JP2016183860A - Measuring device and measuring method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To determine the presence or absence of a substance to be measured contained in a sample or quantitatively analyze the amount of substance when measuring a very small amount of signal contained in the sample by a mixed reaction of a sample and a reagent. As one aspect of the present invention for solving at least one of the above-mentioned problems, measurement data points of a signal amount measured from a sample are plotted against a measurement time to generate a graph, and at least. With one curve fitting function, the reaction curve composed of plots of measurement data points is approximated and fitted, and the sample is measured based on the variable value of the curve fitting function obtained by the fitting. .. [Selection diagram] Fig. 1

Description

  The present invention relates to detection of substances contained in a sample. In particular, the present invention relates to a measuring apparatus and a measuring method for detecting an extremely small amount of a substance contained in a sample with high sensitivity and high accuracy.

  In pharmaceutical factories and beverage factories, in recent years, a manufacturing facility called an aseptic system that realizes an aseptic environment is provided, and pharmaceuticals, drinking water, and soft drinks are manufactured. A culture method is used for guaranteeing the sterility environment of the aseptic system and checking the sterility guarantee of the product. However, since the number of colonies generated after culturing the medium in a thermostat for 2-3 days, or in some cases 10 days or more is visually counted, it takes time to obtain the results. I need to wait. From such a background, it is desired to develop a rapid measurement method that enables sterility determination.

Adenosine triphosphate (ATP) bioluminescence method (hereinafter referred to as ATP method), one of the quick and simple bacterial cell detection methods, uses the luminescence reaction of fireflies to measure the intracellular substance ATP into light. To do. The luciferase enzyme incorporates the substrate luciferin and ATP molecules, and measures the amount of luminescence when the oxidized luciferin (oxyluciferin) transitions from the excited state to the ground state along with the consumption of ATP. Corresponds to the generation of one photon (photon), the number of photons generated is proportional to the number of ATP. Since there is an ATP molecule equivalent to 1 atmol (amol = 10 ^-18 mol) as an energy source per viable cell, it is possible to estimate the total number of viable cells from the amount of ATP luminescence contained in the measurement sample. it can. Furthermore, because it has the best quantum efficiency (Φ _BL : ≒ 0.5) of bioluminescence and chemiluminescence, one cell can be detected as a photon equivalent to several hundred thousand, and one cell by luminescence reaction It is in principle possible to detect considerable light.

However, the lower limit of detection of the ATP method is not only the performance of the measuring device, but also ATP is included in the ATP sample present in the environment, luciferin luciferase reagent, so-called luminescent reagent and other reagents necessary for the ATP method. In general, it is reported to be about 10 ² amol (amol = 10 ^-18 mol) due to fluctuations in the data due to the influence. As a method for preventing such fluctuations in data, as disclosed in Patent Documents 1, 2, and 3, in recent years, a dispensing system having a cleaning function for preventing external contamination, and a material type by spectroscopy using an optical filter are used. A bioluminescence detection system in which an identification system and a high-sensitivity photodetector are placed in the same device in a space that is shielded from light and contaminated from outside has been reported, enabling measurement of ATP molecular weight equivalent to 1 amol. It has become to.

  Moreover, in order to improve the performance of the measuring apparatus, for example, as disclosed in Patent Document 4, the random noise component and the number of dark current pulses are reduced, the fluctuation of the signal component is suppressed, and the signal component of the weak light is increased. A means for extracting with accuracy and improving detection sensitivity is adopted.

  Patent Document 5 discloses that fitting is performed using an evaluation function representing an error, and the calculation is terminated when the parameter deviates more than the target value.

JP 2008-268019 A JP 2008-249628 A Japanese Patent Application Laid-Open No. 2012- 211785 Japanese Patent Laid-Open No. 11-142242 JP 2012-43055 A

  As described in Patent Documents 1, 2, and 3, in order to obtain high detection sensitivity and high detection accuracy by the ATP method, ATP, bacterial cells, and other impurities existing in the environment are added to the measurement sample. Prevention, grasping the status of reagents such as whether the activity of the luminescent reagent is maintained, contamination of reagents by contamination with ATP, bacterial cells, and other impurities, contamination of other consumables, and failure of the measuring device It is important to make sure that the measuring device and its surroundings are always stable, taking into account the state management of whether or not this occurs.

  In particular, regarding reagents, ATP is often mixed during production, and these are ATP that cannot be removed in advance, and always exist as a background during luminescence measurement. If the difference between the background and the ATP derived from the microbial cells of the measurement sample is not accurately indicated as a measurement value, there is a risk of determining false positive or false negative in measurement close to a level indicating sterility.

  As one aspect of the present invention for solving at least one of the above-described problems, a graph is generated by plotting measurement data points of a signal amount measured from a sample against a measurement time, and at least one curve file is generated. A fitting function is used to approximate and fit a reaction curve composed of plots of measurement data points, and a sample is measured based on a variable value of the curve fitting function obtained by the fitting.

  The present invention provides a sample analyzer that can reliably measure the amount of a substance in a sample. Problems, configurations, and effects other than those described above will be clarified by the following description of embodiments.

