JP3300456B2 - Medical image processing equipment - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a medical image processing apparatus, and more particularly, to a method of performing dynamic scan by injecting a contrast medium into a subject and performing blood flow dynamics based on time-series data obtained by the scan. Perfusion: perfusio
The present invention relates to a processing device that calculates a functional parameter reflecting n) and displays the calculation result as a numerical value or an image.

[0002]

2. Description of the Related Art Conventionally, a dynamic scan technique has been useful as a part of functional morphology, and has been used for X-ray CT, MRI, and SPEC.
Actively implemented in fields such as T and PET.

For example, in MRI, Gd-DTPA (Gad
A contrast agent such as olinium-diethylenetriaminepentaacetic acid) is injected from the vein of the subject, and dynamic scanning is performed using a pulse sequence that emphasizes longitudinal relaxation (T ₁ ) and susceptibility effect (T ₂ ^* ). From the time series data obtained by this dynamic scan, the contrast agent concentration-time characteristics (Time Intensity Curve: TIC)
Is created, and its characteristics are analyzed to calculate functional parameters reflecting blood flow dynamics, and the results are displayed as numerical values or images. Blood volume (Blood Volume) is a functional parameter.
: BV) and average transit time are used.

[0004] For example, to calculate the blood volume BV, FIG.
(A) As shown in FIG.
A T ₁ -weighted image or a T ₂ ^* -weighted image at each time of ₀ ... T _N is obtained. Then, the curve of the contrast agent concentration / time characteristic of a certain tissue designated by the ROI common to each image is represented by C
_i (t) (see FIG. 3B), the area S under the curve is

(Equation 1) The area S under the curve is adopted as a functional parameter reflecting the blood volume BV. The injection of the contrast agent may be performed immediately before t ₀ (base image acquisition) or within the time required to reach the tissue from the vein (about 5 seconds).

On the other hand, in RI, X-ray CT, and the like, a contrast agent suitable for the characteristics of a modality is used, and dynamic study and imaging for obtaining a functional image are performed.

[0006]

However, in the above-described method of calculating the functional parameters reflecting the blood flow dynamics, the information on which the calculation is based is based on the tissue contrast agent concentration / time curve C _i ( t) Because it was one,
The value of the functional parameter changes depending on the method of injecting the contrast agent (for example, the injection amount and the injection time width) and the individual's cardiac function, and merely shows relative hemodynamic information. It was not a parameter as an absolute value. Therefore, when the blood flow dynamics obtained by the dynamic study are used for diagnosis, the properties of the parameters are different, and the parameters are not stable, and there is a limit to the utility.

The present invention has been made in view of the above-mentioned problems in the prior art, and obtains a functional parameter relating to blood flow dynamics in a state close to an absolute value, and improves the use value of the parameter. The main purpose is to provide a device. It is another object to enable such functional parameters to be calculated relatively easily.

[0008]

In order to achieve the above object, an image processing apparatus according to the present invention scans the same diagnostic site of a subject into which a contrast agent has been injected over time (for example, MR scan).
I) is a device that obtains parameters (such as blood volume) representing blood flow dynamics based on time-series data obtained by performing the first contrast agent concentration based on the time-series data of the tissue at the diagnostic site. A first TIC generating means for generating a time characteristic, and a second TIC for generating a second contrast agent concentration-time characteristic based on the time-series data of one of the arterial blood and the venous blood of the subject Creating means, and the first contrast agent concentration created by the first TIC creating means-
A parameter calculating means for calculating a parameter representing the blood flow dynamics of the diagnostic site based on the data of the time characteristic; and a parameter calculated by the parameter calculating means in a second TIC generated by the second TIC generating means. Parameter correction means for correcting using the contrast agent concentration-time characteristic.

In the image processing apparatus according to another aspect of the present invention, the parameter calculating means is means for obtaining an area under the curve of the first contrast agent concentration-time characteristic by an integration calculation,
The parameter correction means calculates the area under the curve of the second contrast agent concentration-time characteristic by integral calculation, and calculates the ratio of the integrated calculated value to the integrated calculated value by the parameter calculating means as the blood volume. Means. Then, the first and second TIC creating means create the first TIC created by the first and second TIC creating means.
And an integral range in which the delay of the rise time between the second contrast agent concentration-time characteristics is corrected is determined, and an integral range correcting means for giving the integral range to the parameter calculating means and the parameter correcting means is added.

[0010] In an image processing apparatus according to still another aspect of the present invention, at least one of the first and second TIC creating means includes at least one of a contrast agent in a tissue and a blood at a diagnostic site before the injection of the contrast agent. Means for adopting the concentration as reference data, and averaging at least one contrast agent concentration in the tissue and blood from a plurality of sets of data obtained by scanning the diagnostic site a plurality of times before the contrast agent injection, to obtain the reference data. A reference data setting means was added.

