Abstract: Techniques and system configurations to enable the automated collection and evaluation of clinical data within an automated insights information system are disclosed herein. In an example, the information system is adapted to continuously monitor clinical systems for new patient data, process patient data into a standardized and structured format, selectively run algorithms to classify and characterize data, and stores the results of algorithms (such as findings, predictions, and recommendations) that can be used as input to other algorithms, or sent to clinical systems and presented to end users. In a specific example, a method performed in a computing system may include: requesting and obtaining a first and second set of clinical data, analyzing the first and second set of clinical data with respective algorithms, identifying a clinical finding, and generating output from the computing system based on the identified clinical finding.

Abstract: Techniques for training a deep neural network from user interaction workflow activities occurring among distributed computing devices are disclosed herein. In an example, processing of input data (such as input medical imaging data) is performed at a client computing device with the execution of an algorithm of a deep neural network. A set of updated training parameters are generated to update the algorithm of the deep neural network, based on user interaction activities (such as user acceptance and user modification in a graphical user interface) that occur with the results of the executed algorithm. The generation and collection of the updated training parameters at a server, received from a plurality of distributed client sites, can be used to refine, improve, and train the algorithm of the deep neural network for subsequent processing and execution.

Abstract: A method for simulating magnetic resonance signals is proposed. A lattice array where each point in the array has several magnetic resonance sensitive particles is provided. Statistic property of each point is set. A raw magnetic resonance imaging data is calculated based on statistic property of each point and a magnetic resonance imaging sequence to be applied. A system for simulating magnetic resonance signals is further proposed. By considering statistic property of each point, it can distinguish every part of the object to be scanned and really reflect the structure of object without using a real magnetic resonance imaging device. It saves time and costs for avoiding several scanning by the real a magnetic resonance imaging device.

Abstract: In a method and apparatus for correcting the uniformity of a magnetic field, an active shim shell is placed in the magnetic field, a magnetic resonance image of the active shim shell obtained, and the placement position of the active shim shell is determined by analyzing the magnetic resonance image. The value of a shim current in the active shim shell is determined so as to meet the uniformity requirements of the magnetic field according to the placement position. The value of the shim current in said active shim shell is set to the determined value of the shim current.

Abstract: A method is provided for displaying a concatenated file such as an image or text file that has been generated by combining information from other files. The concatenated file includes a file structure that can be shown on a display unit and that contains basic information and reference information. The reference information includes information about links of the concatenated file with source reference files which are also files of the file structure and are used for generating the concatenated file. The source reference files are automatically identified with the assistance of the reference information after the concatenated file has been selected. The source reference files and the concatenated file are subsequently displayed and accentuated at the display unit in an intelligent tree structure.

Abstract: A method for simulating magnetic resonance signals is proposed. A lattice array where each point in the array has several magnetic resonance sensitive particles is provided. Statistic property of each point is set. A raw magnetic resonance imaging data is calculated based on statistic property of each point and a magnetic resonance imaging sequence to be applied. A system for simulating magnetic resonance signals is further proposed. By considering statistic property of each point, it can distinguish every part of the object to be scanned and really reflect the structure of object without using a real magnetic resonance imaging device. It saves time and costs for avoiding several scanning by the real a magnetic resonance imaging device.

Abstract: In a method and apparatus for correcting the uniformity of a magnetic field, an active shim shell is placed in the magnetic field, a magnetic resonance image of the active shim shell obtained, and the placement position of the active shim shell is determined by analyzing the magnetic resonance image. The value of a shim current in the active shim shell is determined so as to meet the uniformity requirements of the magnetic field according to the placement position. The value of the shim current in said active shim shell is set to the determined value of the shim current.

Abstract: The present utility model provides a locating device for a magnetic resonance system comprising an image sensor, an image display for displaying images acquired by the abovementioned image sensor, and a locator, which locator has at least one locating mark. There is no need for the abovementioned locating device for a magnetic resonance system to use a laser for locating and, therefore, the case where an operator is hurt by the laser will not occur. On the other hand, due to the use of the image sensor and the image display, the remote control of adjustment conditions can be accomplished, therefore there is no need to repeatedly enter into a magnetic resonance examination room to carry out operations during the adjustment process, which saves time and costs for the adjustments.

Abstract: In a method and magnetic resonance apparatus for determination of the position of the apparatus scanner, three localizer exposures situated orthogonally to one another are obtained during or after the insertion of the patient into the scanner. Using these three localizer exposures, an image analyzer automatically determines anatomical markers shown in the localizer exposures, which anatomical markers are dependent on the patient position in the scanner terms of their position in the localizer exposure. The spatial position of the patient is automatically determined by the image analyzer using the markers.

Abstract: In a method for representation of flow in a magnetic resonance image, a first magnetic resonance image of an examination subject is acquired, wherein the flow occurring in the examination subject is not compensated in a first spatial direction; a second magnetic resonance image is acquired, wherein the flow occurring in the first spatial direction is compensated, the phase of the magnetization in each of the first and the second magnetic resonance images is calculated of the phase difference between the first phase image and the second phase image, which is calculated is a measure for the flow along the first spatial direction. A third magnetic resonance image is acquired, wherein the flow in a second spatial direction perpendicular to the first spatial direction is not compensated, and a fourth magnetic resonance image is acquired, wherein the flow occurring in the second spatial direction is compensated.

