Management device, management system, and accuracy management method for automatic analysis device

Abstract

Provided is a management device for an automatic analysis device having a measurement unit. The management device stores therein calibration curve data corresponding to a reagent, a measurement section of a reaction process between a sample and the reagent, a plurality of predetermined determination sections set in advance in the measurement section, and a determination reference value for use in individual determination for each determination section, and executes a first step for selecting a sample for use in accuracy management and a second step for performing accuracy management on the basis of a reagent type corresponding to calibration curve data registered for a measurement item of the sample and reaction process data measured by the measurement unit. In the second step, the management device performs the individual determination by using the determination reference value on the basis of measurement data for each determination section, determines the state of at least one of the measurement unit, the reagent, and the sample on the basis of a combination pattern of the results of the individual determination, and outputs the result of the accuracy management.

Description

Management Device, Management System, and Accuracy Management Method for Automated Analyzer

[0001] The present invention relates to a management device, a management system, and an accuracy management method for an automated analyzer that analyzes biological samples such as blood and urine.

[0002] The automated analyzer measures the absorbance or luminescence obtained by the reaction between a specific component contained in a biological sample such as blood or urine and a reagent, and performs qualitative and quantitative analysis. In the automated analyzer, in order to ensure the reliability of the analysis results, accuracy management is performed to confirm that the state of the reagent and the device used for measurement is kept constant.

[0003] As a method of accuracy management, data of the results of measuring a sample (hereinafter referred to as an accuracy management sample) prepared specifically for accuracy management, whose components and concentrations are known and whose quality is guaranteed, is used as an accuracy management value, and an accuracy management value and its transition (trend) are confirmed. This method is common.

[0004] On the other hand, an accuracy management method using a biological sample derived from a patient (hereinafter referred to as a patient specimen) without using the above accuracy management sample has also been studied. In this method, since an accuracy management sample is not required, there is an advantage that the man-hours and costs involved in the preparation work of the accuracy management sample can be reduced.

[0005] Japanese Patent Application Laid-Open No. 2019-174423 (Patent Document 1) discloses a technique for obtaining the results of determining whether the test results of patient specimens are positive or negative from a plurality of analyzers, and generating an accuracy management index based on the ratio of patient specimens determined to be positive and patient specimens determined to be negative.

[0006] Japanese Patent Application Laid-Open No. 2006-292698 (Patent Document 2) discloses a technique for performing analysis of a predetermined measurement item on the same patient specimen a plurality of times at predetermined time intervals, and determining whether the variation in the analysis values obtained by the plurality of executions is within a predetermined standard.

[0007] Japanese Patent Application Laid-Open No. 2019-174423Japanese Patent Application Laid-Open No. 2006-292698

[0008] However, in Patent Document 1, the quality control index is created by analyzing a predetermined number of patient samples for each analytical instrument or facility and averaging them out by percentages such as positive rate and negative rate. As a result, the quality control index is created by averaging outliers that may be included in individual analysis results.

[0009] Furthermore, although Patent Document 2 describes that users can select patient samples to be used for quality control, there are no clear criteria for selecting patient samples, and if the selection of patient samples is inappropriate, bias may occur in quality control.

[0010] The present invention aims to provide a control device, a management system, and a quality control method for an automated analyzer that can accurately perform quality control using patient samples.

[0011] To achieve the above objective, the present invention provides a management device for an automated analyzer having a measurement unit, comprising a storage device and a calculation device, wherein the storage device stores calibration curve data corresponding to a reagent, measurement intervals for the reaction process of the sample and the reagent, a predetermined number of judgment categories set in advance within the measurement interval, and judgment criterion values ​​used for individual judgment of the judgment categories, the calculation device performs a first step of selecting a sample to be used for quality control, and a second step of performing quality control based on the reagent type corresponding to the calibration curve data registered for the measurement items of the sample and reaction process data which is time-series measurement data measured by the measurement unit in the measurement interval, and in the second step, the calculation device performs the individual judgment for each predetermined judgment category using the judgment criterion values ​​based on the measurement data, determines at least one state of the measurement unit, the reagent and the sample based on the combination pattern of the results of the individual judgment for each judgment category, and provides a management device that outputs quality control results based on the combination pattern of the judgment results of the individual judgment.

[0012] According to the present invention, quality control using patient samples can be performed with high accuracy.

[0013] This figure shows an example configuration of a system within a testing facility equipped with one automated analyzer managed by a control device according to one embodiment of the present invention. This figure shows an example configuration of a system within a testing facility equipped with multiple automated analyzers managed by a control device according to one embodiment of the present invention. This figure shows an example configuration of a control device and management system according to one embodiment of the present invention. This flowchart shows the flow of the process for acquiring common reference value data corresponding to reagents installed in the measurement unit. This flowchart shows an example of the correction process for reference value data (calibration curve data) based on blank measurement values, which is performed after the process in Figure 4. This flowchart shows an example of the correction process for reference value data (calibration curve data) based on correction results, which is performed after the process in step S408 of Figure 4. This flowchart shows an example of the correction process for reference value data (judgment criteria value) based on reaction process data, which is performed after the process in step S408 of Figure 4. This flowchart shows an example of the correction process for reference value data (judgment criteria value) based on correction results, which is performed after the process in step S408 of Figure 4. This figure shows an example of reference value data stored in the storage device of the management server. This figure shows the condition setting screen. This figure shows the calibration curve factor display screen. This figure shows the judgment criteria value display screen. This flowchart shows an example of the procedure for quality control processing by the management server. This is a flowchart showing another example of the procedure for quality control processing by the management server. This is a flowchart showing yet another example of the procedure for quality control processing by the management server. This is a flowchart showing the procedure for the analysis of reaction process data. This is a diagram showing examples of individual judgment results and overall judgment results in the analysis of reaction process data. This is a diagram showing examples of judgment categories for reaction process data. This is a diagram showing the check contents performed in individual judgment. This is a diagram showing an example of the display screen for quality control results. This is a diagram showing the display screen for the results of the analysis of reaction process data. This is a diagram showing a list of data stored in the unified standard information storage unit (Figure 3). This is a diagram showing an example of the reaction process horizontal axis correspondence table stored in the unified standard information storage unit (Figure 3). This is a diagram showing reaction process data before and after unified standard conversion.

[0014] Embodiments of the present invention will be described below with reference to the drawings.

[0015] -Terminology- "Reaction process" A reaction process is the sequence of events in which a reaction occurs between a mixture of reagents and a sample.

[0016] "Reaction Process Data" Reaction process data is time-series measurement data measured by a measurement unit at predetermined time intervals (measurement points) within a measurement section that includes the reaction process. Reaction process data can be obtained, for example, as a graph plotting the measurement data in a two-dimensional coordinate system with the measurement points (photometric time) on the horizontal axis and the absorbance value on the vertical axis.

[0017] "Measurement Interval" A measurement interval is a predetermined time period from the first measurement point (start point) to the last measurement point (end point) for measuring reaction process data.

[0018] "Judgment Category" A judgment category is a category consisting of one measurement point or multiple consecutive measurement points that have been pre-selected within a measurement interval. Judgment categories can be set arbitrarily; for example, a measurement point belonging to judgment category A may also belong to judgment category B, and there may be measurement points within the measurement interval that do not belong to any judgment category.

[0019] "Analysis of reaction process data" Analysis of reaction process data involves classifying and determining combination patterns of individual judgment results for each judgment category of reaction process data. This is equivalent to checking the shape of the reaction process data, and the analysis results correlate with quality control and the status of the measurement unit.

[0020] "Judgment Criteria Value" A judgment criterion value is a reference value (e.g., a threshold value) used for a predetermined individual judgment performed for each judgment category in the analysis of reaction process data.

[0021] "Reference Value Data" Reference value data is data used to evaluate the output of the measurement unit of the automated analyzer. Specifically, it includes calibration curve data related to the creation of a calibration curve, such as the calibration curve factor, and the judgment criteria value mentioned above. The calibration curve factor is a parameter related to the definition of the calibration curve, such as the slope of the calibration curve and the vertical intercept, when the calibration curve is represented as a straight line in a two-dimensional coordinate system. Reference value data is set for each reagent type and measurement item. In this embodiment, common reference value data (common calibration curve data and common judgment criteria value) is prepared in advance and shared by multiple measurement units under the control of the control device. For each measurement unit, customized reference value data can be used by correcting the common reference value data as needed. Examples of reference value data items will be shown later in Figure 9.

[0022] "Blank Sample" A blank sample is a liquid that does not react with the reagents to be used. For example, a sample that does not contain any components that react with the reagents to be mixed, specifically water or physiological saline. Biological samples that do not contain any components of the measurement item using the reagent can also be used as blank samples. Blank measurements, which involve measuring a mixture of a blank sample and a reagent, allow for the evaluation of individual differences in the measurement output of the measurement unit of an automated analyzer. In blank measurements, it is also possible to measure only the blank sample or only the reagent.

[0023] "Standard Sample" A standard sample is a sample commonly used to calculate a calibration curve, which is a relationship between the output of a measurement unit and the concentration of its components. It is a sample whose component concentration for the measurement item is known. Calibration curves are created for each reagent according to the measurement item. In this embodiment, standard samples are not necessarily used to calibrate the calibration curve.

[0024] "Quality control sample" A quality control sample is a sample used to confirm the accuracy of the measured values ​​of a measurement item based on a calibration curve, and is a sample in which the component concentration of the measurement item is known. In this embodiment, a quality control sample is not necessarily used for quality control.

[0025] "Normal Sample" A normal sample is a biological sample (patient sample) such as blood or urine collected from a patient, which meets predetermined conditions that guarantee a certain level of quality, such as the concentration of the measured components being within a predetermined range. The predetermined conditions for a normal sample can be changed or added if the quality can be evaluated. For example, if multiple measurements are performed on a patient sample contained in the same sample container, and the difference between those measurements falls within a predetermined range, the patient sample can be determined to be a normal sample. Other predetermined conditions may include that the results of the analysis of reaction process data fall within a predetermined distribution, that the deviation in shape within a specific section is not large from the previous measurement, and that the degree of abnormality determined from the pattern in which the deviation occurs is not large. In this embodiment, quality control can be performed using normal samples.

