Method for analyzing blood coagulation reaction

Abstract

Provided is a blood coagulation reaction analysis method. This method comprises measuring a blood coagulation reaction of a specimen under analysis, thereby obtaining first data for calculating a blood coagulation time of the specimen under analysis and second data for estimating an abnormal blood coagulation factor of the specimen under analysis. The obtaining the second data includes: taking a first derivative of V(i) of a coagulation reaction curve R(i); and determining a point pk where V(i) reaches Xk before reaching a maximum value Vmax and a point qk where V(i) reaches Xk after reaching Vmax.

Description

Technical Field

[0001] The present invention relates to a method for analyzing a blood coagulation reaction.

Background Art

[0002] A blood coagulation test is a test for diagnosing a patient's blood coagulation ability by adding a predetermined reagent to a patient's blood sample and measuring the blood coagulation time or the like. Typical examples of the blood coagulation time include prothrombin time (PT), activated partial thromboplastin time (APTT), thrombin time, and the like. An abnormality in blood coagulation ability causes an extension of the coagulation time. Causes of the extension of the coagulation time include the influence of a coagulation inhibitor, a decrease in coagulation-related components, a congenital deficiency of a blood coagulation factor, and an acquired appearance of an autoantibody that inhibits the coagulation reaction.

[0003] When an extension of the coagulation time, for example, APTT, is observed in a blood coagulation test, generally, a crossmixing test is further performed to determine whether the extension of APTT is due to a coagulation factor inhibitor (anticoagulation factor), lupus anticoagulant (LA), or a coagulation factor deficiency such as hemophilia. In the crossmixing test, the APTT (immediate reaction) immediately after preparation and the APTT (delayed reaction) after incubation at 37°C for 2 hours of normal plasma, test plasma, and mixed plasma containing the test plasma and normal plasma at various volume ratios are measured. The measured values of the crossmixing test are graphed with the APTT (seconds) on the vertical axis and the volume ratio of the test plasma to the normal plasma on the horizontal axis. The graphs of the immediate reaction and the delayed reaction thus created show "concave downward", "linear", or "convex upward" patterns depending on the APTT extension factor. Based on these patterns of the immediate reaction and the delayed reaction, the APTT extension factor is determined.

[0004] In blood coagulation tests, a coagulation reaction curve can be obtained by measuring the amount of blood coagulation reaction over time after adding a reagent to a blood sample. This coagulation reaction curve has different shapes depending on the type of abnormality in the blood coagulation system (Non-Patent Document 1). Therefore, methods for determining abnormalities in the blood coagulation system based on the coagulation reaction curve have been disclosed. For example, in Patent Documents 1 to 3 and Non-Patent Documents 2 to 4, parameters regarding the first derivative curve and the second derivative curve of the coagulation reaction curve for a patient's blood, such as the maximum coagulation rate, the maximum coagulation acceleration, the maximum coagulation deceleration, and the time to reach them, are used to evaluate the presence or absence of abnormalities in the coagulation factors in the patient. Patent Document 4 describes a method for determining the severity of hemophilia based on the average rate of change in coagulation rate until the time when the patient's coagulation reaction reaches the maximum coagulation rate or the maximum coagulation acceleration. Patent Documents 5 and 6 describe methods for calculating the peak width at a predetermined height of the coagulation reaction rate curve and using the information based on the peak width to determine the presence or absence of coagulation factor abnormalities, the concentration of coagulation factors, factors prolonging the coagulation time, and the like.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Patent Document 2

Patent Document 3

Patent Document 4

Patent Document 5

Patent Document 6

Non-Patent Documents

[0006]

Non-Patent Document 1

Non-Patent Document 2

Non-Patent Document 3

Non-Patent Document 4

Summary of the Invention

Problems to be Solved by the Invention

[0007] The present invention relates to a method for analyzing a blood coagulation reaction that enables acquisition of data useful for estimating coagulation abnormality factors in a blood specimen, together with data for measuring the coagulation time of the blood specimen.

Means for Solving the Problems

[0008] That is, the present invention provides the following. 〔1〕A method for analyzing a blood coagulation reaction, comprising: measuring the blood coagulation reaction of a test subject to obtain first data for calculating the blood coagulation time of the test subject and second data for estimating blood coagulation abnormality factors of the test subject; wherein the acquisition of the second data includes: obtaining the first derivative V(i) of the coagulation reaction curve R(i), where i represents the number of measurement points or time; determining a point p where V(i) becomes X k before reaching the maximum value Vmax, k and a point q where V(i) becomes X k after reaching Vmax, k where k represents a series of integers from 1 to n, n is an integer of 2 or more, and 0 < X k < Vmax, a method including the above. 〔2〕Said X kis Vmax×S k %(S k is the method described in [1] defined by 0.5 to 99). 〔3〕The acquisition of the second data is p k or q k or calculating at least one of their statistical values as the second data, the method described in [1] or [2]. 〔4〕The acquisition of the second data further includes calculating at least one selected from the group consisting of the front Ave, the back Ave, the front SD, the back SD, the front CV, the back CV, the front-back average difference, the front-back SD ratio, M k 、W k 、the distortion index, and the sharpness index as the second data, where the front Ave, the front SD, and the front CV respectively represent the average value, the standard deviation, and the coefficient of variation of p k ; the back Ave, the back SD, and the back CV respectively represent the average value, the standard deviation, and the coefficient of variation of q k ; the front-back average difference represents (the back Ave - the front Ave) / (the average value of p k and q k ); the front-back SD ratio represents the back SD / the front SD; M k represents (p k +q k ) / 2; W k represents q k -p k ; the distortion index represents the coefficient of variation of M k ; the sharpness index represents (the sum or average value of W k for the lower part of the peak of V(i)) / (the sum or average value of W k for the upper part of the peak of V(i)), the method described in [1] or [2]. 〔5〕The acquisition of the second data further includes calculating the Standard Deviation Interval (SDI) of the target parameter for the test subject as the second data, where The SDI of the target parameter for the subject to be examined is SDI = (α - β) ÷ γ α: The value of the target parameter from the subject to be examined β: The reference value of the value of the target parameter based on the data of the normal subject group γ: The standard deviation of the value of the target parameter based on the data of the normal subject group and the target parameter is any two selected from the pre-CV, post-CV, distortion index, sharpness index, pre-post average difference, and pre-post SD ratio, the method described in [4]. 〔6〕The method according to any one of 〔1〕~〔5〕, further comprising calculating the blood coagulation time using the first data. 〔7〕The method according to any one of 〔1〕~〔6〕, wherein the first data includes the point R(E) (E is the end point of the coagulation reaction) on the coagulation reaction curve R(i), or the maximum value Vmax of V(i). 〔8〕The method according to any one of 〔1〕~〔7〕, further comprising obtaining the second data after continuing the measurement of the blood coagulation reaction until the end of the coagulation reaction. 〔9〕The method according to any one of 〔1〕~〔8〕, further comprising estimating the blood coagulation abnormality factor of the subject to be examined based on the second data. 〔10〕The estimation of the blood coagulation abnormality factor includes estimating the type of the blood coagulation abnormality factor of the subject to be examined, and the type of the blood coagulation abnormality factor is selected from the group consisting of coagulation factor deficiency, lupus anticoagulant positive, coagulation factor inhibitor, and heparin positive, according to the method described in 〔9〕. 〔11〕The estimation of the blood coagulation abnormality factor includes estimating the presence or absence of the blood coagulation abnormality factor of the subject to be examined, according to the method described in 〔9〕 or 〔10〕. 〔12〕The estimation of the blood coagulation abnormality factor includes estimating the blood coagulation abnormality factor of the subject to be examined according to an estimation model constructed by machine learning, the estimation model is constructed by machine learning using the feature amount representing the blood coagulation reaction of each specimen in the teacher specimen group as the explanatory variable, and the data on the presence or absence of coagulation abnormality or the coagulation abnormality factor of each specimen in the teacher specimen group as the target variable, The group of teacher specimens includes a blood specimen without coagulation abnormalities and blood specimens each having different coagulation abnormality factors. The feature amount includes the second data. The estimation model estimates the presence or absence of coagulation abnormalities or coagulation abnormality factors in the test specimen from the feature amount of the test specimen. The method according to any one of [9] to

[11] .

Advantages of the Invention

[0009] According to the method of the present invention, useful data for estimating coagulation abnormality factors in a blood specimen can be obtained together with the measurement of the coagulation time of the blood specimen. According to the present invention, not only the presence or absence of coagulation abnormality factors in a blood specimen can be estimated, but also the coagulation abnormality factors in an abnormal specimen can be estimated without performing a time-consuming cross-mixing test as in the conventional method. Further, according to the present invention, a specimen that has no prolongation of the coagulation time and appears normal but actually has coagulation abnormality factors can be detected by a simple procedure.

Brief Description of the Drawings

[0010]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Mode for Carrying Out the Invention

[0011] In a blood coagulation test, a predetermined reagent is added to a blood specimen, and the subsequent blood coagulation reaction is measured, and the blood coagulation time is measured from the coagulation reaction. In the following description of this specification, the blood specimen may sometimes be simply referred to as a specimen. For the measurement of the blood coagulation reaction, general means are used, for example, optical means for measuring the amount of scattered light, transmittance, absorbance, etc., or mechanical means for measuring the viscosity of plasma. The blood coagulation reaction is generally represented by a coagulation reaction curve showing the change over time of the amount of the coagulation reaction. The coagulation reaction curve of a normal specimen without coagulation abnormality factors depends on the measurement means, but basically shows a sigmoid shape. For example, the coagulation reaction curve based on the amount of scattered light of a normal specimen usually rises sharply due to the progress of coagulation at a certain time after the addition of the reagent, and then reaches a plateau as the coagulation reaction approaches completion. On the other hand, the coagulation reaction curves of abnormal specimens having coagulation abnormality factors show various shapes depending on the cause of the abnormality, such as a delay in the rise time of the curve and a gentle rise.

[0012] In the measurement of the blood coagulation time of a specimen, data can be collected until the end of the coagulation reaction, that is, until the coagulation reaction curve reaches a plateau, and the coagulation time can be calculated based on the data. For example, taking the reaction amount from the start to the end of the reaction as 100%, the time until the reaction amount reaches 50% can be calculated as the coagulation time. Alternatively, the coagulation time can be calculated based on the rate of change of the coagulation reaction curve, for example, the peak of the coagulation reaction rate or the change over time of the integrated value of the coagulation reaction in a minute time zone (see Japanese Patent Laid-Open No. 6-249855). In the latter method, since the coagulation time can be calculated before the end of the coagulation reaction, it is possible to measure the coagulation time in a shorter time. In specimens having coagulation abnormality factors, in many cases, the coagulation time is prolonged compared to normal specimens. The prolongation of the coagulation time is an indicator of the presence or absence of coagulation abnormality factors. On the other hand, the type of coagulation abnormality factor (factor causing prolongation of the coagulation time) cannot be estimated from the coagulation time.

