Patent Grant
8390285
1. Field of the Invention
The present invention relates to magnetic resonance spectroscopic imaging technology, and more particularly to a side-band suppression method and a side-band suppression device.
2. Description of the Prior Art
As medical imaging develops, a new noninvasive examination method emerges in the magnetic resonance spectroscopic imaging field. In normal tissues, metabolites can be present in normal concentration. When pathological changes occur in the tissue, the concentration of the metabolites will change. Magnetic resonance spectroscopic imaging can obtain the metabolite concentration by observing the spectrum of the metabolites and further analyze the metabolite concentration to achieve the goal of disease diagnosis.
However, if there is imperfection in the design of the gradient system, the gradient switching can lead to mechanical vibration of the magnet, causing the main magnet field (B0 field) to oscillate over time, which leads to frequency modulation of the spectrum signal. In the spectrum, it appears as weak signal peaks, or side bands, in both sides of the normal signal peak. As the side-band intensity is proportional to the intensity of the modulated signal, the side bands will have a high intensity when the modulated signal has a high intensity and the high intensity side bands overlap and mix with the metabolite spectrum signal, causing difficulty in observation of the metabolite spectrum signal.
In order to suppress the side bands, the positive and negative gradient scanning method are commonly used in the prior art. Supposing the number of scans is N, wherein N is an even number, the gradient polarity used for the first N/2 scans is opposite to that used for the subsequent N/2 scans. For example, a positive gradient is used for the first N/2 scans and a negative gradient is used for the subsequent N/2 scans. Then the signals obtained from each scan are added up and the average value of the sum is calculated. The signal obtained by this scanning method is regarded as a signal without side bands. The theoretical foundation of this method is that the phase of the side-band signal is correlated to the polarity of the gradients and that changing the polarity of the gradients can change the phase of the side-band signal by 180 degrees, without affecting the phase of the metabolite signal. Two side-band signals with phase difference of 180 degrees can offset each other by addition. Theoretically the method can eliminate the side bands.
However, in real application, due to the existence of eddy current, the phase difference between the side-band signals in positive and negative gradient scanning is not exactly 180 degrees. Therefore the two side-band signals cannot completely offset each other. What is worse, the eddy current can also cause the initial phase deviation of the positive and negative gradient scanning signals, which leads to reduced signal-to-noise ratio (SNR) of the added up spectrum signal. For these reasons, the method provided by the prior art cannot effectively suppress the side bands.
In view of this, the present invention is intended to provide a side-band suppression method which can effectively suppress side bands.
Another objective of the present invention is to provide a side-band suppression device which can effectively suppress side bands.
Therefore, the present invention provides a side-band suppression method, including: determining the positive eddy current correction factor and negative eddy current correction factor, scanning N/2 times by a positive gradient and collecting the positive gradient scanning signal during each scan, scanning N/2 times by a negative gradient, and collecting the negative gradient scanning signal during each scan, wherein N is an even number, performing an eddy current correction of the N/2 positive gradient scanning signals collected according to said positive eddy current correction factor, and performing an eddy current correction of the N/2 negative gradient scanning signals collected according to the negative eddy current correction factor.
The method further includes calculating the side-band suppressed spectrum signal according to the N/2 positive gradient scanning signals that have undergone the eddy current correction and the N/2 negative gradient scanning signals that have undergone the eddy current correction.
Preferably, the determination of the positive eddy current correction factor includes: positive reference scanning P times by a positive gradient and collecting the positive reference signal during the Pth positive reference scan, wherein P is a positive integer greater than 1, and then filtering said positive reference signal and taking the filtered signal phase as the positive eddy current correction factor.
According to an embodiment, the filtering includes: acquiring the sampling interval of the positive reference scan to acquire the mechanical vibration frequency of the magnet; calculating the product of the positive reference scanning sampling interval and the mechanical vibration frequency of the magnet, and then calculating the reciprocal of the product and taking the result as the length of the sliding window, designing the mean sliding filter according to the length of the sliding window; and performing mean sliding filtering of the positive reference signal with said mean sliding filter.
Preferably, the acquisition of the mechanical vibration frequency of the magnet includes: taking a spectrum measurement of the water modeling using a Point-Resolved Spectroscopy (PRESS) sequence, to get the spectrum of the water modeling signal, calculating the absolute value of the frequency difference between the water peak and any nearest side band in said spectrum, and taking the absolute value as the mechanical vibration frequency of the magnet.
According to another embodiment, the filtering includes: calculating the positive filtering bandwidth, designing a Gaussian band-pass filter according to said positive filtering bandwidth, and band-pass filtering the positive reference signal with the Gaussian band-pass filter.