[0026] "Calibration" In the following embodiments, calibration means calibration of reference value data and includes updating calibration curve data and updating judgment criteria values.

[0027] - Overview - 1. The management device according to this embodiment is a management device for an automated analyzer having a measurement unit installed in a facility such as a hospital or medical institution, and has a function to set reference value data used for evaluating the output of the measurement unit of the automated analyzer. The management device is, for example, a cloud server including one or more computers, and is equipped with a storage device and a computing device, and is able to communicate with the measurement unit via a network (wide-area network such as the Internet). Various data are stored in the storage device of the management device, for example, the above measurement interval and predetermined judgment categories set within it, and common reference value data for each reagent used in the measurement unit according to the measurement item (common calibration curve data, common judgment reference value), etc.

[0028] Furthermore, the management device, together with a clinical laboratory information system (e.g., an information management PC) installed at the facility where the automated analyzer is installed or at the facility of the user operating the automated analyzer, constitutes the management system. The clinical laboratory information system is connected to the measurement unit by wired or wireless communication means (e.g., LAN) and can communicate with the management device via the network. It manages measurements and other operations performed by the measurement unit, as well as various data acquired during the operation process, and exchanges various data with the management device. In this embodiment, the automated analyzer and the management device are configured to exchange information via the clinical laboratory information system. However, it is also conceivable to configure the management device with a computer connected to the measurement unit within the facility (e.g., the analyzer PC 111, information management PC 112, or client PC 145, etc., described later), such as the clinical laboratory information system. Examples of the configuration of the automated analyzer, management device, and management system will be described later with reference to Figures 1 to 3.

[0029] 2. An example of a characteristic function that can be performed by the calculation unit of the control device is a function that includes the following steps 1 to 4, which distributes the above reference value data to the measurement unit of the automatic analyzer.

[0030] (2-1) In the first step, the control device acquires common reference value data corresponding to the reagents used in the analysis of patient specimens in the measurement unit according to the measurement items from the storage device and transmits it to the measurement unit as necessary. Reagent information including the measurement items (e.g., reagent type, manufacturing lot, bottle number, etc.) is read by the automated analyzer using a barcode or the like when the reagent is placed in the measurement unit and transmitted from the automated analyzer to the control device, for example, via a clinical laboratory information system. In the first step, based on this reagent information, the corresponding common reference value data is retrieved from the data stored in the storage device. This first step is performed for reagents when at least one of the lot or bottle of the reagent used in the measurement unit is new (for example, when the lot or bottle of the reagent loaded into the automated analyzer is determined to be new from the reading information). A new lot / bottle means that it does not match any of the reagent lots / bottles currently registered in the measurement unit. For example, a lot / bottle that matches previously registered data but does not match currently registered data is considered new. Whether a reagent lot / bottle is new is determined, for example, when the management device receives information read from the reagent installed in the measurement unit, or subsequently in the order in which the information is received. An example of the first step will be described later with reference to Figure 4.

[0031] In the second step, the control device acquires measurement data measured by the target measurement unit (i.e., the measurement unit using the target reagent) using the target reagent, which was determined in the first step to be new in lot / bottle, or the corresponding reagent with the same reagent type and measurement item. In the third step, the control device corrects the common reference value data acquired in the first step based on the measurement data acquired in the second step. In the fourth step, the control device transmits the reference value data obtained by correcting the common reference value data in the third step to the measurement unit, for example, via the corresponding clinical laboratory information system, as reference value data to be applied to the measurement of patient samples in the target measurement unit. Examples of the second to fourth steps will be described later with reference to Figures 5 to 8.

[0032] With the above distribution process, reference value data is automatically set for each measurement unit when the reagent lot or bottle is changed, eliminating the need to measure standard samples and create a new calibration curve each time the reagent lot or bottle is changed. This reduces the burden on the user of the automated analyzer, such as the preparation of standard samples and the reacquisition of reference data, which are associated with updating reference value data. Furthermore, the time required to update reference value data is shortened, and the cost of preparing standard samples is also reduced. Thus, it is possible to reduce the burden on the user and lower the cost of calibration.

[0033] (2-2) The common reference value data obtained in the first step above includes common calibration curve data (common calibration curve factor, etc.) relating to the calibration curve corresponding to the target reagent for which the lot or bottle has been determined to be new, and common judgment reference values ​​used for individual determination of reaction process data measured during the reaction process between the target reagent and the patient sample.

[0034] When the purpose is to update calibration curve data from reference value data, the measurement data acquired in the second step above is, for example, a blank measurement value obtained by measuring at least one of the target reagent and a blank sample with the target measurement unit. In this case, the control device uses its calculation unit to calculate the difference between the 0 concentration value (concentration of the measurement item is 0) of the common calibration curve data corresponding to the target reagent and the blank measurement value. If this difference is less than or equal to a preset threshold, the control device corrects the common calibration curve data corresponding to the target reagent based on the blank measurement value to adjust the calibration curve data used in the target measurement unit. Furthermore, if a blank measurement value using the target reagent cannot be obtained from the target measurement unit, the control device uses its calculation unit to determine, for example, whether the measurement item to be measured using the target reagent is new, based on the above reading information. If the measurement item is not new, the control device searches the data stored in the storage device for correction records of the common calibration curve data corresponding to the reagent type and measurement item of the target reagent for the target measurement unit, and corrects the common calibration curve data based on the retrieved correction records to adjust the calibration curve data used in the target measurement unit. However, if a blank measurement is performed on the target measurement unit after the calibration curve data has been corrected based on the correction results, the calibration curve data may be corrected based on the blank measurement values.

[0035] Thus, when updating calibration curve data as reference value data, blank measurements that have a large correlation with individual differences in measurement units can be obtained, and based on these blank measurements, pre-prepared common calibration curve data can be corrected to adjust unique calibration curve data for each measurement unit. In this way, patient samples can be measured using each measurement unit with their own specially customized unique calibration curve data. Furthermore, by correcting the common calibration curve data using past correction results for each measurement unit to adjust the calibration curve data, patient samples can be measured using customized calibration curve data for each measurement unit even when blank measurements are not available.

[0036] (2-3) On the other hand, when the purpose is to correct the judgment criteria used for individual judgment of reaction process data among the reference value data, the measurement data acquired in the second step above is, for example, reaction process data measured using the target reagent in the target measurement unit by applying the common reference value data acquired in the first step. In this case, the control device uses a calculation unit to calculate the distribution of parameters of the approximation formula created for each of a predetermined number of reaction process data, and calculates a judgment criteria value specific to the target measurement unit based on this parameter distribution. Furthermore, the control device calculates the difference between the common judgment criteria value stored in the memory device corresponding to the target reagent and the judgment criteria value specific to the target measurement unit, and if the difference is less than or equal to a preset threshold, it corrects the judgment criteria value related to the target reagent for the target measurement unit based on the judgment criteria value specific to that measurement unit. Furthermore, if a predetermined number of reaction process data using the target reagent has not yet been obtained from the target measurement unit, the control device, using its calculation unit, determines, for example, whether the measurement item measured using the target reagent is new, based on the above-mentioned reading information. If the measurement item is not new, the control device searches the data stored in the storage device for correction records of the judgment criterion value corresponding to the target reagent for the target measurement unit, and adjusts the judgment criterion value used in the target measurement unit by correcting the common judgment criterion value based on the retrieved correction records. However, after correcting the judgment criterion value based on the correction records, if a predetermined number of reaction process data has been obtained from the target measurement unit, a correction of the judgment criterion value based on the obtained reaction process data may be performed.

[0037] In this way, when updating the judgment criteria as reference value data, by actually measuring a predetermined number of reaction process data points with the measurement unit, a judgment criterion value that reflects the individual differences of the measurement unit is calculated based on these measured reaction process data points, and its validity is confirmed by comparing it with the common judgment criterion value, thereby adjusting a unique judgment criterion value for each measurement unit. Using this specially customized unique judgment criterion value, it is possible to perform analysis of the reaction process data of patient samples obtained from each measurement unit, for example, quality control. Furthermore, by correcting the common judgment criterion value using past correction results for each measurement unit and adjusting the judgment criterion value, it is possible to perform analysis of the reaction process data of patient samples using a judgment criterion value customized for each measurement unit, even when a predetermined number of reaction process data points have not been obtained, and to perform quality control, etc.

[0038] (2-3) It is also possible to configure whether steps 2 through 4 are executed automatically. An example of the configuration screen will be shown later in Figure 10.

[0039] By disabling automatic updates of reference value data, the automatic updating of reference value data applied to the measurement unit for new reagent lots or bottles is prohibited, and the user can use the current reference value data or reference value data they have created themselves.

[0040] 3. Another example of a characteristic function that can be performed by the control device's computing unit is a precision control function that uses the analysis of reaction process data, which includes the following first and second steps. This precision control function is used for precision control of measurements performed by the measurement unit to which the reference value data distributed from the control device has been delivered, as well as for determining the state of the measurement unit and detecting signs of component failure.

[0041] (3-1) In the first step, the control device uses a calculation unit to select samples to be used for quality control. The samples selected in this first step are, for example, multiple patient samples (a predetermined number of populations). In this case, the calculation unit selects a patient sample as a sample to be used for quality control (a normal sample) based on the measurement data of the patient sample, if the measurement data satisfies predetermined conditions, and a predetermined number of normal samples that satisfy the predetermined conditions are selected in the first step. As reaction process data for the predetermined number of normal samples, for example, the reaction process data of the first predetermined number of normal samples obtained after starting up the automated analyzer and beginning the measurement operation of the patient sample can be used. The calculation unit performs individual judgments on these predetermined number of normal samples and then outputs the quality control results. However, it is also possible to measure quality control samples and perform quality control based on their reaction process data. Since the quality of quality control samples is guaranteed, it is not always necessary to prepare multiple quality control samples when using quality control samples.

[0042] In the second step, the control device uses a calculation unit to perform quality control of the measurement unit and other components based on the reagent type corresponding to the calibration curve data registered for the measurement items of the sample selected in the first step, and the reaction process data measured by the measurement unit in the measurement interval. In this second step, the calculation unit performs individual judgments for each of a predetermined set number of judgment categories within the measurement interval using judgment criteria values ​​based on the behavior patterns of the measurement data. The calculation unit then determines the state of at least one of the measurement unit, reagents, and sample based on the combination patterns of the results of the individual judgments for each judgment category, and outputs the quality control result of the measurement unit based on the individual judgment results. Individual judgment is a judgment using judgment criteria values ​​of the waveform of the measurement data for the target judgment category. Examples of quality control based on reaction process data are explained in Figures 13A to 13C and Figures 14 to 19.