[0013] Conventionally, the determination of factors prolonging the clotting time (types of clotting abnormality factors) has mainly been carried out by the crossmixing test. Therefore, with the conventional method, without separately conducting a crossmixing test apart from the clotting time measurement, it is impossible to determine the factors prolonging the clotting time. Furthermore, since the crossmixing test requires measurement of the immediate reaction and the delayed reaction after 2-hour incubation for the mixed specimen of the test specimen and the normal specimen, it is time-consuming and laborious.

[0014] Also, with the conventional method, the crossmixing test is usually applied only to specimens in which an extension of the clotting time is observed in the clotting reaction measurement. On the other hand, specimens with a clotting time within the normal range have conventionally been regarded as normal specimens. However, from the research of the present inventors, it has been found that actually, specimens having a clotting abnormality factor may show a clotting time within the normal range due to, for example, the magnitude of sensitivity to the reagent for clotting measurement (see FIG. 8). Such specimens that actually have an abnormality factor but do not show an obvious extension of the clotting time have not been detected as abnormal specimens because they have not been subjected to a crossmixing test or other inspections with the conventional method.

[0015] In the present invention, at the time of measuring the clotting reaction, data for calculating the clotting time is acquired, and data for estimating the clotting abnormality factor is also acquired. Therefore, the present invention enables acquisition of both the clotting time of the specimen and the data for estimating the clotting abnormality factor in a single measurement. Further, the present invention enables estimation of the clotting abnormality factor or acquisition of data therefor at the site where the clotting reaction measurement is performed without conducting a time-consuming crossmixing test. Also, according to the present invention, a specimen that does not show an extension of the clotting time but actually has a clotting abnormality factor can be detected by a simple procedure.

[0016] 〔Method for analyzing blood clotting reaction〕 The present invention provides a method for analyzing a blood coagulation reaction. In the method for analyzing a blood coagulation reaction according to the present invention (hereinafter also referred to as the method of the present invention), the blood coagulation reaction of a test blood sample (hereinafter also referred to as a test sample) is measured, and based on the time-series data of the coagulation reaction obtained by the measurement, data for calculating the blood coagulation time of the test sample (first data) and data for estimating the factors causing blood coagulation abnormalities in the test sample (second data) are obtained.

[0017] Examples of the blood coagulation time that can be calculated according to the present invention include prothrombin time (PT), activated partial thromboplastin time (APTT), and coagulation time in fibrinogen (Fbg) concentration measurement. In the following description of this specification, the method of the present invention will be mainly described by taking the activated partial thromboplastin time (APTT) as an example of the coagulation time. Modification of the method of the present invention to other coagulation times (for example, prothrombin time (PT)) can be implemented by those skilled in the art.

[0018] Hereinafter, the method of the present invention will be described with reference to the basic flow of an embodiment of the method of the present invention shown in FIG. 1.

[0019] 1. Coagulation reaction measurement In the method of the present invention, the plasma of the subject is preferably used as the test sample. An anticoagulant usually used in a coagulation test may be added to the sample. For example, after blood is collected using a blood collection tube containing sodium citrate and then centrifuged, plasma can be obtained.

[0020] In the measurement of blood coagulation reaction, a coagulation time measurement reagent is added to a test specimen, and the blood coagulation reaction is initiated. The coagulation reaction of the mixed solution containing the reagent and the test specimen can be measured. The coagulation time measurement reagent to be used can be arbitrarily selected according to the measurement purpose. Reagents for various coagulation time measurements are commercially available (for example, APTT reagent CoaguPia APTT-N; manufactured by Sekisui Medical Co., Ltd.). For the measurement of the coagulation reaction, general means, such as optical means for measuring the amount of scattered light, transmittance, absorbance, etc., or mechanical means for measuring the viscosity of plasma, etc. may be used. In the following description of this specification, the method of the present invention will be described by taking the measurement of the coagulation reaction based on the amount of scattered light as an example.

[0021] The reaction start point of the coagulation reaction can typically be defined as the time point when the reagent is mixed with the specimen to initiate the coagulation reaction, but other timings may be defined as the reaction start point. The time for continuing the measurement of the coagulation reaction can be, for example, about several tens of seconds to 7 minutes from the time point of mixing the specimen and the reagent. This measurement time may be an arbitrarily determined fixed value, or may be until the time point when the completion of the coagulation reaction of each specimen is detected. During this measurement time, the progress of the coagulation reaction (photometry in the case of optical detection) can be repeatedly measured at predetermined intervals. For example, the measurement may be performed at intervals of 0.1 seconds. The temperature of the mixed solution during the measurement is under normal conditions, for example, 30°C or higher and 40°C or lower, preferably 35°C or higher and 39°C or lower. Also, various conditions for the measurement can be appropriately set according to the test specimen, reagent, measurement means, etc.

[0022] The series of operations in the above-described coagulation reaction measurement can be performed using an automatic analyzer. As an example of an automatic analyzer, the automated blood coagulation analyzer CP3000 (manufactured by Sekisui Medical Co., Ltd.) can be mentioned. Alternatively, some operations may be performed manually. For example, a human can prepare the test specimen, and subsequent operations can be performed using an automatic analyzer.

[0023] Based on the above-described coagulation reaction measurement, measurement data D(i) (the photometric value of the scattered light amount) is sequentially acquired. Here, "i" represents the number of measurement points or the time from the start of the coagulation reaction (also simply referred to as time). For example, if the measurement (photometric) interval is 0.1 second, the time is represented by time = 0.1 × the number of measurement points. An example of the measurement data is shown in FIG. 2. The horizontal axis in FIG. 2 indicates time, and the vertical axis indicates the scattered light amount. As the coagulation reaction of the mixed solution progresses over time, the scattered light amount increases. The coagulation reaction curve based on the scattered light amount as shown in FIG. 2 is usually sigmoid-shaped.

[0024] 2. Acquisition of Reaction R(i) and First Derivative V(i) Next, reaction R(i) is acquired from the measurement data D(i) (step 1). Since the measurement data D(i) includes noise during photometry and fluctuations unrelated to the reaction that appear immediately after the start of photometry, it is preferable to perform smoothing processing on the measured value by a known method. Also, when measuring the coagulation reaction by the scattered light amount, it is preferable to perform zero-point adjustment processing to subtract the scattered light amount derived from the specimen mixed solution before the reaction. For the smoothing processing of the measurement data, any of various known methods related to noise removal can be used. For example, as the smoothing processing, filtering processing, or processing of obtaining a differential value by calculating a difference value or an average slope within an interval described later and then integrating it, and the like can be mentioned. In zero-point adjustment, for example, the smoothed measurement data may be adjusted so that the value at the start of measurement becomes 0. Further, initial fluctuation removal processing may be performed on the measurement data D(i). The initial fluctuation removal processing may be performed so that all values from the start of photometry to a predetermined initial fluctuation removal time become 0. Preferably, the measurement data D(i) is smoothed or zero-point adjusted to obtain the reaction R(i). More preferably, the measurement data D(i) is smoothed and zero-point adjusted to obtain the reaction R(i). Alternatively, after smoothing and zero-point adjusting the measurement data D(i), it may be further converted to a relative value to obtain the reaction R(i). For example, D(i) may be converted so that D(i) at the start of measurement is 0 and D(i) at the end of the coagulation reaction is a predetermined value, for example, 100, to obtain the reaction R(i). The reaction R(i) constitutes the coagulation reaction curve.

[0025] From the obtained reaction R(i), the first derivative V(i) is acquired (step 2). The differentiation process for obtaining V(i) from R(i) can be carried out by any method. For example, it can be performed by calculating the average slope value within an interval. In calculating the average slope value within an interval, a fixed number of measurement points before and after each measurement point i, for example, 2K + 1 measurement points from i - K to i + K, can be used. Here, K is an arbitrary integer. For example, when K is 2, the measurement points of the 5 points of the (i - 2), (i - 1), i, (i + 1), (i + 2)th can be used. The average slope value means the slope value when these multiple measurement points are linearly approximated. Fixed methods such as the least squares method can be used for the operation method of linear approximation. The average slope value of these measurement points can be regarded as the first derivative at the measurement point i. Further, the relative value of the first derivative of R(i) may be acquired as V(i). For example, the first derivative value of R(i) may be converted so that the value at the measurement start time is 0 and the maximum value is a predetermined value, for example, 100, to acquire V(i). The first derivative V(i) constitutes a curve representing the rate or change rate of the coagulation reaction.

[0026] From R(i) or V(i) obtained above, data for calculating the blood coagulation time of the test subject (first data) and data for estimating the blood coagulation abnormality factor of the test subject (second data) are acquired. In the method of the present invention, the acquisition of R(i) and V(i) for the test subject may be performed in parallel with the measurement of the coagulation reaction of the test subject, or may be performed after the measurement of the coagulation reaction is completed. In the method of the present invention, the measurement of the coagulation reaction of the test subject is preferably performed until the coagulation reaction ends. The end of the coagulation reaction can be determined according to any criterion such as the time when R(i) reaches a plateau, the time when V(i) decreases to 0 or a constant value (for example, S% or less of the maximum peak) after reaching the peak.

[0027] As described above, D(i), R(i), and V(i) in this specification may be functions of the number of measurement points or functions of time, respectively. The first data and the second data described later may also be data based on the number of measurement points or data based on time. In the following specification, R(i) and V(i) may be simply abbreviated as R and V, respectively.

[0028] In one embodiment, while measuring the coagulation reaction of the test subject, R and V for the test subject are sequentially obtained. At an appropriate time during the coagulation reaction measurement, the first data is obtained from the R or V, and if necessary, the coagulation time of the test subject is calculated from the first data. Then, preferably after continuing to obtain R and V until the end of the coagulation reaction, the second data is obtained, and if necessary, the coagulation abnormality factor of the test subject is estimated from the second data. In another embodiment, while measuring the coagulation reaction of the test subject, R for the test subject is sequentially obtained. At an appropriate time during the coagulation reaction measurement, the first data is obtained from the R, and if necessary, the coagulation time of the test subject is calculated from the first data. Then, preferably after continuing to obtain R until the end of the coagulation reaction, V is obtained from the R, the second data is obtained, and if necessary, the coagulation abnormality factor of the test subject is estimated from the second data. In another embodiment, preferably while measuring the coagulation reaction of the test subject until the end of the coagulation reaction, R for the test subject is sequentially obtained. Then, V, the first data, and the second data are obtained. If necessary, the coagulation time of the test subject is calculated from the first data, and the coagulation abnormality factor of the test subject is estimated from the second data. In another embodiment, preferably after continuing the coagulation reaction measurement of the test subject until the end of the coagulation reaction, R and V are obtained, and the first data and the second data are obtained. If necessary, the coagulation time of the test subject is calculated from the first data, and the coagulation abnormality factor of the test subject is estimated from the second data. In each of the above-described embodiments, as long as the measured values necessary for obtaining the first data and the second data are obtained, the coagulation reaction measurement may be terminated at a timing earlier than the end of the coagulation reaction.

