JP2003000564A - Device and method for processing image and magnetic resonance imaging system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To surely remove spike noises on raw data without affecting a tomographic image. SOLUTION: When there are stripe patterns in the tomographic image, raw data information in an image storage part 204 are displayed on a display part 12, an extracting area including the spike noises is set and information in that extracting area is replaced with information not to affect the tomographic image. Such processing is performed in a high speed image processing part 202 so that the stripe patterns can be surely removed from the tomographic image after the raw data are reconfigured in the high speed image processing part 202.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus, an image processing method and a magnetic resonance imaging apparatus for removing noise generated in a tomographic image acquired when a human body is imaged by the magnetic resonance imaging apparatus. Regarding

[0002]

2. Description of the Related Art In a magnetic resonance imaging apparatus, the magnetic resonance frequency of hydrogen atoms existing inside the human body is changed by a gradient magnetic field at the position where the hydrogen atoms are present, and a magnetic signal from the excited hydrogen atoms is received by a receiving coil (Coil). Detected by and converted into an electric signal. The amplitude of a specific frequency component of the received plurality of electric signal information (hereinafter, referred to as raw data (Data)) is the magnetic moment of a plurality of hydrogen atoms existing at a specific position determined from the gradient magnetic field ( Moment). Therefore, the frequency component of the electric signal obtained by Fourier transforming the raw data indicates the magnitude of the magnetic moment inside the human body existing at the position corresponding to the frequency, and the frequency distribution of the electric signal is As it is, it shows the distribution of the number of hydrogen atoms existing inside the human body. A two-dimensional tomographic image of the human body can be obtained by converting the amplitude of this frequency distribution into a luminance signal and displaying it.

Since the magnetic resonance imaging apparatus is an electromagnetically highly sensitive apparatus, it is easy to receive external noise. For this reason, the electromagnetically sensitive portion of the magnetic resonance imaging apparatus is an RF (Radio Frequency) shielded room.
It is installed in the room) and is no longer affected by external noise. Also, when transmitting and receiving electrical signals inside and outside the shielded room, a filter is applied at the boundary between them.
To prevent noise from being mixed into the electrical signal.

The external noise includes continuous wave type noise, spike noise, etc., but spike noise often has large power. As the cause of the spike noise, there are automobiles and other transportation equipment that runs in the vicinity of the shield room, but it is often difficult to eliminate the cause. Also, it was difficult to prevent spike noise having a large power from entering even with a filter inside and outside the shielded room.

Therefore, a low-pass type analog filter (Analog Filte) is applied to the electric signal received by the receiving coil.
By using r), spike noise appearing on the raw data is removed. According to this method, the frequency component of noise that exceeds the frequency band of the electric signal can be removed, which is effective in removing spike noise.

[0006]

However, according to the above-mentioned prior art, spike noise has a wide frequency band, and this frequency band overlaps with the frequency band of an electric signal, and spike noise has a large power. Therefore, the spike noise on the raw data could not be completely removed.

In particular, when a low frequency pass type analog filter is used, the cutoff characteristic changes gently near the cutoff frequency, so that it is necessary to completely remove the high frequency component. With difficulty. Furthermore, even if a digital filter with a sharp cutoff characteristic is selected, spike noise remains in a portion where it overlaps with the frequency band of the electric signal because spike noise has a wide frequency band. There is a limit due to the filtering process.

From these facts, spike noise on raw data is removed without using filter processing, and striped patterns of tomographic images obtained by image reconstruction of raw data are eliminated,
It is extremely important how to improve the image quality of the tomographic image and eventually obtain a stable image of the magnetic resonance imaging apparatus.

The present invention has been made in order to solve the above-mentioned problems of the prior art, and eliminates spike noise on raw data and reconstructs the raw data by image reconstruction without using an analog filter. It is an object of the present invention to provide an image processing apparatus, an image processing method, and a magnetic resonance imaging apparatus capable of improving the image quality of a tomographic image by eliminating the noise of the generated tomographic image and eventually obtaining a stable tomographic image.

[0010]

[Means for Solving the Problems]
In order to achieve the object, an image processing apparatus according to the invention of a first aspect is an image processing apparatus for acquiring raw data which is two-dimensional Fourier transform information of two-dimensional image information and displaying the raw data. And setting means for setting an extraction area containing spike noise on the raw data, and replacement means for replacing alternative information of the extraction area extracted by the setting means with information of the extraction area. To do.