[0043] According to the above accuracy management function, the sample used for accuracy management is a selected sample. For example, even when using patient specimens, a plurality of patient specimens that meet the predetermined conditions for ensuring quality are selected as the samples used for accuracy management. Therefore, the reliability of the data that serves as the basis for the accuracy management determination can be ensured. Also, instead of simply comparing measurement data (e.g., the concentration of a measurement item) with a reference value, reaction process data is utilized. Using each determination category of the reaction process data as a checkpoint, for example, a predetermined individual determination is made regarding the characteristics of the waveform of the reaction process data in each determination category, and based on the combined pattern of the results, accuracy management using patient specimens can be accurately implemented. In addition, through the individual determination for each determination category, the cause of an abnormality that affects the measurement accuracy (e.g., suspicion of the accuracy of the reagent dispensing volume, etc.) can be estimated, and it is also possible to present to the user a method for dealing with the abnormality or to detect a sign of component failure.

[0044] (3-2) Further, in the storage device of the management device, correlation data that associates the normal and abnormal states with the above combined pattern for at least one of the measurement unit, reagent, and sample is stored. The management device, by means of the arithmetic device, in the above second step, determines the state of at least one of the measurement unit, reagent, and sample from the combined pattern of the individual determinations for each determination category based on the correlation data pre-stored in the storage device. This correlation data includes the degree of abnormality. An example of the determination result will be given in FIG. 15 later. Also, one of the plurality of determination categories is the timing after the reagent and the sample are mixed and stirred. Examples of the determination categories will be given in FIG. 16 later.

[0045] By setting in advance correlation data between the combined pattern of the individual determinations for each determination category and the state of the measurement unit, etc., the management device can automatically present to the user a method for dealing with an abnormality or detect a sign of component failure from the combined pattern of the determination results of the above individual determinations.

[0046] 4. Another example of a characteristic function executable by the arithmetic unit of the management device is a unified standardization function that converts the outputs of measurement units with different units into data with a common unit, enabling accuracy management to be carried out in the same way even if the manufacturing companies and models of the measurement units are different.

[0047] For example, in the above management system, when the management device or the clinical examination information system (information management PC) receives reaction process data of a sample (such as a patient specimen) from the measurement unit, it performs processing of the reaction process data based on unified standard information unified in a plurality of pre-registered facilities (such as examination rooms). When the processing is performed by the clinical examination information system, the reaction process data after processing is transmitted from the clinical examination information system to the management device via the network.

[0048] For example, if reaction process data γ is obtained from a measurement unit managed by a control device, the units of the vertical and horizontal axes of the graph representing this reaction process data γ are converted based on unified standard information. Reaction process data uses a time element on the horizontal axis and a value corresponding to absorbance, etc., on the vertical axis. However, the time interval of measurement in the reaction process may differ depending on the manufacturer of the automated analyzer, and even between different models from the same manufacturer. For example, the measurement interval for model α may be t1 seconds, while the measurement interval for model β may be t2 seconds (t1 ≠ t2). The unit of the horizontal axis is often expressed in measurement points, which indicate which measurement it is from the start of measurement. Depending on the model, the time interval between measurement points may not be constant. On the other hand, the vertical axis generally represents the magnitude of measurement data such as absorbance converted from the output of the measurement unit based on a calibration curve, but the units of the vertical axis may also differ in terms of magnification, etc., depending on the manufacturer and model. The unified standard information described above includes data such as formulas or tables, which contain parameters for aligning the units of the horizontal and / or vertical axes of a graph for measurement data from multiple measurement units registered under the management of the control device. For example, it is possible to convert the horizontal axis to a universal time axis such as seconds or minutes, and to unify the magnification of the measured values ​​on the vertical axis. Alternatively, one can be selected from multiple registered measurement units, and the unified standard information can be set so that the units of the vertical and horizontal axes of the measurement data of this selected measurement unit are aligned with the units of the measurement data of all registered measurement units. Examples of the unified standardization function will be described later using Figures 20 to 22.

[0049] By converting the units on the horizontal and vertical axes of the reaction process data from each measurement unit to a common, unified standard, reaction process data from measurement units from different manufacturers or models can be handled in the same way, enabling reliable judgments and analysis processes for a variety of measurement units.

[0050] -Example of System Configuration within a Testing Facility 1- Figure 1 shows an example of the configuration of a system within a testing facility equipped with one automated analyzer managed by a management device according to one embodiment of the present invention. The system within the testing facility 100 in Figure 1 comprises one automated analyzer 1, an information management PC 112, and a communication device 130.

[0051] The automated analyzer 1 comprises an analysis operation unit 101, an interface 110, and an analyzer PC 111. The analyzer PC 111 is, for example, a desktop computer and has a control unit that controls the analysis operations of the analysis operation unit 101. This analyzer PC 111 is connected to the analysis operation unit 101 via the interface 110, such as a hub. The analyzer PC 111 also has a monitor device 115 as a display unit that displays the analysis results from the analysis operation unit 101.

[0052] The analysis unit 101 includes various mechanisms for performing the analysis of the sample, such as a sample input unit 104, a sample transport unit 105, measurement units 106a to 106c, and a sample storage unit 109. In the example shown in Figure 1, a configuration in which a single automatic analyzer 1 is equipped with three measurement units 106a to 106c is illustrated, but the number of measurement units that the automatic analyzer 1 has is not limited to three; it can be one or more. In the following description, when measurement units 106a to 106c are not distinguished, the subscript may be omitted and they may be referred to as measurement unit 106.

[0053] Samples such as patient samples, blank samples, standard samples, or quality control samples are dispensed into the sample container 102 (sample cup) depending on the purpose of measurement. In this embodiment, it is also possible to update the calibration curve using patient samples without dispensing standard samples, or to perform quality control without using quality control samples (described later). One or more sample containers 102 are loaded onto the sample rack 103 and placed into the sample input unit 104. When the analysis is requested to start from the analyzer PC 111, the sample transport unit 105 transports the sample rack 103 toward the sample storage unit 109. Measurement units 106a to 106c are connected to the sample transport unit 105, and samples are supplied to one of the measurement units 106a to 106c during the transport process, and the samples are analyzed.

[0054] The measurement unit 106a includes equipment such as a reaction disk 120a, a sample dispensing mechanism 121a, a reagent disk 108a, a reagent dispensing mechanism 122a, and a photometric unit 123a. The reagent disk 108a houses a reagent container 107 containing a reagent that reacts with the sample. Measurement units 106b and 106c have the same configuration as measurement unit 106a and include reaction disks 120b and 120c, sample dispensing mechanisms 121b and 121c, reagent disks 108b and 108c, reagent dispensing mechanisms 122b and 122c, and photometric units 123b and 123c. In the following explanation, when the measurement units 106a to 106c are not distinguished, the subscripts for reaction disks 120a to 120c, sample dispensing mechanisms 121a to 121c, reagent disks 108a to 108c, reagent dispensing mechanisms 122a to 122c, and photometric units 123a to 123c may be omitted.

[0055] The sample, contained in the sample container 102 and transported by the sample transport unit 105, is dispensed along its transport path, for example, in the measurement unit 106a, by the sample dispensing mechanism 121a into the reaction vessel on the reaction disk 120a. The reagent, sealed in the reagent container 107 on the reagent disk 108a, is then dispensed into the reaction vessel containing the sample by the reagent dispensing mechanism 122a. This causes the sample and reagent to react in the reaction vessel. The absorbance and scattered light intensity of the reaction solution in the reaction vessel are measured by the photometric unit 123a. The absorbance and other values ​​measured by the photometric unit 123a are converted into the concentration of the specific component in the sample based on the relative relationship between the concentration of the specific component in the standard sample and its measured value, for example, based on a calibration curve.

[0056] The information management PC 112 is, for example, a Laboratory Information System (LIS), and is connected to the analysis device PC 111 via a communication device 130 within a facility such as a hospital. The LIS responds with test result information to measurement request information sent from the hospital information system (HIS) of the medical department.

[0057] -Example of System Configuration within a Testing Facility 2- Figure 2 shows an example of the configuration of a system within a testing facility equipped with multiple automated analyzers managed by a management device according to one embodiment of the present invention. The system within the testing facility 100 in Figure 2 comprises multiple automated analyzers 1a to 1d (four in this example), an information management PC 112, and a communication device 130. In the following description, when the automated analyzers 1a to 1d are not distinguished, subscripts may be omitted when referring to the automated analyzers 1a to 1d and their components.

[0058] Multiple automated analyzers 1a to 1d have the same configuration as automated analyzer 1 in Figure 1. That is, each automated analyzer 1a to 1d is equipped with an analysis operation unit 101a to 101d, an interface 110a to 110d, and an analyzer PC 111a to 111d. For example, analyzer PC 111a is connected to the analysis operation unit 101a via interface 110a, controls the analysis operation unit 101a, and has a monitor device 115a as a display unit. Analyst PCs 111b to 111d are similar to analyzer PC 111a, and are connected to the analysis operation units 101a to 101d via interfaces 110b to 110d, control the analysis operation units 101a to 101d, and have monitor devices 115a to 115d as display units.

[0059] The information management PC 112 is connected to the analysis device PCs 111a to 111d via the communication device 130 and collects information from each of the automated analysis devices 1a to 1d. Note that the number of automated analysis devices does not need to be more than two, but is not limited to four.

[0060] - Management Device and Management System - Figure 3 shows an example of the configuration of a management device and management system according to one embodiment of the present invention. The management system comprises in-facility systems 100a to 100c, each installed in at least one (three in this example) inspection facility A to C, and a management server (management device) 140. In the following description, when inspection facilities A to C are not distinguished, the in-facility systems 100a to 100c and their components may be written without subscripts.

