Abstract: Two methods are disclosed to remove the image artifacts produced by Maxwell terms arising from the imaging gradients in an echo planar imaging pulse sequence. In the first method, the frequency and phase errors caused by the Maxwell terms are calculated on an individual slice basis and subsequently compensated during data acquisition by dynamically adjusting the receiver frequency and phase. In the second method, two linear phase errors, one in the readout direction and the other in the phase-encoding direction, both of which arise from the Maxwell terms, are calculated on an individual-slice basis. These errors are compensated for in the k-space data after data acquisition.

Abstract: An MRI system performs a scan to acquire NMR signals from which an image can be reconstructed. The NMR signals are compressed prior to digitization by a nonlinear preamplifier, and after digitization, the signal is restored to the correct amplitude by a decompression process. The SNR of the reconstructed image is improved without increasing the size of the analog to digital converter.

Abstract: An MRI scan is conducted in which velocity encoded NMR data is acquired for a slice through the heart. Velocity images and magnitude images are reconstructed at multiple cardiac phases and masks are formed using the magnitude images. The masks are applied to the velocity images to isolate the left ventricle, and rigid body motion is calculated and subtracted from the masked velocity images to indicate deformation of the left ventricle.

Abstract: A multiband RF pulse is produced using the SLR method. A single-band RF pulse is produced and employed to determine correction factors. The SLR polynomial coefficients for each excited band are calculated and corrected, and then combined to form a composite SLR polynomial from which the multiband RF pulse is produced.

Abstract: An MRI system performs a calibration prescan in which NMR projection signals are produced using a calibration pulse sequence. The readout gradient employed during the acquisition of each NMR projection signal is rotated during the signal sampling such that the projection angle changes during the acquisition. The center of the field of view is weighted more heavily in the calculation of the optimal RF transmit power to provide a more accurate setting when using local coils.

Abstract: An MRI system performs a calibration procedure to calculate optimal compensation parameter values for a pre-emphasis filter that alters the shape of gradient waveforms. The gradient field errors are measured using a series of pulse sequences in which different compensation parameter values are sampled. The measured errors are used to calculate a gradient error function and the optimal compensation parameter values are calculated by finding the minimum in the gradient error function.

Abstract: An MRI system measures the movements of a subject during the acquisition of a series of NMR images and automatically updates the scan parameters such that the image plane or volume tracks the movement of the anatomy of interest. An array of tracking coils fasten to the subject and an NMR measurement pulse sequence is interleaved with the image acquisitions to measure the location of the tracking cols.

Abstract: A method is disclosed to remove the image artifacts produced by Maxwell terms arising from the imaging gradients in an echo planar imaging pulse sequence. The frequency and phase errors caused by the Maxwell terms are calculated on an individual slice basis. During the subsequent data acquisition these errors are compensated by dynamically adjusting the receiver frequency and phase.

Abstract: A bandlimited 3D image data set is interpolated before being projected onto a 2D image plane to improve image resolution. The 3D MRA image data set is divided into subvolumes which are separately interpolated and projected onto the 2D image plane. Interpolation is accomplished using a discrete cosine transformation ("DCT"), zero-filling to extend the transformed subvolume and inverse transforming the zero-filled subvolume.

Abstract: Signal fall-off in axial EPI images as well as its variations is corrected by compensating the EPI pulse sequence with gradient pulses that serve to balance the phase dispersion caused by Maxwell terms. Four embodiments are described which employ the slice-selection gradient to compensate the EPI pulse sequence and a fifth embodiment employs the readout gradient.

Abstract: A method is provided for characterizing the spurious, time dependent magnetic field gradient response and for compensating therefor. In accordance with the method, two small sample objects are placed at two positions with respect to the system isocenter where the gradient to be compensated has different values. 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 impulse response of the gradient and are employed to construct compensation filters for both the gradient coils and the polarizing field shim coil.

Abstract: A current regulator for a variable frequency power supply includes a proportional plus integral circuit that produces a composite control output signal for each phase. To improve the performance of the current regulator at higher frequencies, the composite control output signal for each phase also inlcudes a cross coupled component which is produced by multiplying a d.c. signal proportional to frequency times an integrator output signal from another phase. The composite control signals are applied to a voltage source inverter which produces the output currents to a load.

Abstract: A compact generator unit including a cooling system in which the engine and generator are mounted at points outside of the perimeter of the generator fan such that the generator fan serves to cool both the generator and the engine.

Abstract: A laser alignment system includes a transmitter and one or more receivers. The transmitter produces an alignment field by projecting laser energy in a non-planar reference cone, and the receivers locate this reference cone with photodetectors. The laser energy in the alignment field is modulated at 8 kHz and the signals produced by the photodetectors are filtered and amplified to increase the range of the system. Each receiver operates a display which indicates when it is aligned in the reference cone, and it produces an out-of-level indication when the transmitter is not properly aligned.

Abstract: A communications network is comprised of two or more stations which share mastership of the network by passing a token between them. While a station holds the token it may send low or high priority messages. It may also inform other stations of its intent to exit the network and it may solicit for new stations wanting to enter the network. The token is passed from station-to-station in accordance with a predetermined time schedule which insures each station fair access to the network while insuring that messages are conveyed between stations within a predetermined worst case time delay.

Abstract: An NMR zeugmatographic scanner is modified to provide flow images. A motion sensitizing, bi-phasic excitation field gradient is applied during each measurement cycle to produce transverse magnetization in paramagnetic nuclei which are moving in the direction of the gradient. The resulting free induction signal is processed using an inverse Fourier transformation to produce images of flowing paramagnetic nuclei.

Abstract: A laser alignment system includes a remote indicator to display the alignment data when the main receiver is out of sight. The remote indicator receives both operating power and alignment data through a connecting cable to the main receiver. A Liquid Crystal Display (LCD) is used in the remote indicator because of its inherently low power consumption. To make the LCD visible under conditions of reduced ambient lighting, a light source illuminates the LCD from behind. A lighting control circuit turns the light source on only when alignment data is present in order to limit battery drain on the main receiver.

Abstract: An apparatus for rotary manual input utilizes an electromagnetic brake and control circuit to provide tactile feedback to the operator similar to that of a mechanical detent. A shaft connects a handwheel to the brake and to an incremental encoder. The torque generated by the brake when energized is small enough to allow easy rotation of the handwheel. The control circuit keeps the brake normally energized. As the handwheel is turned, the control circuit monitors the encoder position signals and momentarily de-energizes the brake for each incremental movement. The momentary loss of resistance is felt by the operator to verify the incremental movement. The encoder position signals are also made available to an external device, such as for example a Numerical Controller, to implement the commanded movement.

Abstract: The respirometer module consists of an airtight chamber which includes an air pump, a carbon dioxide scrubber, an electronic gas volume transducer, an air vent, and a fill tube. This single module is combined with separate sample chambers of various sizes and applications. The sample chambers are designed for laboratory, on-line, or submersible application to measure respiration of microorganisms, animals, fish, and plants. The respirometer module which provides air circulation, CO.sub.2 absorption, and electronic gas volume sensing can be used interchangeably with any one of these specialized chambers.

Abstract: A programmable controller includes a main processor which executes a user control program. The main processor is interrupted by a support processor which operates as a real time clock. The interval between interrupts is determined by instructions within the user control program and may be dynamically altered when required. The main processor executes a user created interrupt routine when each interrupt occurs.