In an image processing apparatus according to still another aspect of the present invention, the contrast agent is less than a predetermined value with respect to the contrast agent concentration-time characteristic created by at least one of the first and second TIC creating means. A data deleting means for deleting the data of the contrast agent concentration-time characteristic when only the injection is performed is added.

[0012] In an image processing apparatus according to still another aspect of the present invention, the image data is data reflecting information on longitudinal relaxation and information on a susceptibility effect, and converts the image data into a contrast agent density; Data conversion means for adding the conversion result to the first and second TIC creating means is added.

In the image processing apparatus according to still another aspect of the present invention, a display unit for displaying the parameter corrected by the parameter correction unit in at least one of a numerical value and an image is added.

[0014]

In the image processing apparatus according to the present invention, the first contrast agent concentration / time characteristic is created by the first TIC creating means based on the time series data of the tissue at the diagnostic site, and the arterial blood and venous blood of the subject are created. The second contrast agent concentration-time characteristic is created by the second TIC creating means based on one of the time series data. A parameter representing the blood flow dynamics of the diagnostic site is calculated by the parameter calculation means based on the data of the first contrast agent concentration-time characteristic. This parameter is corrected by the parameter correction means using the second contrast agent concentration-time characteristic.

In the apparatus according to another aspect of the present invention, the corrected parameter is displayed in at least one of a numerical value and an image.

In the apparatus according to still another aspect of the present invention, the areas under the curves of the first and second contrast agent concentration-time characteristics are respectively obtained by an integration operation, and the ratio thereof is calculated as a blood volume. An integration range in which the rise time delay between the first and second contrast agent concentration-time characteristics is corrected is determined, and the integration range is provided to the parameter calculation means and the parameter correction means.

In the apparatus according to still another aspect of the present invention, the concentration of at least one of the tissue and blood in the tissue at the diagnostic site before the injection of the contrast agent is adopted as reference data, and a plurality of diagnostic sites are injected before the injection of the contrast agent. At least one contrast agent concentration in tissue or blood is averaged from a plurality of sets of data that have been scanned a number of times, and used as reference data.

[0018] In an apparatus according to still another aspect of the present invention, at least one of the first and second contrast agent concentration-time characteristics, the contrast agent concentration-time when the contrast agent is injected less than a predetermined value is less than the predetermined value. The characteristic data is deleted.

In an apparatus according to still another aspect of the present invention, the image data is data reflecting information on longitudinal relaxation and information on a susceptibility effect. 1st and 2nd TIC
Assigned to creation means.

[0020]

Embodiments of the present invention will be described below with reference to the drawings. The following example describes a magnetic resonance imaging (MR
I) Image processing obtained from the system will be described.

(First Embodiment) A first embodiment will be described with reference to FIGS. 1 to 5 and FIG.

In FIG. 1, reference numeral 1 indicates an MRI system, and reference numeral 2 indicates an image processing apparatus. Among them, the MRI system 1 includes, for example, a normal conducting magnet 1 for generating a static magnetic field.
1 and a static magnetic field power supply 12 for supplying a current to the magnet 11, and a static magnetic field H _{0 in} the z-axis direction of the opening into which the subject P enters.
Generate. In addition, x, y,
The three pairs of gradient magnetic field coils 13... 13 (only some of which are shown) in the z direction are connected to a gradient magnetic field controller 16 via a gradient magnetic field power supply 15. The gradient magnetic field controller 16 operates according to a command from the main controller 17. Thereby, a gradient magnetic field in which a linear gradient magnetic field is superimposed on the static magnetic field H ₀ is formed, and positional information for imaging is provided.

An RF coil 14 is disposed in the opening of the magnet 11 so as to face the subject P.
The coil 14 is individually connected to the transmitter 19 and the receiver 20. The transmitter 19 and the receiver 20 are the RF controller 2
1 is connected to the main controller 17.
As a result, a high-frequency pulse for exciting the NMR is transmitted to the transmitter 19. The receiver 20 detects and amplifies the MR signal obtained by the RF coil 14 and converts the MR signal into an A / A signal.
The data is sent to the data calculator 23 via the D converter 22. The data calculator 23 performs image processing such as Fourier transform on the MR signal to generate image data.

The main controller 17 is further connected to a memory unit 24, an input unit 25 and a display unit 26. The operator can instruct the main controller 17 via the input device 25 to perform a scan of a desired sequence.

On the other hand, as shown in FIG.
An image data storage unit 30 is provided. The image data storage unit 30 can receive and store the image data calculated by the data calculator 23 of the MRI system 1.

The image data storage unit 30 has a bus configuration 3
1, an image processing controller 32, which forms an image processing apparatus 2 together with the unit 30,
It is connected to a processor 33, a memory unit 34, an input device 35, and a TV monitor 36. Input device 35 and TV
The monitor 36 is connected to the bus configuration 31 via interface circuits 37 and 38, respectively.

