Abstract: A computer-implemented method includes, based on scan data defining an input image, determining a reconstructed image using a reconstruction algorithm, and executing a data consistency operation for enforcing consistency between the input image and the reconstructed image. The data consistency operation includes using a norm ball projection that takes into account the available noise level information in order to automatically adjust the balance between the network prediction and the input measurements.

Abstract: Systems and methods for reconstruction for a medical imaging system. An adapter is used to adapt scan data so that different quantities of repetitions or directions may be used to train and implement a single multichannel backbone network.

Abstract: Systems and methods for reconstruction for a medical imaging system. A scaling factor is used during the reconstruction process to adjust a step size of a gradient update. The adjustment of the step size of the gradient provides the ability to adjust a level of denoising by the reconstruction process.

Abstract: For reconstruction in medical imaging, such as reconstruction in MR imaging, scanning is accelerated by under-sampling. In iterative reconstruction, the input to the regularizer is altered provide for correlation of non-local aliasing artifacts. Duplicates of the input image are shifted by different amounts based on the level of acceleration. The resulting shifted images are used to form the input to the regularizer. Providing an input based on shifts allows the regularization to suppress non-local as well as local aliasing artifacts.

Abstract: The disclosure relates to techniques for automatically characterizing liver tissue of a patient, comprising receiving morphological magnetic resonance image data set and at least one magnetic resonance parameter map of an imaging region comprising at least partially the liver of the patient, each acquired by a magnetic resonance imaging device, via a first interface. The techniques further include applying a trained function comprising a neural network to input data comprising at least the image data set and the parameter map. At least one tissue score describing the liver tissue is generated as output data, which is provided using a second interface.

Abstract: Various examples relate to SENSitivity Encoding (SENSE) reconstruction of Magnetic Resonance Imaging (MRI) images. Multiple coil sensitivity maps per coil of a receiver coil array are used, e.g., obtained from an Eigenvalue-based Spatially Constrained Iterative Reconstruction Technique (ESPIRiT) autocalibration protocol.

Abstract: A method for acquiring magnetic resonance imaging data with respiratory motion compensation using one or more motion signals includes acquiring a plurality of gradient-delay-corrected radial readout views of a subject using a free-breathing multi-echo pulse sequence, and sampling a plurality of data points of the gradient-delay-corrected radial readout views to yield a self-gating signal. The self-gating signal is used to determine a plurality of respiratory motion states corresponding to the plurality of gradient-delay-corrected radial readout views. The respiratory motion states are used to correct respiratory motion bias in the gradient-delay-corrected radial readout views, thereby yielding gradient-delay-corrected and motion-compensated multi-echo data. One or more images are reconstructed using the gradient-delay-corrected and motion-compensated multi-echo data.

Abstract: Various techniques of reconstructing multiple Magnetic Resonance Imaging, MRI, images for multiple slices based on an MRI measurement dataset that is acquired using a simultaneous multi-slice protocol and undersampling and K-space are disclosed. A convolutional neural network can be used to implement a regularization operation of an iterative optimization for the reconstruction, i.e., an unrolled neural network or variational neural network. A combination with Dixon imaging, i.e., separation of multiple chemical species, is disclosed.

Abstract: For reconstruction in medical imaging using phase correction, a machine learning model is trained for reconstruction of an image. The reconstruction may be for a sequence without repetitions or may be for a sequence with repetitions. Where repetitions are used, rather than using just a loss for that repetition in training, the loss based on an aggregation of images reconstructed from multiple repetitions may used to train the machine learning model. In either approach, a phase correction is applied in machine training. A phase map is extracted from output of the model in training or extracted from the ground truth of the training data. The phase correction, based on the phase map, is applied to the ground truth and/or the output of the model in training. The resulting machine-learned model may better reconstruct an image as a result of having been trained using phase correction.

