JP2010085229A - Method, device and program for acquiring data for analyzing hepatocellular carcinoma - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve accuracy for detecting hepatocellular carcinoma. <P>SOLUTION: This method for acquiring data for analyzing hepatocellular carcinoma includes: a step for measuring a value X of α-fetoprotein (AFP) from a body fluid of a patient, and storing it into a storage device; a step for measuring a value Y of an AFP lectin fraction (AFP-L3%) from the body fluid of the patient, and storing it into the storage device; a step for measuring a value Z of des-gamma-carboxy prothrombin (DCP) from the body fluid of the patient, and storing it into the storage device; and a score calculation step for calculating a score S for detecting hepatocellular carcinoma by using X, Y and Z stored in the storage device, and storing it into the storage device. In the score calculation step, the score S is calculated by expression: S=α*log(X)+β*log(Y)+γ*log(Z) (α, β and γ are each constant). <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a technique for obtaining data for hepatocellular carcinoma analysis.

  Tumor markers used for the detection of hepatocellular carcinoma include α-fetoprotein (AFP), AFP lectin fraction (AFP-L3%), des-gamma-carboxy prothrombin (DCP) and the like. Are known. However, the positive rate of hepatocellular carcinoma with these tumor markers alone is 30% to 60%, and not all hepatocellular carcinoma patients can be detected. It is also known that these markers give false positives to patients who do not have hepatocellular carcinoma, such as patients with viral chronic hepatitis or cirrhosis. For this reason, in clinical practice, it is usual to make a determination using a plurality of markers without using only one marker, but there is no integrated index for increasing the accuracy of detection of hepatocellular carcinoma.

Although there are papers that propose an integrated index for detecting hepatocellular carcinoma, it cannot be said that sufficient detection accuracy is obtained compared to the case of the marker alone described above.
Volk ML, Hernandez JC, Su GL, Lok AS, Marrero JA (2007). "Risk factors for hepatocellular carcinoma may impair the performance of biomarkers: a comparison of AFP, DCP, and AFP-L3". Cancer Biomark 3 (2) : 79-87 Marrero JA, Su GL, Wei W, Emick D, Conjeevaram HS, Fontana RJ, Lok AS.Des-gamma carboxyprothrombin can differentiate hepatocellular carcinoma from nonmalignantchronic liver disease in american patients.Hepatology 2003; 37: 1114-1121

  Accordingly, it is an object of the present invention to provide an integrated index for improving the accuracy of hepatocellular carcinoma detection.

  The method for obtaining data for hepatocellular carcinoma analysis according to the present invention comprises the steps of measuring α-fetoprotein (AFP) value X from a patient's bodily fluid and storing it in a storage device, and AFP lectin content from the patient's bodily fluid. Measuring the value Y of the image (AFP-L3%) and storing it in a memory device; measuring the value Z of des-gamma-carboxyprothrombin (DCP) from the patient's body fluid and storing it in the memory device; And a score calculation step of calculating a score S for detecting hepatocellular carcinoma using X, Y and Z stored in the storage device and storing the score S in the storage device. In the score calculation step, the score S is calculated as S = α * log (X) + β * log (Y) + γ * log (Z) (α, β, and γ are constants).

  By introducing the score S as described above, AFP, AFP-L3%, and DCP can be converted into integrated index values, and detection accuracy can be improved.

  Furthermore, the method for obtaining data for the present hepatocellular carcinoma analysis compares the value R stored in advance as a threshold value of the score S in the storage device with the score S stored in the storage device, When the score S is equal to or greater than the value R, the method may further include a step of storing a flag in the storage device in association with the score S. In this way, a doctor or the like can easily find a patient to be particularly careful.

  In the case where α is 0.01 or more and 0.34 or less, β is 0.18 or more and 0.66 or less, and γ is 0.8 or more and 1.00 or less. The inventor of the present application has found that there is a remarkable effect as compared with the prior art in terms of the index of AUROC described in the above.

  More preferably, the α is 0.02 or more and 0.28 or less, the β is 0.20 or more and 0.58 or less, and the γ is 0.84 or more and 1.00 or less. . By adopting such a range, there is a more remarkable effect than the prior art in terms of the index AUROC.

  A program for causing a computer to execute the method according to the present invention can be created, and the program is stored in a storage medium or storage device such as a flexible disk, a CD-ROM, a magneto-optical disk, a semiconductor memory, or a hard disk. Is done. Note that data being processed is temporarily stored in a storage device such as a computer memory.