The image processing controller 32 controls the operation of the entire image processing apparatus 2. That is, with the activation, the processing program stored in the memory unit 34 in advance is read out to its own work area, and the overall operation is managed according to the processing program. Upon receiving a command from the image processing controller 32, the perfusion processor 33 reads out a processing program stored in the memory unit 34 in advance into its own work area, and according to the program, performs a function related to blood flow dynamics (perfusion). The operational parameters are calculated by performing, for example, a process described later with reference to FIG.

Command information from the operator is sent to the image processing controller 32 and the perfusion processor 33 via the interface circuit 37. Further, the processing states and calculation results of the image processing controller 32 and the perfusion processor 33 are output to the TV monitor 36 via the interface circuit 38. The memory unit 34 stores programs necessary for the processing of the image processing controller 32 and the perfusion processor 33, and can also store the processing results thereof as needed.

In the MRI system 1, Gd-DTP
A pulse sequence (eg, IR (FLASH) method, FE) that emphasizes longitudinal relaxation (T ₁ ) and susceptibility effect (T ₂ ^* ) while injecting a contrast agent such as A from a vein.
(FLASH) method). The position of the dynamic scan in this embodiment is set at the head of the subject. For this reason, MRI
When a dynamic scan is performed in the system 1, the image data storage unit 30 stores, for example, a scan image S (t ₀ ) of the same portion in the time series t ₀ , t ₁ , t ₂ ... T _N as shown in FIG. , S (t ₁ ), S (t ₂ )... S
(T _N ) data is stored.

Subsequently, the perfusion processor 33
The calculation process of the functional parameters representing the blood flow dynamics according to the above will be described with reference to FIG.

In this embodiment, the blood volume (BV) is calculated as a functional parameter, and the calculation mode is the “ROI mode”. This “ROI mode” is an ROI (region of interest) on the scan image.
Is set, and the functional / ROI for each set ROI is set.
In this mode, parameters are calculated and displayed numerically.

When the perfusion processor 33 receives a command from the image processing controller 32, it starts a series of processes shown in FIG. First, in step 40,
Dynamic scan image data S (t ₀ ), S (t ₁ )... S stored in the image data storage unit 30.
(T _N ) is read.

Next, proceeding to step 41, the perfusion processor 33 displays the scan image S (t ₀ ) at the fitting start time t ₀ (hereinafter referred to as “start time”) corresponding to the start of the injection of the contrast agent on the TV monitor 36. To be displayed.

Then, the process proceeds to a step 42, wherein the current TV
On the scan image (original image) displayed on the monitor 36, an ROI (region of interest): Ri is set to a region of the brain tissue desired by the operator and an R
OI (region of interest): Each Rv is set as shown in FIG. Thereby, the desired tissue and venous blood on the scan plane of the brain are individually selected through the ROI.

Next, the perfusion processor 3
3 shifts the processing to step 43, and from time-series image data S (t ₀ ), S (t ₁ )... S (t _N ),
Image data corresponding to the positions of the ROIs: Ri and Rv set in step 42 are respectively extracted, and based on the extracted image data, the contrast agent concentration-time characteristic (TimeInten
sity Curve: hereinafter referred to as “TIC”): Ci (t)
And TIC of the venous sinus: Cv (t) are created, for example, as shown in FIG.

The TIC can be created in two ways depending on the physical parameters to be detected. For the perfusion measurement in MRI, a pulse sequence sensitive to T ₁ or a pulse sequence sensitive to T ₂ ^* (that is, the susceptibility effect) (see FIGS. 20A to 20C) is used. One method of calculating the contrast agent concentration of the TIC curve is, like X-ray CT, RI, SPECT, PET, image data, that is, each of the original images S (t ₁ )... S (t _N )
Difference from base image S (t ₀ ) at start time t ₀

(Equation 2) Is what you want. Another method is as follows. That is,

(Equation 3) Then, roughly,

(Equation 4) Here, TE: echo time (FLASH method (for example, a normal FLA shown in the pulse sequence of FIG. 20A)
High-speed FLASH with SH method and T ₂ ^* emphasis in FIG.
Method)). Or

(Equation 5) Here, TI: inversion time. k is determined by substituting TI, R ₁ (t ₀ ) in the base image (IR-FLA
SH method (for example, FIG. 20 shows a pulse sequence of (c), T ₁ emphasized Fast IR-FLASH method) adopted). Since it can be expressed in this way, the contrast agent concentration C (t _n ) is calculated from the differences ΔR ₂ and ΔR ₁ . However, the contrast agent concentration is linear only within a certain limited range. However, such an operation has a great merit for T ₂ ^* . That is, in particular, T ₂ ^* is Gd-DT
Since PA and Dy-DTPA are more sensitive than iodine, accurate perfusion measurement can be performed even with a small amount. Alternatively, even if a contrast medium is not injected, PA, Dy-DTPA can be used based on iron in hemoglobin in blood. O ₂ concentration can also be reflected.

