Data processing device and automated analysis device

Abstract

The present application provides a data processing device and an automatic analysis device capable of assisting an operator in quickly or easily determining the cause of an assay abnormality to improve the abnormality. The data processing device refers to data related to an assay of a sample performed by the automatic analysis device, calculates a combination of values of a plurality of factors (F1, F2) based on reaction process data, using, as the factors, predetermined variables or assay values indicating characteristics of a reaction, refers to a correspondence table having a distribution map composed of a normal range region (901) and a plurality of abnormal range regions (902) with combinations of normal / abnormal ranges of the plurality of factors, and information of abnormality cause candidates associated with each abnormal range region, judges which abnormal range region of the correspondence table the combination of values of the plurality of factors matches, and outputs information of the abnormality cause candidates associated with the matched abnormal range region to a user.

Description

Technical Field

[0001] This invention relates to the technology of data processing devices and automatic analysis devices, and to technologies such as data processing and anomaly detection obtained through the determination of samples. Background Technology

[0002] For example, automated analytical devices equipped with functions such as biochemical analysis and immunoassay for clinical examinations process data obtained by measuring biological samples (also known as specimens) such as blood and urine. For instance, an automated analytical device measures the change in absorbance over time in a reaction solution obtained by mixing a certain amount of sample and reagent and stirring it to allow the reaction to proceed. From this, the automated analytical device calculates the concentration, activity value, etc., of the target substance in the sample as the measured value. Furthermore, by measuring the standard solutions used for calibration reagents or managing the state of reagents for checking the status of each device and each analytical item, the automated analytical device monitors the state of the device, samples, etc., and thus prevents a decrease in measurement accuracy.

[0003] As an example of prior art related to the aforementioned automatic analysis device, Japanese Patent Application Publication No. 2009-204448 (Patent Document 1) can be cited. Patent Document 1 discloses that in an automatic analysis device, a reaction rate constant is calculated using an approximation formula based on reaction process data representing absorbance changes stored in a time series, and the presence or absence of an abnormality in the reaction is determined based on the value of the reaction rate constant.

[0004] Existing technical documents

[0005] Patent documents

[0006] Patent Document 1: Japanese Patent Application Publication No. 2009-204448 Summary of the Invention

[0007] The problem that the invention aims to solve

[0008] It is believed that the performance of automated analytical devices, such as their measurement accuracy, is determined by multiple performance factors—in other words, by a combination of factors affecting performance. These performance factors include the setup conditions or state of the automated analytical device, the state of the sample or reagent, and the operator's operational procedures. Examples of performance factors include, for instance, the dispensing volume of the sample or reagent into the reaction vessel, dispensing accuracy, and the homogeneity and stability of the reagent or standard solution.

[0009] Therefore, when abnormal measurement results occur in automated analysis devices and by operators, it is necessary to determine the cause of the abnormality and take measures such as maintenance to address it, thereby improving the abnormal state and suppressing the decrease in measurement accuracy. The aforementioned causes are related to the aforementioned performance governing factors.

[0010] Existing automated analysis devices, for example, can determine / detect abnormalities in measurement results by comparing the values ​​of reaction process data with thresholds representing a predetermined range under normal conditions. However, determining the cause of such abnormalities is difficult due to the consideration of multiple performance governing factors. There are situations where operators find it difficult to determine the cause of the abnormality, and where a significant amount of effort is required for specific tasks. Patent Document 1 does not address the determination of the cause of the abnormality. Particularly in conventional automated analysis devices used for clinical examinations, it is very difficult for operators to quickly or easily determine the presence and cause of abnormalities in large quantities of analyte samples with unknown concentrations, thus hindering the improvement of any abnormalities. There is room for improvement in this regard.

[0011] The purpose of this invention is to provide a technique for automatically analyzing devices that can assist operators in quickly or easily identifying the cause of measurement anomalies and thus improving the situation.

[0012] Methods for solving problems

[0013] A representative embodiment of the present invention has the structure shown below. The data processing device of the embodiment refers to data related to the measurement of a sample performed by an automatic analysis device. Based on reaction process data in the data, it calculates a combination of values ​​for multiple factors, using predetermined variables or measured values ​​representing characteristics of the reaction as factors. Referring to a correspondence table containing a distribution map of normal range regions and multiple abnormal range regions comprised of combinations of normal / abnormal ranges of the multiple factors, and information on candidate abnormal causes associated with each abnormal range region, it determines which abnormal range region in the correspondence table the combination of the multiple factor values ​​falls into, and outputs information on candidate abnormal causes associated with the falling abnormal range region to the user.

[0014] Invention Effects

[0015] According to a representative embodiment of the present invention, the technology of the automatic analysis device can assist the operator in quickly or easily identifying the cause of measurement anomalies and thus improving the anomalies. Attached Figure Description

[0016] Figure 1 This describes the structure of a system that includes the data processing apparatus and the automatic analysis apparatus of Embodiment 1 of this disclosure.

[0017] Figure 2 This illustrates a structural example of the automatic analysis device according to Embodiment 1.

[0018] Figure 3 This section illustrates a structural example of the data processing apparatus described in Embodiment 1.

[0019] Figure 4This describes the main processing flow of the data processing apparatus in Implementation Method 1.

[0020] Figure 5 This illustrates the concept of the correspondence table and the first structural example in Implementation Method 1.

[0021] Figure 6 This illustrates the concept of the correspondence table in Implementation Method 1 and a second structural example.

[0022] Figure 7 This illustrates the concept of the correspondence table in Implementation Method 1 and a third structural example.

[0023] Figure 8 An example representing reaction process data in Implementation Method 1.