[0029] 3. Acquisition of First Data In the method of the present invention, the acquisition of the first data and the calculation of the coagulation time of the test subject using the same (step 3) can be performed according to an arbitrary method. Examples of the method for calculating the coagulation time include a method of calculating the time point when R(i) reaches N% (N is arbitrary, the same hereinafter) of R(E) representing the reaction R at the coagulation reaction end point E as the coagulation time; a method of calculating the time point when V(i) reaches the maximum value Vmax or N% thereof as the coagulation time; a method of calculating the coagulation time based on the temporal change of the integrated value of R(i) in a minute time zone (see Japanese Patent Laid-Open No. 6-249855 and Japanese Patent Application No. 2019-237427); a method of calculating the coagulation time based on the weighted average time of V(i) (see Japanese Patent Application No. 2020-039344); a method of calculating the time point when R(i) reaches N% of R(Te) as the coagulation time with the time point when V(i) reaches a predetermined value after reaching the maximum value Vmax as the calculation start point Te (see Japanese Patent Application No. 2020-068877), and the like.

[0030] Therefore, when the above-described method for calculating the coagulation time is used, examples of the first data acquired in the method of the present invention include R(E) representing the reaction R at the coagulation reaction end point E, or data regarding the time point when R(i) reaches N% of R(E); the maximum value Vmax of V(i), or data regarding the time point when V(i) reaches Vmax or N% thereof; data regarding the temporal change of the integrated value of R(i) in a minute time zone; the weighted average time of V(i); the calculation start point Te representing the point when V(i) reaches a predetermined value after reaching Vmax, or data regarding the time point when R(i) reaches N% of R(Te), and the like. However, the method for calculating the coagulation time in the method of the present invention and the type of the first data used therefor are not limited thereto.

[0031] 4. Acquisition of Second Data In the method of the present invention, the first derivative V is used for obtaining the second data and estimating the coagulation abnormality factor of the test subject using the second data (step 4). As described above, V may be sequentially obtained while performing the coagulation reaction measurement, or may be obtained after the completion of the coagulation reaction measurement. Preferably, in the method of the present invention, after performing the coagulation reaction measurement until the end of the coagulation reaction, the second data is obtained using the obtained V, and if necessary, the coagulation abnormality factor is further estimated. Alternatively, as long as V necessary for obtaining the second data is obtained, the coagulation reaction measurement may be terminated at a timing earlier than the end of the coagulation reaction.

[0032] In the procedure for obtaining the second data, the measurement point number or time i at which V becomes a preset reference value X k (that is, satisfying V(i)=X k ) is determined. Alternatively, when V(i)<X k <V(i + 1), or V(i - 1)<X k <V(i), i can be selected as the measurement point number or time satisfying V(i)=X k . Since V usually has a peak shape with the maximum value Vmax as the peak top, there is at least one i satisfying V(i)=X k before and after the time when V reaches Vmax. In the method of the present invention, for V before and after reaching Vmax, i satisfying V(i)=X k is determined for each. Further, in the method of the present invention, X k is a variable, and i corresponding to each X k is determined. For example, when there are n X k (X1, ···, X n ) (k represents a series of integers from 1 to n, and n is an integer of 2 or more), the point p k at which V(i) becomes X k before V(i) reaches Vmax, and the point q k at which V(i) becomes X k after V(i) reaches Vmax are each determined, and n p k (p1, ···, p n ) and n q k (q1, ···, q n) is obtained. When V shows bimodality, there may be multiple points where V becomes X after reaching Vmax. In that case, the largest among the detected points is selected as q. k Similarly, when there are multiple points where V becomes X before reaching Vmax, the smallest among the detected points (excluding the points during the initial noise) is selected as p. k As described above, p k and q k may be data based on the number of measurement points or may be data based on time. k and q k The value of the variable X

[0033] can be arbitrarily set. Preferably, X k is greater than 0 and less than Vmax. X k can be set based on Vmax. For example, X k is defined as Vmax × S k (%), where S k is greater than 0 and less than 100, and preferably can be set in the range of 0.5 to 99. When setting n X k (X1, ···, X k ), n S n (S1, ···, S k ) are set (where k, n are as described above, and each S n is greater than 0 and less than 100, preferably 0.5 to 99). The number of variables X k to be set is not particularly limited, but preferably 5 to 50, more preferably 10 to 30, that is, k is preferably an integer of 5 to 50, more preferably 10 to 30. k Referring to FIG. 3, X

[0034] , p k and q k are described. In FIG. 3, the first derivative V of the solidification reaction curve is plotted against time. The peak top of V is the maximum value Vmax (100%), and the time when V = Vmax is represented by VmaxT. Twenty X k k ​(i.e., k is an integer from 1 to 20) is set, and each is defined as a value from Vmax×3% (k = 1) to Vmax×98% (k = 20). A line indicating each X is drawn under the curve of V. k For each X k , when V is X k (the intersection point of the line of X k and V), there are two such points, one existing before VmaxT and the other existing after VmaxT. As a result, for 20 Xs k , the times when V = X k are detected at 40 points. Among these times, the 20 points (from t[1] to t

[20] ) before VmaxT correspond to p k , and the 20 points (from t

[40] to t

[21] ) after VmaxT correspond to q k .

[0035] The above-mentioned p k , q k can be obtained as the second data. Alternatively, the second data may or may not include p k , q k , and may include parameters calculated from p k or q k . Examples of the parameters that can be included in the second data include statistical values of p k or q k . Examples of the statistical values of p k or q k include the average value (Ave), standard deviation (SD), and coefficient of variation (CV) of p k (p1, ···, p n ) or q k (q1, ···, q n ). In this specification, the Ave, SD, and CV of the point p k that exists before the point where V reaches Vmax may be referred to as pre - Ave, pre - SD, and pre - CV, respectively. Similarly, the Ave, SD, and CV of the point q k that exists after the point where V reaches Vmax may be referred to as post - Ave, post - SD, and post - CV. p k and q kAs a further example of the statistical value, the difference between the pre-Ave and the post-Ave and p k and q k The ratio to the average value of the whole ([(post-Ave - pre-Ave) / (p k and q k average value)], also called the pre-post average difference), and the ratio of the pre-SD to the post-SD ([post-SD / pre-SD], also called the pre-post SD ratio) can be mentioned.

[0036] As a further example of the parameter that may be included in the second data, the midpoint M of p k and q k , and the peak width W representing the width from p k to q k to k can be mentioned. In the following formula, k represents a series of integers from 1 to n, and n represents an integer of 2 or more. Therefore, n Ms k (M1, ···, M k ) and W k (W1, ···, W n ) can be calculated. The coefficient of variation (CV) of the midpoint M k (M1, ···, M n ) reflects the skewness of the peak shape of V and is also called the skewness index in this specification. That is, the more the peak shape of V is skewed (the greater the asymmetry), the greater the change in M k (M1, ···, M n ), so the CV of M k , that is, the skewness index, becomes larger. k M M k =(p k +q k ) / 2 W k =q k -p k

[0037] As a further example of the parameter that may be included in the second data, a sharpness index reflecting the kurtosis of the peak shape of V can be mentioned. For example, the sharpness index is represented by the ratio of the sum (or average value) of W k (W1, ···, W n ) at the upper and lower parts of the peak of V obtained by the following formula. Peakiness index = (sum or average value of W for the lower part of the peak of V) / (sum or average value of W for the upper part of the peak of V) k For example, when 20 Xs are set at equal intervals as shown in FIG. 3 k the sum or average value of W1 to W represents W for the lower half of the peak of V k and W 10 to W k represents W for the upper half of the peak of V 11 At this time, the peakiness index is obtained by dividing the sum (or average value) of W1 to W 20 by the sum (or average value) of W k to W 10 That is, the sharper the peak of V, the greater the difference in peak width between the upper and lower parts of the peak of V, so the peakiness index increases 11 20 k In one embodiment, the second data obtained by the method of the present invention may include at least p

[0038] and q k and preferably includes at least one selected from the group consisting of p k and q k and a parameter calculated from the above-described p k or q k In one example, the second data includes p k and q k and at least one selected from the group consisting of pre-Ave, post-Ave, pre-SD, post-SD, pre-CV, post-CV, pre-post average difference, pre-post SD ratio, M k W k k the distortion index, and the peakiness index. In a preferred example, the second data includes p k and q k and pre-CV and post-CV. In another preferred example, the second data includes p k and q k and the pre-post average difference and the pre-post SD ratio. In another preferred example, the second data includes p k and q k and M k and W k ​including. In another preferred example, the second data is p k and q k , and includes a distortion index and a sharpness index. In a more preferred example, the second data is p k and q k , and includes the previous CV, the subsequent CV, the average difference between the previous and subsequent, the ratio of the previous and subsequent SDs, the distortion index, and the sharpness index.

[0039] In one embodiment, the second data obtained by the method of the present invention includes at least one selected from the group consisting of parameters calculated from at least the above-mentioned p k or q k . In one example, the second data includes at least one selected from the group consisting of the previous Ave, the subsequent Ave, the previous SD, the subsequent SD, the previous CV, the subsequent CV, the average difference between the previous and subsequent, the ratio of the previous and subsequent SDs, M k , W k , the distortion index, and the sharpness index. In a preferred example, the second data includes the previous CV and the subsequent CV. In another preferred example, the second data includes the average difference between the previous and subsequent and the ratio of the previous and subsequent SDs. In another preferred example, the second data includes M k and W k . In another preferred example, the second data includes the distortion index and the sharpness index. In a more preferred example, the second data includes the previous CV, the subsequent CV, the average difference between the previous and subsequent, the ratio of the previous and subsequent SDs, the distortion index, and the sharpness index.

[0040] The above-mentioned p k , q k , or p k or q k , in addition to the parameters calculated therefrom, V itself, the maximum value Vmax of V, or VmaxT representing the time or the number of measurement points at which V = Vmax may further be included in the second data.

