KR101774888B1 - Pre-treatment apparatus for magnetic resonance spectroscopic data and method of pre-treatment for magnetic resonance spectroscopic data - Google Patents

Abstract

The present invention relates to a magnetic resonance spectroscopic signal pre-processing apparatus which obtains processed data by pre-processing magnetic resonance spectroscopic signal to extract information of materials in a human body from the processed data obtained by pre-processing the magnetic resonance spectroscopic signal obtained with respect to a human body. The magnetic resonance spectroscopic signal pre-processing apparatus comprises: a fourier conversion unit which performs fourier conversion from data in a time area into data in a frequency area of the magnetic resonance spectroscopic signal obtained with respect to a human body; a differential filter application unit which removes a signal, which is not the information of materials in a human body to be analyzed, by applying a differential filter to the magnetic resonance spectroscopic signal of the fourier-converted frequency area; and an abs application unit which applies an absolute value function to the signal to which the differential filter has been applied. Accordingly, the magnetic resonance spectroscopic signal pre-processing apparatus can perform, in an easy manner, acquisition or deletion of data of a metabolism material as an analysis target which a user desires to obtain from a magnetic spectroscopic signal, correct distortion of a base line from a water signal or a fat signal, detect correct data in an easy manner without distortion of the signal due to an absolute value which is not related to a phase value, has an effect of increasing the ratio of noise to a signal, and increases correctness of a metabolism material quantitative analysis with respect to an acquired signal.

Description

TECHNICAL FIELD [0001] The present invention relates to a magnetic resonance spectroscopic signal preprocessing apparatus and a magnetic resonance spectroscopic signal preprocessing method. More particularly, the present invention relates to a magnetic resonance spectroscopic signal preprocessing apparatus and a magnetic resonance spectroscopic signal preprocessing method,

The present invention relates to a magnetic resonance spectroscopic signal preprocessing apparatus and a magnetic resonance spectroscopic signal preprocessing method.

Molecular biological information data in the human body can be obtained through magnetic resonance spectroscopy (MRS) using a magnetic resonance imaging (MRI) imaging device.

The extraction of the signal related to the target substance to obtain information from the acquired magnetic resonance spectroscopy signal is performed by using a peak value analysis based on a real number value obtained by Fourier transforming a time domain magnetic resonance spectroscopic signal into a frequency domain .

MRS is susceptible to distortion in the measurement environment, and the data obtained is typically subjected to a certain level of pretreatment prior to quantitative metabolite analysis. In particular, in the case of MRS (1H MRS) for the signal of the hydrogen nuclei, it is necessary to remove the large signal contained in the obtained magnetic resonance spectroscopic signal, which is processed through an acquisition technique or a signal preprocessing technique.

In order to suppress the large-scale signal included in the magnetic resonance spectroscopic signal, the conventional time-domain filter can be applied only when the water signal is 0 Hz, and the SNR of the metabolite near the water signal is reduced And signal shape distortion occurs in the preprocessing process.

Preprocessing for other purposes (eg, signal-to-noise ratio enhancement, phase correction, fat signal suppression) in a magnetic resonance spectroscopy signal is also the most representative and universal for reasons such as signal distortion or loss of information For the LCModel, a metabolite analysis program for magnetic resonance spectroscopy signals, it is recommended that no signal preprocessing be performed.

Korean Patent Publication No. 10-2004-0044155

The present invention provides a magnetic resonance spectroscopic signal preprocessing apparatus and a magnetic resonance spectroscopic signal preprocessing method for performing a preprocessing operation of a magnetic resonance spectroscopic signal in order to accurately obtain metabolite information to be obtained from a magnetic resonance spectroscopy signal .

The apparatus for pre-processing a magnetic resonance spectroscopic signal according to an embodiment of the present invention includes a magnetic resonance spectroscopic signal processing unit for pre-processing the magnetic resonance spectroscopic signal for extracting substance information in the human body from processed data obtained by pre- A magnetic resonance spectroscopic signal preprocessing apparatus comprising: a Fourier transform unit for Fourier transforming a magnetic resonance spectroscopic signal obtained for a human body into data in a frequency domain from data in a time domain; A differential filter applying unit applying a differential filter to a magnetic resonance spectroscopic signal in a frequency domain of a Fourier transform to remove a signal other than substance information in a human body to be analyzed; And an abs applying part for taking an absolute value function in the signal to which the differential filter is applied, to pre-process the magnetic resonance spectroscopic signal obtained for the human body.