Abstract: A computer-implemented method for creating image data with a trained function from measurement data recorded with a magnetic resonance system may include: providing a trained reconstruction function, which receives magnetic resonance data in a dedicated form as input data, to which the trained reconstruction function is applied and in the process output data comprising image data determines image data, loading recorded measurement data, processing the recorded measurement data into processed magnetic resonance data such that the processed magnetic resonance data is present in a form which corresponds to the dedicated form of the input data, receiving the processed magnetic resonance data as input data, applying the provided trained reconstruction function to the received input data, wherein output data comprising image data is determined, and providing the output data.

Abstract: For reconstruction in medical imaging, such as reconstruction in MR imaging, the number of iterations in deep learning-based reconstruction may be reduced by including a learnable extrapolation in one or more iterations. Regularization may be provided in fewer than all of the iterations of the reconstruction. The result of either approach alone or both together is better quality reconstruction and/or less computationally expensive reconstruction.

Abstract: Techniques are provided for determining magnetic resonance images showing different contrasts in an examination. Magnetic resonance data for all magnetic resonance images are acquired using the same acquisition technique and the magnetic resonance images are reconstructed from their magnetic resonance data sets using at least one reconstruction algorithm. The reconstruction comprises at least one de-noising step. After acquisition of the magnetic resonance data, at least one noise strength measure is determined for the magnetic resonance data sets for each contrast, and de-noising strengths for the de-noising step are chosen individually for each contrast depending on the respective at least one noise strength measure.

Abstract: An adaptive reconstruction of MR data, including acquired MR data of a core region having core segments and simulated MR data of a peripheral region. The method includes ascertaining a peripheral signal based on the MR data of the peripheral region, determining a scaling factor for each core segment by taking into account the peripheral signal and a mean signal intensity of the MR data for the respective core segment, scaling the MR data of the core region by taking into account the MR data of each core segment and that of the scaling factor corresponding to the respective core segment, generating filtered MR data by combining the scaled MR data of the core region with the MR data of the peripheral region, and reconstructing image data from the filtered MR data.

Abstract: In an optimization to obtain spin-species specific magnetic resonance images, the optimization may use a target function that calculates a dephasing of a second spin species with respect to the first spin species based on a sampling trajectory of a respective measurement protocol.

Abstract: Techniques are provided for determining a magnetic resonance imaging (MRI) image using multiple measurement data sets that form a propeller pattern. Partial MRI images are reconstructed for each measurement data set. The partial MRI images are then combined.

Abstract: In a computer-implemented method of training a machine learning based processor, the processor can be trained to derive image data from signal data sets of multiple spin echo sequences. The trained processor can be configured to perform image processing for Magnetic Resonance Imaging (MRI) to derive the image data.

Abstract: For reconstruction in medical imaging, user control of a characteristic (e.g., noise level) of the reconstructed image is provided. A machine-learned model alters the reconstructed image to enhance or reduce the characteristic. The user selected level of characteristic is then provided by combining the reconstructed image with the altered image based on the input level of the characteristic. Personalized or more controllable impression for medical imaging reconstruction is provided without requiring different reconstructions.

Abstract: For reconstruction in medical imaging using a scan protocol with repetition, a machine learning model is trained for reconstruction of an image for each repetition. Rather than using a loss for that repetition in training, the loss based on an aggregation of images reconstructed from multiple repetitions is used to train the machine learning model. This loss for reconstruction of one repetition based on aggregation of reconstructions for multiple repetitions is based on deep set-based deep learning. The resulting machine-learned model may better reconstruct an image from a given repetition and/or a combined image from multiple repetitions than a model learned from a loss per repetition.

Abstract: For reconstruction, a machine-learned model is adapted to allow for reconstruction based on the repetitions available in some scanning. The reconstruction for one or more subsets is performed during the scanning. The machine-learned model is trained to reconstruction separately or independently for each repetition or to use information from previous repetitions without requiring waiting for completion of scanning. The reconstructed image may be displayed much more rapidly after completion of the acquisition since the reconstruction begins during the reconstruction.

Abstract: In a method for generation of a homogenization field suitable for homogenization of magnetic resonance data of an examination object, first magnetic resonance data from an examination region of the examination object is provided, a trained function is provided, a homogenization field is extracted by processing the first magnetic resonance data by way of the trained function, and the homogenization field is provided.