According to the invention of the first aspect, the extraction area containing spike noise on the raw data is set, and the replacement information of the extracted extraction area is replaced with the information of the extraction area, so that it is certain. The spike noise can be removed from the raw data, and the striped pattern can be eliminated from the two-dimensional image information obtained by Fourier transforming the raw data.

The image processing apparatus according to the invention of the second aspect further comprises a selecting means for selecting alternative information of the extraction area extracted by the setting means, and the replacing means is selected by the selecting means. The alternative information is replaced with the information of the extraction area.

According to the invention of the second aspect, since the alternative information of the extraction area is selected and the selected alternative information is replaced with the information of the extraction area, an appropriate alternative is selected from a plurality of alternative information. Information can be selected.

Further, the image processing apparatus according to the invention of the third aspect is characterized in that the setting means sets an area surrounding spike noise on the raw data set by manual operation as an extraction area.

According to the invention of the third aspect, since the area surrounding the spike noise on the raw data set by the manual operation is set as the extraction area, the spike noise can be surely extracted.

Further, in the image processing apparatus according to the invention of the fourth aspect, the setting means manually sets the raw data information into a low frequency side and a high frequency side including spike noise. It is characterized in that the high frequency side of the line is used as the extraction region.

According to the invention of the fourth aspect, the high frequency side of the extraction line set by the manual operation for dividing the raw data information into the low frequency side and the high frequency side including spike noise is used as the extraction region. Since it is decided that the extraction area can be easily set.

The image processing apparatus according to the invention of the fifth aspect is characterized in that zero value information is used as alternative information.

According to the invention of the fifth aspect, since the zero value information is used as the alternative information, the two-dimensional data obtained by Fourier transforming the raw data without changing the raw data information greatly. It is possible to reduce the influence on the image information.

Further, in the image processing apparatus according to the invention of the sixth aspect, as the selection means, there is a relation between the frequency of the extraction region selected by the extraction means from the raw data information as the alternative information and the anti-frequency. It is characterized by using raw data information of the area.

According to the sixth aspect of the invention, the raw data information of the area having the relation between the frequency of the extraction area selected from the raw data information by the extraction means and the anti-frequency is used as the alternative information. Therefore, the raw data information after removing spike noise has the same information as the raw data information before removing spike noise in principle, and the two-dimensional image information obtained by Fourier transforming the raw data is the original information. The image is faithfully reproduced.

In the image processing apparatus according to the seventh aspect, as the setting means, in the raw data information, information on a certain frequency and information having a relationship between the frequency and the anti-frequency are sequentially compared, It is characterized in that an area of information in which a difference between two information values exceeds a predetermined value is set as an extraction area by an automatic operation.

According to the invention of the seventh aspect, in the raw data information, information of a certain frequency and information having a relationship between the frequency and the anti-frequency are successively compared, and the difference between the two information values is calculated. Since the area of the information exceeding the predetermined value is set as the extraction area by automatic operation, the extraction area can be set at high speed without manual operation.

The image processing apparatus according to an eighth aspect is, as the setting means, comparing the information having the same frequency in the raw data information, and extracting a different information area beyond a predetermined value. It is characterized in that the operation is performed automatically.

According to the invention of the eighth aspect, it is possible to perform an automatic operation to compare the information having the same frequency in the raw data information and use the areas of different information exceeding the predetermined value as the extraction areas. Therefore, even when using a pulse sequence such as the half echo method that can obtain only half the information amount of the raw data information of the spin echo method, the extraction area can be set at high speed without manual operation. it can.

An image processing method according to a ninth aspect of the invention is an image processing method for acquiring raw data which is two-dimensional Fourier transform information of two-dimensional image information and displaying the raw data. It is characterized by including a setting step of setting an extraction area including spike noise on the raw data, and a replacement step of replacing the alternative information of the extraction area extracted by the setting step with the information of the extraction area.

According to the ninth aspect of the invention, the extraction area containing spike noise on the raw data is set, and the replacement information of the extracted extraction area is replaced with the information of the extraction area. The spike noise can be removed from the raw data, and the striped pattern can be eliminated from the two-dimensional image information obtained by Fourier transforming the raw data.

The image processing method according to the invention of the tenth aspect further comprises a selecting step of selecting alternative information of the extraction region extracted by the setting step, and the replacing step is selected by the selecting step. The alternative information is replaced with the information of the extraction area.

