JPH09251016A - Data processing method of diagnostic device using liquid chromatography - Google Patents

Abstract

(57) [Abstract] [Problem] When an abnormal chromatogram with an unnatural shape such as a chromatogram is measured, it is quantitatively judged,
In addition, the present invention provides a data processing method capable of determining what causes such an abnormality. In a method for performing various measurements using a chromatogram obtained by liquid chromatography,
The correlation coefficient between the chromatogram obtained by measuring the sample or a part of the chromatogram and the standard chromatogram obtained by measuring the standard sample in advance or a part of the chromatogram is calculated, and the correlation coefficient is calculated. A data processing method for determining whether a chromatogram obtained by measuring a sample based on the above or a part of the chromatogram is normal or abnormal.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data processing method for performing various measurements using liquid chromatography, and particularly to a data processing method suitable for measuring glycated hemoglobin using liquid chromatography. It is about.

[0002]

2. Description of the Related Art When performing various measurements using liquid chromatography, the obtained chromatogram is rarely used as it is. For example, peak information such as peak height, peak area and peak area ratio of chromatogram peaks is used. However, no attention has been paid to the shape of the entire chromatogram or a part of the chromatogram.

For example, when measuring glycated hemoglobin by liquid chromatography, a column packed with a cation exchanger such as silica gel having a carboxymethyl group is used, and a blood sample is diluted with a hemolytic agent and then applied to the column to be measured. Stable glycated hemoglobin is adsorbed, and succinate buffer is used as an eluent to elute the glycated hemoglobin to be eluted.
The absorbance at 5 nm is detected and measured to obtain a chromatogram of the change in the absorbance with respect to the elution time that finally passes. The measurement of glycated hemoglobin based on this chromatogram was carried out by measuring the peak height of the absorbance peak due to glycated hemoglobin, the peak area, etc., as described above, when the same operation was carried out for a sample containing a glycated hemoglobin of known concentration. It is done by comparing with the pod peak area.

[0004]

According to the conventional data processing method, even if the entire chromatogram or a part of the chromatogram is an abnormal chromatogram which is unnatural, the part is not the desired peak. May not be a problem. In addition, there is a problem in that it is necessary to rely solely on the experience of an expert to determine the cause of such an abnormality.

[0005]

Means for Solving the Problems In view of the problems of the prior art, the present inventors have made intensive studies to solve the problems, and as a result, have completed the present invention. An object of the present invention is to provide a data processing method capable of detecting an abnormal chromatogram in which the entire shape of the chromatogram or a part of the chromatogram is unnatural and quantitatively indicating the cause of the abnormality. The present invention made as, that is, in the method of performing various measurements using the chromatogram obtained by liquid chromatography, the chromatogram obtained by measuring the sample or a part of the chromatogram,
A chromatogram obtained by measuring a standard sample in advance or a correlation coefficient with a part of the chromatogram, and measuring the sample based on the correlation coefficient or a part of the chromatogram It is a data processing method that determines whether is normal or abnormal.

Further, the present invention particularly utilizes the above-mentioned data processing method for measuring glycated hemoglobin, in which a glycated hemoglobin is measured using a chromatogram obtained by liquid chromatography. Correlation coefficient is obtained from the chromatogram obtained from the above or a part of the chromatogram and the standard chromatogram obtained by measuring the standard sample in advance or a part of the chromatogram, and the sample is determined based on the correlation coefficient. It is a data processing method in glycated hemoglobin measurement for determining whether a chromatogram obtained by measurement or a part of the chromatogram is normal or abnormal. Hereinafter, the present invention will be described in detail.

The present invention provides a chromatogram or a part of a chromatogram (hereinafter simply referred to as a chromatogram) derived from an object to be measured obtained by liquid chromatography and a chromatogram of a standard sample measured in advance. It is characterized in that a correlation coefficient is obtained, and this is judged to be an abnormal or normal chromatogram, for example, by comparing this with a predetermined threshold value. Here, in the present invention, the standard chromatogram obtained by measuring the standard sample is not limited to a normal chromatogram derived from a normal standard sample, but the sample itself may deteriorate, or some inconvenience may occur during measurement. The abnormal chromatogram obtained as a result of is also used.

In the present invention, an abnormal chromatogram or the like that can be measured when the cause is known, such as poor temperature control or column deterioration, or an abnormal chromatogram obtained due to alteration of the sample, etc. By obtaining the correlation coefficient of the chromatogram or the like obtained by measuring the sample, when the abnormal chromatogram or the like is obtained, the cause thereof can be quantitatively shown. That is, the correlation coefficient is obtained from the obtained chromatogram and the previously obtained abnormal chromatogram, and when this value is high, it is determined that the cause of the abnormal chromatogram is generated. In the present invention, preferably, an abnormal chromatogram or the like that may frequently occur is previously measured, and after measuring a sample, a correlation coefficient with the obtained chromatogram or the like is obtained,
If the correlation coefficients are displayed in descending order, it is possible to make a quantitative suggestion such as estimating the type of cause of abnormality.