Next, at step 44, the original curves of the two TICs Ci (t) and Cv (t) created at step 43 are analyzed. In this analysis, fitting is performed using a certain function (for example, a gamma function) derived from a model, and a fitting curve and fitting parameters are calculated. As for the curve analysis, in addition to the analysis method using the above-described function, a method of analyzing the original curve as it is is also possible.

Further, in step 45, step 44
Using the fitting parameters obtained in the above, for the TIC: Ci (t) of the tissue with the brain, the area A under the curve
Ci

(Equation 6) It is calculated by integration. Similarly, in the next step 46, the area under the curve AV for TIC: Cv (t) of venous blood
v

(Equation 7) It is calculated by integration. When the parameters are obtained by directly analyzing the original curve in step 44, the above-described ACi, A
The integration operation of Vv may follow a conventionally well-known trapezoidal rule, or use higher-order interpolation or fitting by a spline function.

Next, at step 47, step 45,
The ratio of the values of the areas ACi and ACv under the curve obtained at 46, that is, the blood volume BV, is corrected by the following equation.

[0040]

(Equation 8) By this correction calculation, TIC of cerebral venous blood: Cv (t)
Is used as a reference, a blood volume BV (one of the functional parameters representing the blood flow dynamics) of the brain tissue to be diagnosed is obtained. Since this blood volume BV is not a value relying solely on the area under the curve of the TIC of the tissue alone as in the related art, it is a ratio to the change in the concentration of the contrast agent in the blood vessel.
The information has more absolute color than the conventional value.

Next, at step 48, the original curves of TIC: Ci (t) and Cv (t) created at step 43,
A fitting curve based on the result analyzed in step 44, and a functional curve corrected and calculated in step 47.
The blood volume BV as a parameter, the scan image (with the set ROI) displayed in steps 41 and 42, and other accompanying information are displayed on the TV monitor 36, and the data thereof are filed in the memory unit 34.

Thus, based on the time-series image data S (t ₀ ), S (t ₁ )... S (t _N ) after the dynamic scan, the TIC of venous blood: Cv (t) is corrected as a reference. , The blood volume BV of the diagnostic tissue is determined. Therefore, the influence of the difference in the injection amount and the injection speed of the contrast agent and the difference in the heart function of each patient on the blood volume BV is extremely small, and the blood volume BV as a more absolute value index is obtained.
Is required. As a result, it becomes possible to dramatically improve the diagnostic accuracy of the patient regarding the blood flow dynamics.

Further, in obtaining the absolute blood volume BV, it is not necessary to add a special hardware mechanism to the MRI system 1 or the image processing apparatus 2, and it can be dealt with by software processing utilizing a simple method.

(Second Embodiment) FIGS. 6 and 7 show the second embodiment.
It will be described based on. Note that the same reference numerals are used for the same or equivalent hardware configurations and processing contents as in the first embodiment, and description thereof is omitted (the same applies to the following embodiments).

In the second embodiment, TIC: Cv (t) of arterial blood is used instead of TIC: Cv (t) of venous blood of the first embodiment.
Using (t) as a reference, the blood volume BV of the brain tissue subjected to the same correction is obtained. When TIC: Ca (t) of arterial blood is used in this way, since a thick artery cannot be seen at the head except at the level of the occipital fossa, a scan at the cervical or aortic level (for example, the upper lung) is performed by a brain scan. It must be done simultaneously or time-divisionally (alternately).

In this embodiment, as shown in FIG. 6A, slice planes A and B (or B ') are set on the head and neck (or upper lung) of the subject by the MRI system 1. For example, a dynamic scan is performed in a time-division manner. Accordingly, the time-series data S (t ₀₎ of the dynamic scan in accordance with the scanned image SI _A and SI _B slice plane A and B (or B') (or SI _{B '),} S
(T ₁ )... S (t _N ) are obtained and stored in the image data storage unit 30, respectively.

The image processing apparatus 2 performs the same processing as in the first embodiment. Thus, ROI relative to the desired tissue set on the scanned image SI _A: Ri of TIC: C
ROI for i (t) and arterial (internal carotid artery, aorta) blood set on scan image SI _B (or SI _{B ′} ):
The TIC of Ra (or Ra '): Ca (t) is created as shown in FIG. Both TICs: Ci (t) and Ca (t) were analyzed as before and the ratio of their area under the curve,

(Equation 9) Is calculated. As a result, the blood volume BV of the brain tissue closer to the absolute value corrected using the TIC of the arterial blood as a reference can be obtained, which can contribute to the improvement of the diagnostic accuracy.