The differential filter applying unit may perform a convolution operation on the magnetic resonance spectral signal in the frequency domain subjected to the Fourier transform by the differential filter.

The differential filter applying unit performs a calculation of Equation (1) to convolute a differential filter that removes a signal excluding a metabolite to be obtained from a magnetic resonance spectroscopic signal, And the filter coefficient corresponding to the signal excluding the metabolite to be analyzed is applied to perform the calculation of the following equation (1).

[Formula 1]

FD2 (f) = A × [1, -1] * F (f) + B × [-1,2, -1] * F F (f) + E [1,0,0,0, -1] * F (f)

(Where f (f) is real data of the magnetic resonance spectroscopic signals in the frequency domain, and A, B, C, D and E are filter coefficients).

In one example, the abs application unit takes an absolute value function at the T2 value of the signal subjected to the differential filter application by the differential filter application unit.

And the abs applying unit performs an operation of Equation (2) to take an absolute value function for the T2 value of the signal subjected to the differential filter.

[Formula 2]

abs (FD2 (f)) = R (0) * A 1 / Y (f) * sqrt ((A 1 * (2f + 1)) 2 + (1-f 2 * A 1 2 -f * A 1 2 ) ² )

(1 + A ₁ ² * (f + 1) ² ) * (1 + A ² *) where A ₁ is 2pi * T ² , f is a frequency, T2 is a T2 value of a magnetic resonance signal, f ^2), R (0) is real (F (0)) in the sense of an error data on the magnetic resonance signal of the frequency domain transform to the frequency domain by the Fourier transformer, and real (F (0)) is bw * S (0) * T2, and bw is the bandwidth).

A method for pre-processing a magnetic resonance spectroscopic signal according to an embodiment of the present invention includes the steps of pre-processing the magnetic resonance spectroscopic signal for extracting material information in the human body from the processed data obtained by pre- A method for preprocessing a resonance spectroscopic signal, comprising the steps of: obtaining a frequency domain magnetic resonance spectroscopy signal by Fourier transforming a magnetic resonance spectroscopy signal; Convoluting a differential filter with a magnetic resonance spectroscopic signal in a frequency domain; And taking an absolute value of the magnetic resonance spectroscopic signal in the frequency domain to which the differential filter is applied, preprocesses the magnetic resonance spectroscopic signal obtained for the human body.

The convolution operation of the differential filter on the frequency domain magnetic resonance spectroscopy signal may be performed by performing an operation of the following equation 1 to obtain a filter coefficient corresponding to a signal excluding analytes to be obtained from the magnetic resonance spectroscopy signal And the calculation of the following formula (1) is performed.

[Formula 1]

FD2 (f) = A × [1, -1] * F (f) + B × [-1,2, -1] * F F (f) + E [1,0,0,0, -1] * F (f)

(Where f (f) is real data of the magnetic resonance spectroscopic signals in the frequency domain, and A, B, C, D and E are filter coefficients).

Wherein the step of taking the absolute value of the magnetic resonance spectroscopic signal in the frequency domain to which the differential filter is applied comprises the step of convoluting the differential filter with the magnetic resonance spectroscopic signal of the frequency domain to obtain the absolute value function of the T2 value of the signal .

The operation of taking the absolute value function as the T2 value of the signal obtained by performing the convolution operation of the differential filter on the frequency domain magnetic resonance spectroscopic signal is performed from the operation of the following expression (2).

[Formula 2]

abs (FD2 (f)) = R (0) * A 1 / Y (f) * sqrt ((A 1 * (2f + 1)) 2 + (1-f 2 * A 1 2 -f * A 1 2 ) ² )

(Wherein, A ₁ is 2pi * T2, f is the frequency, T2 is the T2 value, Y (f) of the magnetic resonance signal ^{_{(1 + A 1 2 * (}} f + 1) 2) * (1 + A 2 * f ^2), R (0) is real (F (0) Fourier by the converting section sense the real data of the magnetic resonance signal of the frequency domain transform to the frequency domain, and real as) (F (0)) is bw * S (0) * T2, and bw is the bandwidth).