In the present invention, it is particularly preferable that a normal chromatogram and a plurality of typical abnormal chromatograms are measured in advance as standard chromatograms and the like, and the sample is measured, and then all of these standard chromatograms are measured. The correlation coefficient with Gram etc. is calculated, and the values are listed in order. This makes it possible to show at a time whether or not the chromatogram obtained for the sample is normal, and which of the typical abnormal chromatograms is the closest.

With respect to abnormal chromatograms, etc., once the abnormal chromatograms, etc. are obtained, even if it is unknown what causes the abnormal chromatograms, etc. It can be used as a standard chromatogram obtained by measuring a standard sample in the present invention.

As means for obtaining the correlation coefficient in the present invention, for example, the following equation can be used.

[0012]

(Equation 3)

In the formula, C means a correlation coefficient, and n is the number of data points, that is, in the case of adsorbing an object to be measured on a column once, for example, in the case of measuring glycated hemoglobin, the column is adsorbed. The reference time is the time at which the eluent for eluting the target substance starts to be sent, or, for example, when the target substance is not adsorbed to the column, the time at which the sample starts to be sent to the column. Means, for example, the number of data such as absorbance sampled after the elapse of an arbitrary time such as 0.1 seconds, 1 second, or 10 seconds. In the present invention, it is preferable to increase the number of n in order to improve the accuracy of determination of an abnormal chromatogram.

In the formula, X and Y are measurement data, X is data such as absorbance after elapse of an arbitrary time from the reference time in a standard chromatogram, and Y is a chromatogram obtained by measuring a sample. This means data such as absorbance at the same time as in the case of X (the time when the same time has elapsed from the reference time) in gram or the like. Note that X may be data such as a chromatogram obtained by measuring a sample, and Y may be data such as absorbance in a standard chromatogram. X
And Y may be obtained from the entire chromatogram or a part thereof. The latter is particularly effective when the chromatogram derived from the measurement object is only a part of the obtained chromatogram.

As described above, n pieces of data (X)
And based on n sets of data consisting of n pieces of data (Y),
The correlation coefficient C is calculated.

According to the present invention, it is possible to very easily determine whether or not the obtained chromatogram is an abnormal chromatogram by comparing the correlation coefficient with a predetermined threshold value. . Of course, it is also possible to make a decision on a case-by-case basis from the calculated correlation coefficient without using a threshold value, but from the viewpoint of obtaining a decision that does not depend on the skill of the measurer, such as automation of the measuring device. , It is preferable to use a threshold value.

Based on the above equation, the correlation coefficient C takes a value between -1 and 1, especially when X is a constant multiple of Y.
That is, it is -1 or 1 when the standardized shapes of the two chromatograms are the same, and 0 when X and Y are independent, that is, when the shapes of the two chromatograms are completely different.
Therefore, when the absolute value or square value of the correlation coefficient C calculated by the above equation is compared with the magnitude of a predetermined threshold value, and the absolute value or square value of the correlation coefficient C is larger than the threshold value. The abnormal chromatogram can be detected quickly and accurately by determining that the normal chromatogram is the normal chromatogram, and if it is smaller than the threshold value, the abnormal chromatogram.

Here, when an abnormal chromatogram is used as a standard chromatogram or the like used when obtaining the correlation coefficient, for example, the reverse determination may be performed.

Further, the correlation coefficient C is calculated by the above equation, where X is a chromatogram or the like measured on the sample, and Y is one or more typical abnormal chromatograms measured in advance. By arranging them in descending order of absolute value or squared value, it is possible to immediately determine which typical abnormal chromatogram and how close the chromatogram is. In addition, since the degree of similarity can be represented by the numerical value of the correlation coefficient, it is possible to quickly and quantitatively determine / guess out possible abnormalities in descending order of correlation.

[0020]

According to the present invention, a chromatogram obtained by subjecting a sample to liquid chromatography is normally obtained by obtaining a correlation coefficient with a standard chromatogram obtained by measuring a standard sample in advance. It is possible to quickly determine whether or not it is abnormal. In making this determination, it is possible to make an objective determination without using the subjectivity of the measurer by using a threshold value or the like.

According to the present invention, the correlation coefficient between the obtained chromatogram or the like and the previously obtained abnormal chromatogram or the like is obtained, so that the degree of approximation of the chromatogram to the abnormal chromatogram can be immediately determined. Can be determined. Moreover, since the degree of similarity can be represented by the numerical value of the correlation coefficient, it is possible to estimate the cause of the abnormality.

