JP2011067559A - Image processing method for intensity image formed from mr signal, image processing program, and image processor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing method for measuring an image by sub-pixel accuracy, etc. <P>SOLUTION: The image processing method for an intensity image formed from an MR signal includes: a setting step of setting at least N echo times; an acquisition step of acquiring the respective intensity values of pixels in the intensity images imaged by the respectively set echo times; and a first calculation step of calculating the content rate (λ) of a tissue to be measured in the pixel through the use of the obtained at least N intensity values and a theoretical prediction expression having N parameters including the content rate (λ) of the tissue. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an image processing method, an image processing program, and an image processing apparatus for an intensity image generated from an MR signal obtained from a magnetic resonance imaging (MRI) apparatus.

  Conventionally, various image processing methods for processing a signal obtained from an MRI apparatus as an image have been proposed and widely used for medical diagnosis and the like (for example, Patent Document 1).

JP 2009-183689 A

  By the way, since the limit length in the image measurement is the length indicated by one pixel, it is necessary to improve the spatial resolution of the imaging device in order to improve the measurement accuracy. Although it is possible to improve the measurement accuracy by improving the high spatial resolution, medical equipment such as an MRI apparatus is very expensive, so even if existing equipment is improved, a large cost is required. is there. Even if the specific performance is improved, the accuracy of diagnostic imaging may be affected at the expense of other performance. For this reason, manufacturers have had to be cautious about improving medical devices.

  From such a background, the present inventor believes that by performing image processing using physical properties related to an intensity image generated from an MR signal obtained from an MRI apparatus, image measurement with subpixel accuracy becomes possible. It was.

  An object of the present invention is to provide an image processing method, an image processing program, and an image processing apparatus for an intensity image that enables image measurement with sub-pixel accuracy.

  In order to achieve the above object, the invention according to claim 1 is an image processing method for an intensity image generated from an MR signal, comprising: a setting step for setting at least N echo times; and each of the set echoes The acquisition step of acquiring the intensity values of the pixels of the intensity image captured at time, the theoretical prediction having N parameters including the acquired at least N intensity values and the content rate (λ) of the tissue to be measured And a first calculation step of calculating the content (λ) of the tissue in the pixel using an equation.

  The invention according to claim 2 is the image processing method according to claim 1, wherein the first calculation step uses a simultaneous equation in which the acquired at least N intensity values are substituted into the theoretical prediction formula. The content ratio (λ) of the tissue in the pixel is calculated algebraically.

  The invention according to claim 3 is the image processing method according to claim 1, wherein the first calculation step fits the acquired at least N intensity values to the theoretical prediction formula, and The content rate (λ) of the structure in is calculated.

An invention according to claim 4 is the image processing method according to any one of claims 1 to 3, wherein a pixel region including M pixels including a tissue to be measured is specified from the intensity image. A specific calculation step; a second calculation step for calculating the tissue content rate (λ _j ) (1 ≦ j ≦ M) in the pixel using the first calculation step for each pixel; And a third calculation step of calculating the tissue content rate (λ) in the M pixels by using the presence (λ _j ) of the tissue in the pixel.

  The invention according to claim 5 is the image processing method according to any one of claims 1 to 4, wherein a determination step of determining L pixel volume spaces in which a tissue to be measured is included from the intensity image. For each of the L pixel volume spaces, the tissue content rate (λt) (1 ≦ t ≦ L) in the pixel volume space using the first calculation step or the second calculation step. The fifth calculation step of calculating the tissue content rate (λ) in the volume of the L pixel volume spaces is calculated using a fourth calculation step for calculating the tissue content rate (λt) in each intensity image. And a calculating step.

  The invention according to claim 6 is a program for causing a computer to execute processing, wherein the computer includes a setting step for setting at least N echo times, and an intensity image captured at each of the set echo times. Using the obtaining step of obtaining the intensity values of the pixels, the obtained at least N intensity values, and the theoretical prediction formula having N parameters including the content rate (λ) of the tissue to be measured, And a first calculation step of calculating a content ratio (λ) of the tissue in the pixel.

  The invention according to claim 7 is the image processing program according to claim 6, wherein the first calculation step uses a simultaneous equation in which the acquired at least N intensity values are substituted into the theoretical prediction formula. The content ratio (λ) of the tissue in the pixel is calculated algebraically.

  The invention according to claim 8 is the image processing program according to claim 6, wherein the first calculation step fits the acquired at least N intensity values to the theoretical prediction formula, and The content rate (λ) of the structure in is calculated.

The invention according to claim 9 is the image processing program according to any one of claims 6 to 8, and specifies a pixel region including M pixels including a tissue to be measured from the intensity image. A specific calculation step; a second calculation step for calculating the tissue content rate (λ _j ) (1 ≦ j ≦ M) in the pixel using the first calculation step for each pixel; And a third calculation step of calculating the tissue content rate (λ) in the M pixels using the presence (λ _j ) of the tissue in the pixel.