According to this feature, the magnetic resonance spectroscopic signal preprocessing apparatus performs the Fourier transform processing, the differential filter application processing, and the absolute value (abs) application processing on the magnetic resonance spectroscopic signals obtained for the human body, And preprocesses the signal in an easy frequency domain. Thus, there is an effect that the substance information in the human body to be analyzed can be efficiently and accurately extracted from the preprocessed signal.

1 is a schematic block diagram of a magnetic resonance spectroscopic signal preprocessing apparatus according to an embodiment of the present invention.
FIG. 2 is a view showing a form of a signal processed by a magnetic resonance spectroscopic signal preprocessing apparatus according to an embodiment of the present invention.
FIG. 3 is a graph showing a ratio between a signal processed by a magnetic resonance spectroscopic signal preprocessing apparatus and an original signal according to an embodiment of the present invention.
FIG. 4 is a graph showing a ratio between a processed signal and a raw signal when various types of differential filter application units are applied in a magnetic resonance spectroscopic signal preprocessing apparatus according to an embodiment of the present invention.
FIG. 5 is a graph showing a result obtained by processing a specific equation in a magnetic spectroscopic signal preprocessing apparatus according to an embodiment of the present invention, and a graph obtained by applying a differential filter in a time domain to the signal.
FIG. 6 is a graph illustrating a result of applying a signal processed by the magnetic spectroscopic signal preprocessing apparatus according to an embodiment of the present invention to an analysis tool.
7 is a flowchart showing a flow of a method of pre-processing a magnetic spectroscopic signal according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out the present invention. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. In order to clearly illustrate the present invention, parts not related to the description are omitted, and similar parts are denoted by like reference characters throughout the specification.

Hereinafter, a magnetic resonance spectroscopic signal preprocessing apparatus and a magnetic resonance spectroscopic signal preprocessing method according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

First, referring to FIG. 1 to FIG. 6, a magnetic resonance spectroscopic signal preprocessing apparatus according to an embodiment of the present invention includes a Fourier transform unit 100, A differential filter applying section 200, and an abs applying section 300.

The magnetic resonance spectroscopic signal preprocessing apparatus is an apparatus for receiving and pre-processing a magnetic resonance spectroscopy (MRS) signal obtained from a magnetic resonance imaging (MRI) imaging apparatus, and may be a program embedded in a computer apparatus or a computer apparatus equipped with a processor.

At this time, the Fourier transform unit 100 performs Fourier transform processing of the magnetic resonance spectroscopic signal into a frequency domain signal in the time domain.

Accordingly, the time domain magnetic resonance spectroscopic signal S (t) becomes the frequency domain magnetic resonance spectroscopic signal F (w), and the frequency domain magnetic resonance spectroscopic signal includes real number data and imaginary data.

The differential filter applying unit 200 performs a convolution process on the real data of the magnetic resonance spectroscopic signals in the frequency domain.

In one example, the differential filter applied to the real data of the frequency domain magnetic resonance spectroscopy signal in the differential filter applying section 200 includes a differential filter for removing the water signal contained in the frequency domain magnetic resonance spectroscopy signal, It may be a differential filter for removing other non-substance metabolite signals.

Alternatively, the differential filter applying unit 200 may apply a differential filter for removing signals other than the metabolite to be analyzed.

In this case, the differential filters are (1, -1), (-1,2, -1), (1,0,0,1), (1,0,0,1) , -1), or a linear combination thereof, and the differential filter applying unit 200 performs convolution operation on at least one of the differential filters on the frequency domain magnetic resonance spectroscopy signal to generate FD2 (f) .

Here, FD2 (f) can be calculated from the following equation (1).

[Formula 1]

FD2 (f) = A × [1, -1] * F (f) + B × [-1,2, -1] * F F (f) + E [1,0,0,0, -1] * F (f)

In Equation 1, '*' is the convolution operation indication, F (f) is the frequency domain magnetic resonance spectroscopic signal data, and A, B, C, D and E multiplied by each differential filter are filter coefficients .

In this case, Equation 1 may further include a signal obtained by convoluting a new differential filter in the form of a linear combination of the above-described differential filters to F (f) and multiplying the filtered coefficients.

The filter coefficient can be set to an optimal coefficient according to the estimated T2 value and the signal-line width of the acquired signals, in order to remove the water signal, the metabolite signal or the fat signal in the frequency domain magnetic resonance spectroscopy signal have.

