Methods and systems are provided for reconstructing images from measurement data using one or more deep neural networks according to a decimation strategy. In one embodiment, a method for reconstructing an image using measurement data comprises, receiving measurement data acquired by an imaging device, selecting a decimation strategy, producing a reconstructed image from the measurement data using the decimation strategy and one or more deep neural networks, and displaying the reconstructed image via a display device. By decimating measurement data to form one or more decimated measurement data arrays, a computational complexity of mapping the measurement data to image data may be reduced from O(N.sup.4), where N is the size of the measurement data, to O(M.sup.4), where M is the size of an individual decimated measurement data array, wherein M<N.

Imaging systems and methods may facilitate positioning an imaging device in a procedure room. A 3D image of a subject may be obtained, where the subject is to have a procedure performed thereon. A view of the 3D image of the subject may be adjusted to a desired view and an associated 2D image reconstruction at the desired view may be obtained. A position for the imaging device that is associated with the desired view of the 3D image of the subject may be identified. Adjusting a view of the 3D image to a desired view and obtaining a 2D image reconstruction may be performed pre-procedure, such that a user may be able to create a list of desired views pre. A user may adjust a physical position of the imaging device to obtain reconstructed 2D preview images at the adjusted physical position of the imaging device prior to capturing an image.

A method for establishing a 3D medical imaging model of a subject is to be implemented by an X-ray computed tomography (CT) scanner and a processor. The method includes: emitting X-rays on the subject sequentially from plural angles with respect to the subject to obtain M number of X-ray images of the subject in sequence; obtaining, for each pair of consecutive X-ray images, K number of intermediate image(s) by using the pair of consecutive X-ray images as inputs to a convolutional neural network (CNN) model that has been trained for frame interpolation; and establishing the 3D medical imaging model by using a 3D reconstruction technique based on the M number of X-ray images and the intermediate images obtained for the M number of X-ray images.

A method and system for determining a PET image of the scan volume based on one or more PET sub-images is provided. The method may include determining a scan volume of a subject supported by a scan table; dividing the scan volume into one or more scan regions; for each scan region of the one or more scan regions, determining whether there is a physiological motion in the scan region; generating, based on a result of the determination, a PET sub-image of the scan region based on first PET data of the scan region acquired in a first mode or based, at least in part, on second PET data of the scan region acquired in a second mode; and generating a PET image of the scan volume based on one or more PET sub-images.

A method and system are provided for reconstructing a motion-compensated nuclear image of a subject, as well as an arrangement for method. The reconstruction method comprises receiving nuclear image data the acquiring a nuclear image, and a computer program for carrying out the for multiple motion states, reconstructing the data into an image for each motion state, and calculating a deformation vector field for each state for mapping the image onto a reference motion state. Calculating the deformation vector field comprises providing an initial vector field, defining at least one rigid region of the subject, incorporating that rigid region into the initial vector field, and calculating the deformation vector field with the incorporated rigid region. The method further comprises mapping the reconstructed image of each motion state onto the reference state using the deformation vector fields; and combining the mapped images into a motion-compensated nuclear image.

A method and apparatus is provided that uses a deep learning (DL) network together with a multi-resolution detector to perform X-ray projection imaging to provide improved resolution similar to a single-resolution detector but at lower cost and less demand on the communication bandwidth between the rotating and stationary parts of an X-ray gantry. The DL network is trained using a training dataset that includes input data and target data. The input data includes projection data acquired using a multi-resolution detector, and the target data includes projection data acquired using a single-resolution, high-resolution detector. Thus, the DL network is trained to improve the resolution of projection data acquired using a multi-resolution detector. Further, the DL network is can be trained to additional correct other aspects of the projection data (e.g., noise and artifacts).

A method, system, and computer readable medium to compensate for consecutive missing views in Computed Tomography (CT) reconstruction. By utilizing at least one complementary ray from a previous or subsequent view, the missing view(s) can be filled in. When plural complementary rays exist, a linear or non-linear combination of rays can be used to fill in the missing views, and the weights used in the combination may be smoothed to prevent over-emphasis of the replacement views.

Methods for reconstructing images from undersampled k-space data using a machine learning approach to learn non-linear mapping functions from acquired k-space lines to generate unacquired target points across multiple coils are described.

A method for gating in tomographic imaging system includes steps of: (a) performing a tomographic imaging on an object with a target moving periodically along a first axis for acquiring projection images; (b) obtaining projected curves by summing up pixel values along a direction of a second axis perpendicular to the first axis in each projection image; (c) determining a target zone on the projection images, wherein a central position on the first axis of the target zone is corresponding to a position having the largest variation in the projected curves on the first axis; (d) calculating parameter values of pixel values in the target zones and obtaining a curve of a moving cycle of the target according to the parameter values; and (e) selecting the projection images under the same state in the moving cycle for image reconstruction according to the curve of the moving cycle of the target.

The embodiments of the present disclosure disclose methods and systems for determining a ring artifact. The method for determining the ring artifact may include: obtaining an original image; mapping a plurality of pixels in the original image to a polar coordinate image; determining a protection region in the polar coordinate image; obtaining a smooth image by smoothing at least one region in the polar coordinate image other than the protection region; generating a residual image based on the polar coordinate image and the smooth image; determining a location of the ring artifact in the original image based on the residual image. In the present disclosure, the original image may be mapped to a trapezoidal region or a triangular region in the polar coordinate image, and the gradient angle image may be used for image processing, which may reduce the influence of noise. An accurate location of the ring artifact may be determined, and information for imaging device detection and air correction may be provided.