[0024] Figure 9 In Implementation 1, based on Figure 8 An example of a distribution plot of the combination of two factors in the reaction process data.

[0025] Figure 10 The representation of the distribution in Implementation 1 is as follows: Figure 9 An example of the reaction process data in the abnormal range region #1.

[0026] Figure 11 Indicates the output in implementation method 1 and Figure 9 An example of a screen showing the candidate causes associated with the abnormal range area #1.

[0027] Figure 12 This is a screen example from a variation of implementation 1.

[0028] Figure 13 An example representing reaction process data in a variation of Embodiment 1, Example 2.

[0029] Figure 14 In variation example 2, based on Figure 13 An example of a distribution plot of the combination of two factors in the reaction process data.

[0030] Figure 15 This indicates that the distribution in variation example 2 is included in Figure 14 An example of the reaction process data in the abnormal range region #1.

[0031] Figure 16 An example representing reaction process data in variation 3 of embodiment 1.

[0032] Figure 17 In variation example 3, based on Figure 16 An example of a distribution plot of the combination of two factors in the reaction process data.

[0033] Figure 18 Indicates the output in variation example 3 and Figure 17 The abnormal range area #1 is associated with the candidate cause, etc., in the screen example. Detailed Implementation

[0034] Hereinafter, embodiments of the present invention will be described in detail based on the accompanying drawings. Furthermore, in all the drawings, the same reference numerals are used to label the same parts in principle, and repeated descriptions are omitted. The constituent elements (including steps, etc.) in the embodiments can be added, deleted, or replaced, except for the essential elements.

[0035] <Implementation Method 1>

[0036] use Figures 1 to 11 The automatic analysis apparatus and data processing apparatus of Embodiment 1 of this disclosure will be described.

[0037] The data processing apparatus of Embodiment 1 calculates multiple factors based on reaction process data obtained from the measurement of a sample in an automated analysis device, using predetermined variables representing reaction characteristics as factors. The reaction process data is a time series representing the state of the reaction between the sample and reagents, obtained along with absorbance measurements. For example, the reaction process data is a graph with time on the horizontal axis and absorbance on the vertical axis. In this context, a factor refers to a factor related to the determination of the cause of an anomaly. A unique correspondence table is constructed through the combination of multiple factors. Furthermore, this factor is a concept different from the performance governing factors described above.

[0038] The data processing device determines or confirms whether a factor value is normal or abnormal for each factor. In other words, the data processing device classifies factor values ​​into normal or abnormal ranges. Using the determination results of the factor values, the data processing device generates a distribution map of combinations of normal / abnormal ranges for multiple factors (designated as the first combination). This distribution map constitutes a two-dimensional or higher region for two or more factors. The distribution map has normal range regions and multiple abnormal range regions. The data processing device generates a correspondence table containing this distribution map.

[0039] The data processing device associates the information of each candidate cause with the abnormal range region in the first combination of factors in the correspondence table. In other words, in the correspondence table, a combination of candidate causes (denoted as the second combination) is associated with the first combination of factors. This candidate cause is the one presumed to be the cause of the abnormality. The candidate cause has various elements related to the aforementioned performance governing factors. Examples of candidate causes include the setup conditions or status of the automated analysis device, the status of the sample or reagent, and the operator's operating status.

[0040] In addition, the data processing device can also associate the information of each response with the candidate causes in the corresponding table. The response information includes maintenance information used to improve or eliminate abnormal states. The response information also includes operator guidance, troubleshooting, alarms, or manuals indicating how to respond to the causes of abnormalities.

[0041] The aforementioned correspondence table can be pre-set in the data processing device. When the data processing device detects or confirms an anomaly related to the measurement based on the measurement data from the automatic analysis device, it determines the anomaly range area that matches the factor value corresponding to the anomaly based on the correspondence table. Then, the data processing device outputs candidate causes and corresponding response information associated with the anomaly range area to the operator. As a result, the operator can use the output information as an aid to determine and respond to anomalies more quickly and easily than with existing technologies.

[0042] [Automatic analysis device and data processing device]

[0043] Figure 1 This describes the structure of a system (in other words, an automatic analysis system) that includes the data processing apparatus and the automatic analysis apparatus of Embodiment 1. Figure 1 The system mainly consists of a data processing device 1 and an automatic analysis device 10, as described in Embodiment 1, which are connected via wired or wireless communication. The appearance of the automatic analysis device 10 and the data processing device 1 shown in the figure is an example. The operator is a user such as a technician, who operates the automatic analysis device 10 or the data processing device 1 to perform tasks.

[0044] The automated analysis device 10 is an automated analysis device for clinical examination, for example, having the function of performing determination and analysis of biological samples as a biochemical analysis function. The automated analysis device 10, for example, has the function of determining the concentration of components in the sample. The automated analysis device 10 is composed of a single module or a combination of multiple modules. When the automated analysis device 10 is composed of a combination of modules, for example, it can be composed of a combination of one or more analysis modules, operation modules, and delivery modules. A component described later is disposed on the upper surface 10a of the housing of the automated analysis device 10. Figure 2 (various institutions)

[0045] The data processing device 1 has the function of receiving / acquiring data related to sample measurement from the automatic analysis device 10 via communication and processing and analyzing the data. The data includes measurement data such as measurement results. In particular, the data processing device 1 has the function of providing the operator with information such as the causes and countermeasures related to measurement anomalies based on the analysis and confirmation of the measurement results. The data processing device 1 can be installed, for example, via a computer such as a PC. Alternatively, the data processing device 1 can be installed as part of the automatic analysis device 10, such as as part of an operation module or control device. The data processing device 1 can be configured nearby relative to the automatic analysis device 10 as shown in the figure, or it can be remotely configured relative to the automatic analysis device 10 via a LAN or the like.