It is a figure which shows a sample analyzer and its periphery system structure. It is a figure which shows the structure of a sample analyzer. It is a figure which shows an example of the operation | movement flowchart of a sample analyzer. It is a figure which shows the typical example of the data waveform of the ATP measurement result obtained using the sample analyzer. It is a figure which shows the typical example of the data waveform of the ATP measurement result obtained using the sample analyzer. It is a figure which shows the sample-analysis result of ATP light emission of the blank sample which does not contain a living microbe. It is a figure which shows the sample-analysis result of ATP light emission of the blank sample which does not contain a living microbe. It is a figure which shows the sample-analysis result of ATP light emission of the blank sample which does not contain a living microbe. It is a figure which shows the sample-analysis result of ATP light emission of the sample containing about 1 living microbe. It is a figure which shows the sample-analysis result of ATP light emission of the sample containing about 1 living microbe. It is a figure which shows the sample-analysis result of ATP light emission of the sample containing about 1 living microbe. It is a figure which compares the variable value of the fitting process result of a blank sample and an ATP extraction sample. It is a figure which compares the variable value of the fitting process result of a blank sample and an ATP extraction sample. It is a figure which compares the variable value of the fitting process result of a blank sample and an ATP extraction sample. It is a figure which compares the variable value of the fitting process result of a blank sample and an ATP extraction sample. It is a figure which shows the flow which determines whether it is a true signal of a measuring object quantity from the result of the sample analyzer in Example 2, or apparatus abnormality. It is a figure which shows the display screen of a sample analyzer. It is a figure which shows the method of estimating the total number of a living microbe from the calibration curve data of ATP.

  Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. However, it should be noted that this embodiment is merely an example for realizing the present invention and does not limit the present invention. In each drawing, the same reference numerals are assigned to common components.

<Example 1>
FIG. 1A is an example showing a sample analyzer and its peripheral system configuration, and includes a measurement device 1, a controller 2, and a sample analyzer 3. Of course, the sample analyzer 3 may include a function including the function of the control device 2, and may further include an integrated system in which all functions of the measurement device 1, the control device 2, and the sample analyzer 3 are integrated as one device. .

  In this embodiment, the processing function of the measurement result of the measuring apparatus 1 in the sample analyzer 3 of the present invention will be described in detail. Therefore, as an example of the present embodiment, the control device 2 controls the operation of the driving mechanism of the measuring device 1 by the control signal 6, and the sample analyzer 3 outputs the signal output 4 from the measuring instrument in the measuring device 1. The description will be given using the configuration to be processed. In the present embodiment, the control device 2 operates in conjunction with the trigger signal 5 for the signal acquisition start / end of the sample analysis device 3 between the control device 2 and the sample analysis device 3, and The operation is controlled by a control signal 6.

  FIG. 2 is a diagram showing the configuration of the sample analyzer 3. The sample analyzer 3 is mainly composed of a main body 8 and a display 9. In addition, in some cases, manual input means 10 such as a mouse and a keyboard can be connected.

  FIG. 2 shows the processing program of the sample analyzer 8 from the viewpoint of hardware, and the main body 8 has a measurement data acquisition unit that acquires a signal output 4 indicating the amount of chemiluminescence or bioluminescence optical signals from the measuring instrument. 11 and a graph waveform generation unit 13 that plots the measurement data points acquired by the measurement data acquisition unit 11 over time with respect to the measurement time and graphs it as a temporal change in the amount of optical signal, and a graph waveform graphed A curve fitting function storage unit 15 that stores a plurality of curve fitting functions for approximating a plurality of curve fitting functions, a plurality of variables of the plurality of curve fitting functions, and a plurality of curve fitting functions stored in the curve fitting function storage unit 15. A data processing unit 16 that sequentially changes a plurality of variables and performs data waveform fitting processing including measurement data points; A data storage unit 19 that stores the processing result of the data processing unit 16, a variable value stored in the data storage unit 19, a data comparison unit 17 that compares with a variable reference value that can be arbitrarily set, and a measurement data point 12 The data determination unit 20 determines a difference between a variable value obtained by the fitting process and a preset variable reference value.

  Here, the variable reference value is a so-called threshold value, and the magnitude of the deviation is represented by the difference between the variable value stored in the data storage unit 19 and the variable reference value. For example, by setting the average value of the amount of light signal when a reagent for chemiluminescence or bioluminescence is reacted to a blank sample, that is, a solution that does not contain the measurement object, as a variable reference value, the variable value and the variable It is also possible to quantify the amount of the target substance from the difference from the reference value. In this case, a calibration curve is created in advance, which is created with the number of target substances or the target substance concentration as shown in FIG. 10 on the horizontal axis and the amount of light corresponding to the concentration on the vertical axis.

  Each functional block constituting the control function of the main body 8 described above with reference to FIG. 2 may be implemented as a software module or may be realized by hardware. That is, each functional block can be realized by software in the main body 8 by the processor interpreting and executing a program stored in a memory that realizes each function. Each functional block may be realized by hardware by designing a part or all of them, for example, by an integrated circuit. Information such as programs, files, databases, function data, variable data, etc. that realize each function is, for example, a memory, a hard disk, a recording device such as an SSD (Solid State Drive), an IC card, an SD card, a DVD, or the like. It can also be placed on other recording media.