  Therefore, according to the present invention, the accuracy of hepatocellular carcinoma detection can be improved.

[Premise of the embodiment of the present invention]
[About ROC analysis]
In ROC analysis, positive values and negative values are discriminated by cutting off some values for a certain test. The true positive rate (sensitivity) obtained as a result is plotted on the vertical axis and false positive rate (1-specific). This is an analysis based on an ROC curve (receiver operating characteristics curve) plotted on the horizontal axis. FIG. 1 shows an example of the ROC curve. Generally, the more the curve is unevenly distributed in the upper left, the higher the diagnostic ability of the examination, and conversely, if the curve is closer to the diagonal (y = x), it is determined that there is no diagnostic significance of the examination. In addition, since the plotting range of both axes is 0 to 1, and the area is 1, the area at the lower right of the curve is 1 which is the maximum value in the case of perfect inspection. On the other hand, the area under the curve of a test without any separation ability is 0.5 (for details, see Japanese Society for Clinical Laboratory Automation: Manual for evaluating the diagnostic utility of clinical tests. Journal of the Japanese Society for Clinical Laboratory Automation 29: 15-19. Zweig MH, Campbell G: Receiver-operating characteristic (ROC) plots: a fundamental evaluation tool in clinical medicine. Clin Chem 39: 561-577, 1993. I. Yakugyo Jihosha, 1998.).

  Hereinafter, the area under the ROC curve (AUROC: Area under a ROC) is adopted as an evaluation index of clinical usefulness of the test.

[Principle of the embodiment of the present invention]
1. Basic concept The inventor of the present application has intriguedly thought that the score S for detecting hepatocellular carcinoma having the highest utility value in the clinical field should be calculated by the following formula.
S = α * log (X) + β * log (Y) + γ * log (Z) (1)

  Here, X is a measured value for AFP, Y is a measured value for AFP-L3%, and Z is a measured value for DCP. Further, α, β, and γ are coefficients as described later.

  A doctor or the like compares the score S calculated by the equation (1) with a predetermined cut-off value (that is, a threshold value) R, and if the score S is equal to or higher than the cut-off value R, hepatoma is possible. It can be judged that the nature is very high.

  The cut-off value R varies depending on the situation (prevalence) of the clinical test and the clinical significance of false negative / false positive. Therefore, it cannot be determined uniquely, and the set value changes depending on the policy of the facility that conducts the inspection. However, as an example of a general method, there is a method of adjusting the cut-off value by prevalence, false negative, and false positive using a sensitivity / specificity curve (for example, Japanese Society for Clinical Laboratory Automation: Clinical Clin Chem 39: 561- Manual for evaluation of diagnostic usefulness of tests.Journal of Japan Society for Clinical Laboratory Automation 29: 15-19 Zweig MH, Campbell G: Receiver-operating characteristic (ROC) plots: (See 577, 1993.). Alternatively, there is also a method of using, as a cut-off value, an identification value that minimizes the sum of the square value of the false positive rate and the square value of the false negative rate (for example, Ministry of Health and Welfare, Health Policy Bureau, Pharmaceutical Affairs and Economics Research Institute: Clinical Laboratory Drug Information) Personnel Training Text I. See Yakuho Jihosha, 1998.)

2. About specific value ranges of coefficients α, β, γ (1) Calculation of optimum coefficient Here, 93 cases of hepatocellular carcinoma and 399 cases of chronic liver disease collected from 7 institutions are used to calculate the optimum coefficient and coefficient. Consider the range.

  Here, in the formula (1), log (X) = Y1, log (Y) = Y2, and log (Z) = Y3 are simply expressed. And while calculating AUROC by each independently or arbitrary two combinations, the score calculation formula in which AUROC becomes the maximum, and the value of AUROC in that case were put together, and the result as shown in FIG. 2 was obtained.

  From the results shown in FIG. 2, it can be seen that the combination of Y1 and Y3 and the combination of Y2 and Y3 leads to a maximum AUROC value of 0.809.

Therefore, the former score calculation formula S5 = 0.11 * Y1 + Y3 and the latter score calculation formula S6 = 0.51 * Y2 + Y3 are combined to specify the formula for maximizing AUROC and its coefficient in the following four formulas. .
_{S7 = a 1 * Y2 + S5} (2)
_{S8 = Y2 + a 2 * S5} (3)
S9 = a ₃ * Y1 + S6 (4)
_{S10 = Y1 + a 4 * S6} (5)

Then, it was found that when A ₃ = 0.17 and the equation (4) is used, the AUROC is 0.819. More specifically, it is as follows.
S9 = 0.17 * Y1 + S6 = 0.17 * Y1 + 0.51 * Y2 + Y3 (6)
That is, α = 0.17, β = 0.51, and γ = 1.0 are optimum values.