Then, the abs application unit 300 calculates an abs (FD2 (f)) by applying an absolute value to the FD2 (f) signal to which the at least one differential filter is convolution applied in the differential filter applying unit 200. [

As described above, the time domain magnetic resonance spectroscopic signals S (S) processed by the Fourier transform unit 100, the differential filter application unit 200, and the abs application unit 300 of the magnetic resonance spectroscopic signal preprocessing apparatus according to an embodiment of the present invention (t) is converted into a frequency domain magnetic resonance spectroscopic signal F (w) by the Fourier transform unit 100 as shown in FIG. 2, and the signal FD (w), and is converted into an absolute value signal by the abs application unit 300 so that | FD (w) | Signal.

In this case, as an example, the result of applying the differential filter [1, -1] in a magnetic resonance spectroscopic signal obtained from a material having only a single frequency characteristic without phase distortion is calculated by the following Equation 2.

[Formula 2]

abs (FD2 (f)) = R (0) * A 1 / Y (f) * sqrt ((A 1 * (2f + 1)) 2 + (1-f 2 * A 1 2 -f * A 1 2 ) ² )

In formula 2 above, f is the frequency, A ₁ is * 2pi T2, T2 is T2 value of the magnetic resonance signals, and Y (f) is ^{_{(1 + A 1 2 * (}} f + 1) 2) * (1+ A ² * f ² ), respectively.

In the above Equation 2, R (0) is real (F (0)) and real (F (f) is a frequency domain magnetic resonance spectroscopic signal converted into a frequency domain by the Fourier transformer 100) Quot; real data "

In this case, real (F (0)) is bw * S (0) * T2, and bw is bandwidth.

At this time, abs (FD2 (f)) which is data obtained by applying an absolute value by applying a differential filter to the data obtained as in Equation 2, that is, the magnetic resonance spectroscopic signal in the frequency domain, and the real number data Abs (FD2 (f)) / real (F (f)) which is the ratio of real (F (f)) is calculated by the following equation (3).

[Formula 3]

abs (FD2 (f)) / real (F (f)) = A * sqrt ((A * (2f + 1)) 2 + (1-f 2 * A 2 -f * A 2) 2) / (Y (f) * (1 + A ² * f ² ))

The ratio of abs (FD2 (f)) and real (F (f)), calculated in the above equation 3, The ratio is expressed by the graph of the form shown in Fig. 3 (a) according to the frequency and the value of T2.

Referring to the graph of FIG. 3A, the characteristics of the ratio of abs (FD2 (f)) and real (F (f)) according to frequency and T2 value will be described in more detail. The white curve on the graph shows the full-width at tenth-maximum (FWTM) value when the frequency changes along the horizontal axis and the T2 value changes along the vertical axis in the range of 1 to 200 ms.

3 (a), the ratio of abs (FD2 (f)) and real (F (f)) has a low value in the vicinity of the median value and has a substantially constant value in the remaining frequency regions , It can be seen that the shape of abs (FD2 (f)) is almost the same as the real value data real (F (f)) without distortion even though it is absolute value data.

As shown in the graph of FIG. 3 (b), in the graph in which the abscissa is the frequency and the ordinate is the ratio of abs (FD2 (f)) and real (F ) And the blue graph real (F (f)) have a similar shape so that the ratio of abs (FD2 (f)) and real (F (f) The data obtained by applying the filter to the absolute value has almost the same shape as the real data before applying the differential filter.

In the graph of FIG. 3 (b), the T2 value is 200 ms, and the FWTM when the T2 value is 200 ms is about 5 Hz at 3 / (pi * T2).

In this case, however, the size of abs (FD2 (f)) is influenced by the value of abs (FD2 (f)) / real (F (f)) in the center frequency band, the value of abs (FD2 (f)) / real (F (f)) changes as shown in (c) of Fig.

When the differential filter applying unit 200 convolutes a plurality of differential filters and takes an absolute value in the abs applying unit 300, abs (FD2 (f)) / real ( F (f)) is shown in the form shown in FIG.

In the graph of FIG. 4, the blue graph is a graph of abs (FD2 (f)) / real (F (f)) when the differential filter of [1, -1], the black graph is a differential filter of [1,0, -1], the green graph is a differential filter of [1,0,0, -1], and the yellow graph is [1,0 (F (f)) / real (F (f)) in the case where the differential filter of FIG.

