JP2002267756A - Body kinetics analytic method of medicine using compartment model - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method capable of performing a compartment model analysis of body kinetics of a medicine while freely constituting a compartment model. SOLUTION: Compartments are regarded to be entirely equal regardless of the order of arrangement, and all combinations of the compartments are described with only text data. A compartment model analytic program interprets the compartment model from the description method, and calculates and outputs intended speed constant image data from age-based medicine distribution image data obtained by use of a medicine body kinetics imaging technique such as PET or the like and age-based medicine blood concentration data obtained by a method such as arterial blood collection or the like.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

TECHNICAL FIELD The present invention relates to a PET (Positron)
Emission Tomography (positron emission tomography), SP
ECT (Single Photon Emission Computed Tomograph
y; single photon emission tomography, MRI (Magnetic Reso
nance Imaging; Nuclear Magnetic Resonance Imaging), autoradiography, etc., related to the biodistribution of a drug obtained using a technique that can measure the biodistribution of a drug over time and image the pharmacokinetics of the drug. based on the temporal drug distribution image data, to a method of obtaining a rate constant target image data by performing a compartmental model analysis.

[0002]

Pharmacokinetics of the Prior Art] drugs, absorption, distribution, metabolism,
There is a process called excretion, and these four processes are also composed of complicated processes and are influenced by various factors. Furthermore, factors affecting these four processes not only affect one process, but also other processes, and the four processes affect each other. A representative method for quantitatively elucidating these processes is compartment model analysis. The part of interest in the actual living body is approximated by a simple compartment model, and the absorption rate, metabolic rate, and the like in each process are obtained.

[0003] Conventionally, rate constant image data has been obtained from compartmentalized drug distribution image data obtained by PET, SPECT, MRI, or the like, using compartment model analysis. This is because the time-lapse drug distribution image data itself is the absorption, distribution, metabolism,
This is because it is not only a comprehensive result of excretion, but rather, it is often desirable to see functional changes such as absorption rate and metabolic rate in tissues that constitute individual processes. How functional changes is that if needed depends on the purpose such sites or biological reactions of interest, compartment model to be applied are also different. By analyzing with an appropriate compartment model, the activity of a target drug such as a metabolic enzyme and a drug receptor can be examined. Until now, when a compartment model analysis is performed, there is no method for freely describing the compartment model configuration and a program for performing calculations according to the method, and a program limited to a relatively frequently used two or three compartment model has been used. Was. In these programs, for example, in a three-compartment model, only four parameter inputs are assumed, and it is not possible to cope with a more complicated compartment model configuration.

[0004]

[0003] Compartment model analysis has been performed by a diagrammatic method since the days when computers were not developed, but recently, the development of computers has made it possible to perform complicated calculations. Not only a simple compartment model but also a compartment model that is closer to the actual state in a living body is in an environment where calculations can be realistically made. However, the use of computers in this field has not been sufficiently advanced, and calculations with commonly used two or three compartment models are usually performed. However, in the analysis using the analysis program is limited to 2 or 3 compartments, can not correspond to the progress of advanced drug pharmacokinetics imaging technology using PET or the like, become compartment model analysis and neck in research in this field There was a side.

In the analysis using a conventional analysis program limited to a two- or three-compartment model, when a drug exhibiting a pharmacokinetics that deviates from the simple model defined in the program cannot provide good results. There is. For example, if the drug is absorbed from multiple routes, if the drug is metabolized by multiple metabolic processes, if the drug binds to multiple drug receptors, if the drug disappears from multiple excretion routes, etc. In the case where is moved or metabolized along a plurality of pathways in the body at the same time, good results cannot be obtained with the conventional analysis program. Therefore, in the conventional technique, a compartment model analysis of the pharmacokinetics of such a drug could not be performed, and a detailed study of the pharmacokinetics could not be performed. The present invention has been made in view of the present situation of the compartment model analysis, and has as its object to provide a method that allows a compartment model to be freely configured to perform a compartment model analysis of a pharmacokinetics of a drug.