[0046] In Embodiment 1, an automatic analysis device 10 and a data processing device 1 are shown separately, with the data processing device 1 equipped with a specific function for assisting in determining the cause of anomalies; however, this is not a limitation. It is also possible to configure it as an integrated device possessing both the functions of the automatic analysis device 10 and the data processing device 1.

[0047] [Automatic Analysis Device]

[0048] Figure 2 This illustrates a structural example of the automatic analysis device 10. Figure 2 The upper part shows the main elements disposed on the upper surface 10a of the housing of the automatic analysis device 10, and the lower part schematically shows the elements inside the housing connected to these elements. Furthermore, the automatic analysis devices that can be applied are not limited to this structural example.

[0049] The automatic analysis device 10 includes a sample holding mechanism 11, a reagent holding mechanism 12, a reaction vessel holding mechanism 13, a sample dispensing mechanism 14, a reagent dispensing mechanism 15, an optical measurement mechanism 16, a stirring mechanism 17, and a cleaning mechanism 18. The automatic analysis device 10 also includes a sample transport mechanism and a waste disposal mechanism.

[0050] The sample holding mechanism 11 is, for example, a box-shaped device such as a rack, and is a mechanism capable of holding / transporting multiple sample containers. The sample holding mechanism 11 can also be a disc-shaped mechanism, or it can be prepared as an additional module that can be connected. The sample containers contain specimens such as blood as samples. The sample containers can be placed on a sample rack that carries one or more sample containers and transported. The sample holding mechanism 11 also includes, for example, a transport mechanism capable of transporting the sample containers by parallel movement or rotational movement. A drive mechanism 121 is connected to the sample holding mechanism 11. Figure 2 The sample holding mechanism 11 has the following functions: transporting the inserted sample rack to the sample dispensing mechanism 14, holding the sample rack after dispensing until the analysis is completed, and retrieving the sample rack after the analysis is completed to a predetermined position.

[0051] The reagent holding mechanism 12 is, for example, a disc-shaped mechanism capable of holding and transporting multiple reagent containers. The reagent containers hold reagent solutions. The reagent containers may also be composed of reagent bottles or similar containers that centrally hold multiple reagent containers. The reagent holding mechanism 12 may also be a rack-and-pinion mechanism or be prepared as another module. A drive mechanism 122 is connected to the reagent holding mechanism 12. The sample holding mechanism 11 and the reagent holding mechanism 12 may also be configured as a single unit.

[0052] The reaction vessel holding mechanism 13 is, for example, a disc-shaped mechanism capable of holding and transporting multiple reaction vessels. The reaction vessel holding mechanism 13 includes a thermostatic bath. For example, multiple reaction vessels can be mounted on the circumference of the disc. A drive mechanism 123 is connected to the reaction vessel holding mechanism 13. The reaction vessels are made of a light-transmitting material. In the reaction vessels, samples and reagents are dispensed (drawn and discharged) from sample containers and reagent containers, and their mixture is stirred and incubated to become the reaction solution for the analyte. The reaction vessels are maintained at a predetermined temperature by the thermostatic bath.

[0053] The reaction vessel holding mechanism 13, a disc-shaped mechanism, includes a rotating mechanism that enables each container positioned at a location on the circumference to repeatedly rotate and remain stationary in the circumferential direction, thereby moving the container. As a result, each container on the disc moves sequentially to a predetermined position, where a predetermined process is performed by a mechanism corresponding to each position.

[0054] The sample dispensing mechanism 14 is positioned near the sample holding mechanism 11 and the reaction vessel holding mechanism 13, dispensing (drawing and discharging) the sample from the sample container at the dispensing position of the sample holding mechanism 11 into a predetermined reaction vessel of the reaction vessel holding mechanism 13. The sample dispensing mechanism 14 includes, for example, an arm and a probe. A drive mechanism 124 is connected to the sample dispensing mechanism 14.

[0055] The reagent dispensing mechanism 15 is positioned near the reagent holding mechanism 12 and the reaction vessel holding mechanism 13, dispensing (drawing and discharging) the reagent from the reagent container of the reagent holding mechanism 12 into the reaction vessel of the reaction vessel holding mechanism 13. The reagent dispensing mechanism 15 may include, for example, an arm and a pipette nozzle. A drive mechanism 125 is connected to the reagent dispensing mechanism 15.

[0056] The stirring mechanism 17 stirs the mixture of sample and reagent within the reaction vessel held by the holding mechanism 13 to promote the reaction. The reaction vessel containing the reaction solution can be moved to a predetermined area for incubation. The stirring mechanism 17 can be integrated with the dispensing mechanism.

[0057] The optical measurement mechanism 16 includes a light source, a photometer, and measurement circuitry. The optical measurement mechanism 16 is positioned at a predetermined location on the reaction vessel holding mechanism 13. The optical measurement mechanism 16 illuminates the reaction vessel with light from the light source, and the photometer detects the light transmitted or scattered within the reaction vessel. The photometer is, for example, a spectrophotometer capable of measuring absorbance based on spectral detection. The signal of each sample detected by the photometer (e.g., an analog signal of scattered light) is processed by the measurement circuitry, such as through analog-to-digital conversion, and then sent as a measurement signal to the control device 101. A drive mechanism 126 is connected to the optical measurement mechanism 16. The optical measurement mechanism 16 is not limited to the location on the reaction vessel holding mechanism 13 and can be located at other locations.