[0041] The shape of the coagulation reaction curve of the specimen tends to vary depending on its blood coagulation characteristics (i.e., blood coagulation abnormality factors), and that tendency is the above-mentioned p k , q k , and p k or q kIt is reflected in the parameters calculated from. As shown in FIG. 4, an abnormal specimen having a coagulation abnormality factor typically has a different shape of the coagulation reaction curve R and an extended coagulation time compared to a normal specimen. Also, as shown in the lower part of FIG. 4, the V of the abnormal specimen has a smaller peak compared to the normal specimen and tends to show different shapes such as wide, asymmetric, and bimodal depending on the extension factor. On each V in the lower part of FIG. 4, before and after VmaxT, V = X k (k = 1 to 20) at the point (p k , X k )(circle) and (q k , X k )(triangle) are marked, and further, a line connecting (M k , X k ) is shown as a dotted line.

[0042] As a more detailed example, referring to FIG. 5, the parameters of a specimen lacking coagulation factor VIII (FVIII) will be described. In FIG. 5A, the V of the FVIII-deficient specimen is shown, and the point (p k , X k ) on V is marked with a circle, the point (q k , X k ) is marked with a triangle, and a line connecting (M k , X k ) is shown as a dotted line. The width from p k to q k is W k . FIG. 5B is a plot of p k , q k and M k in FIG. 5A against time. FIG. 5C is a diagram showing the change of W k obtained from p k , q k in FIG. 5A with respect to k. This FVIII-deficient specimen shows an extended coagulation time (coagulation time 120 seconds), and the V of this specimen is bimodal as shown in FIG. 5A, rising rapidly to Vmax and then decreasing once, showing a second relatively gentle increase and decrease. Therefore, in this specimen, the change in q k is larger compared to the change in p k , and the change in M k is also large (FIG. 5B). The W kis very small at the part of the first sharp peak (k is 18 - 20), and becomes wider reflecting the gentle peak in the latter half at other parts (Fig. 5C). Therefore, in this specimen, the pre - CV is small, while the post - CV and the distortion index are relatively large, and the sharpness index is also large.

[0043] Fig. 6A shows the V of a normal specimen. The meanings of the circles, triangles and dotted lines in the figure are the same as those in Fig. 5A. Fig. 6B is a plot of p k , q k and M k against time as shown in Fig. 6A. Fig. 6C is a diagram showing the change of W k obtained from p k , q k in Fig. 6A with respect to k. This specimen shows no prolongation of the coagulation time and has a normal coagulation time (coagulation time 26 seconds). The V of this specimen has a unimodal peak shape with a larger Vmax compared to the FVIII - deficient specimen shown in Fig. 5A. Therefore, in this specimen, compared with the FVIII - deficient specimen shown in Fig. 5A, the changes in p k , q k , and M k are small (Fig. 6B). Also, the W k of this specimen is smaller overall and has fewer changes compared to the FVIII - deficient specimen shown in Fig. 5A (Fig. 6C). Therefore, in this normal specimen, the distortion index and the sharpness index are smaller compared to the FVIII - deficient specimen.

[0044] Thus, the parameters calculated from p k , q k , and p k or q k that can be included in the second data reflect the shape of the V of the specimen and thus the blood coagulation characteristics of the specimen. Therefore, based on the second data, the factors causing coagulation abnormalities in the specimen can be estimated.

[0045] 5. Estimation of factors causing coagulation abnormalities In the method of the present invention, based on the second data, it is possible to estimate the coagulation abnormality factor (i.e., the factor for prolonging the coagulation time) of an abnormal specimen with an extended coagulation time. Also, in the method of the present invention, for a specimen in which no prolongation of the coagulation time is observed, it is similarly possible to estimate the coagulation abnormality factor potentially present in the specimen based on the second data. Further, in the method of the present invention, based on the second data, it is possible to estimate the presence or absence of a coagulation abnormality factor in the specimen. In the following description of this specification, the coagulation abnormality factor (i.e., the factor for prolonging the coagulation time) of an abnormal specimen with an extended coagulation time, and the coagulation abnormality factor potentially present in a specimen in which no prolongation of the coagulation time is observed are collectively referred to as the coagulation abnormality factor (or simply the abnormality factor). Also, in the following description of this specification, the estimation of the type of the coagulation abnormality factor and the estimation of the presence or absence of the coagulation abnormality factor are collectively referred to as the estimation of the coagulation abnormality factor.

[0046] Examples of the types of coagulation abnormality factors that can be estimated by the method of the present invention include, for example, coagulation factor deficiency, lupus anticoagulant (LA) positive, coagulation factor inhibitor (inhibitor), and heparin positive (accurately, a specimen containing heparin). Examples of coagulation factor deficiencies include factor V (FV) deficiency, factor VIII (FVIII) deficiency, factor IX (FIX) deficiency, factor X (FX) deficiency, factor XI (FXI) deficiency, and factor XII (FXII) deficiency. Examples of inhibitors include FVIII inhibitor. Preferably, the type of coagulation abnormality factor that can be estimated by the method of the present invention is selected from the group consisting of coagulation factor deficiency, LA positive, inhibitor, and heparin positive, and more preferably, is selected from the group consisting of FVIII deficiency, FIX deficiency, LA positive, FVIII inhibitor, and heparin positive.

[0047] In one embodiment of the method of the present invention, the abnormal factors of the test subject can be estimated based on the relative values of the pre-CV and post-CV of the test subject with respect to the pre-CV and post-CV based on the normal subject group. The pre-CV and post-CV based on the normal subject group are, for example, the average values of the pre-CV and post-CV of a plurality of normal subjects, and are also referred to as reference_pre-CV and reference_post-CV, respectively, in this specification. The data for the normal subjects may be prepared in advance before the measurement of the test subject, or may be obtained by measuring the normal subjects together with the test subject. In one embodiment, the value obtained by subtracting the reference_pre-CV from the pre-CV of the test subject is the relative_pre-CV of the test subject, and the value obtained by subtracting the reference_post-CV from the post-CV of the test subject is the relative_post-CV of the test subject. In another embodiment, the relative_pre-CV of the test subject is represented by (pre-CV of the test subject / reference_pre-CV) - 1, and the relative_post-CV of the test subject is represented by (post-CV of the test subject / reference_post-CV) - 1. The abnormal factors of the test subject can be estimated from the relative_pre-CV and relative_post-CV of the test subject. For example, for each of normal subjects and various abnormal subject types (e.g., FVIII deficiency, FIX deficiency, LA positive, FVIII inhibitor, and heparin positive), standard values (or ranges) of the relative_pre-CV and relative_post-CV are determined in advance, and by examining which standard value of which subject type the relative_pre-CV and relative_post-CV of the test subject are closest to, or which standard range of which subject type they fall into, the abnormal factors of the test subject can be estimated.

[0048] In one embodiment of the method of the present invention, the abnormal factors of the test subject can be estimated based on the relative values of the distortion index and sharpness index of the test subject with respect to the distortion index and sharpness index based on the normal subject group. The distortion index and sharpness index based on the normal subject group are, for example, the average values of the distortion index and sharpness index of a plurality of normal subjects, and are also referred to as reference_distortion index and sharpness index, respectively, in this specification. The data for the normal subjects may be prepared in advance before the measurement of the test subject, or may be obtained by measuring the normal subjects together with the test subject. In one embodiment, a value obtained by subtracting a reference strain index from a strain index of a test subject is a relative strain index of the test subject, and a value obtained by subtracting a reference sharpness index from a sharpness index of the test subject is a relative sharpness index of the test subject. In another embodiment, the relative strain index of the test subject is expressed as (strain index of the test subject / reference strain index) - 1, and the relative sharpness index of the test subject is expressed as (sharpness index of the test subject / reference sharpness index) - 1. An abnormal factor of the test subject can be estimated from the relative strain index and relative sharpness index of the test subject. For example, similar to the case of the relative pre-CV and relative post-CV described above, by examining which standard value of which specimen type the relative strain index and relative sharpness index of the test subject are closest to, or whether they fall within the standard range of which specimen type, the abnormal factor of the test subject can be estimated.

[0049] In one embodiment of the method of the present invention, an abnormal factor of the specimen can be estimated based on relative values of the pre- and post-average difference and pre- and post-SD ratio of the test subject with respect to the pre- and post-average difference and pre- and post-SD ratio based on a normal specimen group. The pre- and post-average difference and pre- and post-SD ratio based on the normal specimen group are, for example, average values of the pre- and post-average difference and pre- and post-SD ratio of a plurality of normal specimens, and are also referred to as a reference pre- and post-average difference and a reference pre- and post-SD ratio, respectively, in this specification. Data for the normal specimens may be prepared in advance before measurement of the test subject, or may be obtained by measuring normal specimens together with the test subject. In one embodiment, a value obtained by subtracting a reference pre- and post-average difference from the pre- and post-average difference of the test subject is a relative pre- and post-average difference of the test subject, and a value obtained by subtracting a reference pre- and post-SD ratio from the pre- and post-SD ratio of the test subject is a relative pre- and post-SD ratio of the test subject. In another embodiment, the relative pre- and post-average difference of the test subject is expressed as (pre- and post-average difference of the test subject / reference pre- and post-average difference) - 1, and the relative pre- and post-SD ratio of the test subject is expressed as (pre- and post-SD ratio of the test subject / reference pre- and post-SD ratio) - 1. From the relative front-back average difference and relative front-back SD ratio of the test subject, abnormal factors of the test subject can be estimated. For example, similar to the case of the above-described relative pre-CV and relative post-CV, by examining which standard value of which specimen type the front-back average difference and relative front-back SD ratio of the test subject are closest to, or which standard range of which specimen type they fall into, abnormal factors of the test subject can be estimated.

[0050] In another embodiment of the method of the present invention, based on the Standard Deviation Interval (SDI) of the parameter calculated from the above-described p k or q k for the test subject, abnormal factors of the specimen can be estimated. The SDI of the target parameter for the test subject is calculated by the following formula. SDI of the target parameter for the test subject = (α - β) ÷ γ α: Value of the target parameter from the test subject β: Reference value of the target parameter based on the data of the normal specimen group γ: Standard deviation of the target parameter based on the data of the normal specimen group In the above formula, the reference value (β) and standard deviation (γ) based on the data of the normal specimen group can be determined in advance.

