JP5256255B2 - Cerebral blood flow quantification device, cerebral blood flow quantification program, and cerebral blood flow quantification method - Google Patents

Description

  The present invention relates to an apparatus, a program, and a method for quantifying cerebral blood flow based on a nuclear medicine image.

  Cerebrovascular disorders represented by cerebral infarction account for the top causes of death, and if the discovery is delayed or the timing of treatment is delayed, serious sequelae may remain. Therefore, early detection of cerebrovascular disorders and diagnosis of severity are very important.

  As main diagnostic methods for cerebrovascular disorders, nuclear medicine image diagnostic methods represented by SPECT and the like are known in addition to MRI and CT. The nuclear medicine imaging method is widely used for early diagnosis and severity diagnosis of cerebrovascular disorders because it is relatively easy to quantify cerebral blood flow at the tissue level.

  The microsphere method and the IMP-ARG method are mainly used as a method for diagnosing a cerebrovascular disorder using a nuclear medicine image diagnostic method. The Microsphere method is a method in which a cerebral blood flow preparation at an early stage after administration is regarded as a so-called Microsphere and a local cerebral blood flow is obtained from nuclear medicine image data based on a one-compartment model. This method has a merit that the theory is simple and calculation is easy, but has a drawback that it is highly invasive because it requires continuous blood collection of arterial blood. In addition, since the washing of the drug from the tissue to the blood vessel is not taken into account, there is a disadvantage that the quantitative accuracy in the later stage after the drug administration is lowered and the data collection time point is limited relatively early after the administration.

  A NIMS method for estimating an input function using a time activity curve based on dynamic SPECT data of the chest instead of performing continuous arterial blood collection has also been proposed, but it is also a method based on a one-compartment model. In the latter period, sufficient quantitative accuracy cannot be obtained.

On the other hand, in the IMP-ARG method, IMP (N-isopropyl-4- [ ¹²³ I] iodoamphetamine hydrochloride) is administered as a radiopharmaceutical, and the count number after a certain period of time in the head nuclear medicine image is expressed by a two-compartment model formula. This is a method for substituting into the cerebral blood flow and obtaining the cerebral blood flow rate by a lookup table method (Non-patent Document 1). The IMP-ARG method has a merit that a method for obtaining one-point arterial blood collection data and correcting the standard input function can be used, and is less invasive than the Microsphere method that requires continuous blood collection data.

Hidehiro Iida et al, "Quantal Mapping of Regional; Cerebral Blood Flow Using Idine-123-IMP and SPECT", The Journal of NuclearMed 94 35, no. 12, pp 2019-2030

As described above, the IMP-ARG method is one of the most excellent diagnostic methods for cerebrovascular disorders. However, arterial blood collection is required even in the IMP-ARG method. In general, arterial blood collection is more invasive than venous blood collection. From such a background, a method for quantifying cerebral blood flow by a less invasive operation has been desired, but such a method has not been known so far. In addition, when venous blood is mixed during arterial blood collection, there is a problem that the reliability of the obtained data is lowered.
The present invention has been made in view of the above circumstances, and an object thereof is to provide a method for accurately quantifying cerebral blood flow by a less invasive operation.

  As a result of intensive studies, the inventor has found that the parameters obtained by fitting the time activity curve based on the head nuclear medicine image (such as cerebral blood flow SPECT) and the two-compartment model formula are actually obtained based on blood sampling. It was found to have a certain relationship with the actually measured value. Then, by using this relationship, it was found that the scaling factor used for correction of the standard input function can be determined using the parameters obtained based on the time activity curve, and the present invention has been completed. It was.

In the present specification, the following terms are used with the following meanings.
Input function: A function that represents the change in the amount of radioactivity flowing into the target organ over time.
Standard input function: A function that represents a temporal change in the inflow of the radiopharmaceutical into the standard brain, that is, an input function that is standardized by using an input function determined based on actual measurement values in a plurality of subjects. For example, it is created by averaging the input functions of a plurality of subjects.
Scaling factor: A correction factor to approximate the standard input function to the input function in the subject.
Cerebral blood partition coefficient (V _d ): ratio (k ₁ / k ₂ ) between cerebral blood flow (k ₁ ) and flux (k ₂ ) of washing out radioactivity counts from brain tissue into blood in a two-compartment model .
Further, in the present invention, the head nuclear medicine image can be a PET image or a planar image in addition to a SPECT image, and is appropriately selected according to the type of radiopharmaceutical used and the purpose of diagnosis.

The parameter determination apparatus according to the present invention includes an interphase expression (mC _f : scaling factor, C _f : scaling factor determined based on a time activity curve of a head nuclear medicine image) representing a relationship between mC _f and C _f · V _d. , V _d : cerebral blood distribution coefficient) and a time activity curve representing a time-dependent change in the count value in the head nuclear medicine image of the subject, the following formula (1):

（Ｂ（ｔ）：時刻ｔにおける頭部核医学画像における放射能カウント、Ｓ（τ）：標準入力関数、ｋ_１：脳血流量、ｋ_２：ｋ_１／Ｖ_ｄ）をフィッティングさせることにより、ｋ_１Ｃ_ｆおよびｋ_２を求める第一のパラメータ算出部と、求めたｋ_１Ｃ_ｆおよびｋ_２とから、下記式（２）：

By fitting (B (t): radioactivity count in the cranial nuclear medicine image at time t, S (τ): standard input function, k ₁ : cerebral blood flow, k ₂ : k ₁ / V _d ), a first parameter calculating section for obtaining a k ₁ C _f and _{k 2,} from _k 1 _{C f} and _{k 2} Metropolitan determined by the following formula (2):

に基づいて、Ｃ_ｆ・Ｖ_ｄを求める第二のパラメータ算出部と、第二のパラメータ算出部にて求めたＣ_ｆ・Ｖ_ｄと前記データベースに格納された前記相間式からｍＣ_ｆを求める第三のパラメータ算出部と、を備える。

Based on, first obtains a second parameter calculating section for obtaining a C f _· V _d, the second C _f · V _d and the mC _f from the phase during formulas stored in the database as determined by the parameter calculator A third parameter calculation unit.