[0058] The cleaning unit 18 is used for cleaning reusable reaction vessels, etc. In clinical examinations, disposable reaction vessels, dispensing pipette tips, etc., are sometimes used. Used disposable reaction vessels, etc., are disposed of in a designated location via a disposal mechanism. In this case, the automated analyzer 10 has a mechanism for storing disposable reaction vessels and dispensing pipette tips.

[0059] The automatic analysis device 10 also includes pumps, tanks, power supply mechanisms, etc. The pumps comprise dispensing mechanisms, etc. The tanks include various containers for storing system water, cleaning fluid, waste liquid, etc.

[0060] As simply shown by the dashed lines in the figure, each component, including the sample holding mechanism 11, is connected to a corresponding drive mechanism. For example, the drive mechanism 121 includes a motor, circuitry, and other components for driving a predetermined action of the sample holding mechanism 11. Each drive mechanism is connected to an interface circuit 150. The interface circuit 150 is a component capable of providing electrical connections and communication between the components. The interface circuit 150 includes a communication interface device under a predetermined communication interface and also connects to… Figure 1 The data processing device 1 is connected.

[0061] A control device 101 (serving as a controller), a storage device 102, an input device 103, an output device 104, and a power supply device (not shown) are connected to the interface circuit 150. The power supply device supplies power to each component. The predetermined components can communicate with each other via the interface circuit 150. The input device 103 and the output device 104 constitute an operation unit for the operator to operate the automatic analysis device 10. The input device 103 may be an operation panel including buttons, etc. The output device 104 may be a display device or a printing device. The operation unit may also integrate the input device 103 and the output device 104 into one unit, for example, by a touch panel. The operation unit may also be configured as an operation module. Alternatively, Figure 1The data processing unit 1 can also function as an operation unit. In addition, the operation unit may also include light-emitting devices such as LEDs and sound output devices such as speakers, which can be used to notify and alarm the operator.

[0062] The operator's tasks include determining the cause of the anomaly and taking appropriate action. When performing analysis tasks, the operator operates the control unit of the automatic analysis device 10 to perform the tasks. In the event of a measurement anomaly, the operator operates the data processing device 1 to determine the cause of the anomaly and take appropriate action, but is not limited to this.

[0063] The control device 101 is composed of an IC board or a computer, etc. The control device 101 controls the entire automatic analysis device 10 to achieve automatic analysis functions. For example, during the determination and analysis of the sample, the control device 101 sends control signals to each drive mechanism based on the operator's operation, setting information, analysis request information, etc., thereby controlling the operation of various parts such as the sample holding mechanism 11. The control device 101 includes, for example, a CPU, ROM, RAM, etc., and implements control functions through circuit or program processing.

[0064] The storage device 102 is composed of an internal or external non-volatile storage device, such as a semiconductor memory or a hard disk. Various data, including programs, setting information, analysis request information, measurement data, and analysis result information, are stored in the storage device 102.

[0065] The operator registers the examination items for each specimen, such as the patient's, through the screen of the operating unit of the automatic analysis device 10 or the screen of the data processing device 1, and registers the analysis request for the corresponding analysis item. The registration information is stored in the storage device 102. The operator inputs instructions for the specimen to be analyzed. The control device 101 starts the measurement and analysis of the specimen according to the instructions and with reference to the registration information.

[0066] The control device 101 controls the optical measurement mechanism 16, etc., to measure the absorbance of each sample (corresponding reaction vessel), obtain a measurement signal, and store it in the storage device 102. The measurement signal contains reaction process data as time-series data. Using the data from the measurement signal, the control device 101 performs analytical processing for the specified analytical items, calculating measured values ​​such as the concentration of the sample components. When performing this measurement / analysis, the control device 101 calculates the concentration value of the components based on a pre-set analytical method and calibration curves in the analytical items.

[0067] The control device 101 stores the measurement / analysis results of each sample, including reaction process data containing absorbance and measured values ​​such as concentration, in the storage device 102. The control device 101 can display these results on, for example, a display screen on the output device 104. The operator can then view the measurement / analysis results on this screen. Furthermore, the control device 101 can also determine whether the measurement results are abnormal. For example, the control device 101 compares the measured values ​​with conditions related to anomaly determination. Alternatively, the control device 101 compares the values ​​of predetermined variables representing reaction characteristics in the reaction process data with conditions related to anomaly determination. These conditions may include, for example, thresholds constituting a normal range. The control device 101 determines, for example, whether the variable value is within the normal or abnormal range. In this case, the measurement / analysis results data from the automatic analysis device 10 include information indicating whether there are any abnormalities.

[0068] The data processing device 1 obtains the measurement / analysis results data as described above from the control device 101 of the automatic analysis device 10, and performs auxiliary processing for determining the cause of the anomaly as a specific process. Alternatively, the data processing device 1 may, instead of the control device 101, make a determination related to the normality / abnormality of the measured values ​​and variable values ​​based on the measurement / analysis results data as described above.

[0069] [Data Processing Device]

[0070] Figure 3 This section describes a computer system structure example including the data processing device 1 of Embodiment 1. The computer system includes the data processing device 1, an input device 205, a display device 206, and a notification device 207 connected thereto. The data processing device 1 consists of a processor 201, a memory 202, a communication interface device 203, an input / output interface device 204, and a bus connecting them. The input / output interface device 204 is connected to the input device 205, the display device 206, and the notification device 207. The input device 205 is, for example, a keyboard or a mouse. The display device 206 is, for example, a liquid crystal display (LCD). The notification device 207 is a light-emitting device, a sound output device, or the like that capable of notifying the operator. However, the input device 205, the display device 206, and the notification device 207 may also be integrated into the main body of the data processing device 10, as in a laptop PC. The communication interface device 203 and... Figure 2 The automatic analysis device 10 is connected to the interface circuit 150, and communicates with the control device 101 of the automatic analysis device 10 through a predetermined communication interface.

