Adaptive coding is constrained to midline locations in a spatial listening task.

Many neurons adapt their spike output to accommodate the prevailing sensory environment. Although such adaptation is thought to improve coding of relevant stimulus features, the relationship between adaptation at the neural and behavioral levels remains to be established. Here we describe improved discrimination performance for an auditory spatial cue (interaural time differences, ITDs) following adaptation to stimulus statistics. Physiological recordings in the midbrain of anesthetized guinea pigs and measurement of discrimination performance in humans both demonstrate improved coding of the most prevalent ITDs in a distribution, but with highest accuracy maintained for ITDs corresponding to frontal locations, suggesting the existence of a fovea for auditory space. A biologically plausible model accounting for the physiological data suggests that neural tuning is stabilized by inhibition to maintain high discriminability for frontal locations. The data support the notion that adaptive coding in the midbrain is a key element of behaviorally efficient sound localization in dynamic acoustic environments.

Concomitant magnetic-field-induced artifacts in axial echo planar imaging.

When a linear magnetic field gradient is used, spatially higher-order magnetic fields are produced to satisfy the Maxwell equations. It has been observed that the higher-order magnetic field produced by the readout gradient causes axial echo planar images acquired with a horizontal solenoid magnet to shift along the phase-encoding direction and lose image intensities. Both the shift and intensity reduction become increasingly severe as the slice offset from the isocenter increases. These phenomena are quantitatively analyzed, and good correlation between experiments and theory has been established. The analysis also predicts a previously unreported Nyquist ghost on images with very large slice offsets. This ghost has been verified with computer simulations. Based on the analysis, several methods have been developed to eliminate the image shift, the intensity reduction, and the ghost. Selected methods have been implemented on a commercial scanner and proved effective in removing these image artifacts.

Real-time acquisition, display, and interactive graphic control of NMR cardiac profiles and images.

A highly interactive MRI scanner interface has been developed that allows, for the first time, real-time graphic control of one-dimensional (1D) and two-dimensional (2D) cardiac MRI exams. The system comprises a Mercury array processor (AP) in a Sun SPARCserver with two connections to the MRI scanner, a data link that passes the NMR data directly to the AP as they are collected, and a control link that passes commands from the Sun to the scanner to redirect the imaging pulse sequence in real time. In the 1D techniques, a cylinder or "pencil" of magnetization is repeatedly excited using gradient-echo or spin-echo line-scan sequences, with the magnetization read out each time along the length of the cylinder, and a scrolling display generated on the Sun monitor. Rubber-band lines drawn on the scout image redirect the pencil or imaging slice to different locations, with the changes immediately visible in the display. M-mode imaging, 1D flow imaging, and 2D fast cardiac imaging have been demonstrated on normal volunteers using this system. This platform represents an operator-"friendly" way of directing real-time imaging of the heart.

Real-time interactive magnetic resonance imaging.

We describe a system for performing interactive MRI in real time. Using a TR/TE 7.1/3.5 ms sequence, the operator may alter a scan parameter and observe the effects of the alteration on the image within a few hundred milliseconds. With this system, we can interactively control the oblique scan slice orientation and, using inversion pulses, the image contrast.

Some useful concepts of matched filtering in intravenous digital subtraction angiography.

Using computer calculations and assumed contrast bolus curves, several aspects of the application of temporal integration methods and matched filtering to intravenous digital subtraction angiography (IV-DSA) were studied. The topics included the improvement in signal-to-noise ratio (SNR) of the final image provided by simple integration, a comparison of the SNR performance of matched filtering and extensive integration, the degradation of SNR caused by the motion of noniodinated objects and the sensitivity of SNR to variations in DSA bolus dynamics from patient to patient. Additionally the dependence of matched filter SNR on exposure position and duration was both estimated and demonstrated with clinical DSA images. The results indicate that a substantial improvement in SNR can be obtained with only moderate integration increasing to a two X improvement for longer durations. Integration methods are able to withstand moderate durations (2 seconds) of motion and still provide image quality superior to more conventional DSA results.

Work in progress: the application of temporal filtering techniques to hybrid subtraction in digital subtraction angiography.

Temporal filtering methods were applied to iodine signal-to-noise ratio (SNR) restoration in intravenous hybrid subtraction digital subtraction angiography (DSA). For equal detected exposure rates hybrid subtraction had approximately 35% of the SNR of temporal subtraction. When matched filtering was applied to a DSA run, the filtered result had approximately two times higher SNR than the peak contrast image in the run. Thus, when matched filtering techniques were applied to the hybrid image sequence, the resultant SNR increased to about 70% of that of temporal subtraction. With an additional factor-of-two increase in exposure rate for the hybrid run, SNR parity with temporal subtraction could be achieved. This compared with a factor-of-nine increase in exposure that would be required if no filtering were performed. Experimental hybrid matched filter results, generated with intravenous canine DSA studies, supported the predictions in SNR performance.

The technical characteristics of matched filtering in digital subtraction angiography.

The technical characteristics of a new digital fluorographic image processing method called matched filtering are presented. This technique, a type of extensive temporal integration, takes a weighted sum of images acquired during passage of a contrast bolus through some area of interest. The weight of each image is governed by the magnitude of the contrast bolus in that image. An essential requirement of the matched filter is that its integral be zero. It is shown for equal exposure rates and typical bolus characteristics that matched filtering provides a factor of two higher signal-to-noise ratio (SNR) than conventional methods for bolus transit times of 10 s or higher. Equilvalently, matched filtering can yield images with quality comparable to conventional digital subtraction angiography (DSA) at a factor of four less patient exposure. The SNR obtained with matched filtering is shown to be within 30% of an ideal bound. Comparisons of matched filtering to standard recursive methods and simple integration are made. Experimental canine studies are presented which compare matched filtering with conventional DSA.