Abstract: A trained deep learning network is for determining a cardiac phase in magnet resonance imaging. In an embodiment, the trained deep learning network includes an input layer; an output layer; and a number of hidden layers between input layer and output layer, the layers processing input data entered into the input layer. In an embodiment, the deep learning network is designed and trained to output a probability or some other label of a certain cardiac phase at a certain time from entered input data. A method for determining a cardiac phase in magnet resonance imaging; a related device; a training method for the deep learning network; a control device and a related magnetic resonance imaging system are also disclosed.

Abstract: Systems and methods for recreating images from undersampled MRI image data includes capturing undersampled MRI data and enhancing it with multiple cascading stages, each including a data consistency block in parallel to a convolutional neural network (CNN). The data consistency block adjusts each input image by a sensitivity map and performs hard replacement of acquired lines in k-space into the image. The CNN estimates a regularizer term that attempts to minimize a difference between a true image and the output of the data consistency block. At each stage, the output of CNN and data consistency block are added to create a set of output images that feed into the next stage.

Abstract: A frequency offset is selected based on similarity measures of multiple MRI images obtained from frequency scout measurements associated with multiple frequency offsets from a reference frequency of a magnetization excitation pulse. The similarity measure for each respective MRI image of the multiple MRI images is determined based on a comparison between at least one reference image and the respective MRI image. The at least one reference image is determined from the multiple MRI images based on spectrum information of each of the multiple MRI images. Such methods facilitate automatically determining/selecting a more optimal frequency offset for an MRI scan following a frequency scout scan, in particular, for an SSFP or a bSSFP pulse sequence, and thereby banding artifacts and/or flow-related artifacts can be reduced for the MRI scan.

Abstract: A computer-implemented method for determining scar segmentation includes receiving a medical image of an object to be segmented acquired after an application of a low-dose of contrast agent, wherein the low-dose of contrast agent comprises less contrast agent than a standard full-dose of contrast agent; and determining a scar segmentation mask by applying a trained artificial neural network to the medical image.

Abstract: Systems and methods for predicting a location for acquiring a target view of an anatomical object of interest in an input image are provided. An input image of an anatomical object of interest of a patient is received. An output image is generated using a machine learning based network. The output image depicts a projection of a 3D image plane for acquiring a target view of the anatomical object of interest identified on the input image. The output image is output.

Abstract: A computer-implemented method for determining an orientation of at least one diagnostically relevant sectional plane for heart imaging in a three-dimensional magnetic resonance imaging image dataset, comprises: providing the three-dimensional image dataset; applying a trained function to the three-dimensional image dataset to determine a position of at least one landmark; determining the orientation of the at least one diagnostically relevant sectional plane as a function of at least one landmark; and providing the orientation of the at least one diagnostically relevant sectional plane.

Abstract: The disclosure relates to techniques for determining a motion parameter of a heart. A subset of a sequence of cardiac MR images is applied as a first input to a first trained convolutional neural network configured to determine, as a first output, a probability distribution of at least 2 anatomical landmarks. The sequence of cardiac MR images is cropped and realigned based on the at least 2 anatomical landmarks to determine a reframed and aligned sequence of new cardiac MR images showing the same orientation of the heart. The reframed and aligned sequence of new cardiac MR images is applied to a second trained convolutional neural network configured to determine, as a second output, a further probability distribution of the at least 2 anatomical landmarks in each new MR image of the reframed and aligned sequence, the motion parameter of the heart is determined based on the second output.

Abstract: In a method for automatic motion detection in medical image-series, a dataset of a series of images is provided. The images can be of a similar region of interest that are recorded at consecutive points of time. The method can further include localizing a target in the images of the dataset and calculating a position of the target in the images to calculate localization data of the target, and calculating movement data of a movement of the target of temporal adjacent images of the images based on the localization data.

Abstract: Systems and methods for predicting a location for acquiring a target view of an anatomical object of interest in an input image are provided. An input image of an anatomical object of interest of a patient is received. An output image is generated using a machine learning based network. The output image depicts a projection of a 3D image plane for acquiring a target view of the anatomical object of interest identified on the input image. The output image is output.

Abstract: A trained deep learning network is for determining a cardiac phase in magnet resonance imaging. In an embodiment, the trained deep learning network includes an input layer; an output layer; and a number of hidden layers between input layer and output layer, the layers processing input data entered into the input layer. In an embodiment, the deep learning network is designed and trained to output a probability or some other label of a certain cardiac phase at a certain time from entered input data. A method for determining a cardiac phase in magnet resonance imaging; a related device; a training method for the deep learning network; a control device and a related magnetic resonance imaging system are also disclosed.

