Abstract: Systems and techniques for determining degree of motion using machine learning to improve medical image quality are presented. In one example, a system generates, based on a convolutional neural network, motion probability data indicative of a probability distribution of a degree of motion for medical imaging data generated by a medical imaging device. The system also determines motion score data for the medical imaging data based on the motion probability data.

Abstract: Systems and techniques for determining degree of motion using machine learning to improve medical image quality are presented. In one example, a system generates, based on a convolutional neural network, motion probability data indicative of a probability distribution of a degree of motion for medical imaging data generated by a medical imaging device. The system also determines motion score data for the medical imaging data based on the motion probability data.

Abstract: An automated image processing technique is disclosed that evaluates pixels of a phase-difference image to determine those pixels corresponding to inflowing phase data and background phase data. Phase-difference images are generated from a first acquisition and a second acquisition. Non-zero spatially varying background phase from the phase-difference images that are due to eddy currents induced by flow encoding gradients used to generate the phase-difference images is determined. This non-zero spatially varying background phase is removed from the phase-difference images to determine phase associated with flowing spins and phase associated with stationary spins.

Abstract: The present invention includes a technique for use with magnetic resonance imaging that includes the application of a non-linear, higher-order gradient field in the presence of a moving object to be scanned. MR data is acquired as the object moves through the non-linear gradient field. Resulting images are contiguous and do not require the patching together of data in either k-space or image space and result in an image with expanded FOV in a longitudinal direction of the moving object.

Abstract: A workstation is programmed to operate as an application development system for a medical imaging system. Objects programmed in an object-oriented language are selected from a component library using a visual component assembler which enables them to be dragged from a framework area on a display to a workspace area. Properties of selected components may be edited, and the resulting collection of components may be saved as an application program.

Abstract: One or more techniques are provided for measuring the motion of an organ in three dimensions. As provided by the technique, the motion of the organ along each dimension may be determined by a suitable methodology. Where sensor-based motion measurements are suitable, one or more sensors may be placed on a patient to measure internal motion of the organ of interest along one or more perpendicular axes. Where image-based techniques are suitable, the motion of the internal organ along a perpendicular axis may determined using pre-acquisition image data or acquisition image data when suitable. Concurrent motion vectors for all three dimensions may be obtained from the motion data acquired for the perpendicular axes by the disparate methodologies. The concurrent motion vectors may be combined to describe the three-dimensional motion of the organ over time.

Abstract: A method is provided for 3-D MR imaging of structure such as a cardiac region of a subject, which is disposed to cyclical motion. The method comprises detecting commencement of each of a succession of motion cycles, and acquiring a set of MR views of the region during each motion cycle. Each view is at a known radial distance from the center of a two-dimensional k-space, and each set includes a view of lowest spatial frequency, which is located proximate to the k-space center. The order in which the views of respective sets are acquired is selected so that the lowest frequency views of respective sets are each acquired at the same specified time, following commencement of their respectively corresponding motion cycles. The views of each set collectively define a segment in the k-space, respective segments being placed in an interleaved spiral or elliptical arrangement with respect to one another.