  The operation of the sample analyzer 3 will be described below using the configuration of the sample analyzer 3 in FIG. 2 and the operation flowchart in FIG. The signal output 4 from the measuring instrument of the measuring apparatus 1 is stored in the measurement data acquisition unit 11 as a measurement data point (S301). Next, the graph waveform generation unit 13 creates a graph waveform from the measurement data points 12 stored in the measurement data acquisition unit 11 (S302). Since the created graph waveform is stored in the data storage unit 19 by a method such as a text file, a blank sample graph display unit 35 or a measurement sample graph display unit 36 on the display screen of the display 9 in FIG. 4A, 4B and the like can be displayed. The measurement data to be displayed is selected from the first graph calling down menu 42 and the second graph calling down menu 43.

  Next, the curve fitting function group 18 determined to be compatible with the data waveform preliminarily estimated from the measurement object and the measurement content is called from the curve fitting function storage unit 15 (S303), and each of the called curve fitting function group 18 is called. Using the above functions, the fitting process is performed in order by the data processing unit 16 (S303). The curve fitting function group 18 is set in advance by the user in the function group setting means 25. In some cases, a fixed variable of the fitting function may be determined by the variable fixed value setting means 26 in order to shorten the fitting processing time. In the fitting process, the graph waveform of the obtained measurement data point 12 is matched with the least-squares criterion while sequentially changing the peak position, peak height, etc., which are waveform variables, repeatedly. For example, for nonlinear curve fitting, the nonlinear least squares problem is solved by the Levenberg-Marquardt method for fitting. The results of a plurality of variable values of all the curve fitting functions in the called curve fitting function group 18 are stored in the data storage unit 19 in association with each curve fitting function (S304).

  It should be noted that all the variable values are stored in the data storage unit 19 in any of the curve fitting function groups 18 without considering the convergence of each curve fitting function (S304). Further, the determination of the presence or absence of convergence can reduce the processing time by limiting the number of times of fitting processing. These can be determined in advance by the user in the fitting process setting means 27. Further, the curve fitting function group used mainly is based on an exponential decay function, a Lorentz function, and a Gaussian function.

  The variable value group stored in the data storage unit 19 is then compared with at least one comparison variable value group 44 preset by the variable threshold value setting unit 28 in the data comparison unit 17 (S305). ). Since at least one threshold value is set for the variable reference value in the comparison variable value group 24, the data determination unit 20 determines whether the value is within the threshold value range or outside the threshold value range (S306). The determination result in the data determination unit is displayed on the determination result display unit 34 of the display 9 (S307).

  Since the created curve fitting curve is stored in the data storage unit 19 by a method such as a text file, the blank sample graph display unit 35 on the display screen of the display 9 or the measurement as shown in FIG. It can be displayed on the graph display section 36 of the sample. The measurement data to be displayed is selected from the pull-down menu 42 for calling the first graph and the pull-down menu 43 for calling the second graph. Further, the graph waveform and the curve fitting curve of the measurement data point 12 should be displayed on one graph as shown in FIGS. 5A, 5B, 5C, 6A, 6B, and 6C to be described later. Is also possible.

  In this embodiment, as an example of determining the presence / absence of a measurement target substance using the sample analyzer 3 and quantitatively analyzing the abundance, a process performed by the sample analyzer 3 taking aseptic test by ATP luminescence measurement as an example. This will be described in detail. However, it goes without saying that the analysis content of the sample analyzer 3 is not limited to the inspection example. The inspection method using the ATP emission method is briefly described below.

In the ATP method, the number of ATP in cells and bacteria is converted into the amount of light using the light emission reaction of fireflies. The principle is that the luciferase enzyme incorporates the substrate luciferin and ATP molecules, and the amount of luminescence when the oxidized luciferin (oxyluciferin) transitions from the excited state to the ground state with the consumption of ATP is measured.

At this time, since the consumption of ATP1 molecule corresponds to the generation of one photon (photon), the number of photons generated is proportional to the number of ATP, and the living bacteria (viable bacteria) endoss ATP of ATP on average. It is known that the total number of viable bacteria contained in the measurement sample can be estimated by grasping in advance the relationship between the number of ATP and the number of photons generated (amount of optical signal). . A method for creating a database for estimating the number of viable bacteria will be described with reference to FIG. First, an ATP standard solution with a controlled ATP number is prepared using a dedicated diluent. Next, measurement is performed using the measurement system 45 of FIG. 1 using the ATP standard solutions controlled by the respective numbers, and the optical signal amount is arranged for each ATP number. Specifically, an ATP calibration curve 39 is created with the ATP number on the horizontal axis and the amount of light on the vertical axis. The slope y is obtained from the calibration curve, and the number of ATPs can be calculated from the optical signal amount by x1 = y / a.

  Next, a viable bacterial standard solution in which the viable cell count is controlled is prepared using a dedicated diluent. Next, measurement is performed using the measurement system 45 in FIG. 1 using the viable standard solution controlled by each number, and the optical signal amount is arranged for each viable count. Specifically, a calibration curve 40 of live bacteria is created with the number of viable bacteria on the horizontal axis and the amount of light on the vertical axis. The slope y is obtained from the calibration curve, and the number of live A bacteria can be calculated from the amount of optical signal by x2 = y / b. From the above, the number of viable bacteria x2 = (a / b) × x1 is obtained.