[0006]

When a diagrammatic method is used to generally describe a compartment model, a program for drawing and interpreting the model is required. A program for drawing on a computer has a drawback in that it requires a program that depends on each computer system itself, and is therefore less versatile. Also, it is difficult to create a program that assumes any combination to interpret and calculate the compartment model.

Therefore, in order to achieve the above object, the present inventor provides a general description method of a compartment model using only text data, and performs a compartment model analysis from time-lapse drug distribution image data to obtain rate constant image data. Method developed.

That is, the method for analyzing the pharmacokinetics of a drug according to the present invention performs a compartment model analysis on the basis of time-dependent drug distribution image data and time-dependent drug concentration data in the blood concerning the biodistribution of the drug, and performs each pixel-by-pixel analysis. In the method for analyzing the pharmacokinetics of a drug for determining a rate constant between compartments, a first step of inputting a compartment model configuration using matrix-format text data representing a binding mode between a predetermined number of compartments, A second step of inputting an initial value of a rate constant between compartments, a third step of inputting time-dependent drug distribution image data and time-dependent blood drug concentration data, and coupling between compartments based on the input compartment model configuration A fourth step of extracting and storing a certain combination; From the compartment model configuration, the initial value of the rate constant between each compartment, the time-dependent drug distribution image data, and the time-dependent blood drug concentration data, the memorized drug concentration transfer amount between the compartments with the binding is successively measured at every minute time. A fifth step of calculating and calculating the temporal drug distribution image data by the compartment model simulation, and a difference between the input temporal drug distribution image data and the temporal drug distribution image data obtained by the compartment model simulation (for example, A sixth step of calculating the sum of squares of the residual), and a seventh step of adjusting the rate constant between the compartments by a non-linear optimization method so that the difference between the two time-lapse drug distribution image data is reduced. Until the difference between the two time-lapse drug distribution image data converges to a specified value or less. Characterized by repeating the seventh step from said fifth step.

Here, the fourth step is added for the purpose of eliminating useless processing from the enormous repetition calculation in the fifth to seventh steps, but may be omitted. In that case, in the fifth step, the amount of drug concentration transferred between all compartments is calculated. In addition, the first to third
May be performed before the fifth step, and the order of execution does not matter. The fourth step may be executed after the first step and before the fifth step, and the execution order is not fixed.

An embodiment in which the execution order of the first to third steps and / or the fourth step is changed within the scope of the object of the present invention, and an embodiment in which the fourth step is omitted are also included in the present invention. Included in the range. There are no flow, no flow, variable velocity constant value,
There is a flow and the value of the rate constant can be specified by one of the fixed. The adjustment of the speed constant in the seventh step is performed only for the speed constant whose value is designated as variable.

A computer-readable recording medium according to the present invention performs a compartment model analysis on the basis of time-dependent drug distribution image data and time-dependent drug concentration data in the blood regarding the distribution of a drug in the body, and calculates a pixel-by-pixel A method for analyzing the pharmacokinetics of a drug for determining a rate constant of a first step of inputting a compartment model configuration using matrix-format text data representing a binding mode between a predetermined number of compartments; A second step of inputting an initial value of a rate constant between compartments, a third step of inputting time-dependent drug distribution image data and time-dependent blood drug concentration data, and coupling between compartments based on the input compartment model configuration A fourth step of extracting and storing a certain combination; From the compartment model configuration, the initial value of the rate constant between each compartment, the time-dependent drug distribution image data and the time-dependent drug concentration data in the blood, the amount of drug concentration transfer between the compartments with the stored bonds is sequentially calculated every minute time. A fifth step of calculating temporal drug distribution image data by compartment model simulation; and a sixth step of calculating a difference between the input temporal drug distribution image data and the temporal drug distribution image data obtained by compartment model simulation. And a seventh step of adjusting a rate constant between the compartments by a non-linear optimization method such that a difference between the two time-lapse drug distribution image data is reduced, and The fifth step until the difference between the distribution image data converges to a specified value or less. The pharmacokinetics analysis method of drugs repeating et seventh step, is obtained by recording a program for causing a computer to execute.