[0051] For example, the average value and standard deviation of the target parameter calculated from a given normal specimen group can be applied to the reference value (β) and standard deviation (γ), respectively. Alternatively, the reference value (β) and standard deviation (γ) can be determined by the following procedure: Calculate the average value of the target parameter from an arbitrarily determined first normal specimen group, and exclude specimens having a target parameter that deviates from the average value (for example, more than ±3SD away) from the first normal specimen group. Then, repeat the same procedure using a second normal specimen group consisting of the remaining specimens. Repeat the above procedure until no specimens having a target parameter that deviates from the average value are detected, and adopt the average value and standard deviation of the target parameter from the normal specimen group consisting of the specimens remaining until the end as the reference value (β) and standard deviation (γ).

[0052] The above SDI represents the bias of the parameter value (α) from the test subject with respect to the reference value (β) based on the normal subject. For example, when the SDI of parameter α1 in the test subject is 3 and the SDI of parameter α2 is -4, in the test subject, the value of parameter α1 is 3 SD higher than the reference value (β), and the value of parameter α2 is 4 SD lower than the reference value (β). By converting the parameter value into SDI, it becomes possible to evaluate the difference between the test subject and the normal subject on the same scale (relative value standard with respect to SD) for all parameters. Furthermore, by obtaining the SDI, it becomes possible to compare different parameters with each other, so that a parameter more suitable for estimating the abnormal factor (enabling more accurate estimation) can be selected. The abnormal factor of the test subject can be estimated from the SDI of the parameter of the test subject. For example, for each of the normal subject and various abnormal subject types (e.g., FVIII deficiency, FIX deficiency, LA positive, FVIII inhibitor, and heparin positive), the standard value (or range) of the SDI of various parameters is determined in advance. By examining which standard value of which subject type the SDI of the same parameter of the test subject is closest to, or whether it falls within the standard range of which subject type, the abnormal factor of the test subject can be estimated.

[0053] In a more detailed example, the abnormal factors of the subject can be estimated using a two-dimensional plot of the parameter or its SDI. In this case, a two-dimensional plot using any two selected from the above-described parameters, or any two selected from the SDI of the above-described parameters is used. In a preferred example, the parameters used for the two-dimensional plot are any two selected from the pre-CV, post-CV, distortion index, sharpness index, pre-post average difference, and pre-post SD ratio. In another preferred example, the parameters used for the two-dimensional plot are any two selected from the relative_pre-CV, relative_post-CV, relative_distortion index, relative_sharpness index, relative_pre-post average difference, and relative_pre-post SD ratio. In another preferred example, the parameters used for the two-dimensional plot are any two selected from the SDI of the pre-CV, the SDI of the post-CV, the SDI of the distortion index, the SDI of the sharpness index, the SDI of the pre-post average difference, and the SDI of the pre-post SD ratio.

[0054] Taking the pre-CV and post-CV as examples, the procedure for estimating abnormal factors by two-dimensional plotting will be described. The pre-CV and post-CV of the subject are plotted as one point (pre-CV, post-CV) or (post-CV, pre-CV) on the two-dimensional plane of the pre-CV and post-CV (see Fig. 9A). On the other hand, the reference_pre-CV and reference_post-CV are also plotted as one point on the two-dimensional plane, and this point is defined as the reference point. By shifting the plot so that the reference point overlaps the origin of the two-dimensional plane, the relative_pre-CV and relative_post-CV of the subject are plotted on the two-dimensional plane. Alternatively, the relative_pre-CV and relative_post-CV obtained above may be two-dimensionally plotted (see Fig. 9B). Based on the positions of the plots of the relative_pre-CV and relative_post-CV on the two-dimensional plane, the abnormal factors of the subject can be estimated. For example, for each of a normal subject and various abnormal subject types (e.g., FVIII deficiency, FIX deficiency, LA positive, FVIII inhibitor, and heparin positive), the standard distribution ranges of the plots of the relative_pre-CV and relative_post-CV on the two-dimensional plane are determined in advance, and by examining which standard distribution range of the subject types the plot of the subject falls into, the abnormal factors of the subject can be estimated. When using SDI, the SDI of the pre-CV and the SDI of the post-CV of the subject to be examined are plotted as one point (SDI of the pre-CV, SDI of the post-CV) or (SDI of the post-CV, SDI of the pre-CV) on a two-dimensional plane (see Fig. 9C). Based on the position of the plot on the two-dimensional plane, the presence or absence of an abnormality or the cause of the abnormality of the subject to be examined can be estimated. For example, for each of a normal subject and various abnormal subject types (e.g., FVIII deficiency, FIX deficiency, LA positive, FVIII inhibitor, and heparin positive), the standard distribution range of the plot of the SDI of the pre-CV and the SDI of the post-CV on the two-dimensional plane is determined in advance (see Fig. 16), and by examining which standard distribution range of the subject type the plot of the subject to be examined falls into, the cause of the abnormality of the subject to be examined can be estimated.

[0055] As an example, referring to Table 1-2, Fig. 9-C, and Fig. 16 described later, the following estimations are possible: When the SDI of the pre-CV of the subject to be examined exceeds ±3, or the SDI of the post-CV exceeds ±3, the subject to be examined is presumed to have a coagulation abnormality factor; When the SDI of the pre-CV of the subject to be examined is 6.2 or more and 25.2 or less, and the SDI of the post-CV is 1.3 or more and 12.5 or less, it is presumed that the cause of the abnormality of the subject to be examined may be FVIII deficiency; Among the subjects to be examined presumed to have a possibility of FVIII deficiency, when the SDI of the pre-CV is 7.8 or more and 17.1 or less, and the SDI of the post-CV is 4.6 or more and 8.0 or less, the cause of the abnormality of the subject to be examined may be LA positive; When the SDI of the pre-CV of the subject to be examined is 2.1 or more and 13.1 or less, and the SDI of the post-CV is -4.2 or more and -1.2 or less, it is presumed that the cause of the abnormality of the subject to be examined is FIX deficiency; When the SDI of the pre-CV of the subject to be examined is -0.2 or more and 4.3 or less, and the SDI of the post-CV is 22.6 or more and 44.7 or less, it is presumed that the cause of the abnormality of the subject to be examined is an FVIII inhibitor; When the SDI of the pre-CV of the test subject is -6.2 or more and -3.9 or less, and the SDI of the post-CV is 9.8 or more and 28.9 or less, the abnormal factor of the test subject is presumed to be LA positive; When the SDI of the pre-CV of the test subject is -1.6 or more and 3.6 or less, and the SDI of the post-CV is -10.4 or more and -3.7 or less, the abnormal factor of the test subject is presumed to be heparin positive.

[0056] The same applies to the distortion index, the sharpness index, and the pre- and post-average difference and the pre- and post-SD ratio. That is, for the plots by the relative_distortion index and the relative_sharpness index, the plots by the relative_pre- and post-average difference and the relative_pre- and post-SD ratio, the plots by the SDI of the distortion index and the SDI of the sharpness index, and the plots by the SDI of the pre- and post-average difference and the SDI of the pre- and post-SD ratio, the same procedure as in the above-mentioned cases of the pre-CV and the post-CV can be used. For example, for each of the normal subjects and various abnormal subject types, the standard distribution area on the two-dimensional plane of the plot by the SDI of the distortion index and the sharpness index, or the plot by the SDI of the pre- and post-average difference and the pre- and post-SD ratio is determined in advance (see FIGS. 17 and 18), and by examining which standard distribution area of which subject type the plot of the test subject falls into, the abnormal factor of the test subject can be estimated.

[0057] In the above-mentioned estimation, when two or more abnormal factors are estimated from the test subject, the estimation in the above procedure can be performed using another parameter, or the estimated abnormal factors of the test subject can be narrowed down by performing another test (for example, a conventionally known extended factor discrimination test such as a crossmixing test). For example, in the estimation using the two-dimensional plot of the SDI of the pre-CV and the SDI of the post-CV as shown in FIG. 16, when both FVIII deficiency and LA positivity are estimated as the abnormal factors of the test subject, by performing the estimation using the two-dimensional plot of the SDI of the pre- and post-average difference and the SDI of the pre- and post-SD ratio as shown in FIG. 18, it is possible to determine whether the estimated abnormal factor of the test subject is FVIII deficiency or LA positivity. Alternatively, the immediate reaction of the crossmixing test of the test subject can be performed to determine whether the estimated abnormal factor is FVIII deficiency or LA positivity.

[0058] 6. Estimation of Coagulation Abnormality Factors by Machine Learning Model In another embodiment of the method of the present invention, according to a coagulation abnormality factor estimation model (machine learning model) constructed by machine learning, the coagulation abnormality factors of a specimen can be estimated. For the training data for machine learning, data on blood coagulation reactions from a training specimen population and data on the presence or absence of coagulation abnormalities or the type of abnormality factors are used. For example, by machine learning with the feature quantities representing the blood coagulation reactions of each specimen in the training specimen population as explanatory variables and the data on the presence or absence of coagulation abnormalities or the type of abnormality factors of each specimen in the training specimen population as objective variables, a machine learning model for estimating the coagulation abnormality factors of a test specimen is constructed.

[0059] As the training specimen population, a blood specimen population in which the blood coagulation reaction and the presence or absence of coagulation abnormalities or the type of abnormality factors are known is used. In one embodiment, the training specimen population includes blood specimens without coagulation abnormalities (normal specimens) and blood specimens with coagulation abnormalities (abnormal specimens). In one embodiment, the training specimen population includes abnormal specimens each having different coagulation abnormality factors (preferably, coagulation factor deficiency, LA positive, inhibitor, heparin positive, etc.). Preferably, the training specimen population includes normal specimens and abnormal specimens each having different coagulation abnormality factors (preferably, coagulation factor deficiency, LA positive, inhibitor, heparin positive, etc.).