According to the invention of the tenth aspect, since the alternative information of the extraction area is selected and the selected alternative information is replaced with the information of the extraction area, an appropriate alternative is selected from the plurality of alternative information. Information can be selected.

The magnetic resonance imaging apparatus according to the invention of the eleventh aspect is a magnetic resonance imaging apparatus for acquiring raw data which is two-dimensional Fourier transform information of two-dimensional tomographic image information and displaying the raw data for image processing. Resonance imaging apparatus, setting means for setting an extraction area containing spike noise on the raw data information, replacement means for replacing the alternative information of the extraction area extracted by the setting means with the information of the extraction area, It is characterized by having.

According to the eleventh aspect of the invention, the extraction area containing spike noise on the raw data information is set, and the alternative information of the extracted extraction area is replaced with the information of the extraction area. It is possible to reliably remove the striped pattern generated on the two-dimensional tomographic image of the resonance imaging device, and further to prevent misdiagnosis when diagnosing the two-dimensional tomographic image.

The magnetic resonance imaging apparatus according to the twelfth aspect of the invention further comprises a selecting means for selecting alternative information of the extraction region extracted by the setting means, and the replacing means is selected by the selecting means. The replacement information is replaced with the information of the extraction area.

According to the invention of the eleventh aspect, since the alternative information of the extraction area is selected and the selected alternative information is replaced with the information of the extraction area, an appropriate alternative information is selected from a plurality of alternative information. Information can be selected.

[0034]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of an image processing apparatus, an image processing method and a magnetic resonance imaging apparatus according to the present invention will be described below with reference to the accompanying drawings.

(Embodiment 1) First, the overall configuration of the magnetic resonance imaging apparatus according to Embodiment 1 will be described. FIG. 1 is a schematic diagram showing the overall configuration of a magnetic resonance imaging apparatus using the magnetic resonance imaging coil according to the first embodiment. In FIG. 1, the magnetic resonance imaging apparatus is roughly composed of a magnet unit 101 and a table (Tabl).
e) It is composed of a unit 102 and a control processing unit 103.

The magnet section 101 has a pair of static magnetic field generating sections 4, a gradient coil section 3 and a transmitting coil section 2 which are vertically opposed to each other. A space in which the subject 5 is arranged is formed between the two transmission coil units 2 that are arranged to face each other. The upper and lower gradient coil units 3 are connected to the scan controller unit 1 via the gradient drive unit 14.
3 is connected. In addition, the upper and lower two transmission coil units 2 are
It is connected to the scan controller unit 13 via the transmission unit 15.

The table section 102 has a cradle section 6 which moves in a space formed between the transmission coil sections 2.
Is provided, and the subject 5 is placed on the cradle portion 6. The movement of the cradle unit 6 is controlled so that the imaging region of the subject 5 is located at the center of the magnet unit 101.

The receiving coil 1 is arranged on the table portion 102 and is located near the center of the magnet portion 101. The receiving coil 1 is connected to a computer section 10 via a receiving section 7, a detecting section 8 and an A / D (Anatog / Digital) converting section 9. The computer unit 10 includes an operation unit 11
The display unit 12 is connected, and the scan controller unit 13 is connected. The scan controller unit 13 includes a receiver 7, an RF oscillator 20, an A / D converter 9,
Each part of the gradient drive part 14 and the transmission part 15 is controlled. R
The F oscillator 20 is connected to the detector 8.

Next, a specific configuration of the computer section 10 shown in FIG. 1 will be described. FIG. 2 is a block diagram showing a specific configuration of the computer unit 10 shown in FIG. Figure 2
In the computer section 10, the computer section 10 includes a central control section 201, a high-speed image processing section 202, a main storage section 203, and an image storage section 2.
04, an input port 205 for inputting an electric signal from the A / D conversion unit 9, an input / output port 206 for transmitting / receiving an electric signal to / from the scan controller unit 13, and a display unit 1
2, a display control unit 207 for performing control of No. 2, an operation unit interface 208 for receiving an input from the operation unit 11, and a shared bus 209 for connecting these and exchanging information.

The central control unit 201 controls the entire control and processing of the computer unit 10, and the main storage unit 203 is also used as a temporary storage location when the central control unit 201 performs control and processing. The image storage unit 204 stores a large amount of raw data and tomographic image information that is two-dimensional image information. The high-speed image processing unit 202 is an arithmetic unit dedicated to image processing that performs image reconstruction at high speed, and has a function of acquiring raw data from the image storage unit 204, sending the tomographic image back to the image storage unit 204 after performing image reconstruction. have.