  For these analyses, JMP (registered trademark) 6 (manufactured by JMP), which is a statistical analysis program, is used.

(2) Specification of specific value range of coefficient In order to specify a preferable value range of coefficient, FIG. 3 shows the change of AUROC when β is fixed at 0.51 and γ = 1.0 and α is changed. . In FIG. 3, the horizontal axis represents α and the vertical axis represents AUROC. Thus, when α is changed from 0 to 1, AUROC increases once, reaches a maximum at 0.17, and then gradually decreases.

  FIG. 4 shows the change in AUROC when β is changed with α = 0.17 and γ = 1.0. In FIG. 4, the horizontal axis represents β and the vertical axis represents AUROC. Thus, when β is changed from 0 to 1, AUROC gradually rises and becomes almost flat, then becomes maximum at 0.51, and then gradually decreases.

  Further, FIG. 5 shows changes in AUROC when γ is changed while α = 0.17 and β = 0.51 are fixed. In FIG. 5, the horizontal axis represents γ and the vertical axis represents AUROC. In this way, when γ is changed between 0 and 1, AUROC rises while gradually saturating and reaches a maximum at 1.0.

  In the present embodiment, from the viewpoint of AUROC, the value of AUROC, which can be said to have a significant effect on the prior art, is considered to be 0.816 in FIGS. Specifically, when 0.816 is set as the threshold value in FIGS. 3 to 5, 0.01 ≦ α ≦ 0.34, 0.18 ≦ β ≦ 0.66, and 0.80 ≦ γ ≦ 1.00. The value range was obtained.

  However, in FIGS. 3 to 5, AUROC is calculated with a fixed value except for the coefficient to be examined, and thus AUROC = 0.816 cannot always be ensured within this range. Therefore, when the value of the coefficient that may have the lowest AUROC in FIGS. 3 to 5 is picked up and AUROC is recalculated together with the sensitivity and specificity, the result shown in FIG. 6 is obtained.

  That is, if α = 0.34, β = 0.18, and γ = 0.80, AUROC is 0.813. Further, if α = 0.34, β = 0.66, and γ = 0.80, AUROC is 0.810. Thus, at least AUROC = 0.810 can be secured, and there is a remarkable effect even when compared with the prior art described below.

  The cut-off value R is set so that the specificity is 90%. The sensitivity is not significantly different from the sensitivity of the optimum score calculation formula (6), and it can be seen that it is more effective than the prior art as described below.

  Furthermore, in the present embodiment, from the viewpoint of AUROC, it is considered that a more preferable AUROC value is 0.815, which can be said to have a remarkable effect with respect to the prior art. Therefore, in FIG. 3 to FIG. 5, when AUROC = 0.817 is set as a threshold and a value range that secures AUROC = 0.815 at least is searched, 0.02 ≦ α ≦ 0.28 and 0.20 ≦ The range of 0.84 ≦ γ ≦ 1.00 was obtained with β ≦ 0.58.

  Again, based on the trends in FIGS. 3-5, FIG. 7 shows the values of coefficients that have the lowest AUROC in the above range, and recalculating AUROC along with sensitivity and specificity. The result was obtained.

  That is, if α = 0.02, β = 0.20, and γ = 0.84, AUROC is 0.816, and if α = 0.02, β = 0.58, and γ = 0.84. , AUROC is 0.815, α = 0.28, β = 0.20, and γ = 0.84, AUROC is 0.816, α = 0.28, β = 0.58, and γ = If 0.84, AUROC will be 0.815. Thus, a better result than the result of FIG. 6 is obtained.

  As in FIG. 6, the cutoff value R is set so that the specificity is 90%. The sensitivity is not significantly different from the sensitivity of the optimum score calculation formula (6), and it can be seen that it is more effective than the prior art as described below.

  Here, FIG. 8 shows data for comparison between the prior art and the present embodiment. In FIG. 8, AFP alone, AFP-L3% alone, DCP alone, and non-patent document 2 in which specific score calculation formulas are presented are compared with the optimal score calculation formula (6). Comparing AUROC, the highest value in the prior art is 0.806 of Non-Patent Document 2, while in the case of the optimal score calculation formula (6), it is 0.819, and the lowest value of FIG. It can be seen that there is a remarkable effect even when compared with 810. Similarly, it can be said that there is a more remarkable effect when compared with the minimum value of 0.815 in FIG.