Preferably, the calculation of the positive filtering bandwidth includes: acquiring the spectrum of the positive reference scan; if there is no overlapping part between the water signal and the metabolite signal in the positive reference scanning spectrum, the waveband of the water signal is acquired and taken as the positive filtering bandwidth; if there is an overlapping part between the water signal and the metabolite signal in the positive reference scanning spectrum, the overlapping waveband between the water signal and the metabolite signal is first divided by 2 to acquire a quotient, and then the difference between the waveband of the water signal and the quotient is calculated and the result is taken as the positive filtering bandwidth.
Preferably, the determination of the negative eddy current correction factor includes: negative reference scanning P times by a negative gradient and collecting the negative reference signal during the Pth negative reference scan, wherein P is a positive integer greater than 1, filtering the negative reference signal and taking the filtered signal phase as the negative eddy current correction factor.
In an embodiment, the filtering includes: acquiring the sampling interval of the negative reference scan to acquire the mechanical vibration frequency of the magnet, calculating the product of said negative reference scanning sampling interval and the mechanical vibration frequency of the magnet, and then calculating the reciprocal of the product and taking the result as the length of the sliding window, designing a mean sliding filter according to the length of the sliding window, and mean sliding filtering the negative reference signal with said mean sliding filter.
Preferably, the acquisition of the mechanical vibration frequency of the magnet includes: taking a spectrum measurement of the water modeling using a Point-Resolved Spectroscopy (PRESS) sequence, to get the spectrum of the water modeling signal, calculating the absolute value of the frequency difference between the water peak and any nearest side band in the spectrum, and taking that absolute value as the mechanical vibration frequency of the magnet.
In another embodiment, the filtering includes: calculating the negative filtering bandwidth, and designing a Gaussian band-pass filter according to the negative filtering bandwidth; band-pass filtering of the negative reference signal with the Gaussian band-pass filter.
Preferably, the calculation of the negative filtering bandwidth includes: acquiring the spectrum of the negative reference scan; if there is no overlapping part between the water signal and the metabolite signal in the negative reference scanning spectrum, the waveband of the water signal is acquired and taken as the negative filtering bandwidth, if there is an overlapping part between the water signal and the metabolite signal in the negative reference scanning spectrum, the overlapping waveband between the water signal and the metabolite signal is first divided by 2 to acquire a quotient, and then the difference between the waveband of the water signal and said quotient is calculated, and the result is taken as the negative filtering bandwidth.
Preferably, the eddy current correction of the positive gradient scanning signal includes: calculating the difference between the phase of said positive gradient scanning signal and said positive eddy current correction factor. The eddy current correction of the negative gradient scanning signal includes: calculating the difference between the phase of the negative gradient scanning signal and the negative eddy current correction factor.
Preferably, the calculation of the side-band suppressed spectrum signal includes: adding the N/2 positive gradient scanning signals that have undergone the eddy current correction and the N/2 negative gradient scanning signals that have undergone the eddy current correction, and then calculating the average value and taking the result as the side-band suppressed spectrum signal.
The present invention also provides a side-band suppression device, which comprises: a positive eddy current correction factor determining module, a negative eddy current correction factor determining module, a positive gradient scanning signal collecting module, a negative gradient scanning signal collecting module, a positive eddy current correcting module, a negative eddy current correcting module, and a side-band suppressing module.
The positive eddy current correction factor determining module is used for determining the positive eddy current correction factor.
The negative eddy current correction factor determining module is used for determining the negative eddy current correction factor,
The positive gradient scanning signal collecting module is used to scan N/2 times by a positive gradient, and to collect the positive gradient scanning signal during each scan, wherein N is an even number.
The negative gradient scanning signal collecting module is used to scan N/2 times by a negative gradient, and to collect the negative gradient scanning signal during each scan.
The positive eddy current correcting module is used for eddy current correcting of the N/2 positive gradient scanning signals collected according to the positive eddy current correction factor.
The negative eddy current correcting module is used for eddy current correcting of the N/2 negative gradient scanning signals collected according to the negative eddy current correction factor.
The side-band suppressing module is used for calculating the side-band suppressed spectrum signal according to the N/2 positive gradient scanning signals that have undergone the eddy current correction and the N/2 negative gradient scanning signals that have undergone the eddy current correction.
The positive eddy current correction factor determining module includes: a positive reference scanning sub-module and a positive filtering sub-module, the positive reference scanning sub-module being used to perform positive reference scanning P times by a positive gradient, and to collect the positive reference signal during the Pth positive reference scan, wherein P is a positive integer greater than 1. The positive filtering sub-module is used for filtering the positive reference signal and taking the filtered signal phase as the positive eddy current correction factor.