Abstract: In a method and a magnetic resonance apparatus for functional imaging, a number of images without and with a designational stimulation of the examination subject are sequentially registered in successive alternation, and an information value that indicates whether the image was registered during a phase with or without stimulation, at least one image-related stimulation value, and at least one image-related evaluation correlation value are obtained and stored for every image.

Abstract: A nuclear-medical image is generated from a nuclear-medical data set that can be acquired with a dual modality tomography apparatus, which has both a scanner for acquisition of magnetic resonance images and a scanner for acquisition of nuclear-medical data sets within a common acquisition volume. A nuclear-medical data set and an MR image are thereby acquired, and a nuclear-medical magnetic resonance atlas is provided with a reference MR data set of the region of a reference patient to be imaged and a corresponding correction data set. A transformation that maps the reference MR data set to the MR image is generated and applied to the correction data set to generate a transformed correction data set that is registered with the nuclear-medical data set. The corrected nuclear-medical image is subsequently calculated from the transformed correction data set and the nuclear-medical data set.

Abstract: In a method and apparatus for classifying plaque in a blood vessel of a living subject using magnetic resonance imaging, a first magnetic resonance image is produced with a first intensity distribution of a cross-section of a vessel containing plaque, and a contrast agent is injected into the vascular system and a second magnetic resonance image of the cross-section of the vessel is produced with a second intensity distribution after a first time duration following the contrast agent injection, and a third magnetic resonance image of the cross-section of the vessel is produced with a third intensity distribution after a second time duration following the injection of the contrast agent, and the plaque is classified dependent on the respective intensity distributions of the magnetic resonance images.

Abstract: In a method and MR apparatus for the automatic segmentation of flow images acquired by magnetic resonance tomography for the depiction of tissue or organs traversed with fluid such as blood, at least one phase image of a selected region of the subject is acquired by magnetic resonance tomography, and fluid-traversed regions in the at least one phase image are automatically segmented.

Abstract: In a method and magnetic resonance apparatus for determination of the position of the apparatus scanner, three localizer exposures situated orthogonally to one another are obtained during or after the insertion of the patient into the scanner. Using these three localizer exposures, an image analyzer automatically determines anatomical markers shown in the localizer exposures, which anatomical markers are dependent on the patient position in the scanner terms of their position in the localizer exposure. The spatial position of the patient is automatically determined by the image analyzer using the markers.

Abstract: In a method for representation of flow in a magnetic resonance image, a first magnetic resonance image of an examination subject is acquired, wherein the flow occurring in the examination subject is not compensated in a first spatial direction; a second magnetic resonance image is acquired, wherein the flow occurring in the first spatial direction is compensated, the phase of the magnetization in each of the first and the second magnetic resonance images is calculated of the phase difference between the first phase image and the second phase image, which is calculated is a measure for the flow along the first spatial direction. A third magnetic resonance image is acquired, wherein the flow in a second spatial direction perpendicular to the first spatial direction is not compensated, and a fourth magnetic resonance image is acquired, wherein the flow occurring in the second spatial direction is compensated.

Abstract: A distortion-corrected magnetic resonance examination with a magnetic resonance device is undertaken by use of a distortion correction algorithm which uses coefficients to describe a magnetic field used in a measurement. In this case, after an examination area has been selected, by including the distortion correction algorithm and the coefficients for description of the magnetic field, a measurement area needed for correction is determined. Subsequently a magnetic resonance measurement is performed in the measurement area. Subsequently a distortion-corrected magnetic resonance image of the examination area is created with the aid of the distortion correction algorithm. In addition to the examination area, the measurement area and also any necessary measurement times can be shown on a display of the magnetic resonance device. This type of magnetic resonance examination does not include in the magnetic resonance image any picture elements which were not measured because of distortion.

Abstract: In a method for the presentation of images of a region that are generated in a chronological sequence the values allocated to the picture element per image are compared to one another for acquiring changes for at least one picture element that represents an identical location per image with respect to the region to be imaged, the acquired changes are set down in a change log for the picture element, a marking is allocated to the picture element dependent on the change log, and the marking is superimposed on the location of the picture element in a selected image of the region.

Abstract: A method is proposed for image conversion of image data with a first contrast range to image data with a second contrast range. Fourier coefficients of a Fourier transform of the image data are entered in a neural network by calculating parameters for carrying out windowing.

Abstract: A nuclear-medical image is generated from a nuclear-medical data set that can be acquired with a dual modality tomography apparatus, which has both a scanner for acquisition of magnetic resonance images and a scanner for acquisition of nuclear-medical data sets within a common acquisition volume. A nuclear-medical data set and an MR image are thereby acquired, and a nuclear-medical magnetic resonance atlas is provided with a reference MR data set of the region of a reference patient to be imaged and a corresponding correction data set. A transformation that maps the reference MR data set to the MR image is generated and applied to the correction data set to generate a transformed correction data set that is registered with the nuclear-medical data set. The corrected nuclear-medical image is subsequently calculated from the transformed correction data set and the nuclear-medical data set.