[0061] The information management PCs 112a to 112c of the in-facility systems 100a to 100c are connected to the communication devices 116a to 116c within the in-facility and transmit information from each automated analyzer 1 (not shown in Figure 3) to the management server 140 via a network 146 such as the Internet. The number of in-facility systems 100 connected to the management server 140 is at least one, and not limited to three.

[0062] Each of the in-facility systems 100a to 100c is equipped with a client PC 145a to 145c. Client PCs 145a to 145c communicate with the management server 140 via communication devices 116a to 116c. Although Figure 3 illustrates a configuration in which client PCs 145a to 145c are installed in in-facility facilities A to C, client PC 145 does not necessarily have to be installed in an in-facility. For example, client PC 145x can be installed in the customer support center of the vendor providing the in-facility systems 100a to 100c. This allows for the remote acquisition of information from the in-facility systems 100a to 100c, facilitating smooth troubleshooting and operational support, including remote guidance, for users such as medical technicians in in-facility facilities A to C.

[0063] The management server 140 is, for example, a cloud server, and is constructed from a group of computers, and includes a storage device 140A and an arithmetic unit 140B. The storage device 140A may consist of storage devices built into multiple computers or storage devices connected directly to computers or via a network. The arithmetic unit 140B may also consist of arithmetic units from multiple computers. However, the management server 140 is not limited to a cloud server and may consist of a single computer. The storage device 140A of the management server 140 includes an information storage unit 142 and a Web application information storage unit 143. The arithmetic unit 140B of the management server 140 includes an information analysis unit 141 and a Web server control unit 144. These information analysis unit 141 and Web server control unit 144 can be implemented as software elements or as hardware elements such as circuits.

[0064] The information analysis unit 141 includes a device data analysis unit 1411 which includes a unified standard conversion unit 14111. It receives information about measurement units from the inspection facility systems 100a to 100c, converts the received information into a predetermined format using the unified standard conversion unit 14111, and stores it in the device data storage unit 1421 of the information storage unit 142. The unified standard conversion unit 14111 appropriately references conversion parameters from the unified standard information storage unit 1431 in the Web application information storage unit 143 and performs the conversion process. However, conversion of the information format is not mandatory. If the information received by the information analysis unit 141 is in file format, for example, it is possible to configure it to store the file directly in the device data storage unit 1421 of the information storage unit 142 without conversion.

[0065] The device data storage unit 1421 stores data about the device itself, such as the IDs and specifications of each automatic analyzer 1, measurement unit 106 and its components, as well as measurement data acquired by each measurement unit, and identification information of samples and reagents read by barcode or RFID (Radio Frequency Identification) readers in each measurement unit 106.

[0066] When a request for information acquisition, display, etc., is received from client PCs 145a-145c, 145x, the Web server control unit 144 analyzes the request using the Web application information management unit 1443, collects the necessary information from the information storage unit 142 and the Web application information storage unit 143, converts it to the display format of the Web application, and then transmits it to the client PCs 145a-145c, 145x. When a request for information registration is received from client PCs 145a-145c, 145x, the Web application information management unit 1443 analyzes the registration content and saves it to the Web application information storage unit 143. Furthermore, the Web server control unit 144 performs reaction process data analysis using judgment criteria values ​​with the reaction process data analysis unit 1441. In addition, the Web server control unit 144 performs reference value data distribution processing processing with the reference value data distribution processing unit 1442 to each measurement unit 106 or each automatic analyzer 1.

[0067] Examples of information registered in the Web application information storage unit 143 include unified standard information, common judgment criteria values, and common calibration curve data. Unified standard information is stored in the unified standard information storage unit 1431 of the Web application information storage unit 143. Judgment criteria values ​​are stored in the judgment criteria value storage unit 1432 of the Web application information storage unit 143. Common calibration curve data is stored in the calibration curve data storage unit 1433 of the Web application information storage unit 143.

[0068] Client PCs 145a-145c and 145x are, for example, desktop computers, smartphones, tablet devices, and notebook PCs, and communicate with the management server 140 via a network. Based on the information transmitted from the management server 140, client PCs 145a-145c and 145x can manage information in the in-facility systems 100a-100c and centrally access the status information of each automated analyzer 1a-1c within each in-facility system 100a-100c by executing a web application. The form and number of client PCs 145a-145c and 145x are not limited as long as they are capable of such communication.

[0069] - Acquisition process of common reference value data - Figure 4 is a flowchart showing the flow of the acquisition process of common reference value data corresponding to the reagents installed in the measurement unit. The process in Figure 4 is performed for calibration curve data and judgment reference values, respectively. In each automated analyzer 1, the reagent information of the target reagent installed in the measurement unit 106 is read, for example, by a reader provided in the automated analyzer 1 and transmitted to the management server 140 via the information management PC 112. In step S401, the management server 140 receives the reagent information of the target reagent installed in the measurement unit 106 from the information management PC 112 and stores and registers it in the device data storage unit 1421 by the device data analysis unit 1411. Here, the installation of reagents means that reagents are loaded onto the reagent disk 108 of the measurement unit 106 of the automated analyzer 1, and this includes, for example, a user manually installing reagents onto the reagent disk 108, or reagents being automatically installed onto the reagent disk 108 by an automated reagent transport mechanism (reagent autoloader), etc. As described above, barcodes or RFID can be applied to read the reagent information. Reagent information includes item information, which specifies the measurement items using the reagent; lot information, which is information about the manufacturing lot of the reagent (e.g., manufacturing lot number); and identification information of the bottle containing the reagent (e.g., bottle number; the bottle number is also called the reagent sequence number).

[0070] In step S402, the management server 140 uses the instrument data analysis unit 1411 to determine whether the registered target reagent is from a new lot. Specifically, if the "lot information" in the reagent information is new, the target reagent is determined to be from a new lot. If the lot information registered in step S401 does not match any of the lot information of reagents currently installed in the automatic analyzer 1, the target reagent is determined to be from a new lot, even if it matches previously stored lot information.

[0071] If the condition in step S402 is met, the management server 140 determines in step S403 whether to update the reference value data using common reference value data for the new lot of the target reagent. For example, if the timing for updating the reference value data corresponding to the target reagent is set to the timing when a new lot of reagent is released for the target measurement unit 106 where the target reagent is installed, the condition in step S403 is met. On the other hand, if the timing for updating the reference value data is set to the timing when a new bottle is released, or if the function to update to common reference value data is disabled, the condition in step S403 is not met. By disabling the function to update to common reference value data, it is possible to prevent the reference value data from being automatically updated by sending it from the management server 140 to the measurement unit 106 when the reagent lot / bottle becomes new, allowing the user to actually look at the measurement data to determine whether or not to update the reference value data and to apply the reference value data created by the user. If the condition in step S403 is met, the management server 140 proceeds to step S406.

[0072] If the determination in step S402 or step S403 is not met, the management server 140 proceeds to step S404, where the device data analysis unit 1411 determines whether the registered reagent is in a new bottle. Specifically, if the "bottle information" in the reagent information is new, the target reagent is determined to be in a new bottle. The determination of whether the bottle is new is performed in the same manner as the determination of whether the lot is new.

[0073] If the condition in step S404 is met, the management server 140 determines in step S405 whether to update the reference value data for the new bottle of target reagent using the common reference value data. In step S405, similar to the process in step S403, if, for example, the timing for updating the reference value data is set to the timing when the reagent bottle becomes new, the condition in step S405 is met. On the other hand, if the timing for updating the reference value data is set to the timing when the lot becomes new, or if the function to update to common reference value data is disabled, the condition in step S405 is not met. By disabling the function to update to common reference value data, it is possible to apply the reference value data created by the user as described above. If the condition in step S404 or step S405 is not met, the management server 140 terminates the process in Figure 4 without searching for the common reference value data related to the target reagent from the calibration curve data storage unit 1433 or the judgment reference value storage unit 1432 for the measurement unit 106.

[0074] If the determinations in steps S402 and S403, or steps S404 and S405 are met, the management server 140 proceeds to step S406, where the reference value data distribution processing unit 1442 starts the process of acquiring common reference value data for the target reagent for which the lot or bottle has been determined to be new. Specifically, in step S406, the management server 140 searches for common calibration curve data and common judgment criteria values ​​corresponding to the item information of the target reagent for which common reference value data is to be acquired, and the lot information or bottle information, from the information registered in the calibration curve data storage unit 1433 and the judgment criteria value storage unit 1432, respectively.

[0075] Next, in step S407, the management server 140 determines whether common reference value data corresponding to the target reagent exists based on the results of the search. If the determination in step S407 is not met, the management server 140 terminates the process shown in Figure 4 without acquiring common reference value data for the target reagent.

[0076] If the judgment in step S407 is met, in step S408, the management server 140 obtains common reference value data corresponding to the item information of the target reagent and lot information or bottle information from the calibration curve data storage unit 1433 and the judgment reference value storage unit 1432, and terminates the process shown in Figure 4.

[0077] Furthermore, in step S408, the acquired common reference value data may be automatically distributed to the automatic analyzer 1, including the target measurement unit 106, via the reference value data distribution processing unit 1442 and the corresponding information management PC 112. Alternatively, the information management PC 112 may automatically access the management server 140 and download the common reference value data acquired in step S408 to the corresponding information management PC 112.

[0078] -Example 1 of calibration curve data correction process- Figure 5 is a flowchart showing an example of a correction process for reference value data (calibration curve data) based on blank measurement values, which is performed after the process in step S408 of Figure 4. This flowchart is executed when a blank measurement using the target reagent from which common reference value data was obtained in the process of Figure 4 is performed in the target measurement unit 106 where the target reagent is installed, and the management server 140 receives the measurement data (blank measurement values). The blank measurement values ​​may be automatically sent from the information management PC 112 to the management server 140, or the management server 140 may request the information management PC 112 to send them.

[0079] In step S501, the management server 140 first acquires the value output from the photometric unit 123a, i.e., the blank measurement value, from the corresponding information management PC 112 via the network 146 at the target measurement unit 106 where the target reagent is installed, using the target reagent.

[0080] In the subsequent step S502, the management server 140 calculates unique feature quantities for the calibration curve related to the target reagent for the target measurement unit 106 based on the blank measurement values ​​obtained in step S501. These feature quantities include, for example, the blank measurement values ​​obtained in step S501, as well as the calibration curve factor included in the calibration curve data. The feature quantities are stored in the device data storage unit 1421.