According to the present invention, it is possible to describe a compartment model of any combination by regarding all compartments as being equal regardless of their arrangement order and describing combinations of all compartments. Also, by describing it with text data only, the description data of the compartment model according to the present invention has high versatility that can be used on any computer system. The program according to the present invention interprets the compartment model from the description method, and the time-dependent drug distribution image data obtained by using a drug pharmacokinetic imaging technique such as PET and the time-dependent drug concentration in blood obtained by arterial blood sampling and the like. From the data, target speed constant image data is calculated and output.

[0013]

Embodiments of the present invention will be described below with reference to the drawings. The method of obtaining the time-dependent drug distribution image data and the time-dependent blood drug concentration data regarding the biodistribution of the drug that can be used in the present invention is not particularly limited. The processing method of the present invention will be described by taking as an example the temporal drug distribution image data relating to the distribution and the temporal blood drug concentration data obtained by arterial blood sampling.

The time from the start to the end of the PET measurement is divided into n _f [frames] at predetermined time intervals, and the radiation emitted from the drug in the local tissue in the body measured during each frame time is counted. Then, the value of the time-integrated average value of the drug concentration in the in-vivo local tissue within each frame time is imaged to obtain n _f sets of time-dependent drug distribution image data. During the period from the start to the end of the PET measurement, blood is collected from the artery at predetermined time intervals to measure the drug concentration in the blood, and the time-dependent change in the drug concentration flowing into local tissues in the body, that is, blood over time is measured. Intermediate drug concentration data IF (t) is obtained.

FIG. 1 is a schematic configuration diagram of a compartment model analysis system according to the present invention. The input device 11 includes a keyboard. The compartment model analysis program 14 according to the present invention is loaded in the main memory of the central processing unit 10. Central processing unit 1
0 further includes an auxiliary storage device 12 for preliminarily storing time-lapse drug distribution image data which is an input of the present program 14, and for storing rate constant image data which is an output of the present program 14. . The output device 13 includes a display device capable of outputting characters. The compartment model analysis program is recorded on a recording medium such as a floppy disk, a CD-ROM,
This system can be realized by a personal computer or the like loaded with the compartment model analysis program.

As an example, FDG (2-fluoro-2-deoxy-D-glucose) is used as a drug, and PET is used.
The composition of a compartment model and a method of describing the compartment model will be described by taking as an example the case of obtaining temporal drug distribution image data in the brain and analyzing the compartment model of the glucose metabolism process in the brain.

FIG. 2 is a diagram showing an example of a compartment model configuration when the drug is FDG. Compartment 1 contains FDG in plasma in blood vessels, compartment 2 contains FDG transferred into brain tissue, and compartment 3
Is FDG phosphorylated by hexokinase in brain tissue at position 6 (FDG-6-P: 2-fluoro-2).
- represents a deoxy -D- glucose-6-phosphate), it has the respective concentrations as values. K ₁₂ and k ₂₁ are the F from the plasma to the brain tissue and from the brain tissue to the plasma, respectively.
4 shows a transfer rate constant of DG. k ₂₃ from FDG by 6 of phosphorylation by hexokinase in the brain tissue FDG-
The production rate constant of 6-P is shown. k ₃₂ shows the dephosphorylation rate constant when FDG is produced from FDG-6-P by 6 position dephosphorylation by hexokinase. The instructions for the compartment model configuration to the compartment analysis program include a “compartment configuration matrix” that determines the coupling mode of the compartments, and a “rate constant initial value matrix” that describes the initial values for estimating each rate constant by the nonlinear optimization method. Is performed.

FIG. 3 is a diagram showing an example of a compartment configuration matrix. As shown in FIG. 3, the compartment configuration matrix has a connection state between the compartment i and the compartment j in an i-th row and a j-th column. The value of the i-th row and the j-th column of the compartment configuration matrix indicates whether or not the drug flows from the compartment i to the compartment j, and is “0”.
Is "does not flow", "1" is "flows (variable value)", "
2 "represents" flow (fixed value) ", respectively. If the drug flows between the compartments, there is a bond between the compartments; if the drug does not flow between the compartments, there is no bond between the compartments. Also,
Compartment 1 contains FDG in blood vessels in the case of FDG.
It is assumed that the compartment is fixed to which the blood drug concentration data with time corresponding to the DG concentration is input.