At this time, since the value of abs (FD2 (f)) / real (F (f)) is determined by the T2 value, abs (FD2 (f)) / real (F When the T2 value is known as shown in the graph of FIG. 4, it is possible to know the differential filter information capable of removing a phase-distorted water signal or other metabolites whose center frequency is not distributed at 0 Hz.

In one example with reference to FIG. 4, when the T2 value of the fat signal is 5 ms or less and the T2 value of the metabolite signal is near 100 ms, filtering is performed by selecting a differential filter of [1,0, -1] Metabolic signals can be selected and extracted.

In performing the filtering using the differential filter of [1,0, -1], it is possible to remove the distorted baseline due to the fat signal and to detect the metabolite to be obtained, that is, the signal of the metabolite to be analyzed It can be preserved without loss.

As described above, the apparatus for pre-processing a magnetic resonance spectroscopic signal of the present invention performs Fourier transform and applies a differential filter to a frequency-domain magnetic spectroscopic signal and then takes an absolute value thereof. Therefore, A filter coefficient to be applied to the filter and the differential filter may be set to easily acquire or remove data of the metabolite to be obtained from the magnetic spectroscopic signal.

Since the magnetic spectroscopic signal abs (FD2 (f)) preliminarily processed by the magnetic spectroscopic signal preprocessing apparatus of the present invention is of an absolute value type independent of the phase value, there is no signal distortion due to phase noise, Do.

This is because, in the conventional method of acquiring metabolite data from a magnetic resonance spectroscopic signal, the signal is distorted due to phase distortion during acquisition of metabolite data from real data, thereby overcoming the difficulty of accurate data detection.

In this case, since the signal is formed to have a short T2 value as a noise type by applying the differential filter at this time, the signal-to-noise ratio is improved as the signal width is formed small.

The signal abs (FD2 (f)) pre-processed by the apparatus for pre-processing a magnetic resonance spectroscopic signal according to an embodiment of the present invention described with reference to Figs. 1 to 4 is obtained from the black graph shown in Fig. To illustrate, a black graph is a signal obtained by applying a differential filter of [-1,2, -1] and a differential filter of [1,0,0, -1]. At this time, a filter coefficient Is a filter coefficient for removing a water signal and a fat signal.

By removing the water signal and the fat signal by applying the differential filter obtained by multiplying the filter coefficients as described above, the value of the high value observed in the real data of the frequency domain magnetic resonance spectroscopic signal shown as the green graph in FIG. 5 (a) It can be seen that the water and fat signals are mitigated and removed from the black graph.

As shown in the example described with reference to FIG. 5A, the magnetic resonance spectroscopic signal abs (FD2 (f)) to which the differential filter is applied to take an absolute value reduces or eliminates a signal to be removed, Can be accurately extracted.

5 (a), it can be seen that the signal widths of the metabolites detected between the water signal and the fat signal in the abs (FD2 (f)) and real (F (f) It is understood by those skilled in the art that there is no error in detecting the data of the metabolites even if a pre-processing is performed by applying a differential filter to the magnetic resonance spectroscopic signal in the frequency domain and taking an absolute value.

In particular, referring to FIG. 5 (b), the conventional method of applying the differential filter in the time domain will be described in comparison with the present invention. In the case of applying the first derivative filter [-1 , 1] and a quadratic differential filter [-1,2, -1] are applied, phase distortion occurs as shown in the left graph in FIG. 5 (b) ), The signal-to-noise ratio is remarkably low as shown in the graph on the right side.

Also, the graph of FIG. 5 (b) can not remove the baseline from the fat signal, resulting in poor data accuracy.

The application of the differential filter in the time domain has a limitation in that it can not be removed when the water signal is detected at a frequency other than 0 Hz. However, the data to which the magnetic resonance spectroscopic signal preprocessing apparatus according to the embodiment of the present invention is applied can secure data efficiently in terms of high signal-to-noise ratio and accuracy of data extraction.

6, data abs (FD2 (f)) obtained by analyzing the metabolite to be analyzed from the data real (F (w)) of the magnetic resonance spectroscopic signal and the absolute value of the magnetic resonance spectroscopic signal using a differential filter, 6 (a), the data of the pre-processed data real (F ((F)), which has not been subjected to the preprocessing, f) can be input into the analysis tool to detect three kinds of metabolites to be analyzed.