[0081] In step S503, the management server 140 calculates the difference between the feature quantities calculated in step S502 and the feature quantities of the common calibration curve data acquired in the processing shown in Figure 4 (for example, the difference between the value at concentration 0 of the measurement item in the common calibration curve data and the blank measurement value), and records it in the calibration curve data storage unit 1433. At this time, the ratio of the feature quantities may be calculated as a value equivalent to the difference between the feature quantities. The difference in feature quantities calculated in step S503 is expected to differ for each measurement unit and will be a value unique to the target measurement unit 106 that reflects the fixed difference.

[0082] In step S504, the management server 140 determines whether the magnitude (absolute value) of the difference in feature quantities of the calibration curve calculated in step S503 is below a preset threshold. If the difference in feature quantities exceeds the threshold, the blank measurement may not be normal. On the other hand, if the difference in feature quantities is below the threshold, it is estimated that the value is valid as unique information of the target measurement unit 106 that reflects the fixed difference.

[0083] If the difference in features is below a threshold, the management server 140 determines in step S505 whether to update the calibration curve data used for the target reagent in the target measurement unit 106. The determination in step S505 is satisfied, for example, if the automatic update function for the calibration curve factor using blank measurements is enabled, and not satisfied if the automatic update function is disabled. By disabling the automatic update function for the calibration curve factor, the calibration curve used for the target reagent in the measurement unit 106 is not automatically updated, and the user can use the current calibration curve or a calibration curve they have created themselves.

[0084] If the conditions in steps S504 and S505 are met, in step S506, the management server 140 corrects the common calibration curve data obtained in the process of Figure 4 by applying the feature quantities calculated in step S502, and transmits the resulting calibration curve data to the corresponding information management PC 112. The management server then updates the calibration curve data used for the target reagent in the target measurement unit 106 and terminates the process of Figure 5. Information indicating that the calibration curve data has been corrected is stored in the calibration curve data storage unit 1433, linked to the applied measurement unit 106, reagent type, measurement item, and the difference in the feature quantities calculated in step S503. The calibration curve data transmitted to the information management PC 112 is stored, for example, in the storage unit of the analyzer PC 111 of the automatic analyzer 1 including the target measurement unit 106, and is used for subsequent measurements using the target reagent in the target measurement unit 106. If the conditions in step S504 or S505 are not met, the management server 140 terminates the process of Figure 5 without performing the calibration curve data update process.

[0085] -Example 2 of calibration curve data correction process- When blank measurements have not been performed using the target reagent in the target measurement unit 106, the calibration curve cannot be customized using the process shown in Figure 5. In such cases, it is also possible to perform calibration curve correction processing based on actual values ​​without performing blank measurements, as in the example below.

[0086] Figure 6 is a flowchart showing an example of the correction process for reference value data (calibration curve data) based on correction results, which is performed after the process in step S408 of Figure 4. In step S601, the management server 140 acquires common reference value data corresponding to the target reagent for which the lot or bottle is new. The process in step S601 is the same as the process in step S408 of Figure 4.

[0087] In step S602, the management server 140 determines whether the measurement item to be measured using the target reagent is a new item based on the data stored in the instrument data storage unit 1421. The determination in step S602 is satisfied if there is data in which the target reagent, reagent type, and measurement item match; otherwise, it is not satisfied. If the determination in step S602 is satisfied, the management server 140 completes the process shown in Figure 6 without performing calibration curve data correction processing.

[0088] If the condition in step S602 is not met, the management server 140 proceeds to step S603 and searches the data stored in the calibration curve data storage unit 1433 for the previous value of the difference in feature quantities calculated in step S503. The previous value calculated in step S603 is the most recent data among the values ​​calculated in step S503 of the processing shown in Figure 5 that is common to the target reagent for the measurement unit 106, reagent type, and measurement item.

[0089] In step S604, the management server 140 determines whether a corresponding previous value exists as a result of the search. If the determination in step S604 is not met, the management server 140 terminates the process shown in Figure 6 without performing the calibration curve correction process.

[0090] If the condition in step S604 is met, the management server 140 determines in step S605 whether to perform calibration curve data correction using the previous value from step S503. The condition in step S605 is met, for example, if the automatic update function for calibration curve data using the previous value is enabled, and not met if the automatic update function is disabled. By disabling the automatic update function, the calibration curve used for the target reagent in the measurement unit 106 is not automatically updated, and the user can use the current calibration curve or a calibration curve created by the user. If the condition in step S605 is not met, the management server 140 completes the process in Figure 6 without performing calibration curve data correction processing.

[0091] If the conditions in step S605 are met, the management server 140 proceeds to step S606 and transmits the calibration curve data obtained by correcting the common calibration curve data acquired in the process of Figure 4 by applying the previous value retrieved in step S503 to the corresponding information management PC 112, updating the calibration curve data to be used for the target reagent in the target measurement unit 106 and ending the process of Figure 6. The calibration curve data transmitted to the information management PC 112 is stored, for example, in the memory of the analyzer PC 111 of the automatic analyzer 1 including the target measurement unit 106, and is used for subsequent measurements using the target reagent in the measurement unit 106. If the determination in step S606 is not met, the management server 140 ends the process of Figure 6 without performing the calibration curve data update process.

[0092] Furthermore, if a blank measurement (step S501) is performed using the target reagent after the process shown in Figure 6, the calibration curve data correction process shown in Figure 5 may be performed for the target reagent even after the calibration curve correction process shown in Figure 6 has been executed.

[0093] -Example 1 of Correction Processing of Judgment Criteria- Figure 7 is a flowchart showing an example of correction processing of reference value data (judgment criteria value) based on reaction process data, which is performed after the processing in step S408 of Figure 4. When accuracy control is performed in each measurement unit 106 by analyzing the reaction process data described later, the results are transmitted from each automatic analyzer 1 to the management server 140 via the information management PC 112. In step S701, the management server 140 determines whether the number of reaction process data measured by applying the common reference value data acquired in the processing of Figure 4 to the target measurement unit 106 that uses the target reagent for which the lot or bottle was determined to be new in the processing of Figure 4 is equal to or greater than a preset threshold.

[0094] If the determination in step S701 is met, the management server 140 calculates the distribution of feature quantities calculated in the analysis of each reaction process data in step S702. Features calculated in step S702 include, for example, measurement data, as well as parameters (coefficient terms) of a specific approximation formula (polynomial) used to represent the reaction process data. For example, when the reaction process data is analyzed using the least squares method, ABS = A0 + A1(1-e -kt When expressed by the approximation formula, A0, A1, k, and Err are the parameters of the approximation formula. A0 is the absorbance at the start of the reaction, A1 is the final reaction absorbance, k is the reaction rate constant, and Err is the sum of the differences between the approximate value and the actual value at each measurement point of the reaction process data. These serve as indicators for evaluating the reaction status of the reaction process. This method is described in detail, for example, in Japanese Patent Publication No. 2009-204448.

[0095] In step S703, the management server 140 calculates a criterion value from the distribution of features confirmed in step S702. As an example, the criterion value can be calculated by creating a distribution diagram of the parameters of the approximation formula for each reaction process data, applying a regression function to obtain a curve (called a regression line) that approximates the distribution of parameters, calculating the residuals from the regression line for all the data constituting the distribution diagram, and then calculating the standard deviation SD of these residual values. For example, the criterion value can be the standard deviation SD × n, where n is, for example, the smallest positive number greater than or equal to |Emax| / SD. Emax is the maximum value of the residuals. However, this is just one example of how to calculate the criterion value from the reaction process data, and the calculation method can be changed.

[0096] In step S704, the management server 140 calculates the difference (absolute value) between the judgment criterion value calculated from the reaction process data measured by the target measurement unit 106 in step S703 and the common judgment criterion value obtained in the process shown in Figure 4. At this time, the ratio of the judgment criterion values ​​may be calculated as a value equivalent to the difference between the judgment criterion values. The difference in judgment criterion values ​​calculated in step S704 is expected to differ for each measurement unit and will be a value unique to the target measurement unit 106 that reflects a fixed difference.

[0097] In step S705, the management server 140 determines whether the difference in the judgment criteria values ​​calculated in step S704 is within a pre-set threshold range.

[0098] If the difference in the judgment criteria values ​​is below the threshold, the management server 140 proceeds to step S706 to determine whether to update the judgment criteria values ​​for the target reagent for the target measurement unit 106. The determination in step S706 is satisfied, for example, if the automatic update function for judgment criteria values ​​using reaction process data is enabled, and not satisfied if the automatic update function is disabled. By disabling the automatic update function for judgment criteria values, the automatic updating of the judgment criteria values ​​used for individual judgment for the target reagent in the target measurement unit 106 is prohibited, and the user can use the current judgment criteria values ​​or judgment criteria values ​​that they have set themselves.

[0099] If the conditions in steps S705 and S706 are met, the management server 140 proceeds to step S707, transmits the judgment criterion value calculated in step S703 to the target measurement unit 106, and then terminates the process in Figure 7 by applying it to the analysis of the reaction process data measured by the measurement unit 106 using the target reagent. If the conditions in steps S701, S705, or S706 are not met, the management server 140 terminates the process in Figure 7 without performing the judgment criterion value update process.

[0100] -Example 2 of Correction Processing of Judgment Criteria- When the number of measurements of reaction process data using the target reagent in the target measurement unit 106 falls below the threshold (the stage where the judgment in step S701 is not met), the process in Figure 7 cannot be executed and the judgment criteria cannot be customized. In such cases, it is possible to perform the correction processing of the judgment criteria based on the actual values ​​without waiting for the number of measurements of reaction process data to reach the threshold, as in the example below.

[0101] Figure 8 is a flowchart showing an example of the correction process for reference value data (judgment criteria value) based on correction results, which is performed after the process in step S408 of Figure 4. In step S801, the management server 140 acquires common reference value data corresponding to the target reagent for which the lot or bottle is new. The process in step S801 is the same as the process in step S408 of Figure 4.

[0102] In step S802, the management server 140 determines whether the measurement item to be measured using the target reagent is a new item, based on the data stored in the device data storage unit 1421. The determination in step S802 is satisfied if there is data in which the target reagent, reagent type, and measurement item match; otherwise, it is not satisfied. If the determination in step S802 is satisfied, the management server 140 completes the process shown in Figure 8 without performing any correction processing of the judgment criteria value.