With this configuration, since the scaling factor (mC _f ) used for correcting the standard input function can be obtained based on the time activity curve of the head nuclear medicine image, the cerebral blood flow using the head nuclear medicine image In quantification, invasive arterial blood collection can be omitted. Thus, not only can the burden on the subject be reduced, but also variations in quantitative values based on human error during arterial blood collection can be prevented.

In the parameter determination device according to the present invention, the time activity curve for fitting with the above equation (1) can be created based on a transverse image (slice) from the parietal lobe to the cerebellar tent (for example, The voxel whose count value is between the upper limit value and the lower limit value is extracted from the cross-sectional image, and is created using the average value or the sum of the radioactivity counts in the extracted voxel), preferably It can be created based on a cross-sectional image (slice) of a region covering the entire ventricle from the parietal lobe to the cerebellar tent. With such a configuration, it is possible to reliably collect radioactivity counts based on cerebral blood flow and create a time activity curve, and to obtain mC _f with higher accuracy.

  In the parameter determination device according to the present invention, the interphase equation stored in the database may be expressed by the following equation (3).

（式中、Ｒは比例定数である。）

(In the formula, R is a proportionality constant.)

As a result of the examination by the inventors, it has been clarified that a proportional relationship is substantially established between the scaling factor (mC _f ) and C _f · V _d . Therefore, by using the equation described in the above equation (3) as the interphase equation, the relationship between the scaling factor (mC _f ) and C _f · V _d can be expressed as a simple linear relationship, and more The scaling factor (mC _f ) can be easily obtained.

The parameter determination program according to the present invention includes a time activity curve representing a change over time of a count value in a nuclear medicine image of a subject who has been administered a radiopharmaceutical and a relationship between mC _f and C _f · V _d stored in a database. A computer program for operating a computer comprising a storage means and a CPU, wherein the computer is executed by the CPU, thereby allowing the computer to display a head nuclear medicine image of a subject. In the time activity curve representing the change over time in the count value at the following formula (1):

をフィッティングさせてｋ_１Ｃ_ｆおよびｋ_２を求める第一のパラメータ算出部と、求めたｋ_１Ｃ_ｆおよびｋ_２とから、下記式（２）：

From the first parameter calculation unit for obtaining k ₁ C _f and k ₂ and the obtained k ₁ C _f and k ₂ , the following equation (2):

に基づいて、Ｃ_ｆ・Ｖ_ｄを求める第二のパラメータ算出部と、第二のパラメータ算出部にて求めたＣ_ｆ・Ｖ_ｄと前記テータベースに格納された前記相間式からｍＣ_ｆを求める第三のパラメータ算出部として動作させる。

Based on the obtained C and the second parameter calculating section for obtaining the f _· V _d, the second C obtained in the parameter calculation section _f · V _d and the stator base on the stored mC _f from the phase between formula Operate as a third parameter calculation unit.

The parameter determination method according to the present invention is a method for determining parameters used by a computer to quantify cerebral blood flow based on a nuclear medicine image of a subject who has been administered a radiopharmaceutical, and the computer uses the computer to determine the parameters of the subject's head. The following formula (1) is applied to the time activity curve representing the time-dependent change of the count value in the nuclear medicine image:

をフィッティングさせてｋ_１Ｃ_ｆおよびｋ_２を求める第一のパラメータ算出処理と、求めたｋ_１Ｃ_ｆおよびｋ_２とから、下記式（２）：

From the first parameter calculation process for obtaining k ₁ C _f and k ₂ and the obtained k ₁ C _f and k ₂ , the following equation (2):

に基づいて、Ｃ_ｆ・Ｖ_ｄを求める第二のパラメータ算出処理と、求めたＣ_ｆ・Ｖ_ｄを利用し、データベースに格納されたｍＣ_ｆとＣ_ｆ・Ｖ_ｄとの関係を表す相間式からｍＣ_ｆを求める第三のパラメータ算出処理と、を順次実行する。

Based _on, C a second parameter calculation processing for calculating the f · _{V d,} the obtained _C f · _{V d} using, between the phases representing the relationship between mC _f and _C f · _{V d} stored in the database formula And a third parameter calculation process for obtaining mC _f from the above.

Also in the parameter determination program and parameter determination method according to the present invention, the time activity curve for fitting with the above equation (1) is based on the image of the portion corresponding to the transverse image (slice) from the parietal lobe to the cerebellar tent. Preferably, it may be created based on a cross-sectional image (slice) of a region covering the entire ventricle from the parietal lobe to the cerebellar tent.
Similarly, also in the parameter determination program and the parameter determination method according to the present invention, the interphase expression stored in the database can be expressed by the following expression (3), similarly to the cerebral blood flow rate quantification device. .