[0071] The processor 201, for example, is composed of a CPU, ROM, RAM, etc., and constitutes the controller of the data processing device 1. The processor 201 implements a predetermined function 201A based on software program processing. This function 201A is a function that assists the operator's work when a measurement anomaly occurs in the automatic analysis device 10, including the function of outputting candidate causes of the anomaly and response information. The processor 201 receives / obtains, for example, measurement / analysis results containing measurement data 23 from the automatic analysis device 10 and stores it in the memory 202. Based on the measurement data 23, the processor 201 refers to condition information such as analytical parameters of the analytical items of the sample and stores it in the memory 202. Based on the measurement data 23, the processor 201 performs anomaly determination, such as whether the measured value or variable value is abnormal, or confirms the anomaly determination result. Based on the reaction process data contained in the measurement data 23, the processor 201 calculates or extracts predetermined factor values. The processor 201 uses predetermined variables that characterize the reaction in the reaction process data as factors and calculates or extracts multiple factor values.

[0072] The processor 201 maintains a pre-set default mapping table as mapping table data 26. In the mapping table, a first combination consisting of normal / abnormal ranges of multiple factor values ​​is defined, along with a second combination consisting of candidate causes associated with each range of the first combination and corresponding response information. When the processor 201 detects or confirms an anomaly based on the measurement data 23, it determines the abnormal range range that the factor value corresponding to the anomaly falls within, based on the mapping table data 26. Then, the processor 201 displays a screen to the operator containing candidate causes and corresponding response information associated with that abnormal range range.

[0073] The memory 202 is composed of a non-volatile storage device, etc., and stores / retains various data and information used by the processor 201, etc. The memory 202 is, for example, a semiconductor memory or a hard disk. The memory 202 stores control program 21, setting information 22, measurement data 23, anomaly judgment result data 24, factor data 25, and correspondence table data 26, etc. The control program 21 is a program used to implement the functions 201A, etc., of the data processing device 1. The processor 201 performs processing corresponding to function 201A based on the control program 21 (described later). Figure 4 The setting information 22 is the setting information of the control program 21 or the user. The operator or manufacturer can operate the input device 205, observe the screen of the display device 206, and set the necessary information, which is reflected in the setting information 22, etc.

[0074] Measurement data 23 is measurement data that can be obtained from the automatic analysis device 1 via communication, including reaction process data during the measurement. Measurement data 23 also includes condition information such as analytical parameters related to the measurement / analysis. The reaction process data and anomaly determination result data 24 of measurement data 23 include values ​​of predetermined variables or measured values, or information that allows factor calculation. Anomaly determination result data 24 is normal / abnormal determination result information obtained by the control device 101 or processor 201 during the measurement. Factor data 25 is data of factor values ​​generated / maintained based on the reaction process data and anomaly determination result data 24 of measurement data 23.

[0075] Furthermore, the anomaly determination results and factor values ​​can be calculated and recorded by the control device 101 in the measurement data 23, etc., but are not limited thereto. The data processing device 1 can also perform anomaly determination and factor value calculation based on the measurement data 23, etc.

[0076] Correspondence table data 26 is a data set that associates a first combination of normal / abnormal ranges of multiple predetermined factors with a second combination of candidate causes and response information. Correspondence table data 26 includes information described later. Figure 5 , Figure 9 A distribution chart consisting of combinations of normal / abnormal ranges of factor values, such as those shown. This distribution chart can also be an image, graph, matrix, etc., with no limitation on its format.

[0077] It can be limited to Figure 3 The diagram illustrates an example of a computer system structure. For instance, a storage device, a server device, etc., can be connected to the data processing device 1, where various types of data can be stored. The data processing device 1 can also be installed by distributing control software 21 from the server device, etc. Alternatively, the data processing device 1 can be connected to an operator-operated client terminal device (e.g., a PC) via a communication network, enabling client-server communication. The functions of the data processing device 1 can also be implemented in a cloud computing system, etc., on a communication network. The computer program can also be stored on a non-transitory, computer-readable recording medium.

[0078] The data processing device 1 is always in communication with the automatic analysis device 10, and performs data processing in real time in cooperation with the control device 101. Thus, the data processing device 1 immediately assists the operator in the event of a measurement anomaly. Not limited to this, the data processing device 1 can also maintain a communication connection with the automatic analysis device 10 for a predetermined period of time when there is an instruction input from the operator. The communication connection can be established by sending a request or instruction from the data processing device 1 to the automatic analysis device 10, or vice versa.

[0079] As by Figure 3Although not shown, the detailed functional blocks implemented by the processor 201 of the data processing device 1 include, for example, the following functional blocks. The data processing device 1 includes a data receiving unit, a data processing unit, a data input unit, a factor determination unit, and a data display unit. The data receiving unit receives / acquires measurement / analysis results data during or after the measurement from the automatic analysis device 10 and stores it in the memory 202 as measurement data 23, anomaly determination result data 24, etc. The data input unit accepts setting information or instructions from users such as operators and stores the input information in the memory 202. The data display unit displays or notifies the operator of candidate causes and response information related to anomalies based on a lookup table.