[0103] If the judgment in step S802 is not met, the management server 140 proceeds to step S803 and searches the data stored in the judgment criterion value storage unit 1432 for the previous value of the difference of the judgment criterion values ​​calculated in step S704. The previous value searched in step S803 is the most recent data among the values ​​calculated in step S704 of the processing shown in Figure 7 that is common to the target reagent for the measurement unit 106, reagent type, and measurement item.

[0104] In step S804, the management server 140 determines whether a corresponding previous value exists as a result of the search. If the determination in step S804 is not met, the management server 140 ends the process shown in Figure 8 without performing any correction processing of the determination criterion value.

[0105] If the condition in step S804 is met, the management server 140 determines in step S805 whether to perform a correction of the judgment criterion value using the previous value from step S704. The condition in step S805 is met, for example, if the automatic update function for the judgment criterion value using the previous value is enabled, and not met if the automatic update function is disabled. By disabling the automatic update function, the automatic update of the judgment criterion value used for the target reagent in the measurement unit 106 is prohibited, and the user can use the current judgment criterion value or the judgment criterion value set by the user. If the condition in step S805 is not met, the management server 140 ends the process in Figure 8 without performing the judgment criterion value correction process.

[0106] If the conditions in step S805 are met, the management server 140 proceeds to step S806 and transmits the judgment criterion value obtained by correcting the common judgment criterion value acquired in the process of Figure 4 by applying the previous value retrieved in step S704 to the corresponding information management PC 112, updating the judgment criterion value to be used for the target reagent in the target measurement unit 106 and ending the process of Figure 8. The judgment criterion value transmitted to the information management PC 112 is stored, for example, in the storage unit of the analyzer PC 111 of the automatic analyzer 1 including the target measurement unit 106, and is used for individual judgment of reaction process data subsequently measured using the target reagent in the measurement unit 106. If the judgment in step S806 is not met, the management server 140 ends the process of Figure 8 without performing the judgment criterion value update process.

[0107] Furthermore, if, after executing the process shown in Figure 8, the number of reaction process data measured using the target reagent reaches a threshold and the judgment in step S701 is satisfied, the judgment criterion value correction process shown in Figure 7 may be executed for the target reagent even after the judgment criterion value correction process shown in Figure 8 has been executed.

[0108] -Example of reference value data- Figure 9 shows an example of reference value data stored in the storage device 140A.

[0109] As shown in Figure 9, the calibration curve data storage unit 1433 stores data related to the reference calibration curve, including the measurement item code, reagent manufacturing lot number, reagent sequence number (bottle number), common calibration curve factor, and the difference between the common calibration curve factor and the calibration curve factor specific to the measurement unit, for each combination of measurement unit and reagent type. In steps S406 and S603, this data stored in the calibration curve data storage unit 1433 is referenced.

[0110] Furthermore, the judgment criterion value storage unit 1432 stores data related to the judgment criterion value, including the measurement item code, reagent manufacturing lot number, reagent sequence number, common judgment criterion value, and the difference between the common judgment criterion value and the judgment criterion value specific to the measurement unit, for each combination of measurement unit and reagent type. In steps S406 and S803, this data stored in the judgment criterion value storage unit 1432 is referenced.

[0111] -Example of a settings screen- Figure 10 shows the condition setting screen. For the processes shown in Figures 4 to 8, conditions can be set, for example, on the analysis parameter screen 1001 shown in Figure 10. The analysis parameter screen 1001 is displayed on the corresponding monitoring device by operating, for example, the analysis device PC 111, the information management PC 112, or the client PC 145.

[0112] The analysis parameter screen 1001 displays an item information field 1002, a calibration curve setting field 1003, a reaction process data analysis setting field 1004, and a registration button 1005. In the item information field 1002, the item name, item code, etc., are set. In the calibration curve setting field 1003, settings are made for whether or not to update to the common calibration curve (see steps S403 and S405), the timing of the calibration curve update (see steps S403 and S405), whether or not to automatically correct the calibration curve factor (see step S505), whether or not to automatically correct the calibration curve factor using the previous value (see step S605), and the threshold for the difference between the common calibration curve factor and the calibration curve factor specific to the measurement unit (see step S504). In the reaction process data analysis settings section 1004, you can set whether or not to update to a common judgment criterion value (see steps S403 and S405), the timing of the judgment criterion value update (see steps S403 and S405), whether or not to automatically correct the judgment criterion value (see step S706), the number of reaction processes required for automatic correction (see step S701), whether or not to use the previous value for automatic correction (see step S805), and the threshold for the difference between the common judgment criterion value and the judgment criterion value specific to the measurement unit (see step S705). The threshold for the difference is set for each judgment criterion value. All of the above settings are selected on the screen (for example, by selecting from a pull-down menu) and applied by pressing the registration button 1005.

[0113] -Example of Calibration Curve Data Display Screen- Figure 11 shows the calibration curve factor display screen. The screen shown in the figure is displayed on the corresponding monitoring device by operating, for example, the analyzer PC 111, the information management PC 112, or the client PC 145. A screen like the one in Figure 11 allows you to check the data regarding the calibration curve factor for each reagent bottle.

[0114] The calibration curve factor column 1101 on the screen displays the calibration curve factor for each reagent bottle for the target measurement unit.

[0115] The "Automatic Correction Date" column 1102 displays the date and time when automatic correction was performed for items where automatic correction of the calibration curve factor was enabled in the "Calibration Curve Setting" column 1003 (Figure 10).

[0116] In the "Applying Previous Value" column 1103, a mark (an asterisk in Figure 11) is displayed for reagent bottles where automatic correction using the previous value is enabled in the calibration curve setting column 1003 (Figure 10), and where the previous value is currently being applied.

[0117] -Example of a display screen for judgment criteria values- Figure 12 shows a display screen for judgment criteria values. The screen shown in this figure is displayed on the corresponding monitoring device by operating, for example, the analyzer PC 111, the information management PC 112, or the client PC 145. A screen like the one in Figure 12 allows you to check the data regarding the judgment criteria values ​​for each reagent bottle.

[0118] The judgment criterion value column 1201 on the screen displays the judgment criterion value for each reagent bottle for the target measurement unit. The criterion values ​​A to C shown in Figure 12 are judgment criterion values ​​that are individually used for individual judgments for each judgment category in the analysis of reaction process data, which will be described later in Figure 15, etc.

[0119] The "Automatic Correction Implementation Date" column 1202 displays the date and time when automatic correction was performed for items for which automatic correction of judgment criteria values ​​was enabled in the "Reaction Process Data Analysis Settings" column 1004 (Figure 10).

[0120] In the "Applying Previous Value" column 1203, reagent bottles to which automatic correction using the previous value is enabled in the reaction process data analysis settings column 1004 (Figure 10) and to which the previous value is currently applied are displayed (marked with an asterisk in Figure 12).

[0121] -Example of Quality Control 1- Figure 13A is a flowchart showing an example of the procedure for quality control processing by the management server 140. The processing in Figure 13A is the quality control processing performed for the measurement of each measurement unit after the processing in Figure 4 and one of the processing in Figures 5 to 8 have been executed.

[0122] In step S10, the management server 140, using the reference value data distribution processing unit 1442, distributes and registers reference value data to be applied to quality control using patient samples to the information management PC 112 corresponding to the target measurement unit 106. The reference value data includes calibration curve data and judgment criteria values. The reference value data is the common reference value data acquired in the process of Figure 4, or, if applicable, the reference value data obtained by correcting the common reference value data in one of the processes of Figures 5 to 8. This data is transmitted to the information management PC corresponding to the target measurement unit 106 as information to be applied to the quality control process from step S20 onward, and is registered by being stored in the storage device of the analyzer PC 111 of the corresponding automatic analyzer 1. The timing of transmission is, for example, when the process of Figure 4 is executed, or when one of the processes of Figures 5 to 8 is executed.

[0123] In step S20, the management server 140 applies the reference value data registered in step S10, and the measurement unit 106 measures normal samples (e.g., absorbance measurement). As mentioned above, normal samples are patient samples that meet predetermined conditions set to guarantee a certain level of quality, and measurements of patient samples whose measurement data does not meet the predetermined conditions are not included in the measurement of normal samples.

[0124] In step S30, the management server 140 receives and acquires the reaction process data of the normal sample measured by the measurement unit 106 in step S20 from the information management PC 112, and records it, for example, in the device data storage unit 1421. The reaction process data is a dataset of photometric points (photometric time) and values ​​corresponding to absorbance, and can be represented as a graph plotted on a planar coordinate system with photometric points on the horizontal axis and values ​​corresponding to absorbance on the vertical axis.

[0125] In step S40, the management server 140 determines whether the number of reaction process data acquired in step S30, that is, the number of reaction process data related to normal samples using the target reagent in the target measurement unit 106, satisfies a predetermined number (for example, 20 to 100 samples) required as the population of normal samples. This determination is performed by the reaction process data analysis unit 1441 shown in Figure 3. If the number of reaction process data related to normal samples is less than the predetermined number, the management server 140 repeats the processing in steps S20 and S30 until the number of reaction process data related to normal samples reaches the predetermined number. By securing a predetermined number of reaction process data for normal samples, the reliability of quality control using patient samples is ensured. Furthermore, the timing for acquiring the predetermined number of reaction process data for normal samples is, for example, when the automated analyzer 1 is started up, and the first predetermined number of reaction process data obtained by the target measurement unit 106 after the patient sample measurement operation has started can be used for quality control.

[0126] Once a predetermined number of reaction process data for normal samples have been obtained, the management server 140 proceeds to step S50, where it analyzes each reaction process data to determine if there are any abnormalities. The process in step S50 will be described later using Figure 14.

[0127] Once the analysis in step S50 is complete, the management server 140 proceeds to step S60 and outputs the quality control results based on the results of the reaction process data analysis performed in step S50.

[0128] -Example 2 of Quality Control- Figure 13B is a flowchart showing the procedure for another example of the quality control process performed by the management server 140. The only difference between the process in Figure 13B and the process in Figure 13A is that the order of steps S40 and S50 is reversed. For processes similar to those in Figure 13A, the same step numbers are used in Figure 13B, and explanations are omitted as appropriate.