The positive filtering sub-module includes: a positive sampling interval acquiring unit, a mechanical vibration frequency acquiring unit, a positive sliding window calculating unit, a mean filter designing unit, and a positive mean sliding filtering unit. The positive sampling interval acquiring unit is used for acquiring the sampling interval of the positive reference scan. The mechanical vibration frequency acquiring unit is used for acquiring the mechanical vibration frequency of the magnet. The positive sliding window calculating unit is used for calculating the product of said sampling interval of the positive reference scan and said mechanical vibration frequency of the magnet, and then calculating the reciprocal of the product and taking the result as the length of the sliding window; said mean sliding filter designing unit is used for designing a mean sliding filter according to the length of the sliding window. The said positive mean sliding filtering unit is used for mean sliding filtering of said positive reference signal with the mean sliding filter.
The mechanical vibration frequency acquiring unit includes: a spectrum measuring sub-unit and a calculating sub-unit. The spectrum measuring sub-unit is used for taking a spectrum measurement of the water modeling using a Point-Resolved Spectroscopy (PRESS) sequence, to get the spectrum of the water modeling signal. The calculating unit is used for calculating the absolute value of the frequency difference between the water peak and any nearest side band in said spectrum, and taking that absolute value as said mechanical vibration frequency of the magnet.
The positive filtering sub-module includes: a positive filtering bandwidth calculating unit, a positive band-pass filter designing unit, and a positive band-pass filtering unit. The positive filtering bandwidth calculating unit is used for calculating the positive filtering bandwidth. The positive band-pass filter designing unit is used for designing a Gaussian band-pass filter according to the positive filtering bandwidth. The positive band-pass filtering unit is used for band-pass filtering of the positive reference signal with said Gaussian band-pass filter.
The positive filtering bandwidth calculating unit includes: a positive spectrum acquiring sub-unit and a positive calculating sub-unit. The positive spectrum acquiring sub-unit is used for acquiring the spectrum of the positive reference scan. The positive calculating sub-unit is used for judging whether there is an overlapping part between the water signal and the metabolite signal in the positive reference scanning spectrum, and if there is an overlapping part, the waveband of the water signal is acquired and taken as the positive filtering bandwidth. Otherwise, the overlapping waveband between the water signal and the metabolite signal is divided by 2 to acquire a quotient, and the difference between the waveband of the water signal and the quotient is calculated, and the result is taken as the positive filtering bandwidth.
The negative eddy current correction factor determining module includes: a negative reference scanning sub-module and a negative filtering sub-module. The negative reference scanning sub-module is used to perform negative reference scanning P times by a negative gradient, and to collect the negative reference signal during the Pth negative reference scan, wherein P is a positive integer greater than 1. The negative filtering sub-module is used for filtering the negative reference signal and taking the filtered signal phase as the negative eddy current correction factor.
The negative filtering sub-module includes: a negative sampling interval acquiring unit, a mechanical vibration frequency acquiring unit, a negative sliding window calculating unit, a mean filter designing unit, and a negative mean sliding filtering unit. The negative sampling interval acquiring unit is used for acquiring the sampling interval of the negative reference scan. The mechanical vibration frequency acquiring unit is used for acquiring the mechanical vibration frequency of the magnet. The negative sliding window calculating unit is used for calculating the product of said sampling interval of the negative reference scan and the mechanical vibration frequency of the magnet, and then calculating the reciprocal of the product and taking the result as the length of the sliding window. The mean sliding filter designing unit is used for designing a mean sliding filter according to said length of the sliding window. The negative mean sliding filtering unit is used for mean sliding filtering of the negative reference signal with the mean sliding filter.
The mechanical vibration frequency acquiring unit includes: a spectrum measuring sub-unit and a calculating unit. The spectrum measuring sub-unit is used for taking a spectrum measurement of the water modeling using a PRESS sequence to get the spectrum of the water modeling signal. The calculating unit is used for calculating the absolute value of the frequency difference between the water peak and any nearest side band in the spectrum, and taking that absolute value as the mechanical vibration frequency of the magnet.
The negative filtering sub-module includes: a negative filtering bandwidth calculating unit, a negative band-pass filter designing unit, and a negative band-pass filtering unit. The negative filtering bandwidth calculating unit is used for calculating the negative filtering bandwidth. The negative band-pass filter designing unit is used for designing a Gaussian band-pass filter according to the negative filtering bandwidth. The said negative band-pass filtering unit is used for band-pass filtering of the negative reference signal with the Gaussian band-pass filter.
The negative filtering bandwidth calculating unit includes: a negative spectrum acquiring sub-unit and a negative calculating sub-unit. The, negative spectrum acquiring sub-unit is used for acquiring the spectrum of the negative reference scan. The negative calculating sub-unit is used for judging whether there is an overlapping part between the water signal and the metabolite signal in the negative reference scanning spectrum, and if there is an overlapping part, the waveband of the water signal is acquired and taken as the negative filtering bandwidth. Otherwise, the overlapping waveband between the water signal and the metabolite signal is divided by 2 to acquire a quotient, and the difference between the waveband of the water signal and the quotient is calculated, and the result is taken as the negative filtering bandwidth.