As described above, by checking the arterial blood, the degree of freedom in obtaining a TIC as a reference is increased, and the variation in the calculation of the functional parameter can be increased. At this time, since the image based on the scan of the neck (or upper lung) where the thicker artery is located is based, the accuracy of creating the TIC itself of arterial blood is improved,
If it is, more accurate blood volume BV can be obtained.

(Third Embodiment) FIGS. 8 and 9 show the third embodiment.
It will be described based on.

The image processing apparatus according to this embodiment is intended to further improve the calculation accuracy by adding a correction mechanism for the rise time difference of the TIC curve to the configuration of the first embodiment.

Here, the TIC of blood venous blood: Cv (t) is used as a reference, and the blood volume BV of the brain tissue corrected in the same manner as in the first embodiment is obtained. The hardware configuration is the same as that of the first embodiment.

The perfusion processor 33 is shown in FIG.
Is performed. During the processing of FIG. 8, the dynamic scan image S (t ₀₎ of the time _{series, S (t 1) ... S} (t N) with fitting curve from step 40 to read the T
The contents up to step 44 of the curve analysis of the IC curve are the same as those described in FIG.

Then, at step 44, the TI of the tissue
After the fitting parameters for C: Ci (t) and TIC: Cv (t) of venous blood are calculated, the process proceeds to step 44a to correct for the rise delay of TIC: Cv (t) of venous blood. .

More specifically, as shown in FIG.
The rise time difference Td of the IC curve is calculated, and the range of time integration for obtaining the area under the curve described later is changed using the time difference Td. That is, the TIC of the organization: Ci
(T) For the curve, set “t _{0 to} Te−Td”,
For the TIC: Cv (t) curve of venous blood, "t ₀ + T
d to Te ”. Te is the fitting end time, which is one of the fitting parameters determined in the curve analysis in step 44.

The time difference Td may be calculated as the difference between the time to reach the peak value of both TIC curves (peak time: obtained as one of the fitting parameters).

The TIC: Ci (t) of the tissue is strictly different depending on which site is selected. To cope with this, a large ROI is set at a desired position, an average TIC: Ci (t) within the ROI is obtained, and the average TI
A rising delay time Td between C: Ci (t) and TIC of certain venous blood: Cv (t) is calculated. The averaged delay time Td may be regarded as a delay between the TIC of each pixel in the ROI and the TIC of venous blood.

Next, at step 45, the T of the desired tissue
The area under the curve for IC

(Equation 10) Is obtained by the integral operation of In step 46, the area under the curve for the TIC of venous blood is

[Equation 11] Is obtained by the integral operation of

[0058]

Further, in step 47,

[Mathematical formula-see original document] By taking the ratio of BV = ACi / ACv, the blood volume BV close to the absolute value is corrected and calculated. In step 48, various numerical values and parameters are displayed and stored as in the first embodiment.

As described above, by calculating the blood volume BV by integration taking into account the rise delay time difference Td of both TIC curves, each area under the curve can be accurately obtained, and the calculation accuracy of the blood volume BV is also improved. . In particular, when the TIC of venous blood is used as a reference, the delay time difference Td is large (it can be considered that there is no delay time in arterial blood).
Therefore, the advantage of the above-mentioned high accuracy becomes particularly remarkable when the TIC of venous blood is used as a reference.

On the other hand, if the fitting end time Te is set by sampling until the TIC curve sufficiently falls, the effect of the delay time difference Td is essentially reduced. , The setting of such a long sampling time can be eliminated.

The following modification can be made to the correction processing of the delay time difference. First, step 44
In FIG. 10A, the delay time difference Td is obtained, and the TIC: Cv (t) of venous blood is converted into a curve of “Cv (t + Td)” as shown in FIG. “t _{0 to} Te−Td” is set as an integration time common to t) and Cv (t). Then, in steps 45 and 46,

(Equation 13) And the same integral range is calculated. In step 47, the ratio BV is calculated.

## EQU14 ## It can also be obtained from BV = ACi / ACv.

(Fourth Embodiment) The fourth embodiment is shown in FIGS.
It will be described based on.

The image processing apparatus according to this embodiment is intended to further improve the calculation accuracy by adding a noise data removing mechanism for the TIC curve to the configuration of the first embodiment.

Here, the blood volume BV of the brain tissue, which has been corrected in the same manner as in the first embodiment, is obtained using TIC: Cv (t) of venous blood as a reference. The hardware configuration is the same as that of the first embodiment.

The perfusion processor 33 is shown in FIG.
1 is performed. During this processing, the contents from step 40 relating to the reading of the dynamic scan image to step 43 relating to the creation of the TIC curve using the fitting curve are the same as those shown in FIG.

