Abstract: A method is provided for characterizing the spurious, time-dependent magnetic field gradient response and for compensating therefor. In accordance with the method a small sample object is placed away from the system isocenter where the gradient to be compensated is non-zero. A magnetic field gradient pulse is applied to the sample, followed by the application after a variable time of a 90.degree. radio frequency pulse. The phase of the resulting free induction decay (FID) signal is monitored as a function of total time since the end of the gradient pulse. These data are related to the gradient impulse response and are fit to a set of (e.g., two or three) exponentials by regression techniques wherein an initial estimate is improved in one embodiment by chi-squared minimization of a Taylor's series expansion about the initial estimate point.

Abstract: A method for reversing residual transverse magnetization due to spatial encoding magnetic field gradient pulses, used in magnetic resonance imaging to encode spatial information, employs a reversing gradient pulse applied in the same direction as the encoding gradient pulse following the observation of the spin-echo signal. The encoding gradient pulse is applied following the 180.degree. RF pulse to avoid the effects of imperfections associated therewith. In one embodiment, the amplitudes of the encoding and reversing gradient pulses are selected to be approximately the negatives of each other so as to substantially cancel the residual magnetization. In another embodiment, the amplitude of the reversing gradient pulse is selected such that the algebraic sum thereof with the corresponding amplitude of the encoding gradient pulse is a constant. In this case, the residual magnetization is not necessarily cancelled, but rather, is left in the same state after each view of the pulse sequence.

Abstract: A method for reducing image artifacts due to signal variations in the course of examining a subject using nuclear magnetic resonance (NMR) techniques includes the acquisition of scan data for imaging the object. The scan data is composed of a number of views. The acquisition of each view includes the implementation of a pulse sequence to generate an NMR signal and application of a magnetic gradient along at least one dimensional axis of the object. The magnetic field gradient is characterized by a parameter (e.g., amplitude or direction) adjustable from view to view to encode spatial information into the NMR signal. The parameter value is selected just prior to implementation of the pulse sequence for that view and depends on the phase of the signal variation at that point in time. The final view order depends on the measured phase during the scan. In this manner, the view order can be continuously adjusted to obtain optimum artifact reduction.

Abstract: The values of pixels in a new image matrix are obtained by mapping pixels of the new matrix in the old matrix and then using polynomial interpolation of a plurality of values of pixels in the old matrix to obtain values of interim pixels in the new matrix and then using polynomial interpolation of a plurality of interim values to obtain final values of pixels in the new matrix. A precalculated table of weighting coefficients based on offset of a new pixel from an adjacent old pixel is used to expedite the polynomial interpolation.

Abstract: Method and apparatus for quickly and efficiently producing ECT (emission computed tomography) images corrected at least in part for attenuation in which a plurality of stored attenuation coefficients are used directly in the projection and back projection operations which produce the ECT image. The evaluation of attentuation factors as needed during the course of projection and back projection avoids the need for precalculation and storage of attenuation factors, thus achieving the reasonably fast reconstruction times needed for efficient diagnostic imaging.

Abstract: A preferred embodiment of a method of removing the effects of nuclear magnetic resonance (NMR) signal baseline error artifacts in imaging applications is disclosed with reference to Fourier transform imaging techniques, of the type frequently referred to as spin warp. In accordance with the method, the phase of the radio frequency (RF) excitation pulse is alternated in adjacent phase-encoding views by 180.degree.. Since the baseline error component is independent of this phase change, the effect of phase reversal of the RF excitation is to modulate the desired image up to the Nyquist frequency in the phase-encoding direction. The artifact due to the baseline error component, however, is unchanged. Views measured with -90.degree. RF excitation are negated prior to reconstruction so that the image is returned to its correct position, while the baseline error artifact is modulated to the edge of the field of view. Since the baseline artifact has limited bandwidth, it does not interfere with the desired image.

Abstract: X-ray energy-difference images are obtained simultaneously by exposing a body to a broad energy spectrum x-ray beam while a radiographic film package is disposed between front and rear x-ray intensifying screens. The film package has two superimposed films with a light-opaque sheet between them. The front screen, on which the polyenergetic image beam emerging from the body is incident, is excited to luminescence by x-ray photons predominantly in one energy band and the rear screen is excited predominantly by photons in another band. The light opaque sheet preferably contains x-ray filter material for additional filtering of photons to increase the fraction of photons at said other band to which the rear screen is sensitive. A shadowgraph of a marker is formed on the film emulsions during x-ray exposure to aid in matching congruent picture elements on each film when the developed films are being read out to obtain signals corresponding to their intensities.

Abstract: A method for correcting image distortion arising from magnetic field gradient nonuniformity in accordance with which the gradient nonuniformity is modeled, the expected distortion in the image calculated, and a geometric correction applied to the distorted image. In the preferred embodiment, polynomial interpolation is used to implement the geometric correction.

Abstract: Method and apparatus for producing CT images in which localized regions ("rub-out" regions) of the original object are not reproduced in order to eliminate artifacts generated by objects within those regions. An operator defines a rub-out region encompassing the object to be removed. The projection set is then modified by creating an average function within the rub-out region which is strongly influenced by the discontinuity created by the object. That information is then utilized to modify the projection set, in effect eliminating the object from the set. The projection set is then used to create a reconstructed image in the normal way.

Abstract: A method is provided for imaging a transverse slice of an object with reduced sensitivity to image artifacts due to object motion during the scan. The method utilizes projection measurements corresponding to at least 180.degree. scan of the slice. The measurements are taken sequentially for views that are adjacent in angle such that measurements which are at the extremes of the scan angle are nevertheless measured at points close in time, reducing inconsistencies therebetween. The method is applicable to a number of modalities utilizing parallel-ray and fan-beam geometries, such as NMR and ultrasound, respectively.

