172 results about "K-space" patented technology

The invention provides a magnetic resonance imaging K space movement artifact correction parallel acquisition reconstruction method. The method comprises the following steps: an imaged object is placed in a magnetic resonance imaging system to acquire magnetic resonance signals for filing the K space; acquisition is performed on the K space in the multiple-blade manner, wherein the inter-blade K space geometrical relationship is that rotation is performed for an angle; correction is performed on the acquired blades according to the PROPELLER algorithm to calculate the inter-blade data change due to rigid body motion and calculate the movement calibration coefficient of each blade; for each blade, other appropriate blades are selected separately, and the other blades are transformed into the Cartesian coordinate system of the blade according to the movement calibration coefficient to calculate the coil consolidation coefficient in the blade data missing direction; the data missed by the blade is filled according to the calculated coefficient; after same operations are performed on all of the blades, the data filling of each blade is completed, then the complete K space is filled according to the PROPELLER algorithm and transformed to the image domain through appropriate transformation to obtain images. By adopting the method, parallel acquisition reconstruction can be performed without acquiring calibration data while effectively removing movement artifact so as to improve the acquisition speed.

The embodiment of the invention discloses a magnetic resonance image reconstruction method and device, equipment and a medium. The method comprises the following steps: acquiring undersampled magneticresonance data; and inputting the magnetic resonance data into an image reconstruction model based on an alternating direction multiplier algorithm to obtain a reconstructed target magnetic resonanceimage, the image reconstruction model being a pre-trained result of a mathematical model in which a data fidelity term and a regular term are both uncertain terms. According to the embodiment of theinvention, the method solves a problem that a data fidelity term in a model suitable for an ADMM-net method is represented by employing a 2-norm between a reconstructed k space and a sampling point ina magnetic resonance image reconstruction process, is not a most effective data consistency guarantee method, and is not suitable for all image reconstruction conditions. The image quality after image reconstruction based on the ADMM algorithm can be improved.

The invention discloses an equal voxel magnetic resonance diffusion imaging method and an equal voxel magnetic resonance diffusion imaging device based on multi-plate simultaneous excitation, whereinthe method comprises the following steps: S1, performing excitation for a tested target for many times through multi-plate simultaneous excitation pulse, and in a process of excitation each time, performing signal acquisition for the tested target through a multi-channel coil, and thereby obtaining k space data acquired through down sampling in excitation each time; S2, by a multi-excitation diffusion imaging reconstruction algorithm uniting the k space and an image domain, recovering k space position data which is not sampled in excitation each time; S3, correcting edge artifacts through an improved NPEN algorithm, thereby obtaining an imaged image. The method can obtain high-resolution images and also can keep relatively high signal-to-noise ratio at the same time, can reduce interference of three-dimensional navigation echo errors and improve quality and stability of reconstructed images, and can improve signal-to-noise ratio efficiency and scanning efficiency on the basis of guaranteeing the quality of the images.

The invention discloses an accelerated magnetic resonance super-resolution imaging method, device and equipment and a medium, and the method comprises the steps: obtaining partial K space frequency domain data of a to-be-detected region of a target patient, and obtaining a target accelerated magnetic resonance image; inputting the target accelerated magnetic resonance image into a target neural network model which is obtained by training an enhanced super-resolution generative adversarial network constructed based on a residual dense network by using a simulation accelerated magnetic resonance image set synthesized based on a standard sequence magnetic resonance image; and performing super-resolution processing on the target accelerated magnetic resonance image through the target neural network model to obtain a super-resolution accelerated magnetic resonance image. According to the method, the pre-created generative adversarial network is trained through the simulation acceleration magnetic resonance image set synthesized based on the standard sequence magnetic resonance image, the real acceleration magnetic resonance image can be simulated, the imaging quality of the acceleration magnetic resonance image is improved, the image noise is reduced, and the image resolution is improved.

A magnetic resonance imaging system detects a magnetic resonance signal generated from a sample, repeatedly acquires data on the detected magnetic resonance signal in an imaging period and arranges the acquired data in k space. The data is acquired in one acquisition pattern of a plurality of different acquisition patterns, which are determined so that an acquisition frequency of data in some areas of k space is different from that in the other areas of k space. An image of the sample is repeatedly reconstructed based on the acquired data and arranged in the k space. The acquisition is controlled to change the acquisition pattern during the imaging period.

The invention discloses an unsupervised training method and an unsupervised training device for a magnetic resonance parameter imaging model. The unsupervised training method comprises the following steps: inputting acquired under-acquisition k space data into a to-be-trained magnetic resonance parameter imaging model to obtain parameter weighted image data; updating a loss function according to the acquired under-acquisition k space data and the parameter weighted image data; and updating the network parameters of the to-be-trained magnetic resonance parameter imaging model according to the updated loss function. According to the method and device, the loss function based on the parameter weighted image is introduced, calculation of the loss function can be completed by utilizing the under-collected k space data and the parameter weighted image data obtained according to the under-collected k space data, full-collected k space data does not need to be adopted, and unsupervised training of the model is achieved.

The invention discloses a dynamic magnetic resonance image reconstruction method based on improved robust tensor principal component analysis. Precision and speed of an existing dynamic magnetic resonance image reconstruction method need to be improved. The method comprises: using tensors for representing K space data obtained through radial sampling, using a tensor robust principal component analysis tool for constructing an image reconstruction model, and guaranteeing the space integrity of high-dimensional data; providing a new tensor nuclear norm to constrain a low-rank part, so that the overall low-rank constraint is improved while the low-rank processing efficiency is ensured; carrying out time-frequency transformation on the sparse part and then carrying out threshold processing toimprove the reconstruction precision; finally, using an iterative soft threshold shrinkage algorithm to solve the optimization problem of the algorithm, and the image reconstruction quality and the accelerated reconstruction efficiency can be effectively improved. According to the invention, a high-quality diagnosis part image can be reconstructed in a short time, and a clear image can still be obtained at an extremely low sampling rate.