[0019] In this example, since the plasma FDG is impossible be changed directly brain tissue FDG-6-P, does not flow from the compartment 1 to 3, i.e. one row and three columns of compartments component matrices corresponding to k ₁₃ The component is described as "0". Also, since the reverse flow is not possible, k
The 3rd row and 1st column component of the compartment configuration matrix corresponding to ₃₁ is described as “0”. k _12, it changes the migration rate constant to the brain tissue in accordance with substantially the bloodstream for each brain topically. Thus, one row and two columns component of k ₁₂ corresponding to the compartment configuration matrix flows with a value variable, is described as i.e. "1". k ₂₃
Is determined according to the enzyme activity of hexokinase in each brain region.
The generation rate constant of DG-6-P changes. Therefore, k ₂₃
The two-row, three-column component of the compartment configuration matrix corresponding to flows in a variable manner, that is, is described as “1”. k ₃₂ is equivalent to the dephosphorylation rate constant in the brain, its speed change per the brain topically so slow is negligible. Therefore it flows with a value fixed in this case, therefore, two columns component 3 rows of compartments component matrices corresponding to k ₃₂ is described as "2".

FIG. 4 is a diagram showing an example of a velocity constant initial value matrix. As shown in FIG. 4, the velocity constant initial value matrix has an initial value of the velocity constant k _ij in the i-th row and the j-th column of the matrix in the same format as the compartment configuration matrix. Drug is FD
In the case of G, it is possible to perform analysis using the compartment model shown in FIGS. 2 to 4 without any problem. For example, in order to examine the pharmacokinetics of a newly developed drug called A, a compartment model analysis is performed. When such a demand arises, the characteristics of the drug A cannot always be approximated by a simple model as in the case of FDG. For example, when it is necessary to consider non-specific binding simultaneously with specific binding of a drug to brain tissue, FIG.
It is necessary to consider the unbound drug A, the non-specifically bound drug A, and the specifically bound drug A compartment in the brain tissue as shown in FIG. Transfer rate of drug A from blood to brain tissue, transfer rate constant from unbound to specific binding, and non-specific binding in brain tissue to k ₁₂ , k ₂₃ , k ₂₄ , respectively the rate constant and _{k 21, k 32, k 42} , these k ₃₂
When considered is not changed by site, also k ₄₂ is negligibly small, the compartment configuration matrix in this case is as shown in FIG.

It is necessary to give each speed constant an initial value, but there is no fixed method for determining it.
For example, an enzyme reaction rate measured in vitro and a rate constant in the whole brain estimated using a different methodology in vivo or in vitro, which data is already known in the literature, are used. In the above-mentioned example of FDG, Sokoloff et al. Exemplified the value obtained in the whole brain from the theoretical formula analytically obtained from the pharmacokinetics of FDG and the corresponding experimental value as the initial value of the rate constant (Reivich, M. , Alavi, A., Wolf,
A., Fowler, J., Russell, J., Arnett, C., MacGrego
r, RR, Shive, CY, Atkins, H., Anand, A., Dan
n, R., Greenberg, JH, Glucose metabolic rate ki
netic model parameter determination in humans: the
lumpedconstants and rate constants for [ ¹⁸ F] fluor
odeoxyglucose and [ ¹¹ C] deoxyglucose, J. Cereb. Bl
ood Flow Metab., Vol. 5, pp. 179-192, 1985).

FIG. 7 shows a series of processes in the compartment model analysis of the present invention. PET actual measurement image data [PE
T _f : f = 1,..., N _f ] is stored in the auxiliary storage device 12 of the central processing unit 10 in advance, and then the compartment model analysis program 14 of the present invention is executed. Input various conditions. In step S1, a compartment configuration written in a matrix as shown in FIG. 8 is input and read into an array variable F (i, j).