  The invention according to claim 10 is the image processing program according to any one of claims 6 to 9, wherein a determination step of determining L pixel volume spaces in which a tissue to be measured is included from the intensity image. For each of the L pixel volume spaces, the tissue content rate (λt) (1 ≦ t ≦ L) in the pixel volume space using the first calculation step or the second calculation step. The fifth calculation step of calculating the tissue content rate (λ) in the volume of the L pixel volume spaces is calculated using a fourth calculation step for calculating the tissue content rate (λt) in each intensity image. And a calculation step.

  According to an eleventh aspect of the present invention, there is provided an image processing apparatus for an intensity image generated from an MR signal, setting means for setting at least N echo times, and an intensity image captured at each of the set echo times. Using the acquisition means for acquiring the intensity values of each of the pixels, the obtained at least N intensity values, and a theoretical prediction formula having N parameters including the content rate (λ) of the tissue to be measured, And a first calculating means for calculating the content (λ) of the tissue in the pixel.

  The invention according to claim 12 is the image processing apparatus according to claim 11, wherein the first calculation unit uses a simultaneous equation in which the acquired at least N intensity values are substituted into the theoretical prediction formula. The content ratio (λ) of the tissue in the pixel is calculated algebraically.

  The invention according to claim 13 is the image processing device according to claim 11, wherein the first calculation unit fits the acquired at least N intensity values to the theoretical prediction formula, and The content rate (λ) of the structure in is calculated.

The invention according to claim 14 is the image processing device according to any one of claims 11 to 13, wherein a pixel region consisting of M pixels including a tissue to be measured is specified from the intensity image. And a second calculating unit that calculates the tissue content rate (λ _j ) (1 ≦ j ≦ M) in the pixel by using the first calculating unit for each pixel, And third calculation means for calculating the tissue content rate (λ) in the M pixels using the presence (λ _j ) of the tissue in each pixel.

  The invention according to claim 15 is the image processing apparatus according to any one of claims 11 to 14, wherein the determining unit determines L pixel volume spaces in which a tissue to be measured is included from the intensity image. For each of the L pixel volume spaces, the tissue content rate (λt) (1 ≦ t ≦ L) in the pixel volume space using the first calculation unit or the second calculation unit A fifth calculation means for calculating the tissue content rate (λ) in the volume of the L pixel volume spaces using a fourth calculation means for calculating the tissue content rate (λt) in each intensity image. And calculating means.

  According to the present invention, image measurement with sub-pixel accuracy is possible, and therefore the size (content ratio) of a structure having a pixel size or less can be measured with high accuracy. Also, the size of an object larger than the pixel size (for example, the diameter of a blood vessel) can be measured with subpixel accuracy. In addition, since the present invention can perform processing based on MR signals acquired from an existing MRI apparatus, the introduction cost can be greatly reduced.

It is a figure which shows the whole structure of an MRI system. It is explanatory drawing which shows the relationship between the structure | tissue of a measuring object, and pixel size. It is a flowchart of an image processing method. It is explanatory drawing which shows the relationship between the structure | tissue of a measuring object, and pixel size. It is a flowchart of an image processing method. It is a flowchart of an image processing method. It is explanatory drawing which shows an experimental result. It is explanatory drawing which shows an experimental result. It is explanatory drawing which shows an experimental result.

[Outline of the present invention]
The signal intensity for each pixel (voxel) of the MR signal obtained from the MRI apparatus depends on the hydrogen atom density of the tissue included in the pixel (voxel). Therefore, it is known that when two or more kinds of substances (structures) having different hydrogen densities are mixed in one pixel (voxel), the signal intensity is determined by the mixing ratio.

  In the present invention, image measurement is performed by processing an intensity image (or an intensity value of each pixel constituting the intensity image) generated from the MR signal by utilizing this technical property. Thereby, image measurement with sub-pixel accuracy is possible, and the existence ratio of a plurality of tissues included in one pixel can be determined with high accuracy. In addition, it is possible to measure the abundance of structures existing in a region defined by a plurality of pixels with sub-pixel accuracy. That is, the present invention is an epoch-making invention that overturns the conventional common sense that the pixel size is considered to be the limit of measurement accuracy.

[First Embodiment]
Hereinafter, a first embodiment of the present invention will be specifically described with reference to the drawings.

[Overall configuration of MRI system]
First, an MRI system to which this embodiment is applied will be described. As shown in FIG. 1, the MRI system 1 includes an MRI apparatus 10 and a signal processing unit 20.