Abstract: In a method for automatic motion detection in medical image-series, a dataset of a series of images is provided. The images can be of a similar region of interest that are recorded at consecutive points of time. The method can further include localizing a target in the images of the dataset and calculating a position of the target in the images to calculate localization data of the target, and calculating movement data of a movement of the target of temporal adjacent images of the images based on the localization data.

Abstract: In a method and magnetic resonance apparatus for recording diagnostic measurement data of a heart of an examination object, the magnetic resonance apparatus is operated by a control sequence wherein an RF pulse excites nuclear spins with a flip angle of at least 60°, the diagnostic measurement data are recorded in a coordinate system independent of the heart, and the basic magnetic field produced by the magnetic resonance apparatus is smaller than 1.0 tesla.

Abstract: In a parameter value determination method, parameter values are determined based on at least two previously determined most similar comparison signal curves. As a result, the parameters for determining can be determined with a resolution greater than the resolution, underlying the comparison signal curves, of the values of the parameters to be determined. Advantageously, the determination of the parameter values are not limited to the values of the comparison signal curves, in other words, are not limited to the lattice/grid of the dictionary.

Abstract: A method for generating at least one acquisition template for an acquisition of magnetic resonance signals, an acquisition template generating unit, a magnetic resonance apparatus and a computer program product. At least one acquisition template is generated with an acquisition template generating unit. The at least one acquisition template has a plurality of spiral-like spokes in a k-space, each spoke having a plurality of spiral points.

Abstract: In a method and apparatus for producing respiration-corrected MR images of an examination volume containing the heart of a patient during respiratory movement of MR signals are recorded continuously during multiple cardiac cycles, each cardiac cycle having multiple time segments. One 2D navigator image data record per cardiac cycle is recorded during a time segment of that cardiac cycle, with k-space being filled along a Cartesian trajectory such that a spatial resolution is achieved in two spatial directions of the examination volume. Also, multiple 3D image data records are recorded in the other time segments of that cardiac cycle, with Cartesian filling of raw data space such that a spatial resolution is achieved in all three spatial directions of the examination volume. The respiratory movement is then determined from these navigator data records. The determined respiratory movement is corrected in the recorded MR signals.

Abstract: In a method and apparatus for producing respiration-corrected MR images of an examination volume containing the heart of a patient during respiratory movement of MR signals are recorded continuously during multiple cardiac cycles, each cardiac cycle having multiple time segments. One 2D navigator image data record per cardiac cycle is recorded during a time segment of that cardiac cycle, with k-space being filled along a Cartesian trajectory such that a spatial resolution is achieved in two spatial directions of the examination volume. Also, multiple 3D image data records are recorded in the other time segments of that cardiac cycle, with Cartesian filling of raw data space such that a spatial resolution is achieved in all three spatial directions of the examination volume. The respiratory movement is then determined from these navigator data records. The determined respiratory movement is corrected in the recorded MR signals.

Abstract: In a method and magnetic resonance apparatus for recording diagnostic measurement data of a heart of an examination object, the magnetic resonance apparatus is operated by a control sequence wherein an RF pulse excites nuclear spins with a flip angle of at least 60°, the diagnostic measurement data are recorded in a coordinate system independent of the heart, and the basic magnetic field produced by the magnetic resonance apparatus is smaller than 1.0 tesla.

Abstract: A method is for determining a type of reconstruction of image data from a magnetic resonance measurement of an examination object by way of a magnetic resonance device. In an embodiment, the method includes determining at least one first parameter which includes information about a breathing process of the examination object; comparing the at least one first parameter with at least one second parameter from a maintained database; and determining a type of reconstruction of image data of the magnetic resonance measurement according to the comparison that has been made. In an example embodiment, at least one of the at least one first parameter and the at least one second parameter is derived from a histogram of respiratory phases and the maintained database includes a self-learning system.

Abstract: A method for generating at least one acquisition template for an acquisition of magnetic resonance signals, an acquisition template generating unit, a magnetic resonance apparatus and a computer program product. At least one acquisition template is generated with an acquisition template generating unit. The at least one acquisition template has a plurality of spiral-like spokes in a k-space, each spoke having a plurality of spiral points.

Abstract: A method is disclosed for determining a type of reconstruction of image data from a magnetic resonance measurement of an examination object by way of a magnetic resonance device. In an embodiment, the method includes determining at least one first parameter which includes information about a breathing process of the examination object; comparing the at least one first parameter with at least one second parameter from a maintained database; and determining a type of reconstruction of image data of the magnetic resonance measurement according to the comparison that has been made. In an example embodiment, at least one of the at least one first parameter and the at least one second parameter is derived from a histogram of respiratory phases and the maintained database includes a self-learning system.