  Thereby, the viable count is obtained from the amount of optical signal. Although the average of 1 amol for live bacteria was previously described, it differs depending on the type of bacteria, so if the target object can be assumed to some extent, the relationship between the number of ATP and the number of live bacteria can be determined using the method of FIG. It is better to keep it clear. These relational expressions are processed by the data comparison unit 17 and the data determination unit 20 in FIG. 2 and displayed on the determination result display unit 34 on the display 9.

Next, a measurement procedure for standard live ATP luminescence measurement will be described:
(X) Removal of foreign ATP molecules other than live bacteria by ATP-degrading enzyme
(Y) Extraction of live ATP molecules with surfactant
(Z) a bioluminescent reaction between an ATP molecule extracted from live bacteria and a luminescent reagent,
It consists of three steps.

  4A, 4B, FIG. 5A, FIG. 5B, FIG. 5C, FIG. 6A, FIG. 6B, and FIG. 6C prepare a predetermined amount of ATP extraction sample that has finished the step (Y), In step (Z), the luciferin-luciferase luminescent reagent was dispensed into the sample, mixed with the ATP extraction sample, and the amount of bioluminescence reaction in the ATP-eluted sample was measured with a photodetector. It is a typical example of the graph waveform which showed the result and the result of the analysis using the sample analyzer 3 of a present Example, ie, a curve fitting.

  The embodiment of the measuring apparatus 1 and the control apparatus 2 described in FIG. 1 having a mechanism for measuring with a photodetector is an example described in Japanese Patent Laid-Open No. 2008-268019 (Patent Document 1). Although the detailed contents are not described in the specification of the present invention, the measuring instrument of the measuring device 1 is a photon counting method for measuring weak light emission, and the signal output 4 from the measuring instrument uses a counter substrate. The total number of processed count values per second is output and input to the sample analyzer 3 as measurement data points.

  4A and 4B, the measurement data acquisition unit 11 stores the signal output 4 from the measurement apparatus 1 (S301 in FIG. 3), and the graph waveform generation unit 13 stores the measurement data points stored in the measurement data acquisition unit 11. A typical example of the result of creating a waveform (ATP emission data waveform) 21 from 12 (S302 in FIG. 3) is shown. These are displayed on the blank sample graph display unit 35 and the measurement sample graph display unit 36 on the display of FIG. FIG. 4A shows the result of an ATP extraction sample containing an ATP amount corresponding to one viable cell, and the result of mixing and reacting a blank sample not containing ATP in the sample, that is, a surfactant extraction reagent. Is FIG. 4B.

  Here, the waveforms (ATP luminescence data waveform 21) are the sample background signal 22 before mixing the sample and the reagent in FIG. 4A and FIG. 4B, and the ATP luminescence signal generated by the reaction after mixing the sample and the luminescent reagent. A distinction is made between two areas. As can be seen from FIG. 4B, even in the blank sample, the signal amount is increased by mixing the luminescent reagent, and a waveform similar to ATP emission appears. This can be considered to indicate that ATP is contained in the luminescent reagent or in other reagents and consumables.

Next, the data processing unit 16 calls the curve fitting function group 18 that can correspond to the waveform 21 preliminarily estimated from the measurement object and the measurement content from the curve fitting function storage unit 15, and calls each of the called curve fitting function groups 18. Are used in order to perform the fitting process. Here, as an example of the fitting process shown in S303 of FIG. 3, as a function group, a curve fitting function 26 including one exponential function attenuation term, a curve fitting function 27 including two exponential function attenuation terms, and three exponential functions. A case where three types of curve fitting functions 28 including an attenuation term are selected will be described.
(Curve fitting function with one exponential decay term)
y = y0 + A1 · EXP (− (x−x0) / t1) (1)
(Curve fitting function including two exponential decay terms)
y = y0 + A1 · EXP (− (x−x0) / t1) + A2 · EXP (− (x−x0) / t2)
... Formula (2)

(Curve fitting function including three exponential decay terms)
y = y0 + A1 · EXP (− (x−x0) / t1) + A2 · EXP (− (x−x0) / t2) + A3EXP (− (x−x0) / t3) (3)
Here, y0 is the average value of the measurement data points of the sample background 22. The average value y0 is preferably calculated from 30 or more measurement data points 12, in other words, measurement data of 30 seconds or more. x0 is a time indicating the maximum value of the ATP light emission signal 23. Since the maximum value of the ATP light emission signal 23 is generally obtained within 5 seconds from the start of light emission, the detection area of the maximum value can be narrowed in advance. Further, y0 may be calculated by extracting only the measurement data point 12 of the sample background 22 and fitting it. These processes can be arbitrarily set by the user using the fitting process setting unit 27 and performed by the data processing unit 16.

  Next, the variable value storage described in S304 of FIG. 3 will be described by way of example. The exponential decay terms A1, A2, and A3 are variables related to the signal intensity at x0 seconds obtained by subtracting the average value of the measurement data points of the sample background 22. In Expression (1), A1 indicates the ATP emission signal value when x0 seconds, A1 + A2 indicates the ATP emission signal value when x0 seconds, and in Expression (3), A1 + A2 + A3 indicates x0 seconds. The ATP emission signal value at the time of is shown. t1, t2, and t3 are variables indicating the attenuation rate of the ATP emission signal 23. T1 in the formula (1) indicates a reaction process of ATP-luciferase luminescence reaction from x0 seconds, that is, a process in which ATP is consumed by luciferase. T2 in the formula (2) also shows a process in which ATP is consumed by luciferase from x0 seconds, and shows another decay rate when it is assumed that the reaction route is different from that of t1. Furthermore, t3 in the equation (3) indicates one more decay rate when it is assumed that another reaction path exists in addition to t1 and t2. Hereinafter, the fitting process using the equations (1), (2), and (3) by the data processing unit 16 will be described in detail.

