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.

Model-based image reconstruction (MBIR) methods using convolutional neural networks (CNNs) as priors have demonstrated superior image quality and robustness compared to conventional methods. Studies have explored MBIR combined with supervised and unsupervised denoising techniques for image reconstruction in magnetic resonance imaging (MRI) and positron emission tomography (PET). Unsupervised methods like the deep image prior (DIP) have shown promising results and are less prone to hallucinations. However, since the noisy image is used as a reference, strategies to prevent overfitting are unclear. Recently, Bayesian DIP (BDIP) networks that model uncertainty tend to prevent overfitting without requiring early stopping. However, BDIP has not been studied with data-fidelity term for image reconstruction. Present disclosure provides systems and method that implement a MBIR framework with a modified BDIP. Specifically, an uncertainty-based penalty is included to the BDIP to improve reconstruction across iterations.

The present disclosure provides methods and systems for X-ray imaging. The methods may include obtaining pre-scan imaging data relating to a target section of a target subject. The methods may include determining, based on the pre-scan imaging data, a chord length of at least one chord of the target section. The methods may also include determining, based on the chord length of the at least one chord of the target section, exposure parameters to be used by the X-ray imaging device in a target scan of the target subject. The methods may further include reconstructing a target image of the target subject based on scan data collected by the X-ray imaging device in the target scan.

Systems and methods of image reconstruction are provided. A system may have a memory and a processor to receive data corresponding to magnetic resonance imaging coils and data corresponding to a region of interest within a field of view of the magnetic resonance imaging machine. By determining different weights to associate with virtualized magnetic resonance imaging coils, images may be reconstructed to favor signals associated with a region of interest and to disfavor interference associated with areas outside the region of interest.

A method for performing real-time magnetic resonance (MR) imaging on a subject is disclosed. A prep pulse sequence is applied to the subject to obtain a high-quality special subspace, and a direct linear mapping from k-space training data to subspace coordinates. A live pulse sequence is then applied to the subject. During the live pulse sequence, real-time images are constructed using a fast matrix multiplication procedure on a single instance of the k-space training readout (e.g., a single k-space line or trajectory), which can be acquired at a high temporal rate.

Embodiments of the present invention disclose a magnetic resonance system, a magnetic resonance image correction method, and a magnetic resonance imaging method. The magnetic resonance image correction method comprises: separately obtaining main magnetic field intensity distribution information and gradient magnetic field information of a magnetic resonance system, the gradient magnetic field information comprising one or more among X-axis gradient magnetic field information, Y-axis gradient magnetic field information, and Z-axis gradient magnetic field information; obtaining a first correction coefficient based on the main magnetic field intensity distribution information; obtaining, based on one or more among the X-axis gradient magnetic field information, the Y-axis gradient magnetic field information, and the Z-axis gradient magnetic field information, one or more corresponding second correction coefficients; and, correcting image data obtained by means of the magnetic resonance system based on the first correction coefficient and the one or more second correction coefficients.

A PET-data correction method, a PET-data correction device, a computer apparatus, and a PET-image reconstruction method. The PET data correction method includes: acquiring single-events during a PET scan, the single-event including a non-scattering event and a scattering event (S101); obtaining a first correction parameter of the non-scattering event, and correcting the non-scattering event according to the first correction parameter (S102); obtaining scattering features of the scattering event, and classifying the scattering event based on the scattering features (S103); and obtaining a second correction parameter of the scattering event of each different classification, and correcting the scattering event according to the second correction parameter (S104). By means of the PET-data correction method, a reconstruction result is more accurate, and an imaging effect is better.

The present disclosure is generally directed to systems and methods for generating de-noised MR images that are reconstructed from a hybridization of two separate image reconstruction pipelines, at least one of which includes the use of a neural network. Further, the amount of influence that the neural network reconstruction has on the hybrid reconstructed image is controlled via a regularization parameter that is selected based on an estimated noise level associated with the initial image acquisition, which can be calculated from pre-scan data.

A nuclear medicine diagnosis apparatus according to an embodiment includes processing circuitry. The processing circuitry is configured: to acquire nuclear medicine data by scanning an examined subject; to specify timing of a body movement of the examined subject during the acquisition of the nuclear medicine data; and to extend an acquisition period of the scan on the basis of the timing.

Described herein is a medical system (100, 300) comprising a memory (110) storing machine executable instructions (120) and a motion estimating neural network (122, 700, 800, 900, 1000) configured for outputting trajectory data (130) in response to receiving a trial motion trajectory (128) as input. The execution of the machine executable instructions causes a computational system (104) to: receive (200) measured k-space data (124) descriptive of a subject (318); perform (202) motion estimation of the subject between the sequence of discrete acquisitions by solving an optimization problem to determine a calculated motion trajectory of the subject in the predefined coordinate system, wherein the optimization problem is modified using the trajectory data; and reconstruct (204) a final motion corrected magnetic resonance image (136) from the measured k-space data and the calculated motion trajectory in the predefined coordinate system.