In step 43, T of tissue and venous blood
When the IC has been created, the process proceeds to step 43a. In step 43a, data having a high probability of being noise data among the data forming the TIC curve is removed. That is, the data at the timing when the contrast agent hardly enters at the rising and falling of the TIC curve becomes almost noise and becomes an error factor in the measurement of the functional parameter. Therefore,
One of the following two removal methods is particularly preferably used. This elimination method will be described by exemplifying the TIC: Cij (t) of a certain pixel Pij of the brain tissue in FIGS. 12 and 13. However, the elimination method can be applied to the TIC of venous blood and arterial blood, and noise can be eliminated similarly. .

In the first method, as shown in FIG.
Using the threshold value Cth empirically determined in advance based on the / N ratio, the constituent data of TIC: Cij (t) is discriminated one by one.
That is, only data satisfying Cij (t _n ) ≧ Cth is extracted,
Data satisfying ij (t _n ) <Cth is regarded as noise and discarded. However, base data (data at start time t = t ₀ ) is included.

In the second method, as shown in FIG. 13, in addition to setting the above-described threshold value Cth, the peak value C of the Cij (t) curve
ij (max) is obtained, and the minimum values Cij (min-L) and Cij (m) exceeding the threshold value Cth on the left and right of the curve of the peak value Cij (max).
in-R) is set for data times Ts and Te. Then, only data within the time width of Ts <t <Te is extracted. However, base data is included.

After data having a high probability of being noise is removed in this way, in step 44, the T
Fitting parameters for IC: Ci (t) and TIC: Cv (t) of venous blood are calculated. afterwards,
In steps 45 and 46, the areas under the curves for both TICs are individually calculated, and in step 47, the blood volume B close to the absolute value is calculated.
V is corrected. In step 48, various numerical values and parameters are displayed and stored as in the first embodiment.

Therefore, according to this embodiment, the noise data is removed before the TIC curve analysis such as the calculation of the area, and the accuracy is determined based only on the data which is considered to truly represent the change in the contrast agent concentration. Is performed, and the calculation accuracy of the required blood volume BV is further improved.

It should be noted that if the noise data removal processing is incorporated together with the third embodiment described above and is performed together with the correction processing of the rise time difference of the TIC curve,
The calculation accuracy of the blood volume BV can be synergistically improved.

(Fifth Embodiment) FIGS. 14 to 1 show the fifth embodiment.
6 will be described.

The image processing apparatus according to this embodiment is obtained by adding a mechanism for improving the accuracy of determining the base data in the TIC curve to the configuration of the first embodiment.

Here, the blood volume BV of the brain tissue, which has been corrected in the same manner as in the first embodiment, is obtained using TIC: Cv (t) of venous blood as a reference. The hardware configuration is the same as that of the first embodiment. However, due to the operation of the MRI system 1, the start time t = t ₀ is stored in the image data storage unit 30 as shown in FIG.
Time series t _-4 from time t _-4 before (ie, before the contrast enhancement),
, T ₀ ,..., T _N dynamic scan images S (t
_-4 ),..., S (t ₀ ),..., S (t _N ).

The perfusion processor 33 is shown in FIG.
Step 5 is performed. During this processing, after Step 40 relating to reading of the dynamic scan image, Step 40a
, S (t _-4 ),..., S
(T ₀ ) are bundled. That is, the processor 33 averages the pixel values of the plurality of images before contrast for each pixel, and
A new base image S ′ (t ₀ ) is created. This base image S ′ (t ₀ ) is incorporated into the time series t ₀ ,..., T _N for calculating the blood flow BV as shown in FIG.

Thereafter, the base image S in step 41
The processes from the display of '(t ₀ ) to the display of the processing result in step 48 and the file are the same as those shown in FIG.

As a result, the base data of the TIC curve (t
Reference data) at = t ₀ is common to the plurality of scan images S of the pre-contrast _{(t -4) ... S (t} 0) shown in FIG. 16 (a), the average value of the pixel values of the desired pixel (See FIG. 3B). This averaging process increases the accuracy of determining the base data, which is advantageous in terms of calculating the S / N ratio and blood volume BV.

In terms of the S / N ratio, as in this embodiment, 5
On average the images, the S / N ratio is such becomes ^"51/2" times, the image averaging of n sheets can achieve the "n ^1/2" times the S / N ratio improvement. For example, when the contrast agent concentration is obtained by subtraction of the original image, C (t _n ) = S (t _n ) −S
(T ₀ ), the S / N ratio is equal to the S / N of the original image S (t _n ).
Assuming that the / N ratio = 1, “1 / (2 ^1/2 )” is obtained for only one base data, and “1 / (1 + 1 / n) ^1/2 ” is obtained for the average of n base data. ".

If the error of the base data, that is, the error of the zero level of the contrast agent concentration is large, the calculation of the area under the curve and the calculation of the blood volume BV become inaccurate. Blood volume B by improving the accuracy of
The calculation accuracy of V is also improved.