[0060] Examples of the feature quantities for the blood coagulation reaction used as explanatory variables include at least one selected from the group consisting of the parameters obtained in the above-described second data acquisition step. In a preferred example, the explanatory variables include at least one selected from the group consisting of pre-CV, post-CV, average difference between before and after, ratio of SD before and after, distortion index, and sharpness index, and more preferably, at least one selected from the group consisting of a combination of pre-CV and post-CV, a combination of average difference between before and after and ratio of SD before and after, and a combination of distortion index and sharpness index. Alternatively, all of pre-CV, post-CV, average difference between before and after, ratio of SD before and after, distortion index, and sharpness index may be used as explanatory variables. In another preferred example, the explanatory variable includes at least one selected from the group consisting of relative pre-CV, relative post-CV, relative pre-post average difference, relative pre-post SD ratio, relative distortion index, and relative sharpness index. More preferably, it includes at least one selected from the group consisting of a combination of relative pre-CV and relative post-CV, a combination of relative pre-post average difference and relative pre-post SD ratio, and a combination of relative distortion index and relative sharpness index. Alternatively, all of relative pre-CV, relative post-CV, relative pre-post average difference, relative pre-post SD ratio, relative distortion index, and relative sharpness index may be used as explanatory variables. In another preferred example, the explanatory variable includes at least one selected from the group consisting of SDI of pre-CV, SDI of post-CV, SDI of pre-post average difference, SDI of pre-post SD ratio, SDI of distortion index, and SDI of sharpness index. More preferably, it includes at least one selected from the group consisting of a combination of SDI of pre-CV and SDI of post-CV, a combination of SDI of pre-post average difference and SDI of pre-post SD ratio, and a combination of SDI of distortion index and SDI of sharpness index. Alternatively, all of SDI of pre-CV, SDI of post-CV, SDI of pre-post average difference, SDI of pre-post SD ratio, SDI of distortion index, and SDI of sharpness index may be used as explanatory variables. Together with the above, at least one other parameter, such as the above-mentioned p k , q k , pre-Ave, post-Ave, pre-SD, post-SD, M k , W k , Vmax, VmaxT, etc., selected from the group consisting of may be added to the explanatory variable.

[0061] Examples of data on the presence or absence of coagulation abnormalities or the type of abnormal factor include data indicating the presence or absence of coagulation abnormalities or data indicating the type of abnormal factor (e.g., coagulation factor deficiency, LA positive, inhibitor, heparin positive, etc.).

[0062] Examples of machine learning algorithms used to build a machine learning model include well-known machine learning algorithms such as support vector machine (SVM), neural network (NN), decision tree, random forest, and k-nearest neighbor method.

[0063] Input the verification data into the constructed model to calculate the estimation result of the coagulation abnormality factor. A model whose estimation result best matches the actual result can be selected as the optimal model. For example, a model with the highest correct rate of the estimation result against the actual result, or a model with the smallest error between the estimation result and the actual result.

[0064] The constructed machine learning model outputs an estimation result of the presence or absence of coagulation abnormality or the type of abnormality factor (e.g., coagulation factor deficiency, LA, inhibitor, heparin positive, etc.) of the test subject from the feature quantities (i.e., data corresponding to the above-described explanatory variables) regarding the blood coagulation reaction of the test subject. In one embodiment, the machine learning model is a model for estimating the presence or absence of coagulation abnormality. By inputting the feature quantities regarding the blood coagulation reaction of the test subject, the presence or absence of coagulation abnormality (e.g., normal or having coagulation abnormality) of the test subject is estimated. In another embodiment, the machine learning model is a model for estimating the coagulation abnormality factor. By inputting the feature quantities regarding the blood coagulation reaction of the test subject, the type of coagulation abnormality factor (e.g., coagulation factor deficiency, LA positive, inhibitor, heparin positive, etc.) of the test subject is estimated. Preferably, the machine learning model is a model for estimating the presence or absence of coagulation abnormality and the type of abnormality factor. By inputting the feature quantities regarding the blood coagulation reaction of the test subject, the presence or absence of coagulation abnormality of the test subject and, if there is an abnormality, the type of the abnormality factor are estimated.

[0065] 7. Application to Other Blood Coagulation Reaction Measurement Methods The blood coagulation time measurement method of the present invention has been described above by taking the case of coagulation reaction measurement based on scattered light amount as an example. However, those skilled in the art can apply the method of the present invention to a blood coagulation time measurement method using other coagulation reaction measurement methods (for example, blood coagulation reaction measurement methods based on transmittance, absorbance, viscosity, etc.). Therefore, such an application is included in the scope of the present invention. For example, the reaction R(i) obtained from an inverse sigmoid-shaped coagulation reaction curve based on the transmitted light amount has the opposite sign to that based on the scattered light amount described above. In such a case, in Steps 1 to 4 described above, the signs of R(i) and V(i) are reversed, and instead of the maximum value Vmax of V, the minimum value Vmin is determined, and p k and q k are determined as the points where X k becomes before and after V(i) reaches Vmin, respectively, which is obvious to those skilled in the art.

[0066] 8. Program and Device The above-described analysis method of the blood coagulation reaction of the present invention can be automatically performed using a computer program. Therefore, one aspect of the present invention is a program for performing the above-described analysis method of the blood coagulation reaction of the present invention. In addition, a series of steps of the method of the present invention described above can be automatically performed by an automatic analyzer. Therefore, one aspect of the present invention is a device for performing the above-described analysis method of the blood coagulation reaction of the present invention.

[0067] An embodiment of the device of the present invention will be described below. An embodiment of the device of the present invention is an automatic analyzer 1 as shown in FIG. 7. The automatic analyzer 1 includes a control unit 10, an operation unit 20, a measurement unit 30, and an output unit 40.

[0068] The control unit 10 controls the overall operation of the automatic analyzer 1. The control unit 10 can be constituted by, for example, a personal computer (PC). The control unit 10 includes a CPU, a memory, a storage, a communication interface (I / F), etc., and performs processing of commands from the operation unit 20, control of the operation of the measurement unit 30, storage and data analysis of measurement data received from the measurement unit 30, storage of analysis results, control of output of measurement data and analysis results by the output unit 40, etc. Further, the control unit 10 may be connected to other devices such as an external medium and a host computer. Note that in the control unit 10, the PC that controls the operation of the measurement unit 30 and the PC that analyzes the measurement data may be the same or different.

[0069] The operation unit 20 acquires an input from the operator and transmits the obtained input information to the control unit 10. For example, the operation unit 20 includes a user interface (UI) such as a keyboard and a touch panel. The output unit 40 outputs measurement data of the measurement unit 30, first data, second data based thereon, and analysis results such as an estimated result of the coagulation time or coagulation abnormality factor as necessary under the control of the control unit 10. For example, the output unit 40 includes a display device such as a display.

[0070] The measurement unit 30 executes a series of operations for a blood coagulation test and acquires measurement data of the coagulation reaction of a sample containing a blood specimen. The measurement unit 30 includes various instruments and analysis modules necessary for a blood coagulation test, for example, a specimen container for storing a blood specimen, a reagent container for storing a test reagent, a reaction container for the reaction of the specimen and the reagent, a probe for dispensing the blood specimen and the reagent into the reaction container, a light source, a detector for detecting scattered light or transmitted light from the sample in the reaction container, a data processing circuit for sending data from the detector to the control unit 10, a control circuit for controlling the operation of the measurement unit 30 in response to a command from the control unit 10, etc.

[0071] The control unit 10 analyzes the coagulation reaction of the specimen based on the data measured by the measurement unit 30. This analysis may include obtaining the coagulation reaction curve R and the first derivative V described above, obtaining the first data and the second data, calculating the coagulation time using the first data, and estimating the coagulation abnormality factor using the second data. Alternatively, the coagulation reaction curve R or the first derivative V may be created by the control unit 10 based on the measurement data from the measurement unit 30, or may be created by another device, such as the measurement unit 30, and sent to the control unit 10. The control unit 10 may store reference values such as the reference pre-CV and reference post-CV used for estimating the coagulation abnormality factor, the reference values (β) and reference deviations (γ) of various parameters, etc., or the control unit 10 may capture the reference values stored on an external device or network during analysis.

[0072] The estimation of the coagulation abnormality factor in the control unit 10 may be performed based on the machine learning model for estimating the coagulation abnormality factor described above. In this case, preferably, the control unit 10 may store the machine learning model for estimating the coagulation abnormality factor. The machine learning model may be constructed externally, sent to the control unit 10, and stored or used, or the machine learning model may be constructed, stored, or used in the control unit 10.

[0073] The above analysis can be implemented by a program for performing the method of the present invention. Therefore, the control unit 10 may include a program for performing the method for analyzing the blood coagulation reaction of the present invention.

[0074] The analysis result obtained by the control unit 10 is sent to the output unit 40 and output. The output can take any form, such as display on a screen, transmission to a host computer, printing, etc. The output information from the output unit includes waveform data of R or V, coagulation time, estimation result of coagulation abnormality factor, information regarding the determination criteria of coagulation time included in the first data, Vmax, VmaxT, p k 、q k, or information about parameters calculated therefrom, a two-dimensional plot image of the parameters, etc. The type of output information from the output unit can be controlled by the program of the present invention.

[0075] In one embodiment of the device of the present invention, the measurement unit 30 continues to measure the test sample until the coagulation reaction is completed, and the data is sent to the control unit 10 in sequence. The control unit 10 continues to perform calculations to obtain first data from the coagulation reaction curve R or the first derivative V in sequence, and calculates the coagulation time at the appropriate time. Meanwhile, in parallel with the acquisition of the first data, the coagulation reaction curve R or the first derivative V continues to be acquired, and R and V are acquired until the coagulation reaction is completed. Next, the control unit 10 acquires second data and performs calculations to estimate the coagulation abnormality factor. The obtained analysis result is sent to the output unit and output. For example, R and V are output in parallel with the measurement in sequence until the coagulation reaction is completed, the coagulation time is output during that time, and the estimated result of the coagulation abnormality factor is output after the measurement is completed. EXAMPLES

[0076] The present invention will be described in more detail below by way of examples, but the present invention is not limited to these examples.

[0077] Example 1 1. Method 1.1) Sample The six normal samples (PNP) were prepared using two types of citrated pooled plasma obtained from healthy subjects (N=1 each), Coagpia Calibrator N (N=1) from Sekisui Medical Co., Ltd., Coagtrol N (N=1) from Sysmex Corporation, Pooled Normal Plasma (N=1) from George King Bio-Medical, Inc., and CRYOcheck Pooled Normal Plasma (N=1) from Precision BioLogic Incorporated. · For the LA positive specimens (LA), Positive Lupus Anticoagulant Plasma from George King Bio-Medical, Inc. was used (N = 10). · For the FVIII inhibitor specimens (VIII#Inh), Factor VIII Deficient with Inhibitor from George King Bio-Medical, Inc. was used (N = 6). · For the FVIII-deficient plasma (HA) and FIX-deficient plasma (HB), Factor VIII Deficient and Factor IX Deficient from George King Bio-Medical, Inc. were used (N = 1 for each). Considering the FVIII activities of PNP (citrated pooled plasma obtained from healthy individuals) and HA as 100% and 0% respectively, PNP and HA were mixed to prepare FVIII activity series specimens with FVIII activities of 50%, 25%, 10%, 5%, 2.5%, 1%, 0.75%, 0.5%, 0.25%, 0.1% (N = 1 for each activity). Similarly, considering the FIX activities of PNP (citrated pooled plasma obtained from healthy individuals) and HB as 100% and 0% respectively, PNP and HB were mixed to prepare FIX activity series specimens with FIX activities of 50%, 25%, 10%, 5%, 2.5%, 1%, 0.75%, 0.5%, 0.25%, 0.1% (N = 1 for each activity). · Heparin (Heparin Na Injection 5000 units / 5 mL "Motida" from Mochida Pharmaceutical Co., Ltd.) was added to PNP (citrated pooled plasma obtained from healthy individuals) to prepare heparin concentration series specimens with 10 concentration steps at intervals of 0.1 unit / mL from 0.1 unit / mL to 1.0 unit / mL (N = 1 for each concentration).