Abstract: A method is provided for reducing streak artifacts in images reconstructed from projections having significant discrepancies between the first and last scan views due to subject motion or to scan geometry aberrations during a typical 360.degree. scan such as that utilized in computerized tomography. The views taken at the beginning and end of the 360.degree. scan are taken far apart in time, but in the image reconstruction process, they are treated as being adjacent. The method recognizes that in a 360.degree. scan each ray in a projection is scanned twice so that the data set contains redundant information. To reduce the inconsistencies, weights less than the nominal weight are assigned to original projections at the beginning and end of the scan, and the views near the middle of the scan containing corresponding redundant data are compensated so that the combined weights of all ray pairs are constant.

Abstract: A CT scanner in which the amount of x-ray information acquired per unit time is substantially increased by using a continuous-on x-ray source and a sampled data system with the detector. An analog filter is used in the sampling system for band limiting the detector signal below the highest frequency of interest, but is a practically realizable filter and is therefore non-ideal. A digital filter is applied to the detector data after digitization to compensate for the characteristics of the analog filter, and to provide an overall filter characteristic more nearly like the ideal.

Abstract: A method and apparatus for compensating CT images for truncated projections created when an object extends beyond the field of view during the course of collecting a set of projections. Projections which have reference channel information influenced by the object are identified, and the reference channel information thereof replaced by more accurate information derived from unaffected projections. The zero and first order moments of the projection set are analyzed to better estimate the expected moments had the truncated projections not been truncated, and to use that information in completing the truncated projections. A closed form solution is used for convolution of the extensions of the truncated projections to avoid substantial delays in processing the final CT image.

Abstract: Film packages for obtaining x-ray energy-difference images simultaneously by exposing a body to a broad energy spectrum x-ray beam. One package embodiment has two radiographic films joined at at least two edges to form a pocket in which a light-opaque sheet is inserted during the x-ray exposure. The package is disposed between front and rear x-ray intensifying screens during the exposure. The front screen on which the broad energy image is incident is excited to luminescence predominantly by x-ray photons in one energy band and the rear screen by photons in another band to thereby produce the difference images. The opaque sheet may contain x-ray filter material. A reflective sheet is substituted in the pocket for said opaque sheet during readout of the intensities of corresponding picture elements. Another film package embodiment is laminated and unitary and has a double-sided reflective sheet in the middle, a light-opaque coating on each reflective surface and a photosensitive emulsion on each coating.

Abstract: A zero crossing detector including timing and logic circuitry for ascertaining and eliminating undesirable waveform characteristics. If a zero crossing indication occurs within a set time period after a trigger event due to a phase inversion or noise, then the indication is ignored. Conversely, if a zero crossing does not occur within a second set time period due to a phase reversal, an artificial crossing will be indicated. By establishing these bounds, the system will function within a specified frequency band.

Abstract: X-ray energy-difference images are obtained simultaneously by exposing a body to a broad energy spectrum X-ray beam while a radiographic film package is disposed between front and rear X-ray intensifying screens. The film package has two superimposed films with a light-opaque sheet between them. The front screen, on which the polyenergetic image beam emerging from the body is incident, is excited to luminescence by X-ray photons predominantly in one energy band and the rear screen is excited predominantly by photons in another band. The light opaque sheet preferably contains X-ray filter material for additional filtering of photons to increase the fraction of photons at said other band to which the rear screen is sensitive. A shadowgraph of a marker is formed on the film emulsions during X-ray exposure to aid in matching congruent picture elements on each film when the developed films are being read out to obtain signals corresponding to their intensities.

Abstract: A matched filtering method for X-ray image subtraction procedures in which an X-ray contrast medium is injected intravenously. A sequence of pre-contrast X-ray images are made during a period before the medium arrives in a blood vessel that is to be imaged and the sequence is continued through the post-contrast period and in some cases, an after-contrast period. A curve or plot of how projected intensity of the contrast medium varies as a function, h, at times (t) is determined at some prior time based on characteristic patients or, in the alternative, by using the post-contrast image data obtained from the patient presently undergoing examination. In either case a sequence of coefficients or weighting factors are produced that are proportioned to the value of the function h at the time (t) at which the corresponding image in the post-contrast sequence was acquired.

Abstract: A time gain control (TGC) signal for a variable gain amplifier in an ultrasonic scanner is derived by determining the attenuation at many levels in the object under examination and then constructing the TGC signal based on the cumulative attenuation of an ultrasonic signal through the object. A zero crossing detector such as a Schmitt triggered monostable multivibrator receives a reflected ultrasonic wave signal and generates pulses which are counted during time intervals corresponding to a depth level in the object. By comparing the zero crossing density at one level to the zero crossing density at another level a measure of attenuation between the two levels is obtained.

Abstract: X-ray pulses at two different photon energy levels are produced with an x-ray tube that has an adjustable resistor element in a circuit that connects the anode and filament of the tube to a high voltage supply. Two bias voltage supplies are connected in additive fashion between a control grid element of the tube and its filament to enable control of voltage drop and current through the tube. Switch means are provided to selectively shunt either or both bias voltage supplies in a repeatable order to thereby alter the bias voltage on the control grid and enable producing x-ray pulses having energies corresponding with the voltage drop and current through the tube at a selected bias voltage in accordance with the load line on the anode characteristic curve plot of the tube as determined by the value of the adjustable resistor element.