[0080] The data processing department performs predetermined data processing on the acquired data, particularly processing related to assisting in the determination of anomaly causes using a correspondence table. Based on measurement data 23, the data processing department determines or confirms the presence or absence of anomalies. The data processing department uses predetermined variables based on the reaction process data as factors and calculates combinations of factor values. In the event of an anomaly, the data processing department compares it with the correspondence table to determine the anomaly range region that corresponds to the combination of factor values ​​for that anomaly. Based on the correspondence table, the data processing department determines candidate causes and responses associated with that anomaly range region and generates output information.

[0081] [Processing Flow]

[0082] Figure 4 Indicates by Figure 3 The main processing flow performed by the processor 201 of the data processing device 1. Figure 4 The process has steps S1-S8. Specifically, the process includes the following steps: calculating a combination of values ​​for two factors based on measurement data 23 obtained from the automated analysis device 10, comparing it with a correspondence table (correspondence table data 26), and outputting candidate causes. In this example, predetermined variables based on reaction process data are used as factors.

[0083] In step S1, the data processing device 1 receives / obtains the measurement / analysis results data from the automatic analysis device 10 and stores them in the memory 202 as measurement data 23, etc.

[0084] In step S2, the data processing device 1 bases its data on the reaction process data (described later) contained in the measurement data 23. Figure 8 ), using multiple predetermined variables as multiple factors to calculate the value of each factor, as factor data 25 ( Figure 3 The data is stored in memory 202. In this example, data processing device 1 calculates the values ​​of two factors (denoted as F1 and F2). Furthermore, if the measurement data 23 contains factor values, the calculation can be omitted.

[0085] In step S3, the data processing device 1 reads the threshold values ​​for the normal / abnormal range of each factor constituting the anomaly determination condition from the memory 202 based on the anomaly determination result data 24. Examples of threshold values ​​include upper and lower threshold values ​​constituting the normal range, but the device is not limited to these. The anomaly determination result data 24 obtained from the automatic analysis device 10 contains information about the threshold values ​​for such ranges. Furthermore, if the data from the automatic analysis device 10 does not contain such threshold information, the data processing device 1 can obtain information for the same condition by referring to the setting information of the automatic analysis device 10 or the setting information of the data processing device 1.

[0086] In this example, as will be described later ( Figure 5 In this way, the thresholds for the normal / abnormal ranges of the two factor values ​​have upper and lower thresholds related to the normal range for each factor. That is, each factor value has one normal range and two abnormal ranges (above and below). However, this is not a limitation; for each factor, the structure of the normal / abnormal ranges can have various structures (e.g., as described later). Figure 6 , Figure 7 ).

[0087] In step S4, the data processing device 1 uses the factor values ​​from step S2 and the threshold values ​​of the range from step S3 to determine or confirm the anomaly determination results for the values ​​of each factor calculated based on the measurement results of the same specimen as the target sample. If the anomaly determination result data 24 contains the anomaly determination result, the data processing device 1 confirms the anomaly determination result. If it does not contain the anomaly determination result, the data processing device 1 performs the anomaly determination in step S4.

[0088] In step S5, the data processing device 1 determines a combination of multiple factor values ​​based on the determination results of the multiple factors in step S4 (e.g., as described later). Figure 5 In that case, the pair of factors F1 and F2 is determined, and it is determined which range of the distribution map in the corresponding table corresponds to the combination.

[0089] In step S6, the data processing device 1 confirms the combination of factor values ​​(described later) as the judgment result of step S5. Figure 9 The process involves checking whether the drawn graph matches a specific abnormal range in the corresponding table. In step S6, if the graph matches (Yes), proceed to the next step S7; otherwise, the process ends.

[0090] In step S7, the data processing device 1 reads the candidate causes and corresponding response information associated with the matching anomaly range region from the memory 202 based on the correspondence table. In the correspondence table, which is pre-set as the default by the manufacturer of the automatic analysis device 10, candidate causes and corresponding response information are associated with each anomaly range region.

[0091] In addition, the corresponding table may contain candidate cause information for each anomaly range and may not contain candidate cause information, and may also contain response information for each anomaly range and may not contain response information.

[0092] In step S8, the data processing device 1 uses the candidate causes and response information read in step S7 to generate output information for the operator, specifically generating screen data. The screen, for example, is used for... Figure 3 The data displayed on the display screen of the display device 206 may be in the form of, for example, a web page. For example, as described later... Figure 11 The screen is displayed accordingly. This screen contains information about the anomaly related to the measurement of the target sample, information on one or more potential causes based on the correspondence table, and response information. After step S8, the process ends. The above-described process is repeated for each target sample or each time an anomaly occurs. The operator uses the information from the screen in step S8 as guidance to perform tasks such as determining the cause of the anomaly and maintenance. As a result, when all causes of the anomaly are eliminated and the combination of factor values ​​falls within the normal range of the correspondence table, the data processing device 1 outputs a message indicating that the condition is normal on the screen.

[0093] [Anomaly Detection Function]

[0094] The automatic analysis device 10 has the function of determining / detecting abnormalities in the measurement / analysis process or results (sometimes referred to as an abnormality detection function). The data processing device 1 has the function of cooperating with the abnormality detection function of the automatic analysis device 10. Figure 2 When the control device 101 determines / detects a measured anomaly, it sends data containing anomaly determination result information to the data processing device 1 as an anomaly notification. The data processing device 1 stores the anomaly determination result data 24 based on the received data. Thus, the data processing device 1 performs... Figure 4 Such processing involves communicating the meaning of the abnormality to the operator through the screen and urging them to take action.

[0095] When the automatic analysis device 10 detects an abnormality during measurement, it activates the device 10 as needed (e.g., along with an abnormality notification) along with an abnormality notification. Figure 2 The operation of each drive mechanism is temporarily stopped. After the operator performs maintenance or other measures and the abnormality is eliminated, the control device 101 of the automatic analysis device 10 restarts the temporarily stopped device operation.