[0078] 1.2) Coagulation reaction measurement Coagpia APTT-N (manufactured by Sekisui Medical Co., Ltd.), an APTT measuring reagent, was used as the measuring reagent, and Coagpia APTT-N calcium chloride solution (manufactured by Sekisui Medical Co., Ltd.) was used as the calcium chloride solution. The coagulation reaction measurement of the specimen containing the specimen was performed using a blood coagulation automatic analyzer CP3000 (manufactured by Sekisui Medical Co., Ltd.). After 50 μL of the specimen was heated in a cuvette at 37° C. for 45 seconds, 50 μL of measuring reagent at about 37° C. was added, and after 171 seconds, 50 μL of 25 mM calcium chloride solution was added to start the coagulation reaction. The reaction was performed at 37° C. In the measurement of the coagulation reaction, the cuvette was irradiated with light having a wavelength of 660 nm using an LED as a light source, and the amount of scattered light at 90 degrees side scattered light was measured at 0.1 second intervals. The measurement time was 360 seconds.

[0079] 1.3) Obtaining reaction R(i) and reaction rate V(i) The photometric data from each sample was subjected to smoothing processing including noise removal, and then zero-point adjustment processing was performed so that the amount of scattered light at the start of photometry was 0 to create the response R(i). The first derivative V(i) was calculated from R(i).

[0080] 1.4) First data acquisition and APTT measurement The APTT of each sample was measured by the percentage method. That is, the time when R(i) reached its maximum value Rmax within the measurement time was detected as the end point E of the coagulation reaction, and the time when R(i) reached 50% of R(E) was calculated and determined as the APTT.

[0081] Figure 8 shows the APTT of each specimen. The data in the figure are shifted vertically according to the type of specimen and the specimen number. The dotted line in the figure represents the upper limit (39 seconds) of the range within which the clotting time is judged to be normal (no prolongation). All normal specimens (PNP) had normal clotting times. All LA-positive specimens and FVIII inhibitor specimens (LA and VIII#Inh in the figure) had prolonged clotting times. For the FVIII activity series specimens, FIX activity series specimens, and heparin concentration series specimens (FVIII, FIX, and Heparin in the figure), the clotting time tended to prolong with a decrease in activity or an increase in heparin concentration, but specimens with high activity or low heparin concentration had clotting times within the normal range. From these results, it was shown that among the specimens that were conventionally judged to be normal based on clotting time, there were specimens with mild coagulation abnormality factors. With the conventional method of selectively subjecting specimens with abnormal clotting times to crossmixing tests and other differentiations, it is not possible to detect these specimens with mild coagulation abnormality factors.

[0082] 1.5) Second data acquisition The maximum value Vmax of V(i) for each specimen and the time VmaxT when V(i) = Vmax were determined. X k was defined as Vmax × S% (S increases in 20 steps of 5% from 3% to 98%), and 20 p k (p1, ···, p k ) that satisfy V(i) = X before VmaxT 20 and 20 q k (q1, ···, q k ) that satisfy V(i) = X after VmaxT 20 were detected (see Figure 3). From the obtained p k and q k , the following parameters were calculated. Pre-Ave: p k (p1, ···, p 20 ) average value Post-Ave: q kk (q1, ···, q 20 ) average value Total-Ave: p k (p1, ···, p 20and q kk (q1, ···, q 20 ) average value Pre - SD: p k (p1, ···, p 20 ) standard deviation Post - SD: q kk (q1, ···, q 20 ) standard deviation Pre - CV: p k (p1, ···, p 20 ) coefficient of variation (%) Post - CV: q kk (q1, ···, q 20 ) coefficient of variation (%) M k : (p k + q k ) / 2 (k is any integer from 1 to 20) W k : q k - p k (k is any integer from 1 to 20) Pre - post average difference: (Post - Ave - Pre - Ave) / Total Ave Pre - post SD ratio: Post - SD / Pre - SD Skewness index: M k (M1, ···, M 20 ) coefficient of variation (%) Kurtosis index: (Sum of W1 to W 10 ) / (Sum of W 11 ~ W 20 )

[0083] 1.6) Relationship between solidification abnormality and parameters 1.6.1) Pre - CV and Post - CV Figure 9A is a two - dimensional plot of the pre - CV and post - CV of each specimen. Depending on the specimen type, that is, the type of solidification abnormality factor of the specimen, the plots tended to show different distributions. To relative - value the plot positions, the average values of the pre - CV and post - CV were obtained for six normal specimens (PNP) respectively as reference_pre - CV and reference_post - CV, and then the relative_pre - CV and relative_post - CV of each specimen were obtained by the following formulae. Relative_pre - CV = (Pre - CV of the specimen under test / reference_pre - CV) - 1 Relative_post - CV = (Post - CV of the specimen under test / reference_post - CV) - 1 The two-dimensional plot of the obtained relative pre-CV and relative post-CV is shown in Fig. 9B. In Fig. 9B, the reference point (reference pre-CV, reference post-CV) is located at the origin (0, 0). Therefore, Fig. 9B corresponds to a figure in which the entire plot is shifted so that the reference point overlaps the origin. Each plot in Fig. 9B shows the relative positions of the pre-CV and post-CV of each specimen with respect to the reference point.

[0084] Next, the Standard Deviation Intervals (SDIs) of the pre-CV and post-CV of each specimen were obtained using the following formula. SDI = (α - β) ÷ γ α: The value of the target parameter (pre-CV or post-CV) from the test specimen β: The average value of the values of the target parameter (pre-CV or post-CV) from six normal specimens (PNPs) γ: The standard deviation of the values of the target parameter (pre-CV or post-CV) from six normal specimens (PNPs) The two-dimensional plot of the obtained SDI of the pre-CV and SDI of the post-CV is shown in Fig. 9C.

[0085] Fig. 10 shows the relationship between the factor activity (logarithmic value) and the relative pre-CV and relative post-CV in the FVIII activity series specimens and FIX activity series specimens (FVIII and FIX in the figure). As shown in Figs. 10A and B, the FVIII activity had a high correlation with both the relative pre-CV and relative post-CV, and the FIX activity had a high correlation with the relative pre-CV. Also, as shown in Fig. 10C, the distance from the origin of the point (relative pre-CV, relative post-CV) on the two-dimensional plot shown in Fig. 9B had a high correlation with both the FVIII activity and the FIX activity.

[0086] Table 1-1 shows the APTT, pre-CV, post-CV, relative_pre-CV, and relative_post-CV for each specimen type. In Table 1-1, relative_pre-CV is represented by *pre-CV and relative_post-CV is represented by *post-CV. In Table 1-1, when the APTT exceeds the upper limit of the normal range (39 seconds), it is represented in light gray background with bold black text. Also, for pre-CV, post-CV, relative_pre-CV, and relative_post-CV in Table 1-1, when the value of the specimen is lower than the lowest value of PNP, it is represented in dark gray background with white text, when it is higher than the highest value of PNP, it is represented in light gray background with bold black text, and when the value of the specimen is within the range of values from PNP, it is represented in white background with black text. Table 1-2 shows the APTT, pre-CV, post-CV, SDI of pre-CV, and SDI of post-CV for each specimen type. In Table 1-2, SDI of pre-CV is represented by *pre-CV and SDI of post-CV is represented by *post-CV. The font styles of APTT, pre-CV, and post-CV in Table 1-2 have the same meaning as in Table 1-1. Also, for *pre-CV and *post-CV in Table 1-2, when the SDI is lower than -3, it is represented in dark gray background with white text, when the SDI is higher than 3, it is represented in light gray background with bold black text, and when the SDI is within ±3, it is represented in white background with black text.

[0087]

Table 1-1

[0088]

Table 1-2

[0089] 1.6.2) Distortion Index and Sharpness Index Figure 11A is a two-dimensional plot of the distortion index and sharpness index for each specimen. Depending on the specimen type, i.e., the type of coagulation abnormality factor of the specimen, the plots tend to show different distributions. To relativeize the plot positions, the average values of the distortion index and sharpness index were obtained for six normal specimens (PNP) as the reference_distortion index and reference_sharpness index respectively, and then the relative_distortion index and relative_sharpness index of each specimen were obtained using the following formula. Relative_distortion index = (distortion index of the specimen under test / reference_distortion index) - 1 Relative sharpness index = (sharpness index of the test subject / reference sharpness index) - 1 A two-dimensional plot of the obtained relative distortion index and relative sharpness index is shown in FIG. 11B. In FIG. 11B, the reference point (reference distortion index, reference sharpness index) is located at the origin (0, 0). Therefore, FIG. 11B corresponds to a figure in which all plots are shifted so that the reference point overlaps the origin. Each plot in FIG. 11B shows the relative position of the distortion index and sharpness index of each specimen with respect to the reference point. Next, the SDI of the distortion index and sharpness index of each specimen was obtained in the same procedure as in 1.6.1). A two-dimensional plot of the obtained SDI of the distortion index and SDI of the sharpness index is shown in FIG. 11C.

[0090] FIG. 12 shows the relationship between the factor activity (logarithmic value) and the relative distortion index and relative sharpness index in the FVIII activity series specimens and FIX activity series specimens (FVIII and FIX in the figure). As shown in FIGS. 12A and B, the FVIII activity had a high correlation with the relative distortion index. Also, as shown in FIG. 12C, the distance of the point (relative distortion index, relative sharpness index) from the origin on the two-dimensional plot shown in FIG. 11B had a high correlation with the FVIII activity.

[0091] Table 2-1 shows the APTT, distortion index, sharpness index, relative distortion index, and relative sharpness index of each specimen type. In Table 2-1, the relative distortion index is indicated as *distortion index, and the relative sharpness index is indicated as *sharpness index. Table 2-2 shows the APTT, distortion index, sharpness index, SDI of the distortion index, and SDI of the sharpness index of each specimen type. In Table 2-2, the SDI of the distortion index is indicated as *distortion index, and the SDI of the sharpness index is indicated as *sharpness index. The fonts in Tables 2-1 and 2-2 have the same meaning as those in Tables 1-1 and 1-2, respectively.

