis and reconstruction method of the static CT based on the arc-shaped detector and the arc-shaped ray source is carried out through the following five steps, and finally a final CT image is obtained. The method has the beneficial effects that in a filtering back projection algorithm, a variable optimization method of geometric parameter transformation and traditional equiangular fan beam reconstruction is introduced, an accurate arc-shaped geometric scanning model is established, and an analysis reconstruction algorithm of linear multi-source static CT is referred to; meanwhile, the influence of the geometric structure characteristics of the arc ray source-arc detector and the projection data completeness on the reconstruction effect is considered, so that stripe artifacts and motion artifacts can be effectively reduced, and the CT image quality is improved. Meanwhile, the method keeps a frame of filtering back projection, and is low in calculation complexity and rapid in reconstruction.

The application discloses a kind of based on the sparse helical CT reconstruction method, system, equipment, medium and program product of back projection tensor interpolation, belong to computer tomography (CT) imaging technical field, specifically includes the following steps: generating the back projection tensor data under sparse angle, the back projection tensor data indicates the intermediate data of back projection process in filter back projection algorithm;Interpolation is carried out to the back projection tensor data of the sparse angle by depth learning network, generates the back projection tensor data of full angle;Image reconstruction is carried out based on the back projection tensor data of full angle, and the final CT image is obtained.The reconstruction method of the application suppresses the generation of strip artifact from the source and directly repairs the key intermediate data for reconstruction, which can better retain and restore the true anatomical structure information than post-processing in the image domain, avoids excessive smoothing or filtering, and has generalization ability, suitable for different CT scanning geometry.

This invention provides a low-dose CT reconstruction method assisted by pre-log domain Voronoi decomposition. The method includes obtaining a pre-log sinusoidal graph of low-dose CT, performing Voronoi decomposition on the pre-log sinusoidal graph using the K-means clustering algorithm to obtain several feature clusters, mapping each feature cluster to its corresponding latent space to obtain a multi-channel input feature map, inputting the multi-channel input feature map into a pre-trained diffusion transformer model for iterative denoising and optimization to obtain optimized cluster features, fusing and performing logarithmic transformation on the optimized cluster features to obtain a post-log sinusoidal graph, and reconstructing the post-log sinusoidal graph using a filtered back-projection algorithm to obtain a high-quality CT image. This invention solves the problems of large dynamic range, uneven gradient, and progressively amplified noise in the pre-log domain by decoupling features from the pre-log sinusoidal graph, modeling the latent space, and optimizing the diffusion transformer, thus achieving high-fidelity low-dose CT image reconstruction.

The invention discloses an SPECT low-dose dynamic reconstruction method based on a deformable convolution and deep expansion network, and relates to the technical field of data processing, and the method comprises the following steps: carrying out the normalization of original projection data, and carrying out the grouping according to a time window; generating an initial image sequence by using a filtered back projection algorithm; constructing a deep expansion network framework comprising a plurality of iteration layers; calculating a projection residual error in each layer and updating the image to match measured data; performing spatial denoising and feature enhancement on the image by using a convolutional neural network; learning inter-frame motion through deformable convolution and carrying out time alignment on features; repeating the processing steps in the network for a plurality of times; and performing cutting and intensity scaling on the final output image to generate a final result. According to the method, the deformable time convolution module is integrated into each iteration layer of the deep expansion network, so that modeling can be effectively carried out, and non-rigid motion in a dynamic sequence can be compensated, and therefore, the space-time consistency and quality of a reconstructed image sequence can be improved.

The invention discloses a self-adaptive weighted filtering reconstruction method and device for double-circle trajectory imaging of a radiation source and a detector. The method comprises the following steps: acquiring original projection data; constructing a composite weight matrix by obtaining the dynamic offset of the virtual projection point; generating weighted projection data according to the original projection data and the composite weight matrix; performing periphery filling and image rotation operation on the weighted projection data to obtain an intermediate processing image; performing full-angle self-adaptive frequency spectrum continuous compensation filtering on the intermediate processing image to obtain a compensated image; performing image rotation and intelligent center cutting operation on the compensated image to obtain a final projection; and based on an FDK back projection formula, performing weighted back projection reconstruction on the final projection to obtain a three-dimensional cross-sectional image. The method can realize high-efficiency, high-quality and isotropic-resolution image reconstruction, and can be widely applied to the technical field of three-dimensional cross-sectional image reconstruction.

This invention relates to the field of PCB blind via defect detection technology and discloses an X-ray-based method for detecting PCB blind via defects. The method includes the following steps: first, the PCB is irradiated with X-rays; then, data is acquired and an initial image is established using CT projection; subsequently, the initial image is reconstructed using filtered back projection to obtain a reconstructed image; then, the blind via region is extracted using the reconstructed image; next, a two-stream neural network architecture is constructed based on a deep residual network; finally, the blind via defect is detected and classified using the two-stream neural network architecture. This invention, through a two-stream neural network architecture constructed based on a deep residual network, comprehensively utilizes geometric and material features, enabling comprehensive and accurate detection and classification of various defects in PCB blind vias, such as cracks, wrinkles, dents, and voids. This improves the efficiency and accuracy of defect detection, helps to promptly identify quality problems in the PCB manufacturing process, and ensures product quality.