[0023] FIG. 8 is obtained by assuming a compartment model consisting of three compartments as shown in FIG. 9, the number of compartments [n _c] can be arbitrarily specified. In FIG. 8, the value in the i-th row and the j-th column indicates whether or not the drug flows from compartment i to j, and is “0”, “0”.
1 "and" 2 "do not flow (there is no change in state due to drug movement or metabolic process, etc.), and flow at variable values (state change due to drug movement, metabolic process, etc., and their speed changes) ), Flowing at a fixed value (there is a change in state due to drug movement or metabolic processes, but the speed does not change)
Means The compartment 1 is fixed to a compartment for inputting blood drug concentration data over time.

In step S2, the initial values of the respective speed constants similarly written in a matrix are read into an array variable k (i, j).
The initial value of each rate constant is the same as in the case of conventional compartment model analysis, such as a value experimentally obtained from the whole brain or a part of tissue sample, a value estimated from some experimental result, etc. It is customary to use a value that does not seem to be significantly different from the final solution on some basis, but it is basically left to the user's judgment.

[0025] reads the n _f sets of PET actual image data stored on an auxiliary storage device (PET _f) In step S3, stores the main memory. Further, the time-dependent blood drug concentration data IF (t) is input from the input device. In step S4, based on the compartment configuration input in step S1, a combination having a combination is extracted from all the combinations of compartments, and the information is stored in an array variable. Here, the combination having a combination corresponds to the compartment configuration matrix shown in FIG. 8 having matrix components of 1 or 2. The purpose of extracting the combination having the combination first in this way is to eliminate useless processing from the enormous repetition calculation in steps S5 to S11.

The processing from step S6 to step S10 is performed for all pixels on the input PET measured image data. In step S6, steps S1 to S
When the time-dependent drug concentration data in blood (IF (t)) is input to the compartment model composed of the compartment configuration and the initial value of the rate constant set in 2, the value of the pixel becomes n _f during the PET measurement time. Theoretically calculate what changes the frame will follow.

[0027] FIG. 10 a detail of PET theoretical data calculated in step S6. Steps S602 to S609 are PE
In step S608, the process is repeated by changing the time t at intervals of Δt in step S608 from the T measurement start time (t = 0) to the measurement end time. In step S603, the value IF (t) of the time-dependent blood drug concentration data at the corresponding time t is put into the compartment 1.

In steps S604 to S606, a repetitive process is performed for all combinations of the combined compartments extracted in step S4 and stored in the array variables. In step S605, the drug concentration transfer speed between the compartments at time t is calculated and stored in the main memory. The rate of drug concentration transfer from compartment i to j is k _ij according to the principle of the compartment model.
× represented by C _i. Where k _ij is the rate constant from compartment i to j, and C _i is the drug concentration in compartment i.

In step S607, step S604
The change in the drug concentration in each compartment is calculated from the balance of the movement amount of the drug concentration in each compartment calculated in S606. The inflow rate of the drug concentration from compartment j to compartment i at time t is the product of the rate constant k _ji from compartment j to compartment i and the drug concentration C _j (t) in compartment j at time t, ie k _ji × C _j (t). Therefore, the sum of the drug concentration inflow rates from each compartment into the compartment i at the time t can be represented by [Equation 1].

[0030]

(Equation 1)

On the other hand, compartment i at time t
Total drug concentration outflow rate from is represented by expression (2).

[0032]

(Equation 2)

Therefore, the differential equation for obtaining the function C _i (t) of the drug concentration change in the compartment i is represented by the following [Equation 3]. From this, the drug concentration in each compartment after Δt from the current time t is obtained by a numerical solution of a differential equation, and the current time is calculated as Δ in step S608.
proceed by t by repeating the operation of S603~S608 Similarly, solution C _i (t) is obtained as a result.

[0034]

(Equation 3)

From the drug concentration changes C _i (t) in the individual compartments obtained from the above [Equation 3], the total drug concentration C _Total (t) is obtained from the following [Equation 4]. Where C
_Since calculating the _total (t) does not exist in a state where each compartment is physically separated, it is merely a virtual division of the state change of the drug in the living body,
This is because time-dependent drug distribution image data measured by PET or the like is also measured as the sum of drug concentration changes in each compartment.