[MRI apparatus 10]
The MRI apparatus 10 includes a static magnetic field coil unit 11, a gradient magnetic field coil unit 12, and an RF (Radio Frequency) coil unit 13. These have, for example, a substantially cylindrical shape, and are arranged so that their central axes (not shown) are coaxial with each other. In addition, a bed portion 15 that supports the subject 16 is provided in a plane including the central axis. The bed unit 15 is installed in the bore 14 (internal space) of the MRI apparatus 10. The subject 16 on the bed unit 15 is carried into and out of the bore 14 by the movement of the bed unit 15 by a transport unit (not shown). The configurations of the static magnetic field coil unit 11, the gradient magnetic field coil unit 12, and the RF coil unit 13 are well-known and are not relevant to the understanding of the present invention, and thus detailed description thereof is omitted.

[Signal Processing Unit 20]
The signal processing unit 20 includes a static magnetic field power source 21, a gradient magnetic field power source 22, a transmission unit 23, a reception unit 24, a sequence control unit 25, and an information processing unit 30.
The static magnetic field power source 21, the gradient magnetic field power source 22, and the transmission unit 23 drive the respective elements by transmitting signals to the static magnetic field coil unit 11, the gradient magnetic field coil unit 12, and the RF coil unit 13, respectively. The receiving unit 24 detects the MR signal received by the RF coil unit 13 and generates complex data (raw data). Note that the configurations of the static magnetic field power supply 21, the gradient magnetic field power supply 22, the transmission unit 23, and the reception unit 24 are well-known and are not relevant to the understanding of the present invention, and thus detailed description thereof is omitted.
The sequence control unit 25 outputs a control signal for controlling the gradient magnetic field power source 22, the transmission unit 23, and the reception unit 24 based on a pulse sequence from the information processing unit 30 described later. The complex data acquired from the receiving unit 24 is output to the information processing unit 30.

  In the present embodiment, various pulse sequences can be used as long as they can cause a change for each TE (echo time) due to the partial volume effect. In other words, any gradient echo pulse sequence may be used. For example, it is considered that the present embodiment can be applied to images having various contrasts created by various sequences, such as images created using SSFP (Steady State Free Precession). That is, it is considered that the technique of this embodiment can be applied to various MR medical images.

  Note that it is considered that the spin-echo sequence is not preferable because the phase effect may be canceled. However, if a spin echo system sequence or the like in which an incidental pulse capable of having a phase effect is inserted in the future can be realized, this embodiment can be used even when a spin echo system sequence is used. Can be applied.

  That is, in this embodiment, it is possible to use a pulse sequence that can create any phase (difference).

  The information processing unit 30 includes a control unit 31, an input unit 32, a display unit 33, and a storage unit 34. The control unit 31 controls the entire information processing unit 30 and includes a CPU (Central Processing Unit), an internal memory, and the like. The control unit 31 reads and executes a program (image processing program) stored in the internal memory of the CPU or the storage unit 34, thereby executing setting means, acquisition means, first calculation means, identification means, and second calculation means. , Third calculation means, determination means, fourth calculation means, fifth calculation means, and the like.

  The input unit 32 is a device that captures information from the user as digital data into the information processing unit 30, and includes, for example, a keyboard, a mouse, a scanner, and the like. The display unit 33 displays a result processed by the control unit 31, a dialog for inputting data such as an imaging condition, and the like, and is configured by an LCD, for example. The storage unit 34 stores various programs for controlling the MRI apparatus 10 and the signal processing unit 20 (for example, an image processing program for an image processing method executed by the control unit 31).

  In this embodiment, a plurality of intensity images based on MR signals obtained by setting different TE (echo times) from the MRI apparatus 10 are acquired, and temporal transitions of intensity values of pixels of the intensity images are used. The content rate of the tissue to be measured included in the pixel is obtained. In the present embodiment, the content ratio λ of the tissue to be measured included in the pixel can be calculated. For example, as shown in FIG. 2, when it is known that the shape of the tissue to be measured on the two-dimensional plane is a circle, the diameter of the tissue to be measured can be calculated based on the calculated λ. It becomes. Note that the pixel value of the pixel used in the present embodiment is, for example, 12 bits (4096 gradations).

[Principle of this embodiment]
Next, the principle of this embodiment will be described. In the following description, the content rate λ on the two-dimensional plane of the tissue to be measured is calculated. Homogeneous tissue A and tissue B exist in a region of area ξ _p (region indicated by one pixel), and the area of tissue B in the region is σ. At this time, the tissue B is included in this region (pixel) at a ratio of λ≡ (σ / ξ _p ) (0 ≦ λ ≦ 1).

Here, it is included in one target pixel,
The hydrogen atom densities of the two types of structures A and B are expressed as ρ _A , ρ _B ,
The time (T ^* ₂ ) representing the standard of signal attenuation during imaging of the tissues A and B is expressed as T ^* _2A , T ^* _2B ,
The phase values of the tissues A and B are expressed as θ _A , θ _B ,
The content of tissue B is λ,
When
The complex data S (t) at time t after the RF pulse is applied to the tissues A and B is

It can be expressed as (Note that j ² = −1 in “Equation 1”.)