  First, y0 is determined from the measurement data point 12 of the sample background 22 in FIGS. 4A and 4B. Specifically, the average value of the measurement data points 12 for 30 seconds from 80 seconds to 110 seconds is calculated by the data processing unit 16 by the fitting process setting unit 27 in FIG. The y0 value is stored in the data storage unit 19.

  Next, the maximum value is extracted from the ATP emission signal 23, the time x0 at that time is extracted, and stored in the data storage unit 19 as the fixed parameter x0 in the fitting process setting means 27. The fixed values y0 and x0 stored in the data storage unit 19 are called by the fitting process setting unit 27, and the called y0 and x0 are sent to the data processing unit 16, and the fitting process setting unit 27 is used to formula (1) ), (2), and (3) are selected from the three types of curve fitting functions, and each of them is fitted by the data processor 16. The waveform variables A1, A2, A3, t1, t2, and t3 are made to coincide with each other using the least-squares criterion while iteratively changing. For example, the nonlinear least squares problem is solved by the Levenberg-Marquardt method for fitting. The results of the variable values of A1, A2, A3, t1, t2, and t3 of all the curve fitting functions of the called curve fitting function group 18 are stored in the data storage unit 19 in association with the respective curve fitting functions.

5A, FIG. 5B, and FIG. 5C show graphs of results obtained by fitting the data processing unit 16 with the formula (1), the formula (2), and the formula (3), respectively, in the blank sample. 6A, FIG. 6B, and FIG. 6C show the results of ATP luminescence of a sample containing about 1 viable ATP, and the data processing unit 16 uses the equations (1), (2), and (3) for the fitting results. Is shown.
In the case of using Expression (1), the fitting process is converged (FIG. 5A), but in the case of using Expression (2) and Expression (3), the fitting process is not converged (FIG. 5B, 5C), t2 in the expression (2), and t3 in the expression (3) are close to infinity (FIGS. 5B and 5C). That is, in the measurement data of the blank sample, the curve fitting function (equation (1)) including one exponential decay term is suitable as the selection of the fitting function, but includes other two exponential decay terms. The present results show that the curve fitting function including the curve fitting function (equation (2)) and the curve fitting function including the three exponential decay terms (equation (3)) are not suitable as the fitting function selection. The variable values indicating the inappropriateness are t2 in FIG. 5B and t3 in FIG. 5C, which are 9.881 × 10 83 power and 4.791 × 10 85 power, respectively, and diverge almost infinitely. ing. This fitting result indicates that the term including t2 and t3 is preferably almost zero, and the term including t2 and t3 is unnecessary for the curve fitting.

  Even if the curve fitting function is not infinite, it may be determined that a term that gives less than 1 CPS after 100 seconds from x0 is unnecessary for determining the inappropriateness of the selected curve fitting function and the necessity of t2 and t3. . Taking FIG. 5B as an example, when A2 = 53.65076 is quoted as it is, t2 = 5400 (5.4 × 10 4), and only 1008 seconds later gives a change amount of 0.98 CPS. .4 × 10 4 or more may be set as the determination threshold. Similarly, if A3 = 27.65634 is quoted as it is in FIG. 5C, if t2 = 2800 (2.8 × 10 4), only a change amount of 0.96 CPS is given after 100 seconds. What is necessary is just to set 8 * 10 <4> or more as a threshold value. That is, from the relationship of An · EXP (− ((x0 + 100) −x0) / tn) <1, it is possible to determine the threshold value of tn used to determine the inappropriateness of the curve fitting function and the necessity of t2 and t3. In the case of the present embodiment, it is sufficient to determine that the curve fitting function is unsuitable and t2 or t3 is unnecessary when the likelihood is higher than 1 × 10 5 with more likelihood. The variable threshold setting means 28 is responsible for determining these determination thresholds.

  On the other hand, FIG. 6A, FIG. 6B, and FIG. 6C show that the fitting process converges in any sample containing about 1 viable ATP using any of the curve fitting functions of equations (1), (2), and (3). An example of the results is shown. From this result, the variable value groups A1, A1 + A2, and A1 + A2 + A3 corresponding to the signal strength of the curve fitting function are about one viable cell than the blank sample described above with reference to FIGS. 5A, 5B, and 5C. It can be seen that the sample containing ATP is larger. Moreover, there is a difference in the values of t1, t2, and t3. Regarding t2 and t3, it can be seen that these variable values differ greatly between the blank sample and the sample containing about 1 viable ATP depending on the convergence of the fitting. .

  Based on this result, the data determination unit 20 determines whether one or more live bacteria are present in the sample according to the following determination criteria. In this case, the determination of S306 is performed as shown in (a) and (b) below, for example, without performing the comparison process of S305 of FIG.

  (A) If the curve fitting curve of y = y0 + A1 · EXP (− (x−x0) / t1) + A2 · EXP (− (x−x0) / t2) does not converge, it is aseptic.