However, if the magnetic resonance spectroscopic signal data abs (FD2 (f)) which has performed the operation of the apparatus for pre-processing a magnetic resonance spectroscopic signal according to an embodiment of the present invention is input to the analysis tool, Five types of target metabolites can be detected.

In one example, the magnetic resonance spectroscopic signal data abs (FD2 (f)) shown in Fig. 6B is [1, -1], [-1,2, -1], and [ The signal-to-noise ratio (SNR) of the data obtained by taking the absolute value by applying the differential filter of FIG. 6A, FIG. 6A, and FIG. 6A is higher than that of FIG. It can be seen that the performance is excellent.

At this time, as an example of the analysis tool used in Fig. 6, a commercial program LCModel can be applied.

Next, referring to FIG. 7, a method for pre-processing a magnetic resonance spectroscopic signal according to an embodiment of the present invention will be described in detail. A method for pre-processing a magnetic resonance spectroscopic signal comprises: A step S200 of convoluting the differential filter with the magnetic resonance spectroscopic signal in the frequency domain, and a step S300 of taking the absolute value of the magnetic resonance spectroscopy signal in the frequency domain to which the differential filter is applied .

The method for pre-processing a magnetic resonance spectroscopic signal including the series of steps may be performed by the magnetic resonance spectroscopic signal preprocessing apparatus of the present invention described above with reference to FIGS. 1 to 6. The flowchart of FIG. 7 is referred to as a magnetic resonance spectroscopy First, the Fourier transform section performs Fourier transform on a time domain magnetic resonance spectroscopy signal to generate a frequency domain magnetic resonance spectroscopy signal.

At this time, the Fourier transform unit receives the magnetic resonance spectroscopic signal obtained from the MRI apparatus, regards it as an input signal, and performs Fourier transform.

From this step S100, the magnetic resonance spectroscopic signal S (t) in the time domain changes into the magnetic resonance spectroscopic signal F (w) in the frequency domain.

Then, the step S200 of convoluting the differential filter with the magnetic resonance spectroscopic signal in the frequency domain causes the differential filter applying unit 200 to apply a differential filter to the real data of the frequency domain magnetic resonance spectroscopic signal F (w) And performs convolution operation to calculate FD (w), which is performed according to Equation (1).

At this time, in Equation 1, the T2 value of the data obtained by applying the differential filter is represented by FD2 (f).

At this time, in the corresponding step S200 of applying the differential filter, it is possible to adjust the filter coefficient to be multiplied by the differential filter and the differential filter applied according to the data to be acquired or the data to be removed.

From this step S200, it is possible to obtain only the data to be acquired from the real number data of the magnetic resonance spectroscopic signals in the frequency domain, that is, the signal from which the data to be removed is removed.

Finally, the step S300 of taking the absolute value of the magnetic resonance spectroscopic signal in the frequency domain to which the differential filter is applied is to determine the absolute value of the absolute value of the magnetic resonance spectroscopic signal FD2 (f) to which the abs filter 300 is applied from the equation To obtain abs (FD2 (f)).

From the above steps, it is possible to extract the real data of the frequency domain magnetic resonance spectroscopy signal, in particular the accurate signal from which the signal to be removed is removed so as to extract only the analytes to be obtained, The quantitative analysis of the metabolite to be analyzed can be accurately performed.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, It belongs to the scope of right.

100: Fourier transform unit 200: Differential filter application unit
300: abs application part

Claims (9)