Here, the technical properties of each parameter of “Equation 1” will be described.
The value of S (t) tends to increase as the values of ρ _A and ρ _B increase (that is, as the density increases).
The smaller the values of T ^* _2A and T ^* _2B , the greater the signal attenuation over time. Therefore, the smaller the values of T ^* _2A and T ^* _2B , the more difficult it is to detect over time.
θ _A and θ _B are theoretically considered to be proportional to TE (that is, a linear function of time and TE) and are related to the response (susceptibility) of the tissue to magnetism. For this reason, since the phase value changes if the structure is different, in the present embodiment, the difference λ is calculated as a partial volume effect on the phase, and the content rate λ is calculated. It is considered that the difference in structure is largely reflected as the difference in magnetic susceptibility, and it becomes possible to recognize the difference in structure more effectively than the conventional method using T ^* ₂ .

  In addition, “Equation 1” is an expression that assumes that two types of tissues exist in one pixel, and thus the number of parameters indicating the content rate is one. When three or more types of tissues exist in the pixel, the number of parameters indicating the content rate of each tissue is “number of types included in pixel−1”.

  Since there are a plurality of parameters in “Equation 1”, in principle, it is possible to obtain λ as a solution by collecting data by the number of parameters and solving simultaneous equations based on them. However, since it is non-linear with respect to each parameter of “Equation 1”, it is considered difficult to obtain the solution. Therefore, in this embodiment, the solution λ is obtained by the following method.

In “Expression 1”, the square value | S (t) | ² of the absolute value of the complex data S (t) (hereinafter referred to as the square intensity value) is

It becomes.

  In “Equation 2”,

It is said.

As “Equation 2” indicates, only the phase difference between tissues affects the intensity value, so the number of parameters is six.
That is, the square intensity value | S (t) | ² is an equation having six parameters of “T ^* _2A ”, “λ”, “ρ _A ”, “ρ _B ”, “ΔT ^* ₂ ”, and “θ”. Is expressed as
Here, ΔT ^* ₂ is a value related to the difference in signal attenuation rate between the tissues A and B (see “Equation 3”). θ is a value called a relative phase, and is a value that affects the change of the square intensity value | S (t) | ² . Since the phase value depends on TE as described above, this relative phase also depends on TE. The relative phase can also be considered as a function of the difference in the response (susceptibility) of the tissue to magnetism.

In the present embodiment, the square intensity values | S (t) | ² at a plurality of different times “t” are acquired, and the values and theoretical prediction formulas (for example, “Equation 2” described above and “Equation 8” described later) are obtained. Can be used to obtain the value of each parameter. The above parameters may be calculated algebraically by setting simultaneous equations using a theoretical prediction formula, or may be calculated by fitting to the theoretical prediction formula.

In the case of calculating algebraically by setting simultaneous equations based on “Equation 2”, since “Equation 2” has 6 parameters, the square intensity value | S (t) | ² at least at 6 times is required. . Note that when the algebra calculation is omitted and the calculation is performed by fitting the theoretical prediction formula of “Equation 2”, the six square intensity values | S (t) | ² are not necessarily required.

Also, assuming that there is no significant difference in the time (T ^* ₂ ) representing the signal attenuation between tissues,

Can be considered.
Also, assuming that tissue A and tissue B are biological tissues and there is no significant difference in hydrogen atom density (water molecule density),

Can be considered. If this assumption is satisfied, square intensity value | S (t) | ² is

It can be expressed as Therefore, when measuring a living tissue (a blood vessel contained in the brain parenchyma, a fat contained in the liver, or the like having another tissue in the tissue, etc.), “Equation 7” As shown in FIG. 4, the number of parameters is four.
That is, the square intensity value | S (t) | ² is expressed as an equation having four parameters “T ^* _2A ”, “λ”, “ρ ₀ ”, and “θ”. That is, in this case, each parameter can be acquired by acquiring the intensity value | S (t) | ² acquired at four different times.

However, even when simultaneous equations are established with a reduced number of parameters in this way, the acquired data itself may have a large error. The content rate λ may be calculated by fitting with a prediction formula.
In other words, the square intensity value | a ^2, | S (t)

The four parameters including “λ” can be calculated using “Equation 8” as a theoretical prediction formula. Note that k is a proportionality constant, and that θ is a linear function of TE. The content rate (λ) can be calculated with subpixel accuracy.

In this case, if algebraically calculate make a simultaneous equation based on the "number 8", because there are four parameters in the "number 8" square intensity value at least four times | S (t) | ² is Necessary. When the algebraic calculation is omitted and the calculation is performed by fitting the theoretical prediction formula of “Equation 8”, the four square intensity values | S (t) | ² are not necessarily required.

[Image Processing Method Performed by Control Unit 31]
Next, an image processing method executed by the control unit 31 will be described with reference to the flowchart of FIG.