  (B) When the curve fitting curve of y = y0 + A1 · EXP (− (x−x0) / t1) + A2 · EXP (− (x−x0) / t2) + A3EXP (− (x−x0) / t3) does not converge Is sterile.

  It is also possible to perform the determination by comparing the variable value with the comparison variable value (S305) before the determination process (S306). That is, the threshold values are provided for t2 and t3 of the variable value group for comparison, and the determination is made as shown in (c) and (d) below. However, the criteria for determining the values of t2 and t3 are not limited to this, and can be arbitrarily set.

  (C) As a result of fitting with a curve fitting curve of y = y0 + A1 · EXP (− (x−x0) / t1) + A2 · EXP (− (x−x0) / t2), the value of t2 is 1 × 10 5 If it is more than a power, it is aseptic.

  (D) Result of fitting with a curve fitting curve of y = y0 + A1 · EXP (− (x−x0) / t1) + A2 • EXP (− (x−x0) / t2) + A3EXP (− (x−x0) / t3) If the value of t3 is 1 × 10 5 or more, it is aseptic.

  On the other hand, if it does not correspond to the above (a) to (d), that is, if all of the curve fitting curves converge, or if t2 or t3 is less than 1 × 10 5, the viable bacteria is 1 It is possible to determine that there are more than one.

  Furthermore, it is also possible to calculate the signal intensity from the sum of A1 and A2 or the sum of A1, A2 and A3, and calculate the ATP amount.

  Furthermore, in addition to (a) to (d), threshold values may be provided for A1, A2, and A3, and the determination accuracy of whether or not viable bacteria are present in the sample can be obtained by comparing the threshold values of a plurality of variables. Can be improved.

  7A, FIG. 7B, FIG. 7C, and FIG. 7D each measure a blank sample and an ATP extraction sample equivalent to one viable cell three times, and the sample analyzer 3 performs a fitting process using equation (2). From the results, the variable values of A1 and A2, A = A1 + A2, and t2 are graphically displayed. Although the difference between the blank sample and the ATP extracted sample cannot be distinguished from the variable value of A1 shown in FIG. 7A, A2 shown in FIG. 7B, A2A = A1 + A2 shown in FIG. 7C, It was confirmed by the variable comparison of t2 shown in 7D that it is possible to clearly determine the difference in signal amount between the blank sample and the ATP-extracted sample from viable bacteria and whether it is bacterial or sterile.

  Thus, the presence / absence determination of viable bacteria by the data determination unit 20 can also be determined using A2 or A = A1 + A2 without using t1, t2, and t3. However, when determining using A2 or A = A1 + A2 of the blank sample, since it is necessary to pay attention to false positives, it is necessary to determine the threshold for determining the presence or absence of viable bacteria based on the detection limit and the quantification limit. Although it is determined based on various ideas, it is preferable to adopt the lower detection limit and the lower limit of quantification determined by the currently accepted IUPAC and ISO (International Organization for Standardization). Specifically, 3.3 times the standard deviation (Standard Deviation: SD) of the signal amount of the Planck sample + the signal amount of the blank sample is set as the lower limit of detection, and 10 times or more of the standard deviation is set as the lower limit of quantification. . For example, the standard deviation (SD) of A2 is calculated from the measurement data of a plurality of blank samples, and the value of A2 + 3.3SD (A2) is stored in the variable reference value of the variable value group 24 for data comparison. Keep it.

  Similarly, SD of A = A1 + A2 is calculated, and the value of A + 3.3SD (A) is stored in the variable reference value of the data comparison variable group 24. By comparing with these stored variable reference values, it can be determined whether or not viable bacteria existed in the ATP extracted sample based on whether or not the threshold value is exceeded. In addition, one variable reference value is set as the average value of the blank sample, and from the amount of difference from the signal value obtained with the measurement sample, the method for calculating the number of viable bacteria from the calibration curve described with reference to FIG. The number of viable bacteria may be displayed as a determination result in addition to the determination of the presence or absence of viable bacteria based on whether or not the lower limit of detection is calculated. These are all displayed on the determination result display unit 34.

  As described above, in this embodiment, the variable threshold value setting unit 28 of the data comparison unit 17 of the sample analyzer 3 selects t2, A2, and A as variables, and each variable reference value, for example, t2 = 1 × 10 5th power, By setting A2 = 70 and A = 140, the presence or absence of viable bacteria can be determined with high accuracy. Further, the number of viable bacteria can be quantified by adding an average value of the plank sample as the threshold value of A.

  In addition, the numerical value illustrated in the present embodiment is a value found from a typical example of the experimental result to the last, and it goes without saying that the value varies depending on the condition of the reagent and the performance of the measuring apparatus 1.

<Example 2>
In this embodiment, the sample analyzer 3 detects mismeasurements due to malfunctions of the measurement apparatus 1 or contamination due to consumables or malfunctions due to reagent activity malfunctions in addition to determining the presence or absence of bacteria. An embodiment that also performs detection will be described.