  Each of the above inventions can also be applied to a cerebral blood flow quantification device, a cerebral blood flow quantification program, and a cerebral blood flow quantification method that quantitate cerebral blood flow based on a nuclear medicine image.

The cerebral blood flow rate quantification device according to the present invention includes at least the following (a) and (b):
(A) Standard input function S (t): Standard function representing time variation of inflow of radiopharmaceutical into the brain (a) Interphase expression: Interphase expression representing the relationship between mC _f and C _f · V _d ,
(MC _f : scaling factor, C _f : scaling factor obtained based on time activity curve of head nuclear medicine image, V _d : brain blood distribution coefficient) A nuclear medicine image data acquisition unit that acquires an image, a time activity curve generation unit that generates a time activity curve that represents a temporal change in the count value in the head nuclear medicine image, and the time activity curve represented by the following formula (1):

（Ｂ（ｔ）：時刻ｔにおける放射能カウント、ｋ_１：脳血流量、ｋ_２：ｋ_１／Ｖ_ｄ）をフィッティングさせることにより、ｋ_１Ｃ_ｆおよびｋ_２を求める第一のパラメータ算出部と、求めたｋ_１Ｃ_ｆおよびｋ_２とから、下記式（２）：

(B (t): Radioactivity count at time t, k ₁ : cerebral blood flow, k ₂ : k ₁ / V _d ) by fitting, a first parameter calculation unit for obtaining k ₁ C _f and k ₂ From the obtained k ₁ C _f and k ₂ , the following formula (2):

に基づいて、Ｃ_ｆ・Ｖ_ｄを求める第二のパラメータ算出部と、第二のパラメータ算出部にて求めたＣ_ｆ・Ｖ_ｄと前記データベースに格納された前記相間式からｍＣ_ｆを求める第三のパラメータ算出部と、求めた該ｍＣ_ｆを用い、下記式（４）または（５）：

Based on, first obtains a second parameter calculating section for obtaining a C f _· V _d, the second C _f · V _d and the mC _f from the phase during formulas stored in the database as determined by the parameter calculator Using the three parameter calculation units and the calculated mC _f , the following formula (4) or (5):

（ｔ_１：頭部核医学画像の撮像開始時刻、ｔ_２：頭部核医学画像の撮像終了時刻）に基づいて単位領域あたりの脳血流量を求める脳血流量算出部と、得られた脳血流量を対応する単位領域と関連づけて表示する表示部と、を備える。

A cerebral blood flow rate calculation unit that obtains a cerebral blood flow rate per unit region based on (t ₁ : imaging start time of the head nuclear medicine image, t ₂ : imaging end time of the head nuclear medicine image), and the obtained brain And a display unit that displays the blood flow volume in association with the corresponding unit area.

With this configuration, the scaling factor (mC _f ) used for correction of the standard input function can be obtained from the parameters based on the time activity curve of the cranial nuclear medicine image, so there is no need to collect arterial blood from the subject. It becomes possible to carry out quantification of the less invasive cerebral blood flow.

In the cerebral blood flow rate quantification device according to the present invention, the time activity curve for fitting with the above equation (1) is created based on an image of a portion corresponding to a transverse image (slice) from the parietal lobe to the cerebellar tent. Preferably, it may be created based on a cross-sectional image (slice) of a region covering the entire ventricle from the parietal lobe to the cerebellar tent. With such a configuration, it is possible to reliably collect radioactivity counts based on cerebral blood flow and create a time activity curve, and to obtain mC _f with higher accuracy.

  In the cerebral blood flow rate quantification device according to the present invention, the interphase equation stored in the database may be expressed by the following equation (3).

As described above, as a result of examination by the inventors, it has been clarified that a proportional relationship is substantially established between the scaling factor (mC _f ) and C _f · V _d . Therefore, by using the equation described in the above equation (3) as the interphase equation, the relationship between the scaling factor (mC _f ) and C _f · V _d can be expressed as a simple linear relationship. The scaling factor (mC _f ) used in the cerebral blood flow rate calculation unit can be obtained more easily.

The cerebral blood flow quantification program according to the present invention can use a nuclear medicine image of a subject and an interphase expression and a standard input function expressing a relationship between mC _f and C _f · V _d stored in a database. And a computer program for operating a computer comprising a CPU, which is executed by the CPU to cause the computer to acquire a nuclear medicine image data acquisition unit for acquiring a head nuclear medicine image of a subject; A time activity curve creation unit that creates a time activity curve that represents a time-dependent change in the count value in the nuclear medicine image, and the following formula (1) in the time activity curve:

をフィッティングさせることにより、ｋ_１Ｃ_ｆおよびｋ_２を求める第一のパラメータ算出部と、
求めたｋ_１Ｃ_ｆおよびｋ_２とから、下記式（２）：

A first parameter calculation unit for obtaining k ₁ C _f and k ₂ by fitting
From the determined k ₁ C _f and k ₂ , the following formula (2):

に基づいて、Ｃ_ｆ・Ｖ_ｄを求める第二のパラメータ算出部と、第二のパラメータ算出部にて求めたＣ_ｆ・Ｖ_ｄと前記データベースに格納された前記相間式からｍＣ_ｆを求める第三のパラメータ算出部と、下記式（４）または（５）：

Based on, first obtains a second parameter calculating section for obtaining a C f _· V _d, the second C _f · V _d and the mC _f from the phase during formulas stored in the database as determined by the parameter calculator Three parameter calculation units and the following formula (4) or (5):

に基づいて単位領域あたりの脳血流量を求める脳血流量算出部と、得られた脳血流量を対応する単位領域と関連づけて表示する表示部として動作させる。

The cerebral blood flow rate calculation unit for obtaining the cerebral blood flow rate per unit region based on the above, and the display unit for displaying the obtained cerebral blood flow rate in association with the corresponding unit region.