[0129] In the example shown in Figure 13B, the management server 140 proceeds to step S50 each time it acquires reaction process data of a normal sample in step S30 to analyze the reaction process data acquired in step S30. After analysis, the management server 140 determines in step S40' whether the number of analysis results has reached a predetermined number, and repeats the processes in steps S20, S30, and S50 until the determination in step S40' is satisfied, thereby obtaining the analysis results of a predetermined number of reaction process data of normal samples and outputting the quality control results (step S60).

[0130] -Example 3 of Quality Control- Figure 13C is a flowchart showing the procedure for yet another example of quality control processing by the management server 140. The difference between the processing in Figure 13C and the processing in Figures 13A and 13B is that, when the number of reaction process data for normal samples falls below a predetermined number, the reaction process data is analyzed using substitute data. The same step numbers are used in Figure 13C for the same processing as in Figure 13A, and explanations are omitted as appropriate.

[0131] In the example shown in Figure 13C, the management server 140 outputs the quality control results without waiting for a predetermined number of reaction process data to be obtained, for example, from the time the automated analyzer 1 is started up until a predetermined number of reaction process data for normal samples are obtained. Specifically, in the determination in step S40, if the number of reaction process data for normal samples is less than a predetermined number, the management server 140 moves to step S45 and performs data supplementation processing. In the data supplementation processing, the number of reaction process data analysis results obtained at this point is supplemented by adding the analysis results of reaction process data performed in the past to the analysis results of reaction process data obtained at this point. For the data supplementation processing, reaction process data of normal samples used in quality control processing of past lots or past sequences of the reagent performed by the measurement unit 106 may be used, or reaction process data of normal samples used in quality control processing of the same lot or same sequence of the reagent performed by a different measurement unit 106 may be used. After a predetermined number of reaction process data for normal samples have been obtained, the processing shown in Figure 13A or Figure 13B may be executed.

[0132] -Analysis of reaction process data- Figure 14 is a flowchart showing the procedure for analyzing reaction process data, and details the process of step S50 in each process of Figures 13A to 13C. The management server 140 executes the following processes using the reaction process data analysis unit 1441. Note that the reaction process data analysis process described below is not necessarily required for carrying out the aforementioned processes such as distribution, correction, and setting of reference value data, and the reaction process data analysis process can also be carried out by appropriately combining known accuracy control methods, such as those described in Patent Document 2.

[0133] In step S1401, the management server 140 sorts the reaction process data into a predetermined number of determination categories.

[0134] In step S1402, the management server 140 performs individual judgments for each judgment category based on the behavior pattern of the measurement data using the judgment criterion value. The behavior pattern is the waveform pattern of the reaction process data for the judgment category to be judged.

[0135] In step S1403, the management server 140 determines whether individual determinations for all determination categories of the reaction process data have been completed. If individual determinations for all determination categories have not been completed, the management server 140 returns to step S1402 and performs individual determinations for all determination categories of the reaction process data.

[0136] After performing individual judgments for all judgment categories of the reaction process data, the management server 140 proceeds to step S1404 and obtains an overall judgment result (analysis result) of the reaction process data based on the individual judgment results for each judgment category. Specifically, as will be described later using Figure 15, the presence or absence of abnormalities, the magnitude of the abnormality, and the estimated cause are referred to in advance to the combination pattern of individual judgment results, and this is used as the overall judgment result.

[0137] In step S1405, the management server 140 outputs the accuracy control results based on the overall judgment result. The accuracy control results can be displayed on the corresponding monitoring device by operating, for example, the analysis device PC 111, the information management PC 112, or the client PC 145.

[0138] -Example of Analysis Results of Reaction Process Data- Figure 15 shows examples of individual judgment results and overall judgment results in the analysis of reaction process data. Individual judgment result 1501 in Figure 15 shows the individual judgment results for each judgment category of the reaction process data for each reaction process data.

[0139] In the example in Figure 15, a check mark is placed next to the judgment category that exceeds the judgment threshold as a result of the individual judgment. For example, the judgment category labeled "Category 1" for the reaction process data labeled "Reaction Process 2" has a check mark, indicating that the individual judgment found that the judgment threshold was exceeded. On the other hand, "Category 2" for "Reaction Process 2" does not have a check mark, indicating that the individual judgment for this judgment category did not find that the judgment threshold was exceeded. Individual judgment results 1501 shows the individual judgment results for multiple judgment categories (Category 1 to Category N) for each reaction process data.

[0140] The overall judgment result 1502 shows the judgment result for the entire reaction process data based on the individual judgment result 1501, for each reaction process data. The overall judgment result 1502 is the analysis result for each reaction process data based on the combination pattern 1502a of judgment categories that were checked in the individual judgment result 1501. The combination pattern 1502a is a type of combination in which, for example, only category 1, only category 2, only categories 1, 2, and 3, or only categories 2 and 4 were checked out of categories 1 to N. For each combination pattern 1502a, the estimated cause of the abnormality (estimated cause 1502c), judgment 1502b, and countermeasures for the estimated cause 1502c (countermeasure 1503) are described. For example, in the example in Figure 15, for "Pattern 1", in which no categories are checked out, the judgment 1502b is determined to be normal. On the other hand, for "Pattern 3," where only categories 1, 3, and 5 have check marks, the estimated cause 1502c is estimated to be "Reagent dispensing failure," which is judged to be highly abnormal in judgment 1502b, and the recommended countermeasure 1503 is "Contact Service."

[0141] Each combination pattern (pattern 1, 2, 3…) is pre-associated with a judgment 1502b, an estimated cause 1502c, and a countermeasure 1503. Data defining this correlation (table data, etc.) is stored in the storage device 140A, and the management server 140 refers to this correlation data to identify the estimated cause 1502c, etc., based on the combination pattern 1502a of the individual judgment results. The correlation between the combination pattern 1502a of the individual judgment results and the estimated cause 1502c, etc., may be, for example, artificially set in advance based on empirical rules, or it may be set by machine learning based on maintenance data of each measurement unit 106 under the management of the management server 140. The estimated cause is not limited to failures or malfunctions of parts of the measurement unit 106, but also includes the condition of reagents and samples.

[0142] Furthermore, judgment 1502b correlates with the estimated cause 1502c. In the example in Figure 15, if the estimated cause 1502c is a reagent dispensing error, it is considered to have a significant direct impact on the measurement data, and judgment 1502b is set to a high degree of abnormality. If the estimated cause 1502c is an air bubble generated within the cell, it is considered to have a small impact on the measurement data, and judgment 1502b is set to a low degree of abnormality.

[0143] -Example of Judgment Categories- Figure 16 shows an example of judgment categories for reaction process data. In Figure 16, the reaction process data is represented graphically. The horizontal axis is the measurement point, and the vertical axis is absorbance. During the reaction process, absorbance is measured at measurement points at predetermined time intervals, and these datasets are obtained as reaction process data. A judgment category is a category consisting of one measurement point or multiple consecutive measurement points of the reaction process data. The reaction process can be broadly divided into a pretreatment reaction process and the main reaction process. The reaction process data exemplified in Figure 16 is to which the two-point end method is applied, and its waveform is the aforementioned ABS = A0 + A1(1 - e -ktIt is approximated by the formula ). In the two-point end method, the concentration of the target component is determined from two measurement points: measurement point P2 before the reaction or immediately after the start of the reaction, and measurement point P4 at the end of the reaction. However, the measurement method is not limited to the two-point end method; other methods such as the rate method may also be used. Note that measurement point P1 is the first measurement point after adding the reagent to the sample and stirring, and measurement point P3 is the first measurement point after stirring the reaction solution after the reaction has started.

[0144] The judgment categories can be arbitrarily set from the measurement points of such reaction process data. In the example in Figure 16, the case where measurement points P2 and P4 are set to judgment categories H and K from judgment categories 1 to N (N≧2) is illustrated. Of course, other measurement points other than measurement points P2 and P4 can also be used as judgment categories, and multiple measurement points from any first measurement point to any second measurement point can be used as a single judgment category. Furthermore, two measurement points may partially overlap with different judgment categories (one measurement point may belong to multiple judgment categories), and there may be measurement points that do not belong to any reaction category.

[0145] -Example of Individual Judgment Algorithm- Figure 17 shows the checks performed in individual judgment. As shown in Figure 17, the contents to be checked in individual judgment are defined for each judgment category. For example, in judgment category H illustrated in Figure 16, items such as whether the measured value is experiencing a positive spike, a negative spike, a gradual increase, or a gradual decrease are checked. For example, in judgment category H (measurement point P2), if the absorbance has risen above the corresponding judgment criterion value relative to the previous measurement point, it is determined that a positive spike has occurred, and a check mark is placed in the individual judgment for judgment category H. Similarly, in judgment category H, if the absorbance has decreased above the corresponding judgment criterion value relative to the previous measurement point, it is determined that a negative spike has occurred, and a check mark is placed in the individual judgment for judgment category H. Also, ABS = A0 + A1(1 - e -ktIf the reaction process data approximated by ) is upward sloping or downward sloping in judgment category H, and the magnitude of the slope in judgment category H exceeds the judgment threshold value, it is judged as fluctuation (gradual increase or gradual decrease), and a check mark is placed in the individual judgment for judgment category H.

[0146] As described above, the check contents for individual judgments are defined for each judgment category, and based on these check contents, individual judgments are performed for each reaction process data as illustrated in Figure 15. The combination patterns of these judgments suggest the estimated cause of the anomaly and countermeasures, which can be used for quality control and predictive diagnostics of failures.

[0147] -Example of Quality Control Results- Figure 18 shows an example of a display screen for quality control results. The screen shown in the figure is displayed on the corresponding monitoring device by operating, for example, the analyzer PC 111, the information management PC 112, or the client PC 145. Using a screen like the one in Figure 18, you can check the quality control results for any measurement.

[0148] In the screen shown in Figure 18, first, enter the period for which you want to check the quality control results (the first and last days of the period) in the measurement date input field 1801, and then press the search button 1802. Then, a list of measurements performed during the specified period will be displayed in the search results display field 1803. Although not shown in Figure 18, it is also possible to further refine the results by adding other conditions, such as measurement items, reagents used, and measurement unit 106.