The positive eddy current correction module includes: a positive receiving sub-module and a positive subtracting sub-module. The positive receiving sub-module is used for receiving the positive gradient scanning signal and said positive eddy current correction factor. The positive subtracting sub-module is used for calculating the difference between the phase of the positive gradient scanning signal and the positive eddy current correction factor.
The negative eddy current correction module includes: a negative receiving sub-module and a negative subtracting sub-module. The negative receiving sub-module is used for receiving the negative gradient scanning signal and the negative eddy current correction factor. The negative subtracting sub-module is used for calculating the difference between the phase of the positive gradient scanning signal and the positive eddy current correction factor.
The side-band suppression module includes: an adding sub-module and an averaging sub-module. The adding sub-module is used for adding the N/2 positive gradient scanning signals that have undergone the eddy current correction and the N/2 negative gradient scanning signals that have undergone the eddy current correction, to acquire a sum. The averaging sub-module is used for calculating the average value of the sum, and taking the result as the side-band suppressed spectrum signal.
It is shown in the above technical scheme that the side-band suppression method and device provided by the present invention first determines the positive eddy current correction factor and negative eddy current correction factor, and then performs eddy current correction of the N/2 positive gradient scanning signals collected according to the positive eddy current correction factor, and performs eddy current correction of the N/2 negative gradient scanning signals collected according to the negative eddy current correction factor, and thus the impact of eddy current on the positive and negative gradient scanning signals can be eliminated. This can ensure that the positive and negative scanning signals will have consistent initial phases and that the phase difference of side-band signals will be 180 degrees in both positive and negative gradient scanning. Finally the side-band suppressed spectrum signal is calculated according to the eddy current corrected N/2 positive gradient scanning signals and eddy current corrected N/2 negative gradient scanning signals. In other words, the side-band signals can be completely canceled out by calculation, which means side bands are effectively suppressed.
a is a schematic diagram showing how positive filtering bandwidth is calculated when the water signal and metabolite signal do not overlap.
b is a schematic diagram showing how positive filtering bandwidth is calculated when the water signal and metabolite signal overlap.
step 101, determining the positive eddy current correction factor and negative eddy current correction factor.
Step 1011, positive reference scanning P times by a positive gradient and collecting the positive reference signal during the Pth positive reference scan, wherein P is a positive integer greater than 1 and may vary depending on the actual needs.
It should be noted that in the present invention, the positive means adopting a positive gradient, and the negative means adopting a negative gradient.
The positive reference scanning is a scanning method satisfying the following conditions: first, there is no water peak suppression; second, the excitation frequency of the system equals the resonance frequency of water protons. The specific scanning method is a prior art and will not be further described herein.
Step 1012, filtering the positive reference signal and taking the filtered signal phase as the positive eddy current correction factor.
The filtering step can be performed at the time domain of the reference signal or the frequency domain. Filtering methods in the different domains will be introduced below.
First, the time-domain filtering method will be described.
During the positive scanning, the positive reference signal phase is composed of two parts, which can be expressed by φref+(t)=φN+(t)+φP+(t), wherein φref+(t) is the positive reference signal phase, φN+(t) is the phase of the low-frequency a periodic signal during the positive reference scanning, and φP+(t) is the phase of the high-frequency periodic signal during the positive reference scanning, wherein φN+(t)=φ0++φΔf+(t)+φEC+(t), wherein φΔf+(t) is the signal phase caused by the inhomogeneity of the magnetic field in the voxel during the positive reference scanning, φEC+(t) is the signal phase caused by the eddy current during the positive reference scanning, and φ0+ is the positive fixed compensated phase which does not change over time; and φP+(t)=φV+(t)+φm+(t), wherein φV+(t) is the signal phase caused by the oscillation of the main magnet field during the positive reference scanning, and φm+(t) is the phase of the metabolite signal during the positive reference scanning.
In this step, the purpose of filtering is to filter φP+(t) from φref+(t) so that the filtered signal phase contains only φN+(t). The filtering method adopted is the mean sliding filtering method. The following will describe the mean sliding filtering method used in this step.
First, the sliding window length is calculated by l=1/(fv*ΔT), wherein fv is the mechanical vibration frequency of the magnet and ΔT is the sampling interval of the positive reference scanning. In the present invention, the sampling intervals of the positive reference scanning and negative reference scanning are the same, and ΔT can be known before the positive reference scanning. fv is obtained by the following method: the spectrum measurement of the water modeling can use a common Point-Resolved Spectroscopy (PRESS) sequence as the scanning sequence to get the spectrum of the water modeling signal. The absolute value of the frequency difference between the water peak and any nearest side band in the spectrum is fv (taking into consideration the first-order side bands only and ignoring the second or higher order side bands). It should be noted that in the spectrum there are two nearest side bands on both sides of the water peak and that the distance between each side band and the water peak is equal. Therefore, fv can be obtained by selecting any nearest side band and calculating the absolute value of the frequency difference between the water peak and the selected side band. Then, a mean sliding filter is designed according to the sliding window length l, and the designed mean filter is used for filtering φP+(t) from φref+(t). In this case, the design of the mean sliding filter according to the sliding window length and the mean sliding filter filtering method used are part of the prior art and are not further described herein.