The invention relates to the technical field of computed tomography imaging, in particular to a CT reconstruction method and system based on data rearrangement and deep learning angle extrapolation, and the method comprises the steps: data rearrangement: enabling an actual ray source sampling point to be opposite to a detector pixel, constructing virtual scanning geometry, and recombining a truncation projection set into global projection; deep learning angle extrapolation: inputting the rearranged limited angle global projection into a physical perception Transform network, predicting to obtain projection data of a missing angle, and forming a complete full-angle projection set; image reconstruction: carrying out filtering back projection reconstruction on the complete full-angle projection set to obtain an initial reconstruction image; and image post-processing: inputting the initial reconstructed image into a double-domain iterative optimization network, and realizing end-to-end cooperative training through projection domain-image domain joint loss. According to the method, the problems of artifacts and structural distortion generated when an image is reconstructed under the conditions of limited angles and cut-off projection in a traditional method are effectively solved, and the imaging quality and reliability of an STCT system are remarkably improved.

The application provides a pre-log domain Voronoi decomposition assisted low-dose CT reconstruction method, which comprises the following steps: obtaining a pre-log sinogram of low-dose CT; performing Voronoi decomposition on the pre-log sinogram by using a K-means clustering algorithm to obtain a plurality of feature clusters; mapping each feature cluster to a corresponding hidden space to obtain a multi-channel input feature map; inputting the multi-channel input feature map into a pre-trained diffusion transformer model for iterative denoising and optimization to obtain optimized clustering features; fusing and logarithmically transforming the optimized clustering features to obtain a post-log sinogram; and reconstructing the post-log sinogram by using a filtered back-projection algorithm to obtain a high-quality CT image. The application solves the problems of large dynamic range, uneven gradient and noise amplification at each stage in the pre-log domain by performing feature decoupling, hidden space modeling and diffusion transformer optimization on the pre-log sinogram, and realizes high-fidelity low-dose CT image reconstruction.

The application discloses a zooming static CT system and an image reconstruction method, wherein the system comprises a cylindrical detector array arranged in an arc and an X-ray source array arranged in a preset shape; the X-ray source array and the detector array remain static during scanning; projection data is acquired at different angles by sequentially emitting beams of different X-ray sources; the distance from the X-ray source to the detector is variable; a data acquisition system is used for controlling the detector to acquire data; a control system controls the emission of the X-ray source and the acquisition of data by the detector; and an image processing system processes acquired image projection data by using an image reconstruction algorithm. The image reconstruction algorithm in the embodiment of the application adopts a weighted filter back-projection reconstruction algorithm, which can optimize the system geometry, improve the efficiency and accuracy of the reconstructed image, and improve the applicability and reliability of the system.

The invention discloses a CT (Computed Tomography) image hybrid filtering optimization method, which comprises the following steps of: 1) reconstructing projection data of a sample by using a filtering back projection algorithm to obtain an initial CT image of the sample; (2) detecting the initial CT image, determining the position of a ray penetrating through the target structure in projection data, and carrying out classified filtering on the projection data to obtain a CT reconstruction image after artifacts are removed; 3) performing coordinate transformation on the CT reconstruction image, and transforming the CT reconstruction image from a rectangular coordinate to a polar coordinate to obtain an image under the polar coordinate; 4) performing wavelet transformation on the image under the polar coordinates, and decomposing the image into a low-frequency component, a vertical high-frequency component, a horizontal high-frequency component and a diagonal high-frequency component under multiple scales; multiplying the vertical high-frequency component by an enhancement coefficient to enhance edge features in the image under polar coordinates, and then performing wavelet inverse transformation; and 5) converting the processed image to rectangular coordinates to obtain a final optimized sample CT image.

The invention discloses a nonlinear back projection reconstruction method for open structure magnetic particle imaging, and belongs to the technical field of medical image processing and reconstruction, and the method specifically comprises the steps: S1, building a zero field line nonlinear change model; s2, deriving a nonlinear filtering back projection reconstruction formula; and S3, signal acquisition and image reconstruction. According to the method, the nonlinear bending rule of the zero field line in the open structure MPI system in the scanning process is constructed; meanwhile, the nonlinear model is embedded into a filtering back projection reconstruction framework, and a new nonlinear zero field line back projection reconstruction method is deduced; and rebuilt image edge distortion caused by zero field line bending is corrected through an algorithm, and high-precision imaging under an open-structure MPI large view field is realized.