As described above, according to the present invention, (1) abnormal chromatograms can be determined quickly and accurately without relying on empirical judgment by a skilled person, and (2) abnormal chromatograms can be classified. It can be performed quickly and quantitatively without the empirical judgment of a skilled person.

[0023]

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, the present invention will be described in more detail with reference to the drawings, but the present invention is not limited to the invention described in these drawings.

FIG. 1 is a block diagram showing the construction of an embodiment in which the present invention is applied to a fully automatic glycated hemoglobin measuring apparatus, and FIG. 2 is a flow chart of a data processing part.

The components of the blood sample 2 collected by the sampling unit 1 are separated by a liquid chromatography separation system 3 including a column filled with a cation exchange resin, and the components are eluted out of the system. The eluted component is detected by the detector 4 as a change in absorbance. The detected absorbance is A /
After being converted into a digital signal by the D converter 5, the CPU 6
And processed for data processing.

In the data processing unit 6, according to the flow chart of FIG. 2, first, in the data of the measured chromatogram and the chromatogram of the normal standard sample (FIG. 3) measured in advance, the characteristic of glycated hemoglobin. The portion where the change appears is extracted, and the correlation coefficient is calculated using the formula described above. In the chromatogram obtained by the measurement of glycated hemoglobin, the part where the characteristic change in glycated hemoglobin appears is an inner diameter of 4.6 mm and a length of 35 m.
When a column packed with a cation exchange resin of m was used and succinate buffer was used as an eluent at a flow rate of 1.5 ml / min, the elution time was about 0.2 minutes to 1.5 minutes. is there.

The value obtained by squaring the calculated correlation coefficient is compared with a preset threshold value. Generally, it is preferable to set the threshold value to about 0.8 in order to carry out highly accurate measurement, and according to the findings of the present inventors, the threshold value is 0 even in the measurement of glycated hemoglobin. It is preferable to set it as 0.8. As a result of the comparison, if the value is larger than 0.8, it is determined to be a normal chromatogram, and if it is smaller than 0.8, it is determined to be an abnormal chromatogram.

Chromatograms obtained by measuring five blood samples (data 1 to 4 are shown in FIGS. 4 to 7 respectively).
Table 1 shows the results when the data processing of the present invention was applied to

[0029]

[Table 1]

As described above, it was determined that the data 1, data 2 and data 3 were abnormal chromatograms having different shapes from the standard chromatogram. In fact, FIGS. 4 and 5 showing chromatograms of data 1, data 2 and data 3
6 is compared with the normal standard chromatogram shown in FIG. 3, it can be seen that the peak in the portion having an elution time of more than 1 minute is significantly different. On the other hand, comparing FIG. 7, which shows the chromatogram of data 4, with the chromatogram of FIG. 3, which is a normal standard chromatogram, in FIG. Considered to be gram.

Next, with respect to the data determined to be the abnormal chromatogram, the correlation coefficient with the typical abnormal chromatogram A (FIG. 8) and the chromatogram B (FIG. 9) by the abnormal hemoglobin measured in advance, and further Is the correlation coefficient with the chromatogram of the normal standard sample (FIG. 3),
The calculation was carried out by extracting a portion of 0.2 minute to 1.5 minutes where glycated hemoglobin shows a characteristic shape change. After squaring the obtained correlation coefficients, they were arranged in descending order and the candidates for classification of abnormal chromatograms were raised in descending order of correlation. In addition, chromatogram A is a typical abnormal chromatogram observed when blood of a newborn is used as a sample in glycated hemoglobin measurement, and chromatogram B is observed when a sample containing abnormal hemoglobin is measured in glycated hemoglobin measurement. It is a typical abnormal chromatogram.

The results for data 1 are shown in Table 2.

[0033]

[Table 2]

As shown in Table 2, data 1 (chromatogram shown in FIG. 4) shows a very high correlation with typical abnormal chromatogram A, while normal chromatogram (FIG. 3) and typical chromatogram It can be seen that the correlation with the abnormal chromatogram B (FIG. 9) is low. From this result, it is inferred that the sample used to obtain the data 1 has the same disorder as that which gave rise to the typical abnormal chromatogram A.

The results for data 2 are shown in Table 3.

[0036]

[Table 3]

As shown in Table 3, data 2 (the chromatogram shown in FIG. 5) shows a relatively high correlation with the typical abnormal chromatogram B, while the normal chromatogram (FIG. 3) and the typical chromatogram. It can be seen that the correlation with the abnormal chromatogram A (FIG. 8) is relatively low. From this result, the sample used to obtain the data 2 had the same impairments that produced the typical abnormal chromatogram B, or produced the typical abnormal chromatograms A or B. It is estimated that there is a disability.