In the fifth embodiment, instead of the scanning image bundling process (averaging process) in step 40a in FIG. 15, between steps 43 and 44, every pixel set by the ROI, The pixel values of a plurality of scanned images may be averaged to form base data.

The base image bundling process of the fifth embodiment may be performed in combination with the above-described second to fourth embodiments as appropriate.

(Sixth Embodiment) FIGS. 17 to 1 show the sixth embodiment.
9 will be described. The first to fifth embodiments described above are processing in the ROI mode (mode in which functional parameters are calculated for each set ROI and the result is displayed numerically). , Functional parameters for each pixel (matrix)
And processing in an imaging mode for displaying the result as an image.

The hardware configuration, scan position, and scan target are the same as in the sixth embodiment, and the perfusion processor 33 performs the processing shown in FIG.

In the series of processes shown in FIG. 17, the processing contents of steps 40 to 41 are the same as those shown in FIG. At step 42, the base image S '
Analysis pixel range (Cmin, Cma) on (t ₀ )
x) is extracted, for example, as shown in FIG. In this process, unnecessary pixels corresponding to air or the like are determined and excluded. That is, for the pixel value C ₀ (i, j) of each pixel on the base image S ′ (t ₀ ), thresholds Cmin and Cmax capable of discriminating unnecessary pixel values such as air are set (FIG. 18).
(B)),

Equation 15] _{Cmin ≦ C 0 (i, j} ) condition ≦ Cmax extracts the analysis area AN of established (FIG. 18 (a) the hatched portion).

After the TIC is created in step 42 for each pixel in the analysis area AN thus extracted, the same processing as described above is performed in steps 44 to 47.

Then, in step 48, for example, FIG.
As shown in FIG. 9, the calculated functional parameters (AT: Appearance Time, PT: Peak Time, MT1)
E: Effective Mean Transit Time) image and corrected blood volume BV image are displayed as a functional image together with necessary supplementary information (analysis type, analysis time range, points of analysis time range, conversion coefficient, unit, etc.). And file.

Thus, a functional image relating to blood flow dynamics can be accurately displayed. At this time, since the image processing is limited to only the area of the actually required pixel value, the entire processing time is significantly reduced as compared with the case where the extraction processing is not performed.

In the image processing in the imaging mode according to the sixth embodiment, the T according to the third embodiment described above is used.
The correction process for the rise time difference of the IC curve and the noise data elimination process according to the fourth embodiment can be appropriately incorporated and executed.

Although the above-described first to sixth embodiments have been described with reference to the case where the present invention is applied to a dynamic scan image acquired by an MRI system, the X-ray CT,
Digital fluorography (DF), SPECT, P
The same can be applied to post-processing of a dynamic scan image such as ET. Also, as an example of the functional parameter, the blood volume BV is exemplified, and the case where the blood volume BV is calculated has been mainly described. However, the functional parameter of the present invention is not necessarily limited to this. For example, BF (blood flow), MTT A parameter such as (average transit time) may be used. Further, the TIC in the present invention is an X-ray C
As in the case of T or the like, it is of course possible to employ a method in which a difference between the original images is simply obtained to obtain a change in the contrast agent density.

[0091]

As described above, the image processing apparatus according to the present invention comprises a first TIC creating means for creating a first contrast agent concentration / time characteristic based on time-series data of a tissue at a diagnostic site. Second TIC creating means for creating a second contrast agent concentration-time characteristic based on one of time-series data of arterial blood and venous blood of the subject, and data of the first contrast agent concentration-time characteristic And a parameter calculating means for calculating a parameter representing the blood flow dynamics of the diagnosis site based on the parameter, and a parameter correcting means for correcting the parameter using the second contrast agent concentration-time characteristic. Therefore, unlike conventional methods such as adopting the area under the curve of a single contrast agent concentration / time characteristic as the blood volume, it affects the difference in contrast agent injection volume, injection speed, and differences in patient's cardiac function, etc. Hard to be, Ri parameters such as blood volume close to the absolute value (Functional parameters) can be calculated by a simple process. This makes it possible to easily provide the most desired functional parameters for the diagnosis, thereby reducing the diagnosis labor and improving the diagnosis accuracy.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration example of an image processing apparatus and an MRI system connected to the image processing apparatus according to an embodiment of the present invention.

FIG. 2 is a schematic diagram showing a time-series dynamic scan image collected by an MRI system.

FIG. 3 is a flowchart illustrating a processing example of a perfusion processor in the first embodiment.

FIG. 4 is an explanatory diagram showing an example of a setting position of an ROI.

FIG. 5 is a graph showing a TIC curve of tissue in the brain and a TIC curve of venous blood as a reference.

FIGS. 6A and 6B are explanatory diagrams showing an example of an MRI scan position and an ROI setting position on the scan image in the second embodiment.