  As described above, for the optimal score calculation formula (6), the cut-off value R is set so that the specificity is 90% in order to ensure the reliability of the score calculation formula. Usually, the sensitivity decreases as the specificity increases, but even if such a high specificity value is set, the sensitivity value is lower than that of AFP alone, but a sufficiently high value is obtained compared to the conventional technique. It has been. Compared with the value of AFP alone, the sensitivity is low, but the specificity is clearly high, and it can be seen that the optimum score calculation formula (6) is a balance between sensitivity and specificity.

3. Other Effects Of the 399 cases of chronic liver disease described above, a total of 140 patients showed a false positive for one or more markers of AFP, AFP-L3%, and DCP. Among them, in the score calculated by the optimal score calculation formula (6), 104 cases (74%) are negative, and false positives can be largely avoided. In summary, FIG. 9 is obtained.

  In FIG. 9, AFP, AFP-L3% and DCP were all positive, AFP and AFP and AFP-L3% were positive, AFP and DCP were positive, AFP-L3% There are cases in which DCP is positive, cases in which only AFP is positive, cases in which only AFP-L3% is positive, and cases in which only DCP is positive. Although there are cases where there is no effect, in cases where AFP and AFP-L3% are positive, cases where only AFP is positive, and cases where only AFP-L3% is positive, false positives are predominantly avoided and good effects are obtained. Is seen.

[Specific examples of embodiments of the present invention]
FIG. 10 shows a functional block diagram of an analyzer for detecting hepatocellular carcinoma according to the present embodiment. The analyzer 100 is held by the sample holding unit 11 that holds a patient sample, the AFP measurement unit 12 that measures the AFP value of the sample held by the sample holding unit 11, and the sample holding unit 11. An AFP-L3% measurement unit 13 that measures the AFP-L3% value for the sample, a DCP measurement unit 14 that measures the DCP value for the sample held in the sample holding unit 11, and AFP measurement An arithmetic processing unit 16 that processes the measurement results from the unit 12, the AFP-L3% measuring unit 13, and the DCP measuring unit 14, and a touch panel, a keyboard, and other input units 15 for instructing the arithmetic processing unit 16 and the like. A data storage unit 17 for storing data such as various measurement results, coefficient values, and score values of calculation results, and guidance from the arithmetic processing unit 16 to the user and arithmetic results of the arithmetic processing unit 16 Having a Shimesuru display unit 18, and a printing unit 19, etc. is the analysis result outputted from the arithmetic processing unit 16 for printing on paper.

  The sample holder 11 includes a container for holding a patient's body fluid (serum, plasma, etc.). For example, it may be an instrument that can hold a plurality of containers for holding body fluid of a patient.

  The AFP measurement unit 12 is a measurement unit that implements an existing AFP measurement technique. For example, measurement is performed according to the methods described in JP-A-61-292062, JP-A-11-287805, and the like. JP-A-6-66778, WO2007 / 121263, WO2002 / 082083 and the like disclose measurement methods using liquid chromatography and capillary electrophoresis, and measurement is performed using these techniques. You may make it do. This technique is well known and will not be discussed further.

  The AFP-L3% measurement unit 13 is also a measurement unit that implements the existing AFP-L3% measurement technology. However, it may be integrated with the AFP measurement unit 12. As the measurement technique, the technique described in the literature similar to AFP can be used.

  Similarly, the DCP measurement unit 14 is also a measurement unit that implements an existing DCP measurement technique. For example, the measurement is performed according to the methods described in JP-A-60-60557, JP-A-5-43357, JP-A-6-66778, WO2007 / 121263, WO2002 / 082083, and the like. This technique is well known and will not be discussed further.

  The measurement units described above may be configured integrally or may be configured separately as shown in FIG. There are also cases where any two are integrated.

  For example, the user inputs a cut-off value R to the input unit 15 and uses it for the processing of the arithmetic processing unit 16, or the coefficient values α, β, and γ in the equation (1) described above. The data is input and used for processing by the arithmetic processing unit 16. Further, the patient ID (or name in some cases) may be input from the input unit 15 and used as a key for the measurement value or score in the data storage unit 17.