[0096] [Correspondence Table]

[0097] Figure 5This corresponds to the concepts and first structural example in Table 5. Table 5 specifically represents a two-dimensional table consisting of a pair of factors F1 and F2 as two factors. Here, the horizontal axis represents factor F1 as the first factor, and the vertical axis represents factor F2 as the second factor. For each factor (F1, F2), predetermined variables (described later) representing the reaction characteristics in the reaction process data are selected.

[0098] Each factor has a range of possible values. Factor F1 has a range of factor values ​​510. Range 510 is the range from the lower limit m11 to the upper limit m12. Each factor has a normal range and an abnormal range (in other words, the range outside the normal range). Range 510 has a normal range 511, indicated by the solid arrow, and two abnormal ranges 512 and 513, indicated by the dashed arrows. The normal range 511 is the range from the lower threshold h11 to the upper threshold h12. The lower abnormal range 512, as the first abnormal range, is the range from the lower limit m11 to the lower threshold h11. The upper abnormal range 513, as the second abnormal range, is the range from the upper threshold h12 to the upper limit m12. Similarly, factor F2 has a range of factor values ​​520. Range 520 is the range from the lower limit m21 to the upper limit m22. Range 520 has a normal range 521, indicated by the solid arrow, and two abnormal ranges 522 and 523, indicated by the dashed arrows. The normal range 521 is the range from the lower threshold h21 to the upper threshold h22. The lower abnormal range 522 is the range from the lower limit m21 to the lower threshold h21. The upper abnormal range 523 is the range from the upper threshold h22 to the upper threshold m22. Furthermore, each range is defined in a non-repeating manner, for example, a range above the first threshold and below the second threshold. When determining which range a value falls into, the specific range that it falls into is determined.

[0099] This correspondence table 5 considers the combination of normal / abnormal ranges of these factors F1 and F2 as the first combination, as shown in the figure, forming multiple combinations in a 2D region (in... Figure 5 In the example, there are 3×3=9 range regions. As multiple range regions, there is one normal range region 501 and multiple (8) abnormal range regions 502 (denoted as #1~#8) located around the normal range region 501. An abnormal range region is a region where at least one factor value is within an abnormal range.

[0100] In addition, Figure 5In the first structural example, there are abnormal ranges on both the upper and lower sides relative to the normal ranges (511, 521) of each factor (F1, F2), thus forming a total of 8 abnormal range regions. In other words, in this example, the normal range of each factor (F1, F2) lies between the lower and upper limits of the entire range (e.g., in the middle), and does not overlap with the lower or upper limits. Therefore, there are 2 abnormal ranges on both sides of the normal range.

[0101] Normal range region 501 is a 2-dimensional region consisting of the normal range 511 of factor F1 and the normal range 521 of factor F2. Multiple abnormal range regions 502 are assigned identification information (e.g., #1 to #8) to enable the identification of each abnormal range region. For example, abnormal range region #1 is a 2-dimensional region consisting of the abnormal range 512 of factor F1 and the abnormal range 523 of factor F2. Abnormal range region #2 is a 2-dimensional region consisting of the normal range 511 of factor F1 and the abnormal range 523 of factor F2. Abnormal range region #1 corresponds to the case where the values ​​of both factors (F1 and F2) are within the abnormal range. Abnormal range region #2 corresponds to the case where the value of factor F1 is within the normal range, and the value of factor F2 is within the abnormal range.

[0102] Furthermore, in the corresponding table 5, information about candidate causes related to the anomaly is associated with each anomaly range region 502 (#1 to #8). In this example, as shown in the illustration at the four corners of the anomaly range region, there are eight candidate causes: A, B, ..., G, and H. Although not illustrated, response information can be further associated with each candidate cause A to H. Each candidate cause and its response information is assigned identification information to enable identification.

[0103] Furthermore, there are different situations, such as different candidate causes in several abnormal range regions, and there are also similar situations, but in any case, they can be matched using the correspondence table 5. In this example, the correspondence table is represented as a 2-dimensional concept based on 2 factors, but it is not limited to this; it can also be considered as a 3-dimensional or higher concept based on 3 or more factors.

[0104] Here, the following points are important. Conventional methods for anomaly determination in automated analysis devices typically compare the values ​​of predetermined variables representing characteristics of the reaction process with conditions based on a predetermined threshold range to determine whether a state is normal or abnormal. That is, in the prior art, for each predetermined variable, only two general cases are considered: whether it is within the normal range (or the abnormal range). Furthermore, in the past, when a variable value was determined to be within the abnormal range, resulting in a determination of a measurement anomaly, multiple candidates for the cause of the anomaly existed, leading to ambiguity. Therefore, it was often difficult for the operator to determine the cause of the anomaly, or it often required considerable effort.

[0105] In contrast, in Implementation 1, as shown in Table 5 above, for example, a combination of two factors is considered, and multiple abnormal range regions 502 are considered in detail around the normal range region 501. Furthermore, in Implementation 1, for each abnormal range region 502, candidate causes related to the abnormality are considered. For each abnormal range region 502, even if the candidate causes differ, responses can be made through the associations in Table 5. Therefore, in Implementation 1, candidate causes and responses corresponding to the matching abnormal range regions can be suggested, making cause determination by the operator easier than before.