  First, in step S10, the control unit 31 performs an imaging process. In this imaging process, a plurality of images (complex data) captured using different TEs are captured. Specifically, four images (complex data) are captured using a dual echo sequence or a multi-echo sequence by a gradient echo method or the like. The captured image (complex data) is stored in the storage unit 34 in association with the TE. The number of images (complex data) captured in step S10 (the number of TEs) varies depending on the theoretical prediction formula and the calculation method used. For example, when calculating algebraically using “Equation 2”, the number of TEs is at least 6, and when calculating algebraically using “Equation 8”, the number of TEs is at least 4. In addition, when calculating by fitting to a prediction theoretical formula, it may be less than the case of calculating algebraically.

In step S <b> 11, the control unit 31 acquires complex data from the storage unit 34.
In step S12, the control unit 31 acquires the TE corresponding to the complex data acquired in step S11 from the storage unit 34.

In step S13, the control unit 31 generates an intensity image (an image in which each pixel is expressed as an intensity value) based on the acquired complex data. In addition, since the process itself which produces | generates an intensity | strength image from complex data is well-known, detailed description is abbreviate | omitted.
In step S <b> 14, the control unit 31 specifies a pixel whose content rate is to be measured, and stores the intensity value of the pixel in the storage unit 34. In addition, although the pixel which measures a content rate can be specified based on the input instruction from a user, for example, you may comprise so that it may specify automatically.

In step S15, the control unit 31 determines whether or not a predetermined number (the number of TEs in step S10) of data (square intensity value | S (t) | ² ) has been acquired. When it is determined that the predetermined number of data has not been acquired (step S15: NO), the control unit 31 shifts the process to step S11 and repeats the process from step S11. On the other hand, when determining that the predetermined number of data has been acquired (step S15: YES), the control unit 31 shifts the process to step S16.

In step S <b> 16, the control unit 31 calculates the content rate λ using the acquired data and the theoretical prediction formula (the formula shown by “Equation 2” or “Equation 8”). Here, it may be calculated algebraically or may be calculated by fitting to a theoretical prediction formula.
In step S <b> 17, the control unit 31 outputs the calculated λ to the display unit 33 and / or the storage unit 34. The calculated λ may be output to an external device or the like.

As described above, in the present embodiment, the tissue content ratio λ in the pixel can be obtained with sub-pixel accuracy by using the physical property related to the intensity image generated from the MR signal obtained from the MRI apparatus 10. it can. Moreover, since the MRI system 1 according to the present embodiment can be constructed using the existing MRI apparatus 10, the introduction cost can be significantly reduced. Further, when performing image processing by reducing the number of parameters using the physical properties of the living tissue, the load on the computer can be reduced, and as a result, the calculation cost of the information processing unit 30 can be reduced. be able to.
As described above, according to the present embodiment, the epoch-making image processing method, the image processing program, and the image processing apparatus that overturn the conventional common sense that the pixel size is considered to be the limit of the measurement accuracy are mechanically modified. Can be provided inexpensively.

[Second Embodiment]
Next, a second embodiment of the present invention will be described. In the present embodiment, even if the tissue to be measured exists over a plurality of pixels, the content rate of the tissue to be measured can be calculated.

In the above-described embodiment, it is assumed that the tissue is completely included in one pixel. However, in a tissue such as an actual blood vessel, even if the diameter of the blood vessel is equal to or smaller than the pixel size, two or more It may exist across pixels (see FIG. 4). In such a case, the content rate λ cannot be calculated even if the obtained square intensity value | S (t) | ² simply uses the theoretical prediction formulas of “Equation 2” and “Equation 8”.

Therefore, in the present embodiment, a pixel set of a region of interest (ROI, also referred to as a pixel region) that is considered to surely contain the tissue to be measured is extracted, and the area S _R of this pixel set is extracted. The content rate λ of the included tissue B (area σ) is calculated.

  When the number of pixels included in the ROI is M, the content λj of the tissue B (area σj) included in the j-th (0 ≦ j ≦ M) pixel (area Sp) is expressed by “Equation 2” or “Equation 8”. "Can be obtained. And the content ratio λ of the tissue B included in the pixel set is

Can be calculated as That is, the average value of the calculated N λj is the content rate λ. In addition, when calculating the sum of λj without averaging, the sum represents “how many times the area has compared with the size of one pixel”, and has the same meaning as the content rate. It will be the amount you have.

[Image Processing Method Performed by Control Unit 31]
Next, an image processing method executed by the control unit 31 will be described with reference to the flowchart of FIG. Of the following steps, the description of the same processing as that in the first embodiment will be simplified.

  First, in step S30, the control unit 31 executes an imaging process. In this imaging process, a plurality of images (complex data) captured using different TEs are captured. The captured image (complex data) is stored in the storage unit 34 in association with the TE. The number of images (complex data) captured in step S30 (the number of TEs) varies depending on the theoretical prediction formula and calculation method used.