Here, A / A is detected by the receiving coil 1.
Raw data, which is a set of a plurality of electric signals digitized by the D conversion unit 9, and tomographic image information obtained by Fourier transforming spike noise existing on the raw data will be described.

FIG. 3A shows an example in which the amplitude of the electric signal of the raw data is brightness-modulated and displayed as two-dimensional information on the display unit 12. Further, in FIG. 3B, the tomographic image 303 is obtained by performing brightness modulation on the information obtained by performing the two-dimensional Fourier transform on the raw data shown in FIG.
An example displayed on the display unit 12 is shown.

As shown in FIG. 3A, the tomographic image 303
The spatial frequency information of is displayed such that the zero frequency component at the center of the display screen is the center and the higher frequency component is toward the outside. In the case of the tomographic image 303 of the human body,
The raw data has a structure that spreads in concentric circles, and the information of the low frequency portion is displayed as the main image information 302. The spike noise 301 is also displayed on the raw data. Fourier transform of this raw data is shown in Figure 3.
The display image shown in (B) is obtained. As shown in FIG. 3B, a striped pattern resulting from the spike noise 301 appears on the tomographic image 303 of the subject 5. This striped pattern becomes a major obstacle when making a diagnosis using the tomographic image 303.

Next, the operation of the central control unit 201 in the computer unit 10 shown in FIG. 2 will be described using the flowchart shown in FIG. As shown in FIG. 4, the central control unit 201 sets an image scan based on an instruction from the operation unit 11 (step S401). In this setting, information such as Spin Echo method and other pulse sequence information and information such as setting of the acquisition position of the tomographic image are transmitted to the scan controller unit 13.

Then, the central control unit 201 is operated by the operation unit 11
Based on the instruction from, the start information is transmitted to the scan controller unit 13 to start image scanning (step S402). When the image scan is completed, raw data is acquired (step S403). Then, the obtained raw data is converted from an analog signal to a digital signal by the A / D conversion unit 9 and transferred to the image storage unit 204 of the computer unit 10.

The raw data transferred to the image storage unit 204 is
The high-speed image processing unit 202 reconstructs the image using the two-dimensional Fourier transform (step S404). The tomographic image information obtained by image reconstruction is stored in the image storage unit 2.
After being moved to 04, it is displayed on the display unit 12 (step S405).

Here, the central control unit 201 controls the display unit 12
If the tomographic image shown in FIG. 3 does not have the striped pattern shown in FIG. 3B (No in step S406), there is no problem in diagnosing the tomographic image, and the next tomographic image is acquired as it is. It is determined whether or not (step S41)
2).

The tomographic image displayed on the display unit 12 is shown in FIG.
If there is a striped pattern as shown in (B) (step S
406 affirmative), the operator displays the raw data accumulated in the image storage unit 204 before performing the image reconstruction on the display unit 12.
Is displayed (step S407). Spike noise 301 as shown in FIG. 3A is visually recognized in the raw data displayed on the display unit 12.

In order to extract this spike noise 301 (step S408), the operator sets the extraction area on the raw data displayed using the operation unit 11.
The extraction area is set by the operator as shown in FIG.
It is performed by any of the methods shown in (A), (B), and (C).

FIG. 5A shows the main image area 30.
This is a method of surrounding the spike noise 301 with a square so that 2 is not included. Considering the influence on the tomographic image, it is desirable to reduce the extraction area surrounded by the square including the spike noise 301.

FIG. 5B shows the main image area 30.
2 and a region including spike noise 301 are divided by a horizontal extraction line. Considering the influence on the tomographic image, it is desirable to make the extraction area surrounded by the horizontal extraction line including the spike noise 301 small.

FIG. 5C shows the main image area 30.
2 and a region including spike noise 301 are divided by vertical extraction lines. Considering the influence on the tomographic image, it is desirable to reduce the extraction area surrounded by the vertical extraction line including the spike noise 301.

Then, the central control unit 201 replaces the information of the extraction area with new information other than the spike noise 301 (step S409). As the new information, zero-value information or center point symmetry or center line symmetry of the extraction region with respect to the main image information 302 on raw data (hereinafter,
Any one of the pieces of information on the area at the position of "central symmetry" is selected by the operator.