1. A magnetic resonance spectroscopic signal preprocessing apparatus for preprocessing a magnetic resonance spectroscopic signal for extracting material information from a human body, the magnetic resonance spectroscopic signal being obtained for a human body,
A Fourier transform unit for Fourier transforming the magnetic resonance spectroscopic signal obtained for the human body into data in the frequency domain from data in the time domain;
A differential filter applying unit applying a differential filter to a magnetic resonance spectroscopic signal in a frequency domain of a Fourier transform to remove a signal other than substance information in a human body to be analyzed; And
An abs applying unit that takes an absolute value function in a signal to which the differential filter is applied;
A magnetic resonance spectroscopic signal pre-processing device for pre-processing a magnetic resonance spectroscopic signal acquired for a human body. The method according to claim 1,
Wherein the differential filter applying unit performs a convolution operation on the magnetic resonance spectral signal in the frequency domain subjected to the Fourier transform by the differential filter. 3. The method of claim 2,
The differential filter applying unit performs a calculation of Equation (1) to convolute a differential filter that removes a signal excluding a metabolite to be obtained from a magnetic resonance spectroscopic signal, Wherein the calculation of Equation (1) is performed by applying a filter coefficient corresponding to a signal excluding the metabolite to be analyzed.
[Formula 1]
FD2 (f) = A × [1, -1] * F (f) + B × [-1,2, -1] * F F (f) + E [1,0,0,0, -1] * F (f)
(Where f (f) is real data of the magnetic resonance spectroscopic signals in the frequency domain, and A, B, C, D and E are filter coefficients). The method according to claim 1,
Wherein the abs applying unit takes an absolute value function as the T2 value of the signal subjected to the differential filter application by the differential filter applying unit. 5. The method of claim 4,
Wherein the abs applying unit performs an operation of Equation (2) to take an absolute value function as a T2 value of the signal subjected to the differential filter application.
[Formula 2]
abs (FD2 (f)) = R (0) * A 1 / Y (f) * sqrt ((A 1 * (2f + 1)) 2 + (1-f 2 * A 1 2 -f * A 1 2 ) ² )
(Wherein, A ₁ is 2pi * T2, f is the frequency, T2 is the T2 value, Y (f) of the magnetic resonance signal ^{_{(1 + A 1 2 * (}} f + 1) 2) * (1 + A 2 * f ^2), R (0) is real (F (0) Fourier by the converting section sense the real data of the magnetic resonance signal of the frequency domain transform to the frequency domain, and real as) (F (0)) is bw * S (0) * T2, and bw is the bandwidth). A magnetic resonance spectroscopic signal preprocessing method for preprocessing a magnetic resonance spectroscopic signal for extracting material information from a human body from processed data obtained by preprocessing a magnetic resonance spectroscopic signal obtained for a human body,
Obtaining a frequency domain magnetic resonance spectroscopy signal by Fourier transforming the magnetic resonance spectroscopy signal;
Convoluting a differential filter with a magnetic resonance spectroscopic signal in a frequency domain; And,
Taking an absolute value of a magnetic resonance spectroscopic signal in a frequency domain to which a differential filter is applied;
And a magnetic resonance spectroscopic signal pre-processing step of pre-processing the magnetic resonance spectroscopic signal obtained for the human body. The method according to claim 6,
The convolution operation of the differential filter on the frequency domain magnetic resonance spectroscopy signal may be performed by performing an operation of the following equation 1 to obtain a filter coefficient corresponding to a signal excluding analytes to be obtained from the magnetic resonance spectroscopy signal And performing an operation of Equation (1) below.
[Formula 1]
FD2 (f) = A × [1, -1] * F (f) + B × [-1,2, -1] * F ) + D × [1,0,0, -1] * F (f) + E × [1,0,0,0-1] * F (f)
(Where f (f) is real data of the magnetic resonance spectroscopic signals in the frequency domain, and A, B, C, D and E are filter coefficients). The method according to claim 6,
Wherein the step of taking the absolute value of the magnetic resonance spectroscopic signal in the frequency domain to which the differential filter is applied comprises the step of convoluting the differential filter with the magnetic resonance spectroscopic signal of the frequency domain to obtain the absolute value function of the T2 value of the signal Wherein the magnetic resonance spectroscopic signal is obtained by the following method. 9. The method of claim 8,
Wherein an operation of taking an absolute value function as a T2 value of the signal subjected to the convolution operation of the differential filter in the frequency domain magnetic resonance spectroscopic signal is performed from an operation of the following Equation 2: Way.
[Formula 2]
abs (FD2 (f)) = R (0) * A 1 / Y (f) * sqrt ((A 1 * (2f + 1)) 2 + (1-f 2 * A 1 2 -f * A 1 2 ) ² )
(Wherein, A ₁ is 2pi * T2, f is the frequency, T2 is the T2 value, Y (f) of the magnetic resonance signal ^{_{(1 + A 1 2 * (}} f + 1) 2) * (1 + A 2 * f ^2), R (0) is real (F (0) Fourier by the converting section sense the real data of the magnetic resonance signal of the frequency domain transform to the frequency domain, and real as) (F (0)) is bw * S (0) * T2, and bw is the bandwidth).

Images