Techniques for performing iterative MRI image reconstruction by learning complementary multi-prior knowledge from images, k-space data, and calibration data are disclosed. In one method, k-space data is obtained from an MRI scan. Image-space modifications are performed on the k-space data using a first neural network trained to operate on data in image space. The k-space data is converted from the frequency domain to a spatial domain to produce input image-space data. Using the first neural network, output image-space data is generated, which is then converted from the spatial domain to the frequency domain. K-space modifications are performed on the k-space data using a second neural network trained to operate on data in k-space. ACS are encoded using a third neural network to guide the second neural network in learning consistency-aware k-space correlations. The k-space data is converted from the frequency domain to the spatial domain to obtain a reconstructed image.

Metal artifact correction including projecting x-rays to scan a volumetric region of an object, the projecting generates corresponding cone beam computed tomography (CBCT) image data, reconstructing an enlarged CBCT volume from the CBCT image data, the enlarged CBCT volume representative of the volumetric region and a volume outside the volumetric region, generating from the enlarged CBCT volume and a projection geometry, maximum intensity projections on a virtual plane, detecting attenuated image areas in the maximum intensity projections corresponding to metal, corresponding the detected attenuated image areas corresponding to the metal to areas of the CBCT image data, and reconstructing a final CBCT volume using the CBCT image data by suppression of the areas of the CBCT image data corresponding to the detected attenuated image areas of the maximum intensity projections. Systems for metal artifact correction are also disclosed.

A method comprising receiving a plurality of images from a multi-modal imaging system; generating a plurality of filtered measurements by performing multi-modal spectral fusion of the plurality of images; and generating, using a physics-informed neural network (PINN) trained based on one or more physical principles associated with X-ray attenuation or scattering, a reconstructed object image based on the plurality of filtered measurements, wherein generating the reconstructed object image comprises (i) generating, using the PINN, a system matrix for an X-ray imaging forward model by refining one or more coefficients of the system matrix based on a physics-informed loss function, and (ii) generating, using the X-ray imaging forward model and based on the plurality of filtered measurements, the reconstructed object image.

Systems and methods for training a machine-learning model for artifact reduction are provided. Such methods include retrieving a three-dimensional digital phantom reconstructed from CT imaging data. The method then selects a first Z position along the central axis and simulates a first set of forward projections from the digital phantom taken along an axial trajectory at the first Z position along the central axis. The first set of forward projections has a first simulated collimation in the axial direction. The method then reconstructs a first simulated image from the first set of forward projections and identifies a plurality of secondary Z positions along the central axis other than the first Z position. For each of the secondary Z positions and the first Z position itself, the method then simulates a set of secondary forward projections from the digital phantom taken along corresponding axial trajectories at the corresponding secondary Z position.

A computer-implemented method includes receiving volumetric image data generated during an imaging examination of a subject with a medical imaging system and creating, based on rank filtering, a different local projection for each of a plurality of identified regions of tissue of interest of the volumetric image data. A first projection for a region of tissue of interest visually emphasizes a first contrast level, a second projection for a second region of tissue of interest visually emphasizes second contrast level, and the first contrast level and the second contrast level are different contrast levels. The computer-implemented method further includes merging the different local projections, including the first projection and the second projection, with the volumetric image data into a single composite image and displaying the single composite image. Both lower contrast and higher contrast structures in the volumetric image data are visually emphasized in the displayed single composite image.

Disclosed herein is a medical system (100, 300) comprising a memory (110) storing machine executable instructions (120) and an anatomical detection module (122), and a computational system (104). The execution of the machine executable instructions causes the computational system to: receive (200) a set of magnetic resonance images (124) descriptive of a field of view (109) of a subject (318) acquired according to a synthetic magnetic resonance imaging protocol; receive (202) an anomaly indicator (126) from the anomaly detection module in response to inputting at least one of the set of magnetic resonance images into the anomaly detection module; determine (204) a set of reconstruction parameters (128) using the anomaly indicator; and reconstruct (206) a synthesized magnetic resonance image (134) from the set of magnetic resonance images and the set of reconstruction parameters.

An image processing apparatus generates a first image by extracting a first region which is a first artifact generation source from a CT image, generates a second image by subtracting a value determined according to a specific part from the first image, generates a third image by performing forward projection and back projection on the second image, generates a fourth image which is a difference image between the second image and the third image, and generates a fifth image by correcting an artifact in the CT image using the fourth image.

Systems and methods for fast reconstruction of motion-resolved magnetic resonance images using space-time-coil convolutional networks are disclosed. The system can receive a plurality of k-space data sets. The system can detect a motion signal therefrom. The system can classify the k-space data sets according to states of the motion signals. The system can resolve the k-space data set to Euclidean space images. The system can resolve the Euclidean space images to a combined Euclidian space image. For example, the system can use a convolutional network that exploits spatial, temporal and coil correlations without k-space data consistency to minimize computation time.

This disclosure introduces an approach that includes techniques for determining an optimal weighted execution sequence of available reconstruction algorithms using a multi-processor unit. The introduced approach includes executing a series of optimal weighted execution sequence candidates on a representative slice of the image data and comparing their results to select one of the candidates as the optimal weighted execution sequence.

A dynamic pulmonary magnetic resonance imaging method, including: with an ultrashort echo time sequence, acquiring pulmonary magnetic resonance signals in a free-breathing condition; during data acquisition, monitoring a respiratory condition, obtaining a respiration curve; with the respiration curve, performing motion-resolved reconstruction on the acquired pulmonary magnetic resonance data, obtaining motion-resolved lung images; performing motion field estimation on the motion-resolved lung images; performing motion-state weighted motion-compensated reconstruction on results of the motion field estimation, obtaining dynamic pulmonary magnetic resonance images; performing ventilation map estimation on the dynamic pulmonary magnetic resonance images, obtaining a pulmonary ventilation.