Surgery – Diagnostic testing – Detecting nuclear – electromagnetic – or ultrasonic radiation
Reexamination Certificate
2000-09-15
2002-04-09
Lateef, Marvin M. (Department: 3737)
Surgery
Diagnostic testing
Detecting nuclear, electromagnetic, or ultrasonic radiation
Reexamination Certificate
active
06368279
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention involves a system and a related method of operation for an ultrasound imaging system, in particular, for compensating for the effects of different time delays of the elements in an ultrasound transducer array.
2. Description of the Related Art
A modern ultrasound imaging system works by transmitting into a region of interest (ROI or “interrogation region”) of a patient's body a beam of ultrasound, which is formed as the composite of ultrasonic transmit signals from an array (one-or two-dimensional) of piezoelectric elements. By varying the relative phases and amplitudes of the transmit signals, the beam can be focused onto different points within the region of interest. The back-scattered signals then cause the array elements to vibrate and to generate electrical signals corresponding to the mechanical vibrations. This is of course well known.
Human tissue is not homogeneous. Indeed, if it were, then ultrasound imaging would be all but ineffective, because there must be some variation in acoustic impedance for ultrasound to be reflected back to the transducer array. One problem that arises because of this inhomogeneity, however, is that it causes differences in propagation times in the region between the array and the point of focus. These differences are unrelated to the structure of interest and reduce the ability of the system to properly align the return signals in time and thus the quality of the image created from the return signals.
A conventional ultrasound system focuses the transmit and receive beams by assuming the medium to be uniform with a known sound speed (typically 1540 m/s) and timing the firing of the different elements and time-gating the return signals in such a way that the assumed round-trip time to and from any given focal point is the same. In reality, since the tissue is not uniform, the propagation time from each element to the focal point deviates from the values calculated based on the assumption of a uniform speed of sound in the medium.
In order to compensate for propagation time errors, several different techniques have been proposed for creating time delay profiles for adaptively focusing the beams. According to one known method, the system determines the peak position of a time-domain cross-correlation of neighboring waveforms; the peak then gives the relative time delay to be applied. The system then constructs a delay profile by adding up all the individual relative time delays. In other words, this prior art system selects a first element then cross-correlates to get an estimate of the time delay to the adjacent, second element, then applies this delay to adjust the second waveform, then cross-correlates the second waveform with the third to get a correction for it, and so on, until a relative time delay is computed through cross-correlation for all the elements. One drawback of this method is that errors accumulate: Any error in the calculated time-delay for the second element (relative to the first) is also included in the cross-correlation of the second and third waveforms, and the increased error in this calculation propagates to the next cross-correlation calculation, and so on.
Yet another method involves cross-correlation of each return waveform with the beam-sum waveform. According to still another known technique, the system determines a reference waveform by adding several neighboring waveforms that have already been aligned using some other method. The time delay for each new waveform is then calculated by determining the peak of the cross-correlation between the new and reference waveforms. Yet another proposal involves forming redundant pairs of cross-correlations and then solving the resulting set of over-constrained equations using a least-mean-squared error method to synthesize a complete delay profile.
According to yet another proposed method, the delay for each waveform is adjusted in turn to maximize the image brightness. In this technique, the delay is adjusted randomly for one element at a time and the adjustment that resulted in maximum image brightness is retained.
One drawback of all these techniques is that they operate on only one waveform at a time, so that information contained in other, neighboring waveforms is ignored. Moreover, because of inhomogeneities, even adjacent waveforms often differ so much in shape (amplitude profile) that the auto-correlation technique may produce erroneous and unstable results. In the presence of a moderate amount of aberration, waveforms received by individual elements may sometimes be severely distorted with amplitude fluctuations, similar to the scintillation associated with the propagation of radio waves in the atmosphere. Similarly, maximizing the beamsum energy based on adjusting delays for individual elements may result in a spiky delay profile. The delay values determined using these techniques are therefore prone to error, and applying these values to correct beamforming may actually worsen the ultimately displayed image.
Newer techniques involving translating the transmit aperture of the transducer or creating common mid-point signals have also been proposed to acquire data for time-delay estimation. Each of these techniques, however, require multiple transmits and are therefore sensitive to motion artifacts. Moreover, as long as individual elements are used to derive the delay profile, the amplitude variability of individual waveforms will still cause problems in the delay estimation. Furthermore, those that depend on comparison with some reference waveform are also sensitive to the choice of the reference: If the reference is severely distorted, for example by significant and highly localized non-linearities, then all calculations based on it will be affected.
What is needed is therefore a way to estimate a time-delay profile that takes into account the variability of the shape (amplitude profile) of individual waveforms. Such a method should avoid the problems associated with a poor choice of some reference, and the problem of error accumulation inherent in those systems that compare neighbors pair-wise without some common reference. The estimated delay profile should align the waveforms as coherently as possible, that is, with maximum beamsum energy, and should also be smooth, that is, the profile should not display abrupt and therefore physically unlikely jumps. This invention provides a system and a related operating method that provides such a time-delay profile.
SUMMARY OF THE INVENTION
The invention provides an ultrasound imaging system and method of operation in which a transducer is activated to transmit into a region of interest of a patient's body a transmit beam of ultrasound from a plurality of elements in the transducer array. An echo signal is then received back from the region of interest, such that each transducer array element generates a respective receive waveform, the waveforms together constituting a receive beam.
An adaptive delay component or module within the processing system then groups the array elements into a predetermined number of groups and, for each group, it selects a control waveform corresponding to a designated control element of the array. For a first one of the groups, the processing system then selects an initial compensation time for the waveforms in the first group; it time-shifts the control waveform by a local time shift amount; it determines a local time compensation profile by interpolation between the local time shift amount and the initial compensation time; and it then calculates a local waveform similarity factor WSF (preferably a function of the r.m.s. value), of the sum of the waveforms in the group time-shifted according to the local time compensation profile. The processing system then time-shifts or “jitters” the control waveform again, interpolates to get time delays for the other waveforms in the group, and evaluates the corresponding WSF of the beam sum. This procedure is repeated, preferably using at first coarse and then fine trial
Lateef Marvin M.
Patel Maulin
Siemens Medical Solutions USA , Inc.
LandOfFree
Time-delay compensation system and method for adaptive... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Time-delay compensation system and method for adaptive..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Time-delay compensation system and method for adaptive... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2918838