Finally, the filtered signal phase is taken as the positive eddy current correction factor, that is, let the positive eddy current correction factor be q+=φN+(t).
As q+=φN+(t) and φN+(t)=φ0++φΔf+(t)+φEC+(t), the positive eddy current correction factor in the present invention can not only correct the signal phase caused by the eddy current in the positive reference scanning, but can also correct the signal phase and the positive fixed compensated phase caused by inhomogeneity of the magnetic field in the voxel during the positive reference scanning, wherein the correction of the signal phase caused by inhomogeneity of the magnetic field in the voxel during the positive reference scanning can reduce the full width at half maximum (FWHM) of the spectral line, thereby improving the resolution and signal-to-noise ratio (SNR) of the spectrum signals. Furthermore, the correction of the positive fixed compensated phase is necessary. After the signal demodulation, water protons should be in the resonance state and their signal phase should also be zero. If the positive fixed compensated phase exists, the real part signals of water protons in the frequency domain are not of pure absorption linear shape and will have distortion. This also causes difficulty in the observation and evaluation of the spectrum.
Next, the frequency domain filtering method is described.
During the positive reference scanning, the positive reference signal phase is composed of two parts:
Sref+(t)=AW+(t)exp(j*ω0t+j*φe+(t))+Am+(t)exp(j*ωmt+j*φe+(t))
In this case, Sref+(t) is the positive reference signal, AW+(t) is the water signal amplitude during the positive reference scanning, ω0 is the resonance frequency of water protons, φe+(t) is the phase signal caused by the eddy current during the positive reference scanning, Am+(t) is the metabolite signal amplitude during the positive reference scanning, and ωm is the resonance frequency of the metabolites.
Further, it should be noted that AW+(t)exp(j*ω0t+j*φe+(t)) is the water signal SW+(t) during the positive reference scanning, and Am+(t)exp(j*ωm+t+j*φe+(t)) is the metabolite signal Sm+(t) during the positive reference scanning. In this step, the purpose of filtering is to remove Sm+(t) so that the filtered signal phase contains only SW+(t). The details of the filtering method are: first, calculate the positive filtering bandwidth BW+ according to the spectrum of the positive reference scanning, wherein the acquisition of the spectrum of the positive reference scanning is prior art. There are two scenarios of BW+ calculation.
If the excitation frequency of the system equals the resonance frequency of water protons, i.e. ω0=0, then S+W(t)=A+W(t)exp(j*φ+e(t)), from which it can be known that
wherein the method for acquiring the real part and imaginary part of S+W(t) is prior art which will not be further described herein.
Finally, let the positive eddy current correction factor be q+=ω+e(t), that is, the filtered signal phase is taken as the positive eddy current correction factor. This concludes the description of the method for determining the positive eddy current correction factor and next the method for determining the negative eddy current correction factor will be described.
Step 1013, negative reference scanning P times by a negative gradient and collecting the negative reference signal during the Pth negative reference scan, wherein P is a positive integer greater than 1.
Step 1014, filtering the negative reference signal and taking the filtered signal phase as the negative eddy current correction factor.
The technical features of step 1013 correspond to those of step 1011, and the technical features of step 1014 correspond to those of step 1012, thus for the description of step 1013 and step 1014 reference can be made to the corresponding description of step 1011 and step 1012.
In step 101, the process of determining the positive eddy current correction factor and that of determining the negative eddy current correction factor can be carried out simultaneously (for example, steps 1011 and 1012 and steps 1013 and 1014 can be performed alternately) or in tandem, and the present invention is not limited to this.
This concludes the description of step 101.
Step 102 involves scanning N/2 times by a positive gradient, and collecting the positive gradient scanning signal during each scan; scanning N/2 times by a negative gradient, and collecting the negative gradient scanning signal during each scan, wherein, N is an even number.
This step belongs to the prior art which will not be further described herein.
Step 103 involves eddy current correcting of the N/2 positive gradient scanning signals according to the positive eddy current correction factor; eddy current correcting of the N/2 negative gradient scanning signals according to the negative eddy current correction factor.
The method for eddy current correction of the positive gradient scanning signal is: subtracting the positive eddy current correction factor from the phase of the positive gradient scanning signal and taking the difference value obtained as the phase of the positive gradient scanning signal which has undergone the eddy current correction. The amplitude of the positive gradient scanning signal is not changed, i.e. Scorr+(t)=S+(t)*exp(−i*q+) wherein S+(t) is the positive gradient scanning signal and (t) is the positive gradient scanning signal which has undergone the eddy current correction.