  The cerebral blood flow quantification method according to the present invention includes a nuclear medicine image data acquisition process for acquiring a head nuclear medicine image of a subject by a computer, and a time activity curve representing a time-dependent change in the count value in the head nuclear medicine image. Time activity curve creation processing to be created, and the following formula (1) for the time activity curve:

をフィッティングさせてｋ_１Ｃ_ｆおよびｋ_２を求める第一のパラメータ算出処理と、求めたｋ_１Ｃ_ｆおよびｋ_２とから、下記式（２）：

From the first parameter calculation process for obtaining k ₁ C _f and k ₂ and the obtained k ₁ C _f and k ₂ , the following equation (2):

に基づいて、Ｃ_ｆ・Ｖ_ｄを求める第二のパラメータ算出処理と、求めたＣ_ｆ・Ｖ_ｄを利用し、データベースに格納されたｍＣ_ｆとＣ_ｆ・Ｖ_ｄとの関係を表す相間式からｍＣ_ｆを求める第三のパラメータ算出処理と、下記式（４）または（５）：

Based _on, C a second parameter calculation processing for calculating the f · _{V d,} the obtained _C f · _{V d} using, between the phases representing the relationship between mC _f and _C f · _{V d} stored in the database formula And a third parameter calculation process for obtaining mC _f from the following equation (4) or (5):

に基づいて単位領域あたりの脳血流量を求める脳血流量算出処理と、得られた脳血流量を対応する単位領域と関連づけて表示する表示処理と、を順次実行させることにより、被験者の核医学画像に基づいて、コンピュータが脳血流量の値を算出するための方法である。

The cerebral blood flow calculation process for obtaining the cerebral blood flow per unit area based on the above, and the display process for displaying the obtained cerebral blood flow in association with the corresponding unit area are sequentially executed, whereby the subject's nuclear medicine This is a method for a computer to calculate a value of cerebral blood flow based on an image.

  Also in the cerebral blood flow quantification program and the cerebral blood flow quantification method according to the present invention, the time activity curve for fitting with the above equation (1) is created based on a cross-sectional image (slice) from the parietal lobe to the cerebellar tent. Preferably, it may be created based on a cross-sectional image (slice) of a region covering the entire ventricle from the parietal lobe to the cerebellar tent.

  Similarly, in the cerebral blood flow quantification program and the cerebral blood flow quantification method according to the present invention, the interphase equation stored in the database is expressed by the following equation (3), similarly to the cerebral blood flow quantification device. can do.

  By using the cerebral blood flow quantification device, the cerebral blood flow quantification program, and the cerebral blood flow quantification method according to the present invention, cerebral blood flow quantification based on nuclear medicine images can be performed without collecting arterial blood. .

It is a flowchart which shows the outline | summary of the process in the preferable aspect of the cerebral blood flow rate determination apparatus which concerns on this invention. It is a flowchart which shows the outline | summary of the process in the preferable aspect of the cerebral blood flow rate determination apparatus (part which concerns on a cerebral blood flow rate calculation part) which concerns on this invention. It is a functional block diagram in a desirable mode of a cerebral blood flow rate quantification device concerning the present invention. It is a figure which shows the system configuration | structure in the preferable aspect of the cerebral blood flow processing apparatus which concerns on this invention. It is a figure which shows the structure in the preferable aspect of the cerebral blood flow rate determination program which concerns on this invention. The figure which shows the position of the right-and-left MCA area | region which calculated cerebral blood flow. (Outlined portion. Displayed on behalf of one slice.) The figure which shows the correlation between the cerebral blood flow volume calculated | required with the method which concerns on this invention (vertical axis | shaft: non-blood-collecting ARG) and the conventional method (horizontal axis | shaft: One-point arterial blood sampling ARG).

  Hereinafter, the present invention will be described in more detail with reference to the drawings. However, it should be noted that the following description is merely an example of a preferable embodiment and is not intended to limit the contents of the present invention.

(Processing by cerebral blood flow quantification device)
FIG. 1 is a diagram showing an outline of processing of the cerebral blood flow rate quantification apparatus 10 according to the present invention. The cerebral blood flow quantification method according to the present invention can be realized by operating the cerebral blood flow quantification apparatus 10 according to the present invention.

  First, the cerebral blood flow rate quantification apparatus 10 acquires a plurality of temporally continuous nuclear medicine image data (step S11). In a preferred embodiment, the nuclear medicine image data acquired here is used not only to create a time activity curve described later but also to quantitate cerebral blood flow. In this example, the nuclear medicine image data at each time point is composed of a plurality of spatially continuous tomographic images (preferably transverse images). In addition to SPECT data, PET data can be preferably used as nuclear medicine image data.