[0036]

(Equation 4)

Since CBV (Cerebral Blood Volume), which is the ratio of blood present in blood vessels to the whole tissue, is small in many parts of the brain, the drug concentration change component C ₁ in blood is calculated according to [Equation 4]. (t) is neglected, but if you want to include the effect of cerebral blood volume CBV, determine the CBV by an appropriate method,
Alternatively, it is also possible to separately measure in advance and then give the result to the program to perform the calculation after performing the CBV correction by the following [Equation 5]. In [Equation 5], C ₁ (t) is a change in blood drug concentration, and CBV is a ratio (0% to 100%) of blood in the blood vessel to the whole tissue. CBV apportioned the contribution of drug concentration.

[0038]

(Equation 5)

The change in the total drug concentration obtained by [Equation 4] (or [Equation 5]) is expressed as a function of time t.
ET does not theoretically exist essentially those of instantaneous values to a radiation measuring, as an integral average value of the total drug concentration changes in a frame time-separated n _f at predetermined time intervals data can be obtained. In step S610 to step S612, numerical integration of C _Total (t) obtained from [Equation 4] (or [Equation 5]) is performed for each time interval corresponding to each PET frame, and PET theoretical data for each frame is obtained. (PET _cal (f)) is calculated.

Returning to FIG. 7, in step S7, the obtained PET theoretical data and PET measured data PET _obs (f)
And find the difference between the two. In this case, it is necessary to use an optimal calculation method according to the nature and purpose of the data to be processed. In general, a weighted residual sum of squares as shown in [Equation 6] is used. .

[0041]

(Equation 6)

Here, k represents a set of k _ij . In the adjustment of the speed constant by the nonlinear optimization method in step S8, the speed constant is adjusted so that χ ² (k) is reduced. However, what is adjusted here is only the velocity constant for the combination of the values of the compartment configuration matrix shown in FIG. When the adjustment of the rate constant is repeated, the change in 調節² (k) becomes smaller. Therefore, when the change becomes equal to or less than the convergence condition (specified value) previously input from the input device in S9, the nonlinear optimization method is used. The adjustment of the rate constant is stopped, and the current k
Is stored in the main memory as the resulting rate constant.

[0043] Also, the standard deviation of the measurements PET _obs (f) The E _f sigma
Ideally, the reciprocal {σ (f)} ⁻¹ of (f) is used, but in the case of the PET data shown in the example, PET _obs (f) is a measurement value of each point once. The standard deviation of the measured values cannot be calculated from. Therefore, the standard deviation of the measured value of the count rate in the radiation measurement theory expressed by [Equation 7] is used instead.

[0044]

(Equation 7)

Here, t _frame (f) is the measurement time of frame f. In this case, E _f is represented by [Equation 8].

[0046]

(Equation 8)

After performing the processing of steps S5 to S11 for all the pixels, in step S12 the speed constant stored in the main storage is stored as an image data file in the auxiliary storage device, and the program ends. Some corrections can be made in addition to the above description. The time-dependent drug concentration data in the blood is originally the drug concentration in the blood flowing into the body tissue of interest, but in practice, it is usually substituted by collecting blood from peripheral blood vessels. Therefore, there is a time lag between the actually measured input function and the true blood drug concentration data with time. In addition to each rate constant k, by adjusting the shift of this blood drug concentration data over time, PET
It is possible to make the difference between the measured data and the PET theoretical data smaller and make correction. Further, as shown in [Equation 5], the CBV can be corrected. However, the correction can be performed by adjusting the CBV for each pixel in the same manner.

As an example, the FDG shown in FIG.
5 shows an example of a rate constant image obtained by the method of the present invention in an example of a compartment model analysis of cerebral glucose metabolism by using the method. FIG. 11 shows an example of temporal drug distribution image data obtained by PET measurement in the case of FDG. The drug distribution images in the same brain cross section are sequentially arranged in numerical order. The unit shown on the scale is the concentration of drug radioactivity (kBq) per unit volume (1 ml) in the body.