In step S <b> 31, the control unit 31 acquires complex data from the storage unit 34.
In step S32, the control unit 31 acquires the TE corresponding to the complex data acquired in step S31 from the storage unit 34.
In step S33, the control unit 31 generates an intensity image based on the acquired complex data.

In step S <b> 34, the control unit 31 specifies a pixel in the specified ROI and stores the intensity value of the pixel in the storage unit 34. The ROI can be specified based on an input instruction from the user, but may be configured to be specified automatically.
In step S35, the control unit 31 determines whether or not the intensity values of all the pixels in the ROI have been acquired. If it is determined that there is an intensity value that has not been acquired (step S35: NO), the control unit 31 shifts the process to step S34 and repeats the process of step S34. On the other hand, when determining that the intensity values of all the pixels have been acquired (step S35: YES), the control unit 31 shifts the process to step S36.

  In step S36, the control unit 31 determines whether or not a predetermined number of data (the number of TEs in step S30) has been acquired. When determining that the predetermined number of data has not been acquired (step S36: NO), the control unit 31 shifts the process to step S31 and repeats the process from step S31. On the other hand, when determining that a predetermined number of data has been acquired (step S36: YES), the control unit 31 shifts the process to step S37.

In step S <b> 37, the control unit 31 calculates the content rate λj for each pixel included in the ROI, using the acquired data and the prediction theoretical formula (the formula shown by “Formula 2” or “Formula 8”). . Here, it may be calculated algebraically or may be calculated by fitting to a theoretical prediction formula.
In step S <b> 38, the control unit 31 calculates the content rate λ using “Equation 9”.
In step S <b> 39, the control unit 31 outputs the calculated λ to the display unit 33 and / or the storage unit 34. The calculated λ may be output to an external device or the like.

  As described above, in the present embodiment, even when the tissue to be measured exists across a plurality of pixels, the content rate of the tissue can be calculated. Further, according to the present embodiment, since the content rate of the tissue to be measured existing in the ROI can be calculated, it is finer than the spatial resolution depending on the pixel size that could not be imaged by the conventional method. The density of tissues such as capillary blood vessels can be measured. Thereby, new diagnostic image information representing a disease state can be provided.

[Third Embodiment]
Next, a third embodiment of the present invention will be described. In the first and second embodiments described above, the content rate of the tissue to be measured in the area is calculated. By applying the above-described processing to a continuous slice surface (image), the volume can be increased. The content of the existing tissue can be calculated.

  When the volume defined by successive slice planes is V, the tissue content in each target pixel volume space included in the volume is λt, and the number of pixel volume spaces is L, the tissue content in volume V λ is

It can be expressed as When the tissue content rate in each pixel volume space is calculated according to the first embodiment, λt can be calculated using “Equation 2” or “Equation 8”. Further, when the tissue content rate in each pixel volume space is calculated according to the second embodiment, λt can be calculated using “Equation 9”.

[Image Processing Method Performed by Control Unit 31]
Next, an image processing method executed by the control unit 31 will be described with reference to the flowchart of FIG. Of the following steps, the description of the same processing as that of the first embodiment or the second embodiment described above will be simplified.

  First, the control part 31 performs an imaging process in step S51. In this imaging process, a plurality of images (complex data) captured using different TEs are captured. The captured image (complex data) is stored in the storage unit 34 in association with the TE. The number of images (complex data) captured in step S51 (the number of TEs) varies depending on the theoretical prediction formula and the calculation method used. In step S51, complex data of continuous slice planes is acquired.

In step S <b> 52, the control unit 31 acquires complex data from the storage unit 34.
In step S53, the control unit 31 acquires the TE corresponding to the complex data acquired in step S52 from the storage unit 34.
In step S54, the control unit 31 generates an intensity image based on the acquired complex data.

In step S <b> 55, the control unit 31 specifies a pixel in the specified ROI and stores the intensity value of the pixel in the storage unit 34. The ROI can be specified based on an input instruction from the user, but may be configured to be specified automatically.
In step S56, the control unit 31 determines whether or not the intensity values of all the pixels in the ROI have been acquired. If it is determined that there is an intensity value that has not been acquired (step S56: NO), the control unit 31 shifts the process to step S55 and repeats the process of step S57. On the other hand, when determining that the intensity values of all the pixels have been acquired (step S56: YES), the control unit 31 shifts the processing to step S57.

  In step S57, the control unit 31 determines whether a predetermined number of data (the number of TEs in step S51) has been acquired. When determining that the predetermined number of data has not been acquired (step S57: NO), the control unit 31 shifts the process to step S52 and repeats the process from step S52. On the other hand, when determining that a predetermined number of data has been acquired (step S57: YES), the control unit 31 shifts the process to step S58.