When the zero value information is used as new information, the information of the extraction area including the spike noise 301 is replaced with the zero value. At this time, at the boundary between the raw data and the extraction area replaced by the zero value, the information value does not change continuously and a step is generated. In order to eliminate this step, the spatial filter is used for smoothing after replacing the information of the extraction region with a zero value.

When the information of the area at the position centrally symmetrical to the extraction area is used as new information, the information on each point in the extraction area is the information at the point centrally symmetrical with respect to the main image information 302. Replace with. This is the tomographic image information
Since it is originally a real value, the raw data information which is the Fourier transform of the tomographic image information is located at the position symmetrical about the zero frequency, that is, the raw data of the point having the relationship between the frequency and the anti-frequency. Information is based on the mathematical fact that information is equal. In this case, the raw data information of the two complex numbers at the centrally symmetric position has a complex conjugate relationship, but the phase information includes an error due to the magnetic field inhomogeneity of the static magnetic field generation unit 4, so the phase information is calculated in advance. It is necessary to collect and correct error information.

Next, the central control unit 201 reconstructs the image using the raw data in which the extraction area is replaced with new information (step S410), and displays the reconstructed tomographic image on the display unit 12. (Step S411). Then, it is determined whether or not the next tomographic image is to be acquired (step S412). If the tomographic image is to be acquired again, the process returns to step S401, otherwise the process ends.

As described above, in the first embodiment, when the tomographic image 303 has a striped pattern, the raw data is displayed and the spike noise 301 on the raw data is set to the zero value information or the position of central symmetry. Since the raw data information is replaced with certain raw data information and the image is reconstructed and displayed, the striped pattern of the tomographic image 303 can be eliminated.

In the extraction of spike noise 301 in step S408, an example in which a square extraction area is set has been shown (FIG. 5A), but the shape of the extraction area is not limited to a square. The pulse sequence used when acquiring the raw data includes the spin echo method and the field echo method, but it is a half to generate a tomographic image from about half the information on the raw data. In the case of the echo (Half Echo) method and the half next (Half Next) method, the striped pattern of the tomographic image 303 can be removed by replacing the extraction area of the spike noise 301 with zero value information.

Further, in pulse sequences such as the ultra-high speed scanning method and the first spin echo method, the influence of the relaxation time on the strength of the signal on the raw data differs depending on the position on the raw data. Therefore, by correcting the effect of the relaxation time so that the signals on the raw data are all affected by the same relaxation time, the spike noise 301 is removed. The striped pattern of the tomographic image 303 due to the noise 301 can be eliminated.

Further, in the first embodiment, the computer unit 10
Has a control function and an image processing function. However, an image processing device dedicated to the image processing function may be provided separately from the computer unit 10 and connected to the computer unit 10 to remove spike noise. it can. In this case,
It is possible to reduce the burden on the control processing capacity of the computer unit 10.

(Second Embodiment) By the way, in the first embodiment, the existence of spike noise 301 on the raw data is foreseen from the striped pattern appearing on the tomographic image, and the spike noise 301 is displayed after the raw data is displayed. Although it has been decided to manually set the extraction region to be included on the raw data to remove the spike noise 301, the present invention is not limited to this, and the spike noise 30 on the raw data is automatically set.
It can also be applied to the case where 1 is detected and the spike noise 301 is removed.

FIG. 6 is a block diagram showing the configuration of the computer section 60 according to the second embodiment. The computer unit 60 corresponds to the computer unit 10 shown in FIG. 1, and the other configuration is the same as that shown in FIG. 1, so a detailed description thereof will be given here. Omit it.

The central control unit 601 and the high-speed image processing unit 602 shown in FIG. 6 are the same as the central control unit 2 shown in FIG.
01 and the high-speed image processing unit 202, and other configurations are the same as those shown in FIG. 2, and therefore detailed description thereof will be omitted here.

Next, the operations of the central controller 601 and the high speed image processor 602 will be described with reference to the flowchart of FIG. The central control unit 601 sets the image scan based on the instruction from the operation unit 11 (step S70).
1). In this setting, pulse sequence information such as the spin echo method and information such as the setting of the acquisition position of the tomographic image are transmitted to the scan controller unit 13.

The central control unit 601 has the operation unit 11
Based on the instruction from, the start information is transmitted to the scan controller unit 13 to start image scanning (step S702). When the image scanning is completed, raw data is acquired (step S703). Then, the obtained raw data is converted from an analog signal to a digital signal by the A / D conversion unit 9 and transferred to the image storage unit 204 of the computer unit 60.