Next, the cerebral blood flow rate quantification apparatus 10 creates a time activity curve using the acquired nuclear medicine image data (step S12). The time activity curve is a curve representing a change in the count value with time, and can be obtained by plotting the count value against the elapsed time after administration of the radiopharmaceutical. In creating the time activity curve, it is necessary to extract pixels corresponding to the site where the radiopharmaceutical is taken. The extraction of the pixel can be performed by various methods. For example, for nuclear medicine image data at each time point, threshold values based on the maximum value and the minimum value of the count are set, and the threshold values indicate the upper limit and the lower limit, respectively. For example, a count 30% smaller than the maximum value and a count 10% larger than the minimum value are set as the upper and lower thresholds, respectively. Then, the count above the threshold value based on the maximum value and the count below the threshold value based on the minimum value are discarded, and the remaining pixels can be extracted as pixels used to create a time activity curve.
The time activity curve can be obtained by calculating the count average value of the extracted pixels at each time point and plotting it against the elapsed time after administration of the radiopharmaceutical.

  The cerebral blood flow rate quantification apparatus 10 adds the following formula (1) to the created time activity curve:

（Ｂ（ｔ）：時刻ｔにおける放射能カウント、Ｓ（τ）：標準入力関数、ｋ_１：脳血流量、ｋ_２：ｋ_１／Ｖ_ｄ）をフィッティングさせることにより、ｋ_１Ｃ_ｆおよびｋ_２を求める（ステップＳ１３）。

By fitting (B (t): radioactivity count at time t, S (τ): standard input function, k ₁ : cerebral blood flow, k ₂ : k ₁ / V _d ), k ₁ C _f and k ₂ is obtained (step S13).

Then, the cerebral blood flow rate quantification apparatus 10 calculates the following formula (2) from k ₁ C _f and k ₂ calculated above:

に基づいて、Ｃ_ｆ・Ｖ_ｄを求め（ステップＳ１４）、データベース３０からｍＣ_ｆとＣ_ｆ・Ｖ_ｄとの関係を表す相間式を読み出し、上記Ｃ_ｆ・Ｖ_ｄを当該相間式に適用して、ｍＣ_ｆを求める（ステップＳ１５）。

C _f · V _d is obtained based on the equation (step S14), the interphase equation representing the relationship between mC _f and C _f · V _d is read from the database 30, and the above C _f · V _d is applied to the interphase equation. MC _f is obtained (step S15).

Next, the cerebral blood flow rate quantification apparatus 10 reads the standard input function S (t) from the database 30 and uses the mC _f obtained above to obtain the following formula (4) or (5):

に基づいて単位領域あたりの脳血流量を求める（ステップＳ１６）。

Based on the above, the cerebral blood flow per unit area is obtained (step S16).

FIG. 2 is a diagram showing the flow of processing in the step of calculating cerebral blood flow (step S16, hereinafter referred to as “cerebral blood flow calculation step”). The cerebral blood flow rate quantification apparatus 10 substitutes mC _f and the standard input function S (t) read from the database 30 into the above formula (4) or (5), and counts the nuclear medicine image count value B (t) (formula A table representing the cerebral blood flow k ₁ corresponding to the left side of the above equation (5) (when quantifying using equation (5)) is created (step S21).

The table can be created using a known method. For example, it is possible to use a method in which cerebral blood flow k ₁ corresponding to all count values from the minimum count to the maximum count of all pixels in the nuclear medicine image used for quantification is obtained by fitting, and this is collected as a table. . At this time, the value of the partition coefficient V _d, and a value determined in advance known methods, may have been substituted into the above formula (4) or (5). Thus, it is possible to reduce the number of parameters to be used for fitting, can improve the calculation accuracy of k _1.

  In addition, the head nuclear medicine image used for the calculation of the cerebral blood flow can be used by selecting a plurality of temporally continuous images from the nuclear medicine image data acquired in step S11. For example, when quantification is performed based on the above equation (4), the sum of the count values is taken over all the time points for each pixel in the selected image to obtain the radioactivity count B (t). Then, among the imaging time of the selected series of images, a time that is intermediate between the earliest imaging time and the latest imaging time is set as an image data acquisition time t. For example, when quantification is performed using 6 to 15 phase data among a series of image data obtained by performing 15 phase imaging in 1 phase 2 minutes, the data acquisition time t is set to 20 minutes and used for the calculation.

The cerebral blood flow rate quantification apparatus 10 reads the value of the cerebral blood flow rate k ₁ corresponding to the count value B (t) obtained in each pixel from the table (step S22). The cerebral blood flow rate quantification device 10, a quantitative processing of the cerebral blood flow k _1, determines whether performed on all pixels determined to cerebral blood flow k ₁ (step S23), it has been completed for all pixels If not (No in step S23), the remaining pixels are subjected to quantitative processing (step S22). When the quantification process is completed for all pixels (Yes in step S23), the cerebral blood flow calculation process is completed.
In the example has exemplified the case of reading cerebral blood flow k ₁ from the table created, it is of course also possible to use other methods. For example, the cerebral blood flow k ₁ corresponding to the count value for each pixel may be obtained by fitting for all the pixels for which the cerebral blood flow k _{1 is} to be obtained without creating the table.

The cerebral blood flow rate quantification apparatus 10 outputs the obtained cerebral blood flow rate k ₁ by an output device such as a display (step S17). The output can be performed in various ways. For example, it represents the value of the cerebral blood flow k ₁ obtained at each pixel, the corresponding pixel, and a method of displaying by the luminance and color according to the quantified value of the cerebral blood flow, the position of each pixel in the numerical values of the coordinates etc. , it is possible to use a method such as displaying a numerical cerebral blood flow k ₁ in the form which corresponds to that number.

