Provided is a medical image photographing apparatus, including: an X-ray generator configured to radiate an X-ray toward an object that is located in a three-dimensional (3D) virtual grid space which includes a plurality of voxels; an X-ray detector comprising a plurality of detecting elements and configured to detect the X-ray that has propagated through the object; and an image processor configured to process projection image data corresponding to the detected X-ray based on a volume of a first region within the plurality of voxels, through which the X-ray propagates.

Magnetic resonance imaging (MRI) image reconstruction using machine learning is described. A variational or unrolled deep neural network can be used in the context of an iterative optimization. In particular, a regularization operation can be based on a deep neural network. The deep neural network can take, as an input, an aliasing data structure being indicative of aliasing artifacts in one or prior images of the iterative optimization. The deep neural networks can be trained to suppress aliasing artifacts.

Systems, methods, and non-transitory computer readable media are described herein to facilitate generation of high-resolution two-dimensional projection images of an object having minimal artifacts from three-dimensional computed tomography volumes. Direct or iterative image reconstruction techniques can be used in concert with binning to identify and select measurement data subject to a criterion and resampling of the initial volumetric dataset to generate the high-resolution, two-dimensional projection images of at least a portion of the object.

Generally, there is provided a hybrid image reconstruction system. The hybrid image reconstruction system includes a deep learning stage and a compressed sensing stage. The deep learning stage is configured to receive an input data set that includes measured tomographic data and to produce a deep learning stage output. The deep learning stage includes a mapping circuitry, and at least one artificial neural network. The mapping circuitry is configured to map image domain data to a tomographic data domain. The compressed sensing stage is configured to receive the deep learning stage output and to provide refined image data as output.

The present invention relates to a device (100) for iterative reconstruction of images recorded by at least two imaging methods, the device comprising: an extraction module (10), which is configured to extract a first set of patches from a first image recorded by a first imaging method and to extract a second set of patches from a second image recorded by a second imaging method; a generation module (20), which is configured to generate a set of reference patches based on a merging of a first limited number of atoms for the first set of patches and of a second limited number of atoms for the second set of patches; and a regularization module (30), which is configured to perform a regularization of the first image or the second image by means of the generated set of reference patches.

A computing system (116) includes a reconstruction processor (114) configured to execute computer readable instructions, which cause the reconstruction processor to: receive, in electronic format, non-spectral projection data, reconstruct the non-spectral projection data to generate a non-spectral image, retrieve a non-spectral to spectral voxel value map for a basis material of interest from a set of non-spectral to spectral voxel value maps, generate a spectral iterative reconstruction start image based on the non-spectral image and the non-spectral to spectral voxel value map, and reconstruct a spectral image, in electronic format, for the material basis of interest from the non-spectral projection data with a spectral iterative reconstruction algorithm and the spectral iterative reconstruction start image.

An extra-oral dental imaging system comprises an X-ray source (102) and an imaging device (101) suitable for producing multiple frames during at least part of an exposure of an object (200), the imaging device (101) being displaced along a scanning direction (X). A method for creating a cephalometric image of a human skull comprises a step of setting said imaging device (101) with an active area having in an imaging plane a width extending along said scanning direction (X), said width varying along a height direction perpendicular to said scanning direction (X); a step of synchronously displacing the X-ray source (102) and the imaging device (101) along said exposure profile; and a step of registering multiple frames produced by the imaging device (101) during the exposure of said object (200) to be imaged. Using for creating a cephalometric image by digital tomosynthesis.

A CT image reconstructing method, a CT image reconstructing device and a CT system are provided for reducing motion artifacts in CT images in case of motion of an object. The CT image reconstructing method reconstructs CT images from projection data obtained by X-ray scanning, including a moving object position detecting step for detecting a position of a moving object in a CT image; a partial angle selecting step for selecting a view point and an angle range according to said position of the moving object and selecting data of partial angles in said projection data according to said view point and said angle range; a partial angle constraint step for generating partial angles constraint conditions according to the data of said partial angles; and an iterative reconstruction step for generating CT images by iterative reconstruction, thereby improving temporal resolution of CT images of moving objects and reducing motion artifacts.

A system executes efficient computational methods for high quality image reconstructions from a relatively small number of noisy (or degraded) sensor imaging measurements or scans. The system includes a processing device and instructions. The processing device executes the instructions to employ transform learning as a regularizer for solving inverse problems when reconstructing an image from the imaging measurements, the instructions executable to: adapt a transform model to a first set of image patches of a first set of images containing at least a first image, to model the first set of image patches as sparse in a transform domain while allowing deviation from perfect sparsity; reconstruct a second image by minimizing an optimization objective comprising a transform-based regularizer that employs the transform model, and a data fidelity term formed using the imaging measurements; and store the second image in the computer-readable medium, the second image displayable on a display device.

Systems and methods are provided for reprojection and back projection of objects of interest via homographic transforms, and particularly one-dimensional homographic transforms. In one example, a method may include acquiring imaging data corresponding to a plurality of divergent X-rays, assigning a single functional form to the plurality of divergent X-rays, determining, via a homographic transform, weights of interaction between a plurality of distribution samples and a plurality of X-ray detector bins based on the single functional form, and reconstructing an image based on the weights of interaction.