[0149] In the example shown in Figure 18, the search results display area 1803 shows the date and time when the sample was first measured for each sample ID.

[0150] Furthermore, when a measurement of a specific sample ID is selected in the search results display area 1803, the quality control results for each measurement item of that sample ID are displayed in the detailed display area 1804. In the detailed display area 1804, the "measurement result," "quality control result," and "estimated cause" are displayed for each measurement item of the measurement of the sample ID selected in the search results display area 1803. The measurement result is the measurement data for the measurement item of the selected sample ID. The quality control result represents the result of the individual judgment for each judgment category of the reaction process data described above. In other words, in Figure 15, measurements judged to have a high degree of abnormality in the overall judgment result are displayed as NG, and measurements judged to have a low degree of abnormality or measurements for which no check marks were placed in the individual judgment are displayed as OK. The estimated cause corresponds to the estimated cause 1502c explained in Figure 15.

[0151] As shown in Figure 18, the screen allows users to check the accuracy control results for each measurement and when signs of failure began to appear in components that are suspected of malfunctioning.

[0152] -Example of Reaction Process Data Analysis Results- Figure 19 shows the display screen of the reaction process data analysis results. The procedure for analyzing the reaction process data is as described above using Figure 14. Figure 19 is displayed based on the reaction process data analysis results illustrated in Figure 15. The screen shown in the figure is displayed on the corresponding monitoring device by operating, for example, the analysis device PC 111, the information management PC 112, or the client PC 145.

[0153] The screen in Figure 19 displays, for each measurement, the measurement unit 106 used, sample ID, measurement item, measurement data (e.g., concentration of the measurement item), measurement date and time, as well as a reaction process shape check field 1901, a judgment field 1902, and a predicted cause field 1903.

[0154] The reaction process shape check column 1901 displays the results of individual judgments (No. 1, No. 2, etc.) for each judgment category performed on the reaction process data for each measurement. The judgment column 1902 displays the judgment result of the validity of the reaction process data. The display in judgment column 1902 corresponds to the quality control result based on the combination pattern of judgment pattern results. If no check marks are placed in any of the individual judgments in the reaction process shape check column 1901, it is considered "normal." If check marks are placed, it is considered "high degree of abnormality" or "low degree of abnormality" based on the combination pattern. The display in judgment column 1902 corresponds to judgment 1502b in Figure 15. The expected cause column 1903 displays the cause of the abnormality estimated based on the combination pattern of the individual judgment results in the reaction process shape check column 1901. The display in expected cause column 1903 corresponds to estimated cause 1502c in Figure 15.

[0155] In this embodiment, the presence or absence of abnormalities in the measurement unit 106, reagents, samples, etc., is determined based on combination patterns of individual judgment results, and these judgment results are displayed along with the expected cause as shown in Figure 19. This detection of abnormalities in the measurement unit 106, etc., based on combination patterns of individual judgment results is useful not only for accuracy control to confirm the reliability of measurement data, but also for predicting failures of components in the measurement unit 106.

[0156] -Example of standardization of measured values- For example, the units of reaction process data measured by the measurement unit 106 may differ depending on the manufacturer of the automatic analyzer 1. Also, even if the manufacturer is the same, the units of reaction process data may differ depending on the model. Therefore, if the manufacturers and models of the measurement units 106 under management are not standardized, the management server 140 cannot process the received reaction process data in the same way and perform the above-mentioned accuracy control and other processing. Processing reaction process data with different units in the same way as is will not yield reliable processing results, and the management server 140 is forced to process only the measurement data from measurement units 106 with common units for the reaction process data.

[0157] In this embodiment, in order to perform various highly reliable processes, a function is described that converts reaction process data with different units into a unified standardized form to align the units.

[0158] Figure 20 shows a list of data stored in the unified standard information storage unit 1431 (Figure 3). The unified standard information 2001 stored in the unified standard information storage unit 1431 includes a "reaction process horizontal axis correspondence table," a "reaction process vertical axis conversion formula," and a "reaction process vertical axis conversion coefficient." The reaction process vertical axis conversion formula is a conversion formula such as A' = aA + b. Here, A: absorbance before conversion, A': absorbance after conversion, a, b: reaction process vertical axis conversion coefficients. By using this reaction process vertical axis conversion formula, the unit of absorbance A of each measurement unit 106 (e.g., magnification) is converted to the unit of absorbance A' expressed in the unified standard. The unit of absorbance A' in the unified standard may be a new unit that does not correspond to the absorbance unit of the reaction process data of any of the measurement units 106, but it may also be the same unit as absorbance A of any measurement unit 106 selected from under the management of the management server 140.

[0159] Figure 21 shows an example of a reaction process horizontal axis correspondence table stored in the unified standard information storage unit 1431. The photometric timing column 2101 contains the photometric timing time corresponding to the instrument model number of the measurement unit 106 for each manufacturer. The reagent addition timing column 2102 contains the reagent addition timing time corresponding to the instrument model number. The unit of time is minutes or seconds, etc. In the example in Figure 21, for example, even when measuring the same measurement item using the same type of reagent, for example, in the measurement unit 106 (No. 1) of manufacturer A with instrument model number XXXX, the first photometric point is set 15 seconds after the first reagent is added to the sample, whereas in the measurement unit 106 (No. 3) of manufacturer B with instrument model number ZZZZ, the first photometric point is set 25 seconds after the first reagent is added to the sample. Furthermore, in manufacturer A's measurement unit (No. 1), the time intervals between photometric points are equal, whereas in manufacturer B's measurement unit (No. 3), the time intervals between photometric points are unequal. Moreover, even within the same manufacturer A's measurement unit 106, in device model XXXX (No. 1), the photometric points are set at 15-second intervals after the first reagent is added to the sample, whereas in device model YYYY (No. 2), the photometric points are set at 10-second intervals after the first reagent is added to the sample. Therefore, if photometric points are plotted on the horizontal axis, the waveforms of the reaction process data may appear to show similar trends in the graph, but they cannot be evaluated on the same scale.

[0160] Figure 22 shows reaction process data before and after conversion to a unified standard. The reaction process graph 2201 shown in the upper part of Figure 22 shows the reaction process graph before conversion to a unified standard. As shown in the example in the upper part of Figure 22, the reaction process data output by a certain measurement unit is represented by a two-dimensional coordinate system reaction process graph 2201, where the horizontal axis unit is photometric points and the vertical axis unit is a device-specific unit called "A". In this embodiment, the unit of the horizontal axis of the reaction process graph 2201 is converted from photometric points to time (seconds) using the reaction process horizontal axis correspondence table (Figure 21), and the unit of the vertical axis is converted from the device-specific A to a common absorbance A' using the reaction process vertical axis conversion formula (A' = aA + b). The converted reaction process graph 2202 in the lower part of Figure 22 corresponds in waveform to the reaction process graph 2201 in terms of trend, but the units of the horizontal and vertical axes are different.

[0161] By converting the units of the horizontal and vertical axes of the reaction process data from each measurement unit 106 to a common unified standard (in this example, the horizontal axis is seconds and the vertical axis is absorbance), the reaction process data from measurement units 106 from different manufacturers or models can be handled in the same way, enabling highly reliable judgments and analysis processes.

[0162] 1...Automatic analyzer, 106...Measurement unit, 112...Information management PC (clinical laboratory information system), 140...Management server (management device), 140A...Storage device, 140B...Calculation unit, 146...Network, P2, P4P...Measurement points (judgment categories)

Claims

1. A management device for an automated analyzer having a measurement unit, comprising a storage device and a calculation device, wherein the storage device stores calibration curve data corresponding to a reagent, a sample and a measurement interval for the reaction process of the reagent, a predetermined number of judgment categories set in advance within the measurement interval, and a judgment criterion value used for individual judgment of the judgment category, the calculation device performs a first step of selecting the sample to be used for quality control, and a second step of performing quality control based on the reagent type corresponding to the calibration curve data registered for the measurement items of the sample and the reaction process data which is time-series measurement data measured by the measurement unit in the measurement interval, in the second step, the calculation device performs the individual judgment for each predetermined judgment category using the judgment criterion value based on the measurement data, determines at least one state of the measurement unit, the reagent, and the sample based on the combination pattern of the results of the individual judgment for each judgment category, and outputs a quality control result based on the combination pattern of the judgment results of the individual judgment.

2. The management device according to claim 1, wherein the storage device stores correlation data linking the normal and abnormal states of at least one of the measurement unit, the reagent, and the sample with the combination pattern, and the calculation device determines, in the second step, the state of at least one of the measurement unit, the reagent, and the sample from the combination pattern based on the correlation data.

3. The management device according to claim 2, wherein the correlation data includes the degree of the abnormality.

4. The control device according to claim 1, characterized in that one of the plurality of determination categories is the timing after the reagent and the sample have been mixed and stirred.

5. The management device according to claim 1, characterized in that, in the second step, the individual determination is a determination using the determination criterion value of the waveform of the measurement data of the target determination category.

6. A control device according to claim 1, wherein the sample selected in the first step is a plurality of patient samples, and the calculation device outputs the quality control result after performing the individual determination for the plurality of patient samples.

7. A management device according to claim 6, wherein the calculation device selects the patient sample in the first step based on the measurement data of the patient sample if the measurement data satisfies predetermined conditions.

8. A management system comprising the management device described in claim 1, and a clinical laboratory information system connected to the measurement unit and capable of communicating with the management device via a network, wherein the automated analyzer and the management device are configured to exchange information via the clinical laboratory information system.

9. A method for managing the accuracy of a measurement unit of an automated analyzer, comprising: a second step of managing accuracy based on a reagent type corresponding to calibration curve data registered for the measurement items of a sample, and reaction process data which is time-series measurement data measured by the measurement unit, wherein in the second step, individual judgments are performed for each of a predetermined number of predetermined judgment categories set in advance within the measurement interval of the reaction process of the sample and reagent, using predetermined judgment criteria values ​​based on the measurement data; at least one state of the measurement unit, the reagent, and the sample is determined based on a combination pattern of the results of the individual judgments for each judgment category; and an accuracy management result is output based on the combination pattern of the judgment results of the individual judgments.

Images