The method for eddy current correction of the negative gradient scanning signal is: subtracting the negative eddy current correction factor from the phase of the negative gradient scanning signal and taking the difference value obtained as the phase of negative gradient scanning signal which has undergone the eddy current correction. The amplitude of the negative gradient scanning signal is not changed, i.e. Scorr−(t)=S−(t)*exp(−i*q−) wherein S−(t) is the negative gradient scanning signal and Scorrt−[(t) is the negative gradient scanning signal which has undergone the eddy current correction.
In the method of this embodiment, the step of eddy current correction of the N/2 positive gradient scanning signals can also be performed after step 102, collecting the positive gradient scanning signals. The present invention is not limited to the order of steps stated above.
Step 104 involves calculating the side-band suppressed spectrum signal according to the N/2 positive gradient scanning signals that have undergone the eddy current correction and the N/2 negative gradient scanning signals that have undergone the eddy current correction.
The method for calculating the side-band suppressed spectrum signal is: adding up the N/2 positive gradient scanning signals that have undergone the eddy current correction and the N/2 negative gradient scanning signals that have undergone the eddy current correction, and then calculating the average value and taking the result as the side-band suppressed spectrum signal.
Based on the above side-band suppression method,
In this case, the positive eddy current correction factor determining module 11 is used for determining the positive eddy current correction factor; the negative eddy current correction factor determining module 21 is used for determining the negative eddy current correction factor; the positive gradient scanning signal collecting module 12 is used to scan N/2 times by a positive gradient, and to collect the positive gradient scanning signal during each scan, wherein N is an even number; the negative gradient scanning signal collecting module 22 is used to scan N/2 times by a negative gradient, and to collect the negative gradient scanning signal during each scan; the positive eddy current correcting module 13 is used for eddy current correcting of the N/2 positive gradient scanning signals collected according to the positive eddy current correction factor; the negative eddy current correcting module 23 is used for eddy current correcting of the N/2 negative gradient scanning signals collected according to the negative eddy current correction factor; and the side-band suppressing module 30 is used for calculating the side-band suppressed spectrum signal according to the N/2 positive gradient scanning signals that have undergone the eddy current correction and the N/2 negative gradient scanning signals that have undergone the eddy current correction.
The positive eddy current correction factor determining module 11 includes: a positive reference scanning sub-module 111 and a positive filtering sub-module 112.
In this case, the positive reference scanning sub-module 111 is used to perform positive reference scanning P times by a positive gradient, and to collect the positive reference signal during the Pth positive reference scan, wherein P is a positive integer greater than 1; and, the positive filtering sub-module 112 is used for filtering the positive reference signal and taking the filtered signal phase as the positive eddy current correction factor.
The positive filtering sub-module 112 includes: a positive sampling interval acquiring unit 1121, a mechanical vibration frequency acquiring unit 1122, a positive sliding window calculating unit 1123, a mean filter designing unit 1124, and a positive mean sliding filtering unit 1125.
In this case, the positive sampling interval acquiring unit 1121 is used for acquiring the sampling interval of the positive reference scan; the mechanical vibration frequency acquiring unit 1122 is used for acquiring the mechanical vibration frequency of the magnet; the positive sliding window calculating unit 1123 is used for calculating the product of the sampling interval of the positive reference scanning and the mechanical vibration frequency of the magnet, and then calculating the reciprocal of the product and taking the result as the length of the sliding window; the mean sliding filter designing unit 1124 is used for designing a mean sliding filter according to the length of the sliding window; and the positive mean sliding filtering unit 1125 is used for mean sliding filtering of the positive reference signal with the designed mean sliding filter.
The mechanical vibration frequency acquiring unit 1122 includes: a spectrum measuring sub-unit 11221 and a calculating sub-unit 11222.
In this case, the spectrum measuring sub-unit 11221 is used for taking a spectrum measurement of the water modeling using a PRESS sequence, to get the spectrum of the water modeling signal; and the calculating unit 11222 is used for calculating the absolute value of the frequency difference between the water peak and any nearest side band in the spectrum, and taking that absolute value as the mechanical vibration frequency of the magnet.
The negative eddy current correction factor determining module 21 includes: a negative reference scanning sub-module 211 and a negative filtering sub-module 212.
In this case, the negative reference scanning sub-module 211 is used to perform the negative reference scanning P times by a negative gradient, and to collect the negative reference signal during the Pth negative reference scan, wherein P is a positive integer greater than 1; and, the negative filtering sub-module 212 is used for filtering the negative reference signal and taking the filtered signal phase as the negative eddy current correction factor.
The negative filtering sub-module 212 includes: a negative sampling interval acquiring unit 2121, a mechanical vibration frequency acquiring unit 1122, a negative sliding window calculating unit 2123, a mean sliding filter designing unit 1124, and a negative mean sliding filtering unit 2125.