The raw data transferred to the image storage unit 204 is
Since the information corresponding to each frequency is a complex number, it consists of two numerical values. The absolute value of the complex number is calculated from these two numerical values (step S704), and the spike noise 301 is extracted by the high-speed image processing unit 602 using the raw data information converted into the absolute value (step S705).

This extraction is performed on the basis of the mathematical fact that the absolute values of the raw data information at the positions symmetrical about the zero frequency information on the raw data are equal.
The information in the right half of the raw data is compared with the information in the left half of the centrally symmetrical position one by one, and if they differ by more than a predetermined value, a large value of information is taken as spike noise. Figure 8
In the example shown in (A), spike noise 301 (black circle)
And the absolute value information of the point 801 (white circle) existing at the position of center symmetry thereof are compared, and it is determined that the point 301 is spike noise because the absolute value of the spike noise 301 is significantly large. Then, a set of different points that are larger than a predetermined value and are larger is set as an extraction region.

Further, the spike noise 301 can be extracted from the property that the absolute values of the raw data information located on the concentric circles centering on the zero frequency on the raw data are equal. This is because the spike noise 3 is compared by comparing the information on the circle indicated by the dotted line including the position where the spike noise 301 exists as the point on the concentric circle as shown in FIG. 8B.
01 is extracted. Furthermore, the extraction area can be determined by comparing the information on all circles with different radii.

If there is spike noise (Yes at step S706) and the extraction area is confirmed, the information of the extraction area on the raw data is replaced with a zero value (step S707). Then, the discontinuity of the information value generated at the boundary of the extraction area on the raw data is smoothed by the spatial filter, and then the image is reconstructed (step S708). If spike noise does not exist (No at step S706), the information of the extraction region is not replaced, and the image reconstruction of step S708 is performed.

After the image reconstruction is completed, the tomographic image is displayed on the display section 12 (step S709), and it is judged whether or not to acquire the tomographic image again (step S710). The return image scan is set, and if not acquired, the process ends.

As described above, in the second embodiment, the spike noise 301 and its extraction area are automatically extracted by comparing centrosymmetrical information with the zero frequency on the raw data as the central point. Because
Before performing image reconstruction of raw data, spike noise 301 can be replaced with zero-value information at high speed to obtain a tomographic image without stripes.

When the spike noise 301 is extracted in step S705, it is performed by using the mathematical fact that the absolute values of the raw data information at positions symmetrical about the zero frequency information on the raw data are equal. It was In this case, the information on the right half on the raw data is compared with the information on the left half of the centrally symmetric position one by one.
By replacing the large value information with the small value information, the spike noise 301 can be extracted and the extracted area information can be replaced at the same time. According to this, it is possible to simplify the processing and shorten the calculation time.

Further, when the information on the raw data is compared one by one, the information on the right half and the information on the left half are compared, but the information on the upper half and the information on the lower half may be compared one by one.

[0074]

As described above, according to the present invention,
Set the extraction area including spike noise on the raw data,
Since the information of the extraction area is replaced with information that does not affect the tomographic image, for example, zero value information or information at a position of central symmetry on the raw data, the raw data is obtained by performing a two-dimensional Fourier transform. The striped pattern generated in the tomographic image can be reliably removed in a state where the influence on the tomographic image is suppressed to a minimum, or there is no effect on the tomographic image. This has the effect of preventing misdiagnosis and making an accurate diagnosis.

[Brief description of drawings]

FIG. 1 is a block diagram showing an overall configuration of a magnetic resonance imaging apparatus.

FIG. 2 is a block diagram showing a computer unit according to the first embodiment.

FIG. 3 is a diagram showing a relationship between spike noise on raw data and a striped pattern on a tomographic image.

FIG. 4 is a flowchart showing the operation of the computer unit according to the first embodiment.

FIG. 5 is a diagram showing setting of an extraction region including spike noise on raw data.

FIG. 6 is a block diagram showing a computer unit according to the second embodiment.

FIG. 7 is a flowchart showing the operation of the computer unit according to the second embodiment.

FIG. 8 is a diagram showing a principle of automatically extracting spike noise on raw data.