FIG. 12 is a graph showing an example of time-dependent blood drug concentration data obtained by arterial blood sampling.
FIG. 13 shows the time-lapse drug distribution image data of FIG. 11 and FIG.
5 shows rate constant image data obtained according to the processing procedure of the present invention from the time-dependent blood drug concentration data of FIG. 13 (a) is obtained by imaging the transition rate constant k ₁₂ of FDG into the brain tissue from the blood plasma, and FIG. 13 (b)
Is an image of the transfer rate constant k ₂₁ of FDG from brain tissue to plasma. FIG. 13 (c) shows that FDG was converted to FD by phosphorylation at position 6 by hexokinase in brain tissue.
The migration rate constant k ₂₃ of the G-6-P is obtained by imaging. The unit shown on the scale is the drug concentration transfer rate per unit time (1 minute).

From the rate constant image data obtained in this way, information that cannot be seen from the time-dependent drug distribution image data, which is the sum total of various factors, is obtained.
For example, it is assumed that cerebral glucose metabolism is measured by FDG in a disease state such as Alzheimer's disease, and this is lower than that in a healthy person. By further dividing the FDG data into each rate constant, the rate constant k is strongly affected by the blood flow.
_12, it becomes possible to study the rate constant k ₂₃ representing the enzyme activity individually, when cerebral glucose metabolism was a site of reduced, this is what is due to reduction of blood flow, the enzyme activity Since it is possible to determine whether the decrease is due to the decrease, it can be used to elucidate the disease state, and by accumulating this data, it can also be used to diagnose a brain degenerative disease that is difficult to distinguish.

A program for causing a computer or the like to execute a procedure or a calculation algorithm for realizing the processing shown in FIGS. 7 and 10 is a computer-readable recording medium such as a floppy disk or M
It can be recorded on O, CD, DVD-RAM, etc. and stored and distributed.

[0052]

According to the present invention, it is possible to develop a program for freely constructing a compartment model on any computer and performing a compartment model analysis, and to use this program for a drug in a free compartment model. Since pharmacokinetic analysis can be performed, even for those with unknown pharmacokinetics such as new drugs, it is possible to find the optimal compartment model by trial and error, obtain rate constant image data, and conduct detailed studies Become.

[Brief description of the drawings]

FIG. 1 is a schematic configuration diagram of a compartment model analysis system according to the present invention.

FIG. 2 is a diagram showing an example of a compartment model configuration when a drug is FDG.

FIG. 3 shows an example of a compartment arrangement matrix in the case of the example of FIG.

FIG. 4 is a diagram showing an example of a rate constant initial value matrix in the case of the example of FIG. 2;

FIG. 5 is a diagram showing an example of a four-compartment model configuration of virtual drug A.

FIG. 6 is a diagram showing an example of a compartment configuration matrix in the case of the example of FIG. 5;

FIG. 7 is a view for explaining a series of processes in the compartment model analysis of the present invention.

FIG. 8 is a diagram showing an input example of a compartment configuration matrix.

FIG. 9 is a view for explaining a compartment model configuration represented by an input example of a compartment configuration matrix in FIG. 8;

Figure 10 illustrates a predefined process in step S6 in FIG.

FIG. 11 is a diagram showing an example of temporal drug distribution image data in the case of FDG.

FIG. 12 is a diagram showing an example of time-dependent drug concentration data in blood in the case of FDG.

FIG. 13 is a diagram showing an example of speed constant image data in the case of FDG.

[Explanation of symbols]

11 input device, 12 auxiliary memory device, 13 output device, 14 compartment model analysis program

Claims (5)