  FIG. 8 shows a step of determining from the result of the sample analyzer 3 whether the sterility test is sterilized or sterile, and further whether it is an erroneous measurement due to a measurement abnormality. Here, an example is shown in which the determination is made by a curve fitting function including two exponential function attenuation terms of Expression (2). The measurement target is ATP, the time series signal value of ATP bioluminescence is the measurement data point, the measurement data point is measured for 200 seconds, the sample background signal before mixing the sample and the reagent, the sample and the luminescent reagent Two regions of the ATP emission signal generated by the reaction after mixing are measured together.

  Here, description will be made with reference to FIG. 9 showing a display screen of the sample analyzer in this embodiment. First, aseptic sample, that is, blank sample is measured by the measuring device 1. After the measurement is completed, in the sample analyzer 3, the curve fitting function group including the exponential function decay term is selected from the function group selection pull-down menu 29 on the display of FIG. 9 (formula (4)).

y = y0 + ΣAi · EXP (− (x−x0) / ti) (1 ≦ i ≦ n) (4)
Thereafter, the sample analysis start button 30 is clicked to start the analysis. The signal output 4 from the measuring instrument of the measuring apparatus 1 is received, the measurement data points are stored (S801), and a graph waveform is created (S802). Next, the graph waveform fitting process is performed by sequentially increasing i in Expression (4) from 1 (S803). Next, variable values of a plurality of curve fitting processing results are stored (S804). The stored variable value is displayed on the measurement sample variable value display unit 32 on the screen. The value of the variable displayed in the measurement sample variable value display section is saved and can be recalled arbitrarily. Here, as described in Example 1, threshold values are set for the variable values of A and t2 based on the blank sample result. For example, AA is set to 140 and t2 is set to 1E5. When the comparison variable group setting button 38 for saving this as a comparison variable group is clicked, it is stored in the sample analyzer 3.

  When the stored comparison variable group is recalled during sample measurement, it is displayed on the comparison variable group display unit 32 of FIG. In the comparison variable group display unit 32, symbols of variable values are displayed in the first column, and those variable values are displayed in the second column. Based on these variable values, they are input to the threshold value display unit 37 as variable reference values. These variable reference values are set as threshold values. Of course, it is also possible to call and display the stored threshold value, the variable value of the result of the blank sample is merely a reference value, and the set value of the threshold value may be freely selected.

  After this state, measurement of the actual measurement sample is started. The signal output 4 from the measuring instrument of the measuring device 1 is received, the measurement data point is stored (S801), a graph waveform is created (S802), and then i in the equation (4) is sequentially increased from 1. Then, graph waveform fitting processing is performed (S803). Next, variable values of a plurality of curve fitting processing results are stored (S804). The stored variable value group is displayed on the measurement sample variable value display unit 33 on the screen. Next, the displayed measurement sample variable value is compared with the comparison variable value. First, comparison 1 is executed to determine whether A2 is greater than or less than a threshold (S805). Next, comparison 2 is performed to determine whether t2 is greater than or less than a threshold (S806, S807). By the above steps, the presence / absence of bacteria, that is, the presence or absence of bacteria or the measurement abnormality is judged.

  As a specific example, in comparison 1 (S805), when the light signal value of the measurement sample is larger than the threshold value, that is, the average value of the light signal value of the blank sample, or the standard deviation SD of A of the blank sample measurement If the ratio is 3.3 times or more, it is determined that there is a high possibility of being a bacterium, and then the process proceeds to comparison 2 (S806). In comparison 2 (806), if it is less than the threshold value, it is determined as being microbial. If it is determined to be microbial, it is possible to count the amount of microorganisms. From the relationship between the number of ATP, the number of viable bacteria, and the amount of each light signal, the estimated amount of the total number of viable bacteria contained in the measurement sample can be calculated. Can be calculated.

  On the other hand, in order to show that the attenuation curve corresponding to the concentration of ATP emission is not obtained when the value is equal to or greater than the threshold value in comparison 2 (806), although the value is equal to or greater than the threshold value in comparison 1 (S805). It is determined that the measurement is abnormal. Possible causes of measurement anomalies that produce such results include a loss of the light-shielding performance of the measuring device, or a mixture of stationary light substances other than ATP emission due to contamination. A message 38 is displayed and the measurement is stopped.

  In comparison 1 (S805), when it is less than the threshold value, that is, when the light signal value of the measurement sample is equal to or lower than the average value of the light signal value of the blank sample, or the standard deviation SD of A in the blank sample measurement If it is less than 3.3 times, it is determined that there is a high possibility of being sterile, and then the process proceeds to comparison 3 (S807). In comparison 3 (806), when it is equal to or greater than the threshold value, it is determined as sterile. On the other hand, when it is determined that the value is less than the threshold value in comparison 3 (807), it is determined that the measurement is abnormal because a normal attenuation curve of ATP emission is not obtained. The cause of the measurement abnormality may be that the dispensed amount of the reagent is less than the set value, or that the reagent has not been dispensed, and the measurement stop message 38 is displayed, and the measurement is stopped.

  On the screen of FIG. 9, the determination result display unit 34 displays the microbial and aseptic results, the number of viable bacteria or the number of ATP in the case of bacteria, and the measurement abnormality is displayed in the case of measurement abnormality. Furthermore, a graph composed of measurement data points and a fitting curve of the graph are also displayed so that it can be confirmed visually whether the analysis is proceeding smoothly (35, 36).