In step S58, the control unit 31 calculates the tissue content rate λt for each target pixel volume space.
In step S59, the control unit 31 calculates the content rate λ using “Equation 10”.
In step S <b> 60, the control unit 31 outputs the calculated λ to the display unit 33 and / or the storage unit 34. The calculated λ may be output to an external device or the like.

  As described above, in the present embodiment, the content of the tissue existing in the volume can be calculated.

[Experimental result]
Next, as an example of the experimental result of the present invention, the experimental result of the above-described second embodiment will be described below. The experimental environment is as follows.
(1) 3.0TMRI equipment: Achieva 3.0T (Philips)
(2) Gelatin phantom (20% concentration)
(3) 8ch SENSE HEAD COIL
(4) 4096 gradations

  In this experiment, a gelatin phantom including a simulated blood vessel having a known diameter is arranged so that the axial direction of the simulated blood vessel is parallel to the static magnetic field direction, and TR (Repetition Time, signal) is obtained with a plurality of very close TEs. (Interval for applying the RF wave necessary for this) imaging is performed using the shortest TR for each TE and under the same conditions other than that. A square ROI is set around each simulated blood vessel on the captured image. The change due to TE of the image signal in the ROI is measured, and λ of each pixel is calculated from the theoretical prediction. The simulated blood vessel diameter is determined from λ of the entire ROI and the pixel size, and compared with the true value of the simulated blood vessel diameter.

FIG. 7 shows data obtained by manually setting an ROI so that a blood vessel (simulated blood vessel outer diameter: about 0.7 mm) is placed in a square ROI composed of nine pixels in a 512 × 512 matrix (six TEs). (Data obtained in the above). In the figure, “Signal” indicates the square intensity value | S (t) | ² .

FIG. 8 shows each parameter calculated by fitting four values of the square intensity values | S (t) | ² obtained in FIG. 7 to the theoretical prediction formula shown in “Equation 8” for each pixel. ing. In the figure, when the true value is compared with the actual measurement value, the error is 3.7%, which indicates that the blood vessel diameter can be measured with very high accuracy according to the present invention. That is, the experimental results prove that the content λ obtained by the present invention is measured with very high accuracy.

  FIG. 9 shows the results of experiments conducted in the same manner by changing the parameters of the number of pixels and the blood vessel outer diameter. Also in the figure, it is shown that the blood vessel diameter can be measured with very high accuracy according to the present invention (the result shown in FIG. 8 (matrix: 512 × 512, blood vessel diameter true value: 0.674 mm) and Although the results shown in FIG. 9 are different, this difference is due to different experimental data).

  In this way, simulated blood vessel diameters that are smaller than the pixel size and cannot be accurately measured by existing methods due to the effect of the partial volume effect can be measured with high accuracy with an error of 10% or less in almost all cases. It is. The present invention is a useful image measurement method that can measure a blood vessel diameter with high accuracy without requiring a special device and without depending on spatial resolution. In addition, the present invention can be calculated even when there are a plurality of blood vessels in the ROI, and can also be applied to the measurement of the blood vessel area at a site where blood vessels are densely packed in the ROI.

As described above, according to the present invention, it is possible to perform image measurement with sub-pixel accuracy using the technical features of the intensity image. Thereby, it is considered that the density of blood vessels contained in a tissue such as cancer can be determined. In the case of cancer or the like, a large amount of nutritional blood vessels that are necessary for growth are stretched. By observing this amount, it can be expected that it becomes a standard for image diagnosis for judging malignancy. The current determination of the grade of malignancy takes time for tumor markers, follow-up, etc., and thus may burden the patient in terms of time.
According to the present invention, it is possible to provide an innovative diagnostic imaging method that can be determined only by one imaging. In addition, it is considered that there is a certain demand for such diagnostic imaging in conjunction with the expansion of future screening work.

  As described above, the present invention can be applied to various medical treatments. Further, the present invention can be widely used in fields where MRI apparatuses are used, such as making it possible to determine the size of an accurate structure by nondestructive inspection even in fields where MRI apparatuses are used for purposes other than medical treatment.

  Furthermore, since the present invention does not require improvement of an existing MRI apparatus and is processing by software, demand is expected not only for a company that develops MRI apparatuses but also for companies that develop image processing software.

  The present invention is not limited to the above-described embodiment, and it goes without saying that various improvements and modifications can be made without departing from the scope of the present invention.

  Further, each flowchart described above is merely an example, and the processing may be realized by another flowchart as long as a result equivalent to the processing of each flowchart can be obtained. The present invention can also be realized as a recording medium or the like on which an image processing program is recorded.