  When processing a plurality of patients at once, an external interface with a data reading unit (for example, a reading unit of a storage medium such as a flexible disk, a USB (Universal Serial Bus) memory, etc.) included in the input unit 15 Etc.), data such as patient ID may be taken in via Furthermore, the processing results for a plurality of patients may be written in a storage medium or the like as a data reading / writing unit instead of the data reading unit.

  The data storage unit 17 stores the cutoff value R, the coefficient values α, β, and γ in the above-described equation (1), the value of the score S calculated by the arithmetic processing unit 16, and the like. Data necessary for the unit 16 to perform processing is stored.

  Moreover, although not shown, the analysis apparatus 100 may have a network interface. In such a case, the processing result can be transmitted to another terminal or server connected to the network. Furthermore, necessary data may be received from the network interface and stored in the data storage unit 17. Furthermore, if it is possible to connect to an external network such as the Internet, maintenance may be performed by remote operation from a maintenance center. Furthermore, not only maintenance but also a configuration in which data of a large number of patients are accumulated and coefficient values α, β and γ, cutoff value R, etc. are reset to more precise values may be adopted. .

  Next, the processing content of the analyzer 100 will be described with reference to FIG. First, the AFP measurement unit 12, the AFP-L3% measurement unit 13, and the DCP measurement unit 14 respectively detect the AFP value, the AFP-L3% value, and the DCP value from the body fluid of the patient held in the sample holding unit 11. The value is measured (step S1). Then, the AFP measurement unit 12 outputs the AFP measurement value, the AFP-L3% measurement unit 13 outputs the AFP-L3% measurement value, and the DCP measurement unit 14 outputs the DCP measurement value to the arithmetic processing unit 16 for calculation. The processing unit 16 receives these measurement values and records them in the data storage unit 17 (step S3). For example, the data storage unit 17 manages data with a data structure as shown in FIG. That is, a patient ID (or name), an AFP measurement value, an AFP-L3% measurement value, a DCP measurement value, a score value, and a flag are registered. A serial number of the measurement order may be registered instead of the patient ID. Moreover, patient ID is unnecessary if there is no need for management. In step S3, the score value and flag are not set.

  Thereafter, the arithmetic processing unit 16 calculates the value of the score S according to the equation (1) described above using each measurement value recorded in the data storage unit 17 (step S5). As described above, coefficient values such as α, β, and γ are set in advance by the input unit 15 or the like, or are set in the analysis apparatus 100 in advance. As described in the principle of the present embodiment, α, β, and γ are values that fall within the above-described range. This significantly improves the detection accuracy of hepatocellular carcinoma compared to the prior art.

  Further, the arithmetic processing unit 16 records the value of the score S calculated in step S5 in the data storage unit 17 (step S7). A score value is recorded corresponding to the patient ID of the patient involved in the process.

  The arithmetic processing unit 16 determines whether the value of the score S is equal to or greater than a preset cutoff value (threshold value) R (step S9). The cut-off value R is also stored in the data storage unit 17.

  When the value of the score S is equal to or greater than the cutoff value R, the arithmetic processing unit 16 sets a flag in the data storage unit 17 (step S11). A flag is set on corresponding to the patient ID of the patient involved in the process.

  When the value of the score S is less than the cut-off value R, or after step S11, the arithmetic processing unit 16 is data representing each measurement value, the value of the score S, and the flag in response to an instruction from the user or automatically. Is output to the display unit 18 or the like (step S13). For example, a display as shown in FIG. 13 is performed. In the example of FIG. 13, the patient ID, the measured value of AFP, the measured value of AFP-L3%, the measured value of DCP, the value of score S, and remarks are displayed. The patient ID may not be included. In addition, the remark includes data for emphasizing the flag when it is set on. In the example of FIG. 13, a star is added. Instead of a star, an annotation such as “It is greater than or equal to the cut-off value” may be inserted, or in order to emphasize the entire display data, it may be blinked or the color may be changed.

  Furthermore, data as shown in FIG. 13 may be printed on paper from the printing unit 19. Further, it may be output to the outside as electronic data from the external interface as described above.

  By performing the processing as described above, a doctor who is a user can obtain an appropriate index value for hepatocellular carcinoma and can perform a correct diagnosis.

  Although the embodiment has been described above, the present invention is not limited to this. The arithmetic processing unit 16 of the analysis apparatus 100 may be configured by a combination of a processor and a program, for example. In such a case, the program is held in, for example, the data storage unit 17 or a memory circuit built in the processor, and the above processing is performed by the processor executing the program.

  Further, the analyzer 100 may not have the printing unit 19. In such a case, for example, an interface unit that can be connected to the printing apparatus may be held.