DESCRIPTION OF SYMBOLS 1 ... Measuring apparatus, 2 ... Control apparatus, 3 ... Sample analyzer, 4 ... Signal output from measuring instrument, 5 ... Trigger signal, 8 ... Main body, 9 ... Display, 10 ... Manual input means 11 ... Measurement data acquisition part, DESCRIPTION OF SYMBOLS 13 ... Graph waveform generation part, 15 ... Curve fitting function storage part, 16 ... Data processing part, 17 ... Data comparison part, 18 ... Curve fitting function group, 19 ... Data storage part, 20 ... Data determination part, 21 ... Waveform ( ATP emission data waveform), 22 ... sample background, 23 ... ATP emission signal, 24 ... variable value group for comparison, 25 ... function group setting means, 26 ... curve fitting function including one exponential decay term, 27 ... 2 Curve fitting function including two exponential decay terms, 28... Curve fitting function including three exponential decay terms, 29. Pull-down menu, 30 ... Sample analysis start button, 31 ... Comparison variable group call pull-down menu, 32 ... Comparison variable group display section, 33 ... Measurement sample variable value display section, 34 ... Judgment result display section, 35 ... Blank sample Graph display unit 36 ... Graph display unit for measurement sample, 37 ... Threshold display unit, 38 ... Comparison variable group setting button, 38 ... Message for measurement stop, 39 ... Calibration curve for ATP, 40 ... Calibration curve for viable bacteria, 41: Relationship between the number of ATP and the number of viable bacteria, 42: Pro-down menu for first graph call, 43 ... Pro-down menu for second graph call, 44 ... Comparison variable value group, 45 ... Measurement system, 46 ... Curve Fitting curve y = y0 + ΣAi · EXP (-(x-x0) / ti) i setting button

Claims (10)

A measuring device that measures the amount of a substance contained in a sample based on a change in signal amount due to a reaction between the sample and a reagent,
A measurement data acquisition unit for acquiring measurement data points of the signal amount measured from the sample;
A graph generator for generating a graph by plotting the measurement data points against the measurement time;
A data processor for approximating and fitting a response curve composed of plots of the measured data points of the graph with at least one curve fitting function;
A measurement unit that measures the sample based on a variable value of the curve fitting function obtained by the fitting, and
A measuring device characterized by that. The measuring device according to claim 1,
The measuring unit is
The variable value obtained by the fitting is compared with a threshold value of a variable set in advance for the curve fitting function used for the fitting, and the amount of the substance is measured based on the comparison result.
A measuring device characterized by that. The measuring device according to claim 1,
The measuring unit is
Based on the variable value of the curve fitting function,
Measure the amount of the measurement object contained in the sample,
A measuring device characterized by that. The measuring device according to claim 1,
The measuring unit is
Based on the variable value of the curve fitting function,
Determining whether there is an abnormality in the measurement;
A measuring device characterized by that. The measuring device according to claim 1,
The measurement data acquisition unit
Acquire the measurement data points of the signal amount measured from the sample before reagent mixing and the signal amount measured from the sample after mixing the reagent,
The data processing unit
The fitting approximates a change curve that subsequently decreases from the time including the maximum value of the signal amount generated in the reaction after mixing the reagents.
A measuring device characterized by that. The measuring device according to claim 1,
The measurement data acquisition unit
Acquire the measurement data points of the signal amount measured from the sample before reagent mixing and the signal amount measured from the sample after mixing the reagent,
The measurement data point is the time change of chemiluminescence or bioluminescence optical signal amount before and after the reagent mixing.
A measuring device characterized by that. The measuring device according to claim 1,
The data processing unit
Performing the fitting using a curve fitting function including an exponential decay term in at least one of the variables;
A measuring device characterized by that. It is a measuring device of Claim 7, Comprising:
The measurement data acquisition unit
Acquire the measurement data points of the signal amount measured from the sample before reagent mixing and the signal amount measured from the sample after mixing the reagent,
The data processing unit
The fitting is performed using y = y0 + ΣAi · EXP (− (x−x0) / ti) (1 ≦ i ≦ n) as the curve fitting function,
In the curve fitting function, y0 is an average value of the background signal amount before mixing the sample and the reagent, and x0 is a maximum value of the signal amount after mixing the sample and the reagent.
A measuring device characterized by that. It is a measuring device of Claim 8, Comprising:
The data processing unit
At least one of variables t1 to tn, at least one variable Ai of variables A1 to An, or a sum of at least two variables Ai of variables A1 to An Set the threshold to
The variable value obtained by the fitting is compared with the threshold value, and the amount of the substance is measured based on the comparison result.
A measuring device characterized by that. In a method for measuring the amount of a substance contained in a sample based on the time change of the signal amount due to the reaction between the sample and the reagent,
Preparing a blank sample, mixing and reacting reagents, obtaining a first measurement data point, which is a temporal change in signal amount, in a measurement data obtaining unit;
Fitting the first measurement data point with at least one curve fitting curve;
Storing a first variable group of a first curve fitting curve obtained by fitting;
Preparing a measurement sample, mixing and reacting reagents, and obtaining a second measurement data point, which is a temporal change in signal amount, in the measurement data acquisition unit;
Fitting a second measurement data point with at least one curve fitting curve;
Storing a second variable group of the second curve fitting curve obtained by the fitting;
A comparison unit that compares the first variable group and the second variable group stored in the storage unit with a comparison unit.

Images