(Configuration of cerebral blood flow quantification device)
FIG. 3 is a diagram showing a configuration of the most preferable aspect of the cerebral blood flow rate quantification apparatus 10 according to the present invention. The cerebral blood flow rate quantification apparatus according to the present invention can be configured as a computer that reads a cerebral blood flow rate quantification program 300 described later.
In the present embodiment, the SPECT image is used as the head nuclear medicine image, but the head nuclear medicine image in the present invention is not necessarily limited to this. In addition to SPECT images, PET images and planar images can also be used.

In a preferred embodiment, the cerebral blood flow rate quantification device 10 according to the present invention includes a database 30, a nuclear medicine image data acquisition unit 40, a time activity curve creation unit 50, a first parameter calculation unit 60, a second parameter calculation unit 70, A third parameter calculation unit 80, a cerebral blood flow rate calculation unit 90, and an output unit 100 are included. The cerebral blood flow rate quantification device 10 according to the present invention is coupled to a SPECT device 200 via an electric communication line, and the SPECT device 200 includes a γ camera 210 and an image reconstruction unit 220. In this example, the projection data acquired by the γ camera 210 is reconstructed into a series of tomographic images by the image reconstruction unit 220.
In this example, the parameter determination device 20 according to the present invention is incorporated as a part of the cerebral blood flow rate quantification device 10.

  FIG. 4 shows a system configuration in the most preferable mode of the cerebral blood flow rate quantification apparatus 10 according to the present invention. In a preferred embodiment, the cerebral blood flow rate quantification apparatus 10 includes a CPU 230, a memory 240, an output device 250 such as a monitor, a communication interface 260, and an input device 270 such as a keyboard connected via a bus 280. . In addition, the image processing apparatus 10 may include a CD-ROM drive, a USB interface, and the like. The communication interface 260 is used for connecting to the SPECT apparatus 200. The memory 31 stores a cerebral blood flow rate quantification program 300 according to the present invention.

The database 30 stores at least an interphase expression and a standard input function representing the relationship between mC _f and C _f · V _d . The standard input function is obtained by a known method (for example, a method described in a document (Hidehiro Iida et al, The Journal of Nuclear Medicine, December 1994, Vol. 35, No. 12, pp 2019-2030)). Is available. mC _f and _C f · _{V d} and a phase expression representing a relationship obtains the mC _f and _C f · _{V d} in a plurality of _subjects, linear approximation graph plotting mC _f against C f · _{V d} Can be requested. Note that mC _f can be obtained by a known method using arterial blood collection data, and C _f · V _d can be obtained by fitting a time activity curve to a two-compartment model equation.

  The nuclear medicine image data acquisition unit 40 executes the process according to step S11 described above. Specifically, a plurality of temporally continuous nuclear medicine image data is acquired from the SPECT apparatus. The nuclear medicine image data acquired here is used not only for creating a time activity curve but also for quantifying cerebral blood flow. In this example, the nuclear medicine image data at each time point is composed of a plurality of spatially continuous tomographic images (preferably transverse images).

The time activity curve creation unit 50 executes the process according to step S12 described above.
The first parameter calculation unit 60 executes the process according to step S13 described above.
The second parameter calculation unit 70 executes the process related to step S14 described above.
The third parameter calculation unit 80 executes the process according to step S15 described above.
The cerebral blood flow rate calculation unit 90 executes the processes according to Step S16 and Steps S21 to 23 described above.
And the output part 100 performs the process which concerns on step S17 mentioned above.
The most preferable aspect of the cerebral blood flow rate quantification device 10 according to the present invention has been described above.

Next, the cerebral blood flow rate determination program 300 according to the present invention will be described. As described above, the cerebral blood flow rate quantification apparatus 10 according to the present invention can be configured as a computer that has read the cerebral blood flow rate quantification program 300.
FIG. 5 is a diagram showing the configuration of the most preferred embodiment of the cerebral blood flow rate quantification program according to the present invention. In a preferred embodiment, the cerebral blood flow quantification program 300 includes an input module 310, a time activity curve creation module 320, a first parameter calculation module 330, a second parameter calculation module 340, and a third parameter calculation module 350. And a cerebral blood flow calculation module 360 and an output module 370. In a preferred embodiment, the first parameter calculation module 330, the second parameter calculation module 340, and the third parameter calculation module 350 constitute a parameter determination program 380 according to the present invention, and a cerebral blood flow rate determination program 300.

  The input module 310 causes the computer to execute the process according to step S11. The time activity curve creation module 320 causes the computer to execute the process according to step S12. The first parameter calculation module 330 causes the computer to execute the process according to step S13. The second parameter calculation module 340 causes the computer to execute the process according to step S14. The third parameter calculation module 350 causes the computer to execute the process according to step S15. The cerebral blood flow rate calculation module 360 causes the computer to execute the processes according to step S16 and steps S21 to S24. The output module 370 causes the computer to execute the process according to step S17.

EXAMPLES Next, although an Example is described and this invention is demonstrated in more detail, this invention is not limited to this content. In the examples below, the value of V _d was used 38.
(Creation of interphase formula)
Using cerebral blood flow SPECT data of a plurality of subjects (based on subjects who received 222 MBq of N-isopropyl-4-iodoamphetamine ( ¹²³ I) hydrochloride injection solution (manufactured by Nippon Mediphysics)), C _f · V _d MC _f was plotted against and linear regression was performed to obtain the following interphase equation (6).