  In order to simply carry out this embodiment, each measurement value is obtained online or offline from a measuring device having all or part of the AFP measuring unit 12, the AFP-L3% measuring unit 13, and the DCP measuring unit 14. You may make it comprise the analyzer which only acquires and calculates and outputs the value of a score according to (1) Formula mentioned above.

  Further, steps S9 and S11 shown in FIG. 11 may not be performed. This may be determined by a doctor or the like.

  In addition, an additional function may be added to the analysis apparatus 100.

It is a figure which shows an example of a ROC curve. It is a figure for demonstrating the optimal score calculation formula deriving procedure. It is a figure showing the change of AUROC when (beta) and (gamma) in an optimal score calculation formula are fixed, and (alpha) is changed. It is a figure showing change of AUROC when (alpha) and (gamma) in an optimal score calculation formula are fixed, and (beta) is changed. It is a figure showing the change of AUROC when (alpha) and (beta) in an optimal score calculation formula are fixed, and (gamma) is changed. It is a figure showing the data for verifying the 1st range of a score calculation formula. It is a figure showing the data for verifying the 2nd range of a score calculation formula. It is a figure for performing comparison with a prior art. It is a figure for demonstrating another effect. It is a functional block diagram of the analyzer in an embodiment of the invention. It is a figure which shows the processing flow of the analyzer in embodiment of this invention. It is a figure which shows an example of the data stored in a data storage part. It is a figure which shows an example of the data output.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Analysis apparatus 11 Sample holding part 12 AFP measurement part 13 AFP-L3% measurement part 14 DCP measurement part 15 Input part 16 Arithmetic processing part 17 Data storage part 18 Display part 19 Printing part

Claims (6)

Measuring α-fetoprotein (AFP) value X from the body fluid of the patient and storing it in a storage device;
Measuring the value Y of the AFP lectin fraction (AFP-L3%) from the body fluid of the patient and storing it in the storage device;
Measuring the value Z of des-gamma-carboxyprothrombin (DCP) from the body fluid of the patient and storing it in the storage device;
Calculating a score S for hepatocellular carcinoma detection using X, Y and Z stored in the storage device, and storing the score S in the storage device;
Including
In the score calculating step, the score S is
S = α * log (X) + β * log (Y) + γ * log (Z)
(Α, β and γ are constants)
Method for obtaining data for hepatocellular carcinoma analysis calculated in When the value R stored in advance as a threshold value of the score S in the storage device is compared with the score S stored in the storage device, the score S is equal to or greater than the value R. The method according to claim 1, further comprising: storing a flag in the storage device in association with the score S.   The α is 0.01 or more and 0.34 or less, the β is 0.18 or more and 0.66 or less, and the γ is 0.8 or more and 1.00 or less. Item 3. The method according to Item 1 or 2.   The α is 0.02 or more and 0.28 or less, the β is 0.20 or more and 0.58 or less, and the γ is 0.84 or more and 1.00 or less. Item 3. The method for analyzing liver cancer cells according to Item 1 or 2. First measuring means for measuring α-fetoprotein (AFP) value X from the body fluid of the patient and storing it in a storage device;
A second measuring means for measuring the value Y of the AFP lectin fraction (AFP-L3%) from the body fluid of the patient and storing it in the storage device;
Third measurement means for measuring a value Z of des-gamma-carboxyprothrombin (DCP) from the body fluid of the patient and storing it in the storage device;
A score calculation means for calculating a score S for hepatocellular carcinoma detection using X, Y and Z stored in the storage device, and storing the score S in the storage device;
Have
In the score calculation means, the score S is:
S = α * log (X) + β * log (Y) + γ * log (Z)
(Α, β and γ are constants)
A device for obtaining data for hepatocellular carcinoma analysis calculated by Α-fetoprotein (AFP) value X measured from the patient's bodily fluid, AFP lectin fraction (AFP-L3%) value Y measured from the patient's bodily fluid, and measured from the patient's bodily fluid Reading said X, Y and Z from a storage device storing a des-gamma-carboxy prothrombin (DCP) value Z;
A score calculation step for calculating a score S for hepatocellular carcinoma detection using the read X, Y and Z, and storing the score in the storage device;
To the computer,
In the score calculating step, the score S is
S = α * log (X) + β * log (Y) + γ * log (Z)
(Α, β and γ are constants)
Program for obtaining data for hepatocellular carcinoma analysis calculated by

Images