Abstract: A surface coil (26) is positioned against a subject in the image region to detect emanating radio frequency resonance signals. The detected signals are processed (52) into a magnetic resonance images representing the image region which include an image value for each of an array of pixel coordinates. There is a large change in magnitude between image values of pixels corresponding to the coil and adjacent pixels. Magnitudes and directions of pixel value change gradients are calculated (60) from the image values corresponding to pixels in each of a first set of magnetic resonance image slices representing axial and transaxial slices through the surface coil. A coordinate calculator (62) calculates image region coordinates corresponding to segments of the surface coil. From the image region coordinates of the surface coil segments, a surface coil sensitivity calculator (92) calculates coil sensitivity which is in turn used to enhance (94) the reconstructed images.

Abstract: During a cardiac cine examination, a multiplicity of imaging sequences, each about 20 msec long, are applied following the R-wave. Each imaging sequence includes a saturation portion in which a bi-modal pre-saturation pulse (38) and a spoiler gradient (56) are applied. The bi-modal RF pulse has a relatively low tip angle, about 50.degree., but is repeated sufficiently often that blood in regions (30a, 30b) parallel to a selected slice (32) are driven toward saturation. Each imaging sequence further includes a gradient echo or other conventional imaging sequence during an imaging portion to generate magnetic resonance data (60). Each imaging sequence is repeated twice for each temporal interval with the same phase encoding, but once with the relative phase of the signal in the slice and the relative phase of the signals from within the pair of regions (30a, 30b) reversed. These two signals are combined such that the signals from within the slice add and the signals from with the pair of regions subtract.

Abstract: A patient's cardiac cycle is monitored (34) for a characteristic point (54) of the cardiac cycle. Following or in response to the characteristic point, a series of field echo sequences (FIG. 3 ) are applied to generate magnetic resonance echoes (50a, 50b, 50c, etc.) at about 10 millisecond intervals (TR=10 ms) following the characteristic point. The echoes are phase encoded (44) such that echoes phase encoded in a lowest frequency or central most segment (I) of k-space are generated at regular intervals. Between temporal consecutive segment (I) echoes, echoes with higher frequency phase encoding from segments (II) and (III) (FIG. 4 ), segments (II-IV) (FIGS. 5A, 5B), etc. are generated. The views are sorted (60) into data sets (62a, 62b, 62c, etc.) such that the central most views are sent to only a single data set and at least some of the higher frequency views are conveyed to two data sets, i.e., the high frequency views are shared or commonly used by two data sets.

Abstract: A positive portion (52) of k-space and a negative portion (56) are both divided into n segments (FIG. 2 ). In each cardiac cycle, a multiplicity of field echoes (106) are generated, which multiplicity of gradient echoes are divided into groups of n contiguous echoes. Within each group, the echoes are all from either the positive portion of k-space or the negative portion of k-space. Preferably, every group of each cardiac cycle has the same views in the same order to generate like, time displaced frames of a cine image sequence. Within each group, the views from the central-most segment n are collected in the middle of the group, views from progressively less central, higher frequency segments are progressively less centrally located within each group with views of the most peripheral, highest frequency segments being collected at the ends of the group. The total number of segments is an integer multiple of four times the number of views per group to facilitate 0.degree.-180.degree. phase cycling.

Abstract: A first image (20) and a second image (22) are taken through the patient's heart region at small time displaced intervals. The first and second images are subtracted (24) to generate a difference image which is indicative of the tissue which has moved during the short time interval, i.e. the boundary of the ventricles. Voxels from regions outside the boundary are adjusted to remove lung tissue (40), and ventricle boundary or edge voxels (48) and analyzed to generate a non-blood tissue histogram (62). Voxels within the boundary are analyzed to generate a blood tissue histogram (60). The histograms are fit (66, 74) to smooth curves which represent the probability distribution or confidence that each voxel value represents blood or non-blood tissue. Contiguous voxels within the boundary are counted (96) and adjusted for voxel size (98) to create an indication of left and right ventricle volume (100l, 100r).