[Claims] 1. A pharmacokinetics of a drug in which a compartment model analysis is performed based on a time-dependent drug distribution image data and a time-dependent blood drug concentration data concerning a drug distribution in a body, and a rate constant between the compartments is determined for each pixel. In the analysis method, a first step of inputting a compartment model configuration using matrix-format text data representing a coupling mode between a predetermined number of compartments, and a first step of inputting a velocity constant initial value between each compartment using matrix-format text data A second step, a third step of inputting the time-dependent drug distribution image data and the time-dependent blood drug concentration data, and a fourth step of extracting and storing a combination having a combination between the compartments from the input compartment model configuration.
From the input compartment model configuration, the rate constant initial value between each compartment, the time-dependent drug distribution image data and the time-dependent blood drug concentration data, the stored drug concentration transfer amount between the compartments with the binding, A fifth step of sequentially calculating every minute time and calculating time-lapse drug distribution image data by compartment model simulation; and input time-lapse drug distribution image data and time-lapse drug distribution image data obtained by compartment model simulation A sixth step of determining a difference between; and a seventh step of adjusting a rate constant between the compartments by a non-linear optimization method such that a difference between the two time-lapse drug distribution image data is reduced, Until the difference between the two temporal drug distribution image data converges to a specified value or less. Wherein the fifth to seventh steps are repeated. 2. The method for analyzing the pharmacokinetics of a drug according to claim 1, wherein the binding mode between the compartments is such that there is no flow, there is a flow and the value of the rate constant is variable, and there is a flow and the value of the rate constant is fixed. A method for analyzing pharmacokinetics of a drug, wherein the method is designated by one of the methods. 3. The method for analyzing the pharmacokinetics of a drug according to claim 2, wherein the adjustment of the rate constant in the seventh step is performed only for the rate constant whose value is designated as variable. analysis method. 4. A pharmacokinetics of a drug, wherein a compartment model analysis is performed on the basis of time-dependent drug distribution image data and time-dependent drug concentration data in the blood regarding the biodistribution of the drug, and a rate constant between the compartments is determined for each pixel. An analysis method, comprising: a first step of inputting a compartment model configuration using matrix-format text data representing a connection mode between a predetermined number of compartments; and inputting a velocity constant initial value between each compartment using matrix-format text data. A second step of inputting time-dependent drug distribution image data and time-dependent blood drug concentration data;
A fourth combination for extracting and storing a combination having a connection between compartments from the input compartment model configuration
From the input compartment model configuration, the initial value of the rate constant between the compartments, the drug distribution image data over time and the drug concentration data over time blood, from the stored drug concentration transfer amount between compartments with binding, A fifth step of sequentially calculating every minute time and calculating the time-dependent drug distribution image data by the compartment model simulation, and the input time-dependent drug distribution image data and the time-dependent drug distribution image data obtained by the compartment model simulation Find the difference between the sixth
And a seventh step of adjusting the rate constant between the compartments by a non-linear optimization method such that the difference between the two time-lapse drug distribution image data is reduced,
A program for causing a computer to execute a drug pharmacokinetic analysis method in which the fifth to seventh steps are repeated until the difference between the two temporal drug distribution image data converges to a specified value or less is recorded. Computer readable recording medium. 5. A pharmacokinetics of a drug for which a compartment model analysis is performed based on a time-dependent drug distribution image data and a time-dependent drug concentration data in the blood concerning a biodistribution of a drug, and a rate constant between the compartments is determined for each pixel. An analysis method, comprising: a first step of inputting a compartment model configuration using matrix-format text data representing a connection mode between a predetermined number of compartments; and inputting a velocity constant initial value between each compartment using matrix-format text data. A second step of inputting time-dependent drug distribution image data and time-dependent blood drug concentration data;
A fourth combination for extracting and storing a combination having a connection between compartments from the input compartment model configuration
From the input compartment model configuration, the initial value of the rate constant between the compartments, the drug distribution image data over time and the drug concentration data over time blood, from the stored drug concentration transfer amount between compartments with binding, A fifth step of sequentially calculating every minute time and calculating the time-dependent drug distribution image data by the compartment model simulation, and the input time-dependent drug distribution image data and the time-dependent drug distribution image data obtained by the compartment model simulation Find the difference between the sixth
And a seventh step of adjusting the rate constant between the compartments by a non-linear optimization method such that the difference between the two time-lapse drug distribution image data is reduced,
A program for causing a computer to execute a drug pharmacokinetic analysis method in which the fifth to seventh steps are repeated until the difference between the two temporal drug distribution image data converges to a specified value or less.