The results for data 3 are shown in Table 4.

[0039]

[Table 4]

As shown in Table 4, data 3 (the chromatogram shown in FIG. 6) is normal chromatogram (FIG. 3), typical abnormal chromatogram A (FIG. 8) and typical abnormal chromatogram B ( It can be seen that the correlation is almost the same as in FIG. 9). From this result, the sample used to obtain data 3 has a typical abnormal chromatogram A or B
It is highly probable that there is a disorder other than the one that caused

As described above, two examples are given as typical abnormal chromatograms due to abnormal hemoglobin. However, by using more abnormal chromatograms and abnormal chromatograms due to hard factors such as poor suction, abnormal chromatograms can be obtained. It becomes easy to classify when measured, and it is possible to easily investigate the cause and take appropriate measures.

[Brief description of drawings]

FIG. 1 is a diagram showing the configuration of a fully-automatic glycated hemoglobin measuring device in which the present invention is applied to glycated hemoglobin measurement.

FIG. 2 is a diagram showing a flowchart of data processing by a CPU in the fully automatic glycated hemoglobin measurement device shown in FIG.

FIG. 3 is a diagram showing a chromatogram obtained when an appropriate glycated hemoglobin measurement is performed on a normal sample.

FIG. 4 is a diagram showing a chromatogram obtained by performing glycated hemoglobin measurement on an arbitrary blood sample.

FIG. 5 is a diagram showing a chromatogram obtained by performing glycated hemoglobin measurement on an arbitrary blood sample.

FIG. 6 is a diagram showing a chromatogram obtained by performing glycated hemoglobin measurement on an arbitrary blood sample.

FIG. 7 is a diagram showing a chromatogram obtained by performing glycated hemoglobin measurement on an arbitrary blood sample.

FIG. 8 is a diagram showing a typical abnormal chromatogram obtained when measuring blood of a newborn baby in measuring glycated hemoglobin.

FIG. 9 is a diagram showing a typical abnormal chromatogram obtained when measuring abnormal hemoglobin blood in glycated hemoglobin measurement.

[Explanation of symbols]

 1 Sampling part 2 Blood sample 3 Liquid chromatography separation system 4 Detector 5 A / D converter 6 CPU (processing part)

Claims (12)

[Claims] 1. A method for performing various measurements using a chromatogram obtained by liquid chromatography,
The correlation coefficient between the chromatogram obtained by measuring the sample or a part of the chromatogram and the standard chromatogram obtained by measuring the standard sample in advance or a part of the chromatogram is calculated, and the correlation coefficient is calculated. A data processing method for determining whether a chromatogram obtained by measuring a sample based on the above or a part of the chromatogram is normal or abnormal. 2. The data processing method according to claim 1, wherein the standard chromatogram obtained by measuring the standard sample or a part of the chromatogram is a normal chromatogram. 3. The data processing method according to claim 1, wherein the standard chromatogram obtained by measuring the standard sample or a part of the chromatogram is an abnormal chromatogram. 4. The data processing method according to claim 1, wherein the following equation is used to obtain the correlation coefficient. [Equation 1] (Where n is the number of data points, X and Y are standard and measured data, and C is the correlation coefficient). 5. The data processing method according to claim 1, wherein the obtained correlation coefficient is compared with a preset threshold value. 6. The data processing method according to claim 5, wherein the absolute value or square value of the obtained correlation coefficient is compared with a threshold value of 0.8. 7. A method for measuring glycated hemoglobin using a chromatogram obtained by liquid chromatography, the chromatogram obtained by measuring a blood sample or a part of the chromatogram, and a normal standard in advance. Obtain the correlation coefficient from the standard chromatogram obtained by measuring the sample or a part of the chromatogram, and the chromatogram obtained by measuring the sample based on the correlation coefficient or a part of the chromatogram is normal. A method for processing data in measuring glycated hemoglobin, which determines whether there is abnormal or abnormal. 8. The data processing method for measuring glycated hemoglobin according to claim 7, wherein the standard chromatogram obtained by measuring the standard sample or a part of the chromatogram is a normal chromatogram. 9. The data processing method according to claim 7, wherein the standard chromatogram obtained by measuring the standard sample or a part of the chromatogram is an abnormal chromatogram. 10. The data processing method for measuring glycated hemoglobin according to claim 7, wherein the following equation is used to obtain a correlation coefficient. [Equation 2] (Where n is the number of data points, X and Y are standard and measured data, and C is the correlation coefficient). 11. The obtained correlation coefficient is compared with a preset threshold value.
A data processing method for measuring glycated hemoglobin according to any one of items. 12. The absolute value or the squared value of the obtained correlation coefficient is compared with a threshold value of 0.8.
Data Processing Method for Glycated Hemoglobin Measurement in Rats.