In this case, the negative sampling interval acquiring unit 2121 is used for acquiring the sampling interval of the negative reference scan; the mechanical vibration frequency acquiring unit 1122 is used for acquiring the mechanical vibration frequency of the magnet; the negative sliding window calculating unit 2123 is used for calculating the product of the sampling interval of the negative reference scanning and the mechanical vibration frequency of the magnet, and then calculating the reciprocal of the product and taking the result as the length of the sliding window; the mean sliding filter designing unit 1124 is used for designing a mean sliding filter according to the length of the sliding window; and the negative mean sliding filtering unit 2125 is used for mean sliding filtering of the negative reference signal with the designed mean sliding filter.
The mechanical vibration frequency acquiring unit 1122 includes: a spectrum measuring sub-unit 11221 and a calculating unit 1222.
In this case, the spectrum measuring sub-unit 11221 is used for taking a spectrum measurement of the water modeling with a PRESS sequence, to get the spectrum of the water modeling signal; and the calculating unit 11222 is used for calculating the absolute value of the frequency difference between the water peak and any nearest side band in the spectrum, and taking that absolute value as the mechanical vibration frequency of the magnet.
The positive eddy current correcting module 13 includes: a positive receiving sub-module 131 and a positive subtracting sub-module 132.
In this case, the positive receiving sub-module 131 is used for receiving the positive gradient scanning signal and the positive eddy current correction factor; and the positive subtracting sub-module 132 is used for calculating the difference between the phase of the positive gradient scanning signal and the positive eddy current correction factor.
The negative eddy current correction module 23 includes: a negative receiving sub-module 231 and a negative subtracting sub-module 232.
In this case, the negative receiving sub-module 231 is used for receiving the negative gradient scanning signal and the negative eddy current correction factor; and the negative subtracting sub-module 232 is used for calculating the difference between the phase of the positive gradient scanning signal and the positive eddy current correction factor.
The side-band suppression module 30 includes: an adding sub-module 301 and an averaging sub-module 302.
In this case, the adding sub-module 301 is used for adding up the N/2 positive gradient scanning signals that have undergone the eddy current correction and the N/2 negative gradient scanning signals that have undergone the eddy current correction, to acquire the sum; and the averaging sub-module 302 is used for calculating the average value of the sum, and taking the result as the side-band suppressed spectrum signal.
In this case, the positive eddy current correction factor determining module 11 is used for determining the positive eddy current correction factor; the negative eddy current correction factor determining module 21 is used for determining the negative eddy current correction factor; the positive gradient scanning signal collecting module 12 is used to scan N/2 times by a positive gradient, and to collect the positive gradient scanning signal during each scan, wherein N is an even number; the negative gradient scanning signal collecting module 22 is used to scan N/2 times by a negative gradient, and to collect the negative gradient scanning signal during each scan; the positive eddy current correcting module 13 is used for eddy current correction of the N/2 positive gradient scanning signals collected according to the positive eddy current correction factor; the negative eddy current correcting module 23 is used for eddy current correction of the N/2 negative gradient scanning signals collected according to the negative eddy current correction factor; and the side-band suppressing module 30 is used for calculating the side-band suppressed spectrum signal according to the N/2 positive gradient scanning signals that have undergone the eddy current correction and N/2 negative gradient scanning signals that have undergone the eddy current correction.
The positive eddy current correction factor determining module 11 includes: a positive reference scanning sub-module 111 and a positive filtering sub-module 112.
In this case, the positive reference scanning sub-module 111 is used to perform the positive reference scanning P times by a positive gradient, and to collect the positive reference signal during the Pth positive reference scan, wherein P is a positive integer greater than 1; and, the positive filtering sub-module 112 is used for filtering the positive reference signal and taking the filtered signal phase as the positive eddy current correction factor.
The positive filtering sub-module 112 includes: a positive filtering bandwidth calculating unit 1126, a positive band-pass filter designing unit 1127, and a positive band-pass filtering unit 1128.
In this case, the positive filtering bandwidth calculating unit 1126 is used for calculating the positive filtering bandwidth; the positive band-pass filter designing unit 1127 is used for designing a Gaussian band-pass filter according to said positive filtering bandwidth; and the positive band-pass filtering unit 1128 is used for band-pass filtering of said positive reference signal with the Gaussian band-pass filter.
The positive filtering bandwidth calculating unit 1126 includes: a positive spectrum acquiring sub-unit 11261 and a positive calculating sub-unit 11262.
In this case, the positive spectrum acquiring sub-unit 11261 is used for acquiring the spectrum of the positive reference scan; and the positive calculating sub-unit 11262 is used for judging whether there is an overlapping part between the water signal and the metabolite signal in the positive reference scanning spectrum, and if there is an overlapping part, the waveband of the water signal is acquired and taken as the positive filtering bandwidth; otherwise, the overlapping waveband between the water signal and the metabolite signal is divided by 2 to acquire a quotient, and then the difference between the waveband of the water signal and the quotient is calculated, and the result is taken as the positive filtering bandwidth.