FIG. 7 is a graph showing a TIC curve of a tissue and a TIC curve of arterial blood as a reference in the second embodiment.

FIG. 8 is a flowchart illustrating a processing example of a perfusion processor in the third embodiment.

FIG. 9 is a graph illustrating a rise time difference of a TIC curve and a correction process thereof.

FIG. 10 is a graph illustrating a modification of the correction processing for the rise time difference of the TIC curve.

FIG. 11 is a flowchart illustrating a processing example of a perfusion processor in a fourth embodiment.

FIG. 12 is a graph illustrating a process of removing TIC curve noise data.

FIG. 13 is a graph illustrating a modified example of the removal processing for noise data of a TIC curve.

FIGS. 14A and 14B are explanatory diagrams for each time series for explaining a base image bundling process according to a fifth embodiment.

FIG. 15 is a flowchart illustrating a processing example of a perfusion processor in the fifth embodiment.

FIGS. 16A and 16B are explanatory diagrams equivalently explaining a bundling process of a base image for each pixel position.

FIG. 17 is a flowchart illustrating a processing example of a perfusion processor in the sixth embodiment.

FIGS. 18A and 18B are explanatory diagrams illustrating an analysis range extracting process according to a sixth embodiment.

FIG. 19 is a schematic diagram showing a display example in an imaging mode of the sixth embodiment.

FIGS. 20A to 20C are flowcharts illustrating an example of a pulse sequence applied to the embodiment.

FIGS. 21 (a) and (b) are explanatory diagrams illustrating a conventional calculation state of blood volume.

[Explanation of symbols]

 Reference Signs List 1 MRI system 2 Image processing device 30 Image data storage unit 31 Bus 32 Image processing controller 33 Perfusion processor 34 Memory unit 35 Input device 36 TV monitor

Claims (10)

(57) [Claims] 1. A medical image processing apparatus for obtaining parameters representing blood flow dynamics based on time-series data obtained by scanning the same diagnostic site of a subject into which a contrast agent has been injected over time. First TIC creating means for creating a first contrast agent concentration / time characteristic based on the time series data of the tissue, and a first TIC creating means based on the time series data of one of arterial blood and venous blood of the subject Second TIC creating means for creating the second contrast agent concentration-time characteristic, and blood flow dynamics at the diagnostic site based on the data of the first contrast agent concentration-time characteristic created by the first TIC creating means Parameter calculating means for calculating a parameter representing the following, and the parameter calculated by the parameter calculating means is converted to the second TIC generated by the second TIC generating means.
And a parameter correcting means for correcting using the contrast agent concentration-time characteristic. 2. The method according to claim 1, wherein the parameter is a blood volume.
The medical image processing apparatus according to the above. 3. The parameter calculating means is means for calculating an area under a curve of the first contrast agent density-time characteristic by integral calculation, and the parameter correcting means is the second contrast agent density-time characteristic. 3. The medical image processing apparatus according to claim 2, further comprising means for calculating an area under the curve by an integral operation, and means for calculating a ratio of the integral operation value to the integral operation value by the parameter operation means as the blood volume. 4. An integrated range in which a rise time delay between the first and second contrast agent concentration-time characteristics created by the first and second TIC creating means is corrected, and the integrated range is defined by the parameter 4. The medical image processing apparatus according to claim 3, further comprising an integral range correcting means provided to the calculating means and the parameter correcting means. 5. At least one of the first and second TIC creating means is means for employing, as reference data, at least one contrast agent concentration in a tissue or blood at a diagnostic site before injection of the contrast agent. 2. A reference data setting means according to claim 1, further comprising a reference data setting means for averaging at least one of the concentration of the contrast agent in the tissue and the blood from a plurality of sets of data obtained by scanning the diagnostic site a plurality of times before the injection of the contrast agent. Medical image processing device. 6. The contrast agent concentration-time characteristic created by at least one of the first and second TIC creating means, wherein the contrast agent concentration when the contrast agent is injected less than a predetermined value is reduced. 2. The medical image processing apparatus according to claim 1, further comprising a data deletion unit for deleting data of the time characteristic. 7. The time-series data includes the first and second data.
2. The medical image processing apparatus according to claim 1, wherein two diagnostic sites of the subject are individually scanned image data corresponding to the contrast agent concentration-time characteristics created by the TIC creating unit. 8. The medical image processing apparatus according to claim 1, wherein the time-series data is image data scanned by a magnetic resonance imaging apparatus. 9. The image data is data reflecting information on longitudinal relaxation and information on a susceptibility effect. The image data is converted from the image data into a contrast agent density, and the conversion result is converted into the first and second TICs. 9. The medical image processing apparatus according to claim 8, further comprising a data conversion unit provided to the creation unit. 10. The medical image processing apparatus according to claim 1, further comprising a display unit for displaying the parameter corrected by the parameter correction unit in at least one of a numerical value and an image.