DESCRIPTION OF SYMBOLS 1 MRI system 10 MRI apparatus 20 Signal processing part 26 Information processing part 31 Control part

Claims (15)

An image processing method for an intensity image generated from an MR signal,
A setting step for setting at least N echo times;
An acquisition step of acquiring the intensity value of each pixel of the intensity image captured at each set echo time;
Using the acquired at least N intensity values and a theoretical prediction formula having N parameters including the tissue content (λ) to be measured, the tissue content (λ) in the pixel is calculated. A first calculating step to:
An image processing method comprising: The first calculation step algebraically calculates the tissue content rate (λ) in the pixel using simultaneous equations obtained by substituting the acquired at least N intensity values into the theoretical prediction formula.
The image processing method according to claim 1. The first calculation step fits the acquired at least N intensity values to the theoretical prediction formula to calculate the tissue content rate (λ) in the pixel.
The image processing method according to claim 1. A specifying step of specifying a pixel region consisting of M pixels including the tissue to be measured from the intensity image;
A second calculation step for calculating the tissue content rate (λ _j ) (1 ≦ j ≦ M) in the pixel using the first calculation step for each of the pixels;
A third calculation step of calculating the tissue content (λ) in the M pixels using the presence (λ _j ) of the tissue in each pixel;
The image processing method according to claim 1, further comprising: A determination step of determining L pixel volume spaces including the tissue to be measured from the intensity image;
For each of the L pixel volume spaces, the tissue content rate (λt) (1 ≦ t ≦ L) in the pixel volume space is calculated using the first calculation step or the second calculation step. A fourth calculation step,
A fifth calculation step of calculating the tissue content rate (λ) in the volume of the L pixel volume spaces using the tissue content rate (λt) in each intensity image;
The image processing method according to claim 1, further comprising: On the computer,
A setting step for setting at least N echo times;
An acquisition step of acquiring the intensity value of each pixel of the intensity image captured at each set echo time;
Using the acquired at least N intensity values and a theoretical prediction formula having N parameters including the tissue content (λ) to be measured, the tissue content (λ) in the pixel is calculated. A first calculating step to:
An image processing program for executing The first calculation step algebraically calculates the tissue content rate (λ) in the pixel using simultaneous equations obtained by substituting the acquired at least N intensity values into the theoretical prediction formula.
An image processing program according to claim 6. The first calculation step fits the acquired at least N intensity values to the theoretical prediction formula to calculate the tissue content rate (λ) in the pixel.
An image processing program according to claim 6. A specifying step of specifying a pixel region consisting of M pixels including the tissue to be measured from the intensity image;
A second calculation step for calculating the tissue content rate (λ _j ) (1 ≦ j ≦ M) in the pixel using the first calculation step for each of the pixels;
A third calculation step of calculating the tissue content (λ) in the M pixels using the presence (λ _j ) of the tissue in each pixel;
The image processing program according to claim 6, wherein the image processing program is executed. A determination step of determining L pixel volume spaces including the tissue to be measured from the intensity image;
For each of the L pixel volume spaces, the tissue content rate (λt) (1 ≦ t ≦ L) in the pixel volume space is calculated using the first calculation step or the second calculation step. A fourth calculation step,
A fifth calculation step of calculating the tissue content rate (λ) in the volume of the L pixel volume spaces using the tissue content rate (λt) in each intensity image;
The image processing program according to claim 6, wherein the image processing program is executed. An image processing apparatus for an intensity image generated from an MR signal,
Setting means for setting at least N echo times;
Acquisition means for acquiring the intensity value of each pixel of the intensity image captured at each set echo time;
Using the acquired at least N intensity values and a theoretical prediction formula having N parameters including the tissue content (λ) to be measured, the tissue content (λ) in the pixel is calculated. First calculating means for
An image processing apparatus comprising: The first calculation means algebraically calculates the content (λ) of the tissue in the pixel using simultaneous equations obtained by substituting the acquired at least N intensity values into the theoretical prediction formula.
The image processing apparatus according to claim 11. The first calculation means fits the acquired at least N intensity values to the theoretical prediction formula to calculate the tissue content rate (λ) in the pixel.
The image processing apparatus according to claim 11. A specifying means for specifying a pixel region composed of M pixels including the tissue to be measured from the intensity image;
Second calculation means for calculating the tissue content rate (λ _j ) (1 ≦ j ≦ M) in the pixel using the first calculation means for each pixel;
Third calculation means for calculating the tissue content rate (λ) in the M pixels using the presence (λ _j ) of the tissue in each pixel;
The image processing apparatus according to claim 11, further comprising: Determining means for determining L pixel volume spaces including the tissue to be measured from the intensity image;
For each of the L pixel volume spaces, the content ratio (λt) (1 ≦ t ≦ L) of the tissue in the pixel volume space is calculated using the first calculation unit or the second calculation unit. Fourth calculating means for
Fifth calculation means for calculating the tissue content (λ) in the volume of the L pixel volume spaces using the tissue content (λt) in each intensity image;
15. The image processing apparatus according to claim 11, further comprising:

Images