The negative eddy current correction factor determining module 21 includes: a negative reference scanning sub-module 211 and a negative filtering sub-module 212.
In this case, the negative reference scanning sub-module 211 is used to perform the negative reference scanning P times by a negative gradient, and to collect the negative reference signal during the Pth negative reference scan only, wherein P is a positive integer greater than 1; and, the negative filtering sub-module 212 is used for filtering the negative reference signal and taking the filtered signal phase as the negative eddy current correction factor.
The negative filtering sub-module 212 includes: a negative filtering bandwidth calculating unit 2126, a negative band-pass filter designing unit 2127, and a negative band-pass filtering unit 2128.
In this case, the negative filtering bandwidth calculating unit 2126 is used for calculating the negative filtering bandwidth; the negative band-pass filter designing unit 2127 is used for designing a Gaussian band-pass filter according to said negative filtering bandwidth; and the negative band-pass filtering unit 2128 is used for band-pass filtering of said negative reference signal with the Gaussian band-pass filter.
The negative filtering bandwidth calculating unit 2126 includes: a negative spectrum acquiring sub-unit 21261 and a negative calculating sub-unit 21262.
In this case, the negative spectrum acquiring sub-unit 21261 is used for acquiring the spectrum of the negative reference scan; and the negative calculating sub-unit 21262 is used for judging whether there is an overlapping part between the water signal and the metabolite signal in the negative reference scanning spectrum, and if there is an overlapping part, the waveband of the water signal is acquired and taken as the negative filtering bandwidth; otherwise, the overlapping waveband between the water signal and the metabolite signal is divided by 2 to acquire a quotient, and then the difference between the waveband of the water signal and the quotient is calculated, and the result is taken as the negative filtering bandwidth.
The positive eddy current correcting module 13 includes: a positive receiving sub-module 131 and a positive subtracting sub-module 132.
In this case, the positive receiving sub-module 131 is used for receiving the positive gradient scanning signal and the positive eddy current correction factor; and the positive subtracting sub-module 132 is used for calculating the difference between the phase of the positive gradient scanning signal and the positive eddy current correction factor.
The negative eddy current correction module 23 includes: a negative receiving sub-module 231 and a negative subtracting sub-module 232.
In this case, the negative receiving sub-module 231 is used for receiving the negative gradient scanning signal and the negative eddy current correction factor; and the negative subtracting sub-module 232 is used for calculating the difference between the phase of the positive gradient scanning signal and the positive eddy current correction factor.
The side-band suppression module 30 includes: an adding sub-module 301 and an averaging sub-module 302.
In this case, the adding sub-module 301 is used for adding up the N/2 positive gradient scanning signals that have undergone the eddy current correction and the N/2 negative gradient scanning signals that have undergone the eddy current correction, to acquire a sum; and the averaging sub-module 302 is used for calculating the average value of the sum, and taking the result as the side-band suppressed spectrum signal.
For a detailed description of the embodiment of a side-band suppression device provided by the present invention, refer to the corresponding description of the embodiment of the method shown in
It can be seen that based on the above side-band suppression method and device, first the positive eddy current correction factor and negative eddy current correction factor are determined, and then eddy current correction of the N/2 positive gradient scanning signals collected is conducted according to the positive eddy current correction factor, and eddy current correction of the N/2 negative gradient scanning signals collected is conducted according to the negative eddy current correction factor. Thus the impact of eddy current on the positive and negative gradient scanning signals can be eliminated. This ensures that the positive and negative gradient scanning signals have consistent initial phases, and that the phase difference of side-band signals will be 180 degrees during the positive and negative gradient scanning. Finally the side-band suppressed spectrum signals are calculated according to the N/2 positive gradient scanning signals that have undergone the eddy current correction and the N/2 negative gradient scanning signals that have undergone the eddy current correction. Thus, the side-band signals can be completely cancelled out by calculation, meaning that the method and device provided by the present invention can effectively suppress the side bands.
Although modifications and changes may be suggested by those skilled in the art, it is the intention of the inventors to embody within the patent warranted hereon all changes and modifications as reasonably and properly come within the scope of their contribution to the art.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2009 1 0157758 | Jul 2009 | CN | national |
| Number | Name | Date | Kind |
|---|---|---|---|
| 4885542 | Yao et al. | Dec 1989 | A |
| 5451877 | Weissenberger | Sep 1995 | A |
| 6037772 | Karczmar et al. | Mar 2000 | A |
| 6160397 | Washburn et al. | Dec 2000 | A |
| 8258785 | Hirata et al. | Sep 2012 | B2 |
| Entry |
|---|
| U.S. Appl. No. 12/842,459, filed Jul. 23, 2010. |
| Number | Date | Country | |
|---|---|---|---|
| 20110043203 A1 | Feb 2011 | US |