[0092]

Table 2-1

[0093]

Table 2-2

[0094] Figure 13A shows a two-dimensional plot of the relative distortion index and the relative sharpness index of abnormal specimens that have coagulation abnormality factors but whose APTT is within the normal range (39 seconds or less), together with the plot of PNP. Figure 13B shows a two-dimensional plot of the SDI of the distortion index and the SDI of the sharpness index of the same abnormal specimens, together with the plot of PNP. Figures 13C and D are tables showing the APTT, distortion index, sharpness index, *distortion index (relative distortion index or distortion index SDI), and *sharpness index (relative sharpness index or sharpness index SDI) of the abnormal specimens shown in Figures 13A and B. The font in the tables of Figures 13C and D has the same meaning as in Tables 2-1 and 2-2. At least one of the relative distortion index and the relative sharpness index of the abnormal specimens shown in Figure 13 differed from PNP or showed an SDI different from PNP. From these results, it was shown that the relative distortion index and the relative sharpness index, or the SDI of the distortion index and the sharpness index, can be used as indices for distinguishing abnormal specimens from normal specimens, and that these indices are effective for detecting specimens that actually have abnormal factors even though the coagulation time is within the normal range. However, for two specimens (the specimen with 0.1 unit of Heparin and the specimen with 50% active FIX), since the plots were close to PNP, it was difficult to distinguish them from PNP using these indices.

[0095] 1.6.3) Mean difference before and after and ratio of SD before and after Figure 14A is a two-dimensional plot of the mean difference before and after and the ratio of SD before and after for each specimen. Depending on the specimen type, that is, the type of coagulation abnormality factor of the specimen, the plots tended to show different distributions. To relative-value the plot positions, the average values of the mean difference before and after and the ratio of SD before and after were obtained for six normal specimens (PNP) as the reference mean difference before and after and the reference ratio of SD before and after, respectively, and then the relative mean difference before and after and the relative ratio of SD before and after for each specimen were obtained using the following equations. Relative mean difference before and after = (mean difference before and after of the test specimen / reference mean difference before and after) - 1 Relative ratio of SD before and after = (ratio of SD before and after of the test specimen / reference ratio of SD before and after) - 1 A two-dimensional plot of the obtained relative front-back average difference and relative front-back SD ratio is shown in FIG. 14B. In FIG. 14B, the reference point (reference front-back average difference, reference front-back SD ratio) is located at the origin (0, 0). Therefore, FIG. 14B corresponds to a figure in which all plots are shifted so that the reference point overlaps the origin. Each plot in FIG. 14B shows the relative position of the front-back average difference and the front-back SD ratio of each specimen with respect to the reference point. Next, the SDI of the front-back average difference and the front-back SD ratio of each specimen was determined in the same procedure as in 1.6.1). A two-dimensional plot of the obtained SDI of the front-back average difference and the SDI of the front-back SD ratio is shown in FIG. 14C.

[0096] FIG. 15 shows the relationship between the factor activity (logarithmic value) and the relative front-back average difference and relative front-back SD ratio in the FVIII activity series specimens and FIX activity series specimens (FVIII and FIX in the figure). As shown in FIGS. 15A and B, the relative front-back average difference had a high correlation with both FVIII activity and FIX activity. The relative front-back SD ratio had a high correlation with FVIII activity. Also, as shown in FIG. 15C, the distance from the origin of the point (relative front-back average difference, relative front-back SD ratio) on the two-dimensional plot shown in FIG. 14B had a high correlation with both FVIII activity and FIX activity.

[0097] Table 3-1 shows the APTT, front-back average difference (InA), front-back SD ratio (InB), relative front-back average difference (*InA), and relative front-back SD ratio (*InB) of each specimen type. Table 3-2 shows the APTT, front-back average difference (InA), front-back SD ratio (InB), front-back average difference SDI (*InA), and front-back SD ratio SDI (*InB) of each specimen type. The fonts in Tables 3-1 and 3-2 have the same meaning as those in Tables 1-1 and 1-2, respectively.

[0098]

Table 3-1

[0099]

Table 3-2

[0100] 1.6.4) Changes in parameter distribution due to coagulation abnormalities Figure 16 is a diagram showing the tendency of the distribution range of the plots of each specimen type on the two-dimensional plot shown in Fig. 9C. The distribution ranges of the plots of the pre-CV and post-CV SDI showed different tendencies according to the specimen type, that is, the type of coagulation abnormality factor of the specimen. However, the distribution range of LA was divided into two, and one overlapped with FVIII.

[0101] Figure 17 is a diagram showing the tendency of the distribution range of the plots of each specimen type on the two-dimensional plot shown in Fig. 11C. The distribution ranges of the plots of the SDI of the distortion index and the sharpness index showed different tendencies according to the specimen type, that is, the type of coagulation abnormality factor of the specimen. The distribution range of LA was divided into two, and one overlapped with FVIII. Figure 18 is a diagram showing the tendency of the distribution range of the plots of each specimen type on the two-dimensional plot shown in Fig. 14C. The distribution ranges of the plots of the SDI of the pre-post mean difference and the pre-post SD ratio showed different tendencies according to the specimen type, that is, the type of coagulation abnormality factor of the specimen.

[0102] From the above results, it was suggested that the SDI of the pre-CV and post-CV, the SDI of the distortion index and the sharpness index, or the SDI of the pre-post mean difference and the pre-post SD ratio could be used as an index for estimating the coagulation abnormality factor of the test specimen. Similarly, the relative_pre-CV and relative_post-CV, the relative_distortion index and the relative_sharpness index, or the relative_pre-post mean difference and the relative_pre-post SD ratio could also be used as an index for estimating the coagulation abnormality factor of the test specimen.

[0103] Comparative Example 1 Using the same specimen group as in Example 1, the relationships between the parameters Vmax and VmaxT and coagulation abnormalities were investigated. For six normal specimens (PNP), the average values of VmaxT and Vmax were obtained respectively as reference_VmaxT and reference_Vmax, and then the relative_VmaxT and relative_Vmax of each specimen were obtained by the following formula. relative_VmaxT = (VmaxT of the test specimen / reference_VmaxT) - 1 relative_Vmax = (Vmax of the test specimen / reference_Vmax) - 1 A two-dimensional plot of the relative VmaxT and relative Vmax of each specimen is shown in Fig. 19. The plots of all abnormal specimens were located below the lower right of PNP according to the delay of the coagulation reaction. Also, the position of the plot tended to move downward to the right according to the severity (factor concentration or activity) of the specimen. As a result, the distribution ranges of LA, FVIII, and FIX overlapped, and a part of the distribution range of LA overlapped with Heparin.

Claims

1. A method for analyzing a blood coagulation reaction, comprising: measuring the blood coagulation reaction of a test specimen to obtain first data for calculating the blood coagulation time of the test specimen and second data for estimating factors causing blood coagulation abnormalities in the test specimen; including obtaining the second data by determining the first derivative V(i) of the coagulation reaction curve R(i), where i represents the number of measurement points or time; Before V(i) reaches the maximum value Vmax, k Point p k After V(i) reaches Vmax, k Point q k where X k is a variable represented by X k = Vmax x S k %, 0 < S k < 100, k represents a series of integers from 1 to n, n being an integer equal to or greater than 2; and calculating at least one selected from the group consisting of pre-Ave, post-Ave, pre-SD, post-SD, pre-CV, post-CV, pre-post average difference, pre-post SD ratio, distortion index, and sharpness index as the second data; including where pre-Ave, pre-SD, and pre-CV represent the average value, standard deviation, and coefficient of variation of pk, respectively; post-Ave, post-SD, and post-CV represent the average value, standard deviation, and coefficient of variation of qk, respectively; the pre-post average difference represents (post-Ave - pre-Ave) / (average value of pk and qk); the pre-post SD ratio represents post-SD / pre-SD; the distortion index represents the coefficient of variation of Mk, and Mk represents (pk + qk) / 2; the sharpness index represents (sum or average value of Wk for the lower part of the peak of V(i)) / (sum or average value of Wk for the upper part of the peak of V(i)), and Wk represents qk - pk; a method.

2. The S as described in k Claim 1, wherein S is from 0.5 to 99.

3. Obtaining the second data further includes calculating the Standard Deviation Interval (SDI) of the target parameter for the test specimen as the second data, where SDI of the target parameter for the test specimen = (α - β) ÷ γ α: value of the target parameter from the test specimen β: reference value of the value of the target parameter based on the data of the normal specimen group γ: standard deviation of the value of the target parameter based on the data of the normal specimen group and the target parameter is any two selected from pre-CV, post-CV, distortion index, sharpness index, pre-post average difference, and pre-post SD ratio; The method according to claim 1 or 2.

4. The method according to any one of claims 1 to 3, further comprising calculating the blood coagulation time using the first data.

5. The method according to any one of claims 1 to 4, wherein the first data includes a point R(E) (E is the end point of the coagulation reaction) on the coagulation reaction curve R(i) or the maximum value Vmax of V(i).

6. The method according to any one of claims 1 to 5, comprising obtaining the second data after continuing the measurement of the blood coagulation reaction until the end of the coagulation reaction.

7. The method according to any one of claims 1 to 6, further comprising estimating a blood coagulation abnormality factor of the test subject based on the second data.

8. The estimation of the blood coagulation abnormality factor includes estimating the type of the blood coagulation abnormality factor of the test subject, and the type of the blood coagulation abnormality factor is selected from the group consisting of coagulation factor deficiency, lupus anticoagulant positive, coagulation factor inhibitor, and heparin positive. The method according to claim 7.

9. The method according to claim 7 or 8, wherein the estimation of the blood coagulation abnormality factor includes estimating the presence or absence of the blood coagulation abnormality factor of the test subject.

10. The estimation of the blood coagulation abnormality factor includes estimating the blood coagulation abnormality factor of the test subject according to an estimation model constructed by machine learning, the estimation model is constructed by machine learning using, as explanatory variables, feature amounts representing the blood coagulation reactions of the respective specimens in a group of teacher specimens, and using, as objective variables, data on the presence or absence of coagulation abnormalities or coagulation abnormality factors of the respective specimens in the group of teacher specimens, the group of teacher specimens includes blood specimens without coagulation abnormalities and blood specimens each having a different coagulation abnormality factor, the feature amount includes the second data, the estimation model estimates the presence or absence of coagulation abnormalities or the coagulation abnormality factor of the test subject from the feature amount of the test subject, The method according to any one of claims 7 to 9.

Images