(SPECT imaging)
56 example subjects (average 64.5 years old, 11 to 82 years old, male / female = 44/12) to, hydrochloride N- isopropyl-4-iodo-amphetamine ⁽¹²³ I) Injection (manufactured by Nippon Medi-Physics Co., Ltd.) And 222 MBq were administered, 15 SP dynamic images were continuously acquired in 2 minutes per phase, and SPECT images were obtained by performing image reconstruction for each phase (matrix size 64 × 64, pixel size 3.44 mm).

(Quantification of cerebral blood flow using the method according to the present invention)
Using the captured SPECT images of each phase, a slice of the region covering the entire ventricle from the cerebellar tent to the parietal lobe (about 15 slices from the cerebellar tent) is selected, and the average count value +2.5 SD for each pixel is calculated. Pixels were extracted with the upper limit as 10% + 2.5SD of the upper limit and the lower limit. The average value of the count values was calculated for all the extracted pixels. A time activity curve was created using the calculated average count value.

  In the created time activity curve, the following formula (1):

（Ｂ（ｔ）：時刻ｔにおける放射能カウント、Ｓ（τ）：標準入力関数、ｋ_１：脳血流量、ｋ_２：ｋ_１／Ｖ_ｄ）をフィッティングさせることにより、ｋ_１Ｃ_ｆおよびｋ_２を求め、さらに、下記式（２）：

By fitting (B (t): radioactivity count at time t, S (τ): standard input function, k ₁ : cerebral blood flow, k ₂ : k ₁ / V _d ), k ₁ C _f and k ₂ is obtained, and the following formula (2):

に基づいて、Ｃ_ｆ・Ｖ_ｄを求めた。求めたＣ_ｆ・Ｖ_ｄを上記式（６）に代入してｍＣ_ｆを求めた。

Based on the above, C _f · V _d was determined. The obtained C _f · V _d was substituted into the above equation (6) to obtain mC _f .

Using the calculated mC _f , the following formula (4):

The cerebral blood flow (mL / min / 100 g) in the left and right MCA regions was determined according to As the left and right MCA regions for which quantification was performed, a region set in advance on the standard brain was used (FIG. 6). In addition, for calculation of cerebral blood flow, data of 6 to 15 phases (20 minutes from 10 minutes after administration of the radiopharmaceutical) were used.

(Quantification by conventional method (single-point blood sampling ARG method))
On the other hand, in parallel with the dynamic SPECT imaging, arterial blood is collected 10 minutes after the start of imaging, and a known method (for example, the document Hidehiro Iida et al, The Journal of Nuclear Medicine, December 1994, Vol. 35, No. 12, pp2019). The scaling factor (mC _f ) was determined according to the method described in −2030), and the cerebral blood flow (mL / min / 100 g) was determined.

(Comparison of quantitative values of cerebral blood flow between the method according to the present invention and the conventional method)
FIG. 7 shows the correlation between the cerebral blood flow obtained by the method according to the present invention and the cerebral blood flow obtained by the conventional method. As shown in this figure, the cerebral blood flow obtained by the method according to the present invention (indicated as “non-blood-collected ARG” in FIG. 7) and the cerebral blood flow obtained by the conventional method (in FIG. 7, “one-arterial blood-collected ARG”). ")", A very good correlation was obtained, and both values showed good agreement.
Thus, it has been shown that by using the method according to the present invention, the cerebral blood flow can be quantified satisfactorily without using arterial blood collection data.

  The present invention can be used in the field of image processing programs.

10 cerebral blood flow rate quantification device 20 parameter determination device 30 database 40 nuclear medicine image data acquisition unit 50 time activity curve creation unit 60 first parameter calculation unit 70 second parameter calculation unit 80 third parameter calculation unit 90 cerebral blood flow rate Calculation unit 100 Output unit 200 SPECT device 210 Gamma camera 220 Image reconstruction unit 230 CPU
240 memory 250 monitor 260 communication interface 270 keyboard 280 bus 300 cerebral blood flow rate determination program 310 input module 320 time activity curve creation module 330 first parameter calculation module 340 second parameter calculation module 350 third parameter calculation module 360 cerebral blood Flow rate calculation module 370 Output module 380 Parameter determination program

Claims (10)

Interphase expression representing the relationship between mC _f and C _f · V _d (mC _f : scaling factor, C _f : scaling factor determined based on time activity curve of cranial nuclear medicine image, V _d : brain blood distribution coefficient) And a database that stores
The following formula (1) is applied to the time activity curve representing the time-dependent change of the count value in the subject's nuclear medicine image.

By fitting (B (t): radioactivity count at time t, S (τ): standard input function, k ₁ : cerebral blood flow, k ₂ : k ₁ / V _d ), k ₁ C _f and k _A first parameter calculation unit for obtaining ₂ ;
From the determined k ₁ C _f and k ₂ , the following formula (2):

A second parameter calculation unit for _obtaining C _f · V _d based on
A third parameter calculation unit for obtaining mC _f from C _f · V _d obtained by the second parameter calculation unit and the interphase equation stored in the database;
A parameter determination device comprising: The parameter determination device according to claim 1, wherein the time activity curve is created based on an image of a portion corresponding to a transverse image from the parietal lobe to the cerebellar tent. The interphase formula stored in the database is expressed by the following formula (3):

The parameter determination device according to claim 1 or 2, represented by (R: proportionality constant). A time activity curve representing a change over time of a count value in a head nuclear medicine image of a subject and an interphase expression representing a relationship between mC _f and C _f · V _d stored in a database can be used, and a storage unit, a CPU, A computer program for operating a computer comprising: the computer program executed by the CPU,
The following formula (1) is applied to the time activity curve representing the time-dependent change of the count value in the subject's nuclear medicine image.