[Explanation of symbols]

1 receiver coil 2 Transmit coil section 3 gradient coil 4 Static magnetic field generator 5 subject 6 Cradle part 7 Receiver 8 Detection unit 9 A / D converter 10, 60 Computer section 11 Operation part 12 Display 13 Scan controller section 14 Gradient drive 15 Transmitter 20 RF oscillator 101 magnet part 102 table section 103 control processing unit 201 Central control unit 202 High-speed image processing unit 203 main memory 204 image storage unit 205 input port 206 I / O port 207 Display control unit 208 Operation unit interface 209 shared bus 301 Spike noise 302 Main image information 303 tomographic image 601 Central control unit 602 High-speed image processing unit

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁷ Identification symbol FI theme code (reference) G01N 24/02 530Y (72) Inventor Koji Kato 127 GE Yokogawa Medical, 4-7 Asahigaoka, Hino City, Tokyo System Co., Ltd. (72) Inventor Tsunemoto Suzuki 127, 4-7 Asahigaoka, Hino City, Tokyo GE Yokogawa Medical Systems Co., Ltd. (72) Nobuyuki Miura 127, 4-7 Asahigaoka, Hino City, Tokyo GE Yokogawa Medical Systems Co., Ltd. F-term (reference) 4C096 AA20 AB11 AB22 AD02 AD13 AD14 AD25 DA04 DA08 DB07 DB19 DC05 DC27 5B057 AA09 BA06 BA24 BA26 CA08 CA12 CA16 CB08 CB12 CB16 CC03 CE02 DA08 DA17 DB02 DB09 DC30 DC36 5L096 CAA06 CA24 DA01 FA23

Claims (12)

[Claims] 1. An image processing apparatus for acquiring raw data, which is two-dimensional Fourier transform information of two-dimensional image information, and displaying the raw data, the extraction area including spike noise on the displayed raw data. An image processing apparatus comprising: a setting unit configured to set the setting information; and a replacement unit that replaces the replacement information of the extraction region extracted by the setting unit with the information of the extraction region. 2. A selection means for selecting replacement information of the extraction area extracted by the setting means, wherein the replacement means replaces the replacement information selected by the selection means with information of the extraction area. Claim 1 characterized by
The image processing device according to item 1. 3. The image processing apparatus according to claim 1, wherein the setting unit uses an area surrounding the spike noise on the raw data set by manual operation as an extraction area. 4. The high frequency side of an extraction line set by a manual operation that divides the raw data information into a low frequency side and a high frequency side including spike noise is used as an extraction region. The image processing apparatus according to claim 1 or 2. 5. The image processing apparatus according to claim 1, wherein the selection unit uses zero value information as the alternative information. 6. The selection means uses, as alternative information, the raw data information of an area having a relationship between a frequency and an anti-frequency of an extraction area selected by the extraction means from the raw data information. The image processing device according to claim 1. 7. The setting means successively compares, in the raw data information, information of a certain frequency and information having a relationship of the frequency and an anti-frequency, and a difference between the two information values exceeds a predetermined value. The image processing apparatus according to claim 1 or 2, wherein the extraction of the extracted information area is performed automatically. 8. The setting means automatically compares, in the raw data information, information having the same frequency, and sets areas of different information exceeding a predetermined value as extraction areas by an automatic operation. The image processing apparatus according to claim 1 or 2. 9. An image processing method for acquiring raw data, which is two-dimensional Fourier transform information of two-dimensional image information, and displaying the raw data, wherein an extraction area including spike noise on the raw data is set. An image processing method comprising: a setting step; and a replacing step of replacing the alternative information of the extraction area extracted by the setting step with the information of the extraction area. 10. The method further comprises a selecting step of selecting alternative information of the extraction area extracted by the setting step, wherein the replacing step replaces the alternative information selected by the selecting step with the information of the extraction area. The image processing method according to claim 9, which is characterized in that. 11. A magnetic resonance imaging apparatus for acquiring raw data, which is two-dimensional Fourier transform information of two-dimensional tomographic image information, and displaying the raw data for image processing, wherein spikes on the raw data information are provided. A magnetic resonance imaging apparatus comprising: a setting unit that sets an extraction region including noise; and a replacement unit that replaces the replacement information of the extraction region extracted by the setting unit with the information of the extraction region. 12. The selecting means further comprises selecting means for selecting alternative information of the extraction area extracted by the setting means, wherein the replacing means replaces the alternative information selected by the selecting means with the information of the extraction area. The magnetic resonance imaging apparatus according to claim 11, wherein the magnetic resonance imaging apparatus is a magnetic resonance imaging apparatus.