A first parameter calculation unit for obtaining k ₁ C _f and k ₂ by fitting
From the determined k ₁ C _f and k ₂ , the following formula (2):

A second parameter calculation unit for _obtaining C _f · V _d based on
A parameter determination program that operates as a third parameter calculation unit that calculates mC _f from C _f · V _d calculated by the second parameter calculation unit and the interphase equation stored in the database. A method for determining parameters used by a computer to quantify cerebral blood flow based on a nuclear medicine image of a subject comprising:
The following formula (1) is applied to the time activity curve representing the time-dependent change of the count value in the subject's nuclear medicine image.

A first parameter calculation process for obtaining k ₁ C _f and k ₂ by fitting
From the determined k ₁ C _f and k ₂ , the following formula (2):

A second parameter calculation process for _obtaining C _f · V _d based on
A parameter that sequentially executes a third parameter calculation process for obtaining mC _f from an interphase expression representing a relationship between mC _f and C _f · V _d stored in the database, using the obtained C _f · V _d Decision method. At least the following (a) and (b):
(A) Standard input function S (t): a function representing a time change of the inflow of a radiopharmaceutical into a standard brain (a) Interphase equation: an interphase equation representing a relationship between mCf and Cf · Vd,
(MCf: Scaling factor, Cf: Scaling factor obtained based on time activity curve of head nuclear medicine image, Vd: Brain blood distribution coefficient)
A nuclear medicine image data acquisition unit for acquiring a head nuclear medicine image of the subject;
A time activity curve creating unit for creating a time activity curve representing a time-dependent change of the count value in the head nuclear medicine image;
In the time activity curve, the following formula (1):

(B (t): Radioactivity count at time t, k ₁ : cerebral blood flow, k ₂ : k ₁ / V _d ) by fitting, a first parameter calculation unit for obtaining k ₁ C _f and k ₂ When,
From the determined k ₁ C _f and k ₂ , the following formula (2):

A second parameter calculation unit for _obtaining C _f · V _d based on
A third parameter calculation unit for obtaining mC _f from C _f · V _d obtained by the second parameter calculation unit and the interphase equation stored in the database;
Following formula (4) or (5):

A cerebral blood flow rate calculation unit for obtaining a cerebral blood flow rate per unit region based on (t ₁ : imaging start time of the head nuclear medicine image, t ₂ : imaging end time of the head nuclear medicine image);
A display unit that displays the cerebral blood flow obtained in association with the corresponding unit region;
A cerebral blood flow quantification apparatus comprising: The cerebral blood flow rate quantification device according to claim 6, wherein the time activity curve is created based on an image of a portion corresponding to a transverse image from the parietal lobe to the cerebellar tent. The interphase formula stored in the database is expressed by the following formula (3):

The cerebral blood flow rate quantification device according to claim 6 or 7, represented by (R: proportional constant). A nuclear medicine image of a subject to whom a radiopharmaceutical is administered and a correlation formula and a standard input function representing a relationship between mC _f and C _f · V _d stored in a database can be used, and a storage means and a CPU are provided. A computer program for operating the computer, which is executed by the CPU,
A nuclear medicine image data acquisition unit for acquiring a head nuclear medicine image of the subject;
A time activity curve creating unit for creating a time activity curve representing a time-dependent change of the count value in the head nuclear medicine image;
In the time activity curve, the following formula (1):

A first parameter calculation unit for obtaining k ₁ C _f and k ₂ by fitting
From the determined k ₁ C _f and k ₂ , the following formula (2):

A second parameter calculation unit for _obtaining C _f · V _d based on
A third parameter calculation unit for obtaining mC _f from C _f · V _d obtained by the second parameter calculation unit and the interphase equation stored in the database;
Following formula (4) or (5):

Is applied to the head nuclear medicine image of the subject, and the cerebral blood flow calculation unit for obtaining the cerebral blood flow per unit region in the nuclear medicine image,
A cerebral blood flow quantification program that operates as a display unit that displays the cerebral blood flow obtained in association with a corresponding unit region. A computer-based method for quantifying cerebral blood flow based on a nuclear medicine image of a subject,
Nuclear medicine image data acquisition processing for acquiring a subject's head nuclear medicine image;
A time activity curve creating process for creating a time activity curve representing a time-dependent change in a count value in the head nuclear medicine image;
The obtained time activity curve has the following formula (1):

A first parameter calculation process for obtaining k ₁ C _f and k ₂ by fitting
From the determined k ₁ C _f and k ₂ , the following formula (2):

A second parameter calculation process for _obtaining C _f · V _d based on
A third parameter calculation process for obtaining mC _f from an interphase expression representing a relationship between mC _f and C _f · V _d stored in the database using the obtained C _f · V _d ;
Following formula (4) or (5):

Cerebral blood flow calculation processing to apply to the head nuclear medicine image of the subject, to determine the cerebral blood flow per unit region in the head nuclear medicine image,
A cerebral blood flow quantification method that sequentially executes display processing for displaying the obtained cerebral blood flow in association with a corresponding unit region.

Images