[0106] As another example, Figure 6 Similarly, a second structural example corresponding to Table 5 is shown. In Figure 6 In the second structural example of the corresponding table 5, a case is shown where the normal range of a factor overlaps with the lower limit of the range. Specifically, in this corresponding table 5, regarding factor F1, and Figure 5 Similarly, within range 610, there are abnormal ranges 612, normal ranges 611, and abnormal ranges 613. On the other hand, regarding factor F2, within range 620, there are normal ranges 621 and abnormal ranges 622. The normal range 621 overlaps with the lower limit m21 of range 620 and is the range from the lower limit m21 to the threshold h20. The abnormal range 622 is the range from the threshold h20 to the upper limit m22. Thus, this correspondence table 5 has one normal range region 601 and five abnormal range regions 602 (#1 to #5) located around the normal range region 601. Cause candidates (A to E) are associated in each abnormal range region 602. For example, abnormal range region #4 is composed of the abnormal range 612 of factor F1 and the normal range 621 of factor F2.

[0107] As another example, Figure 7 Similarly, the third structural example corresponding to Table 5 is shown. In Figure 7 In the third structural example of the correspondence table 5, the case where the normal range and the lower limit of the range overlap for both factors is shown. Specifically, in this correspondence table 5, for factor F1, there is a normal range 711 and an abnormal range 712 within range 710, divided by a threshold h10. For factor F2, there is a normal range 721 and an abnormal range 722 within range 720, divided by a threshold h20. Thus, this correspondence table 5 has one normal range region 701 and three abnormal range regions 702 (#1 to #3) located around the normal range region 701. Cause candidates (A to C) are associated with each abnormal range region 702.

[0108] Moreover, as a variation, the corresponding table is as follows: Figure 7As shown by the dashed lines, a certain abnormal range region can be further divided into multiple abnormal range regions as a sub-concept. Furthermore, information such as candidate causes can be associated with each of these abnormal range regions. For example, abnormal range region #3 is further divided into four abnormal range regions (#31 to #34). The abnormal range 712 of factor F1 is divided using a threshold h11. The normal range 721 of factor F2 is divided using a threshold h21. For example, abnormal range region #31 consists of the abnormal range from threshold h10 to threshold h11 of factor F1 and the normal range from threshold h21 to threshold h20 of factor F2.

[0109] In Implementation 1, the structure corresponding to Table 5 is set in advance as a design consideration and a fixed value in the data processing device 1. Figure 3 The control program 21, setting information 22, and correspondence table data 26, etc. are included. The manufacturer or other supplier of the automatic analysis device 10 pre-sets the correspondence table to the default value before the product leaves the factory. That is, the factors used, value ranges, thresholds, candidate causes associated with each abnormal range area, and response information in the correspondence table are pre-set. Regarding the setting of this correspondence table, it is not limited to fixed values. In the variant examples described later, the operator can variably set the information of the correspondence table on the device.

[0110] Furthermore, while the first combination of multiple factors constitutes multidimensional information, for ease of understanding when visualizing this information to the operator on a screen, it can also be displayed as two-dimensional information. That is, compared with... Figure 5 Similarly, data processing device 1 can, for example, generate and display a 2D distribution map for each pair of the two selected factors of interest.

[0111] Furthermore, when multiple factors exist that can be used to determine the cause of an anomaly, multiple correspondence tables can be prepared corresponding to the combinations of factors used. For example, if factors F1, F2, and F3 exist, a first correspondence table consisting of combinations of factors F1 and F2, a second correspondence table consisting of combinations of factors F3 and F1, and so on, can be prepared. The operator can select the factor or correspondence table to use from these multiple factors or multiple correspondence tables.

[0112] [Measured Value]

[0113] Measurement data 23 from an automated analysis device 10 for clinical examinations with functions such as biochemical analysis Figure 3 The data includes the measured values ​​at each time point. These "measured values" are values ​​with predetermined units, such as concentration and activity values, calculated based on absorbance data, and can be considered as one factor.

[0114] [Reaction process data]

[0115] The reaction process data included in the measurement data 23 from the automatic analysis device 10 will be explained. The reaction process data represents the time-varying changes in absorbance, etc., during the sample measurement from start to finish. The predetermined variables are variables that can be calculated based on the reaction process data to represent the characteristics of the reaction process.

[0116] Figure 8 This is an example of reaction process data 8. This reaction process data 8 represents the change in absorbance (denoted by the symbol A, in dimensionless quantities) of the sample over time. Figure 8 In the graph, the horizontal axis represents time (measurement time, denoted by T, in minutes), and the vertical axis represents absorbance A. The automatic analysis device 10 measures the absorbance A of the sample in the reaction vessel and calculates the concentration of the sample's components, etc., based on the absorbance A as the measured value. In the graph, the plotted points, represented by black dots, indicate the absorbance A values ​​measured at each time point. For example, at approximately 0 to 5 minutes (T1), the absorbance A is approximately a fixed value a1. At approximately 5 to 10 minutes (T2, the plotted point from the corresponding time point t1 to time point t2), the absorbance A increases approximately linearly from value a2 to value a3.

[0117] Line 801 is an asymptotic line related to the distribution of absorbance A, generated based on the plotted point of absorbance A at time T2. In other words, it is an approximate line with a positive slope. The slope of this asymptotic line (denoted as G) varies depending on the sample, analytical item, and reaction in the measurement, representing the characteristics of the reaction. The difference 802, indicated by the arrow, is the difference between the absorbance A value 803 at a certain time point tx and line 801. A large difference 802, i.e., a large deviation from the asymptotic line, suggests a possible measurement anomaly.

[0118] [factor]

[0119] Next, the factors will be explained. Factors consist of measured values ​​or variables from the reaction process data 8 as described above. In the example of Embodiment 1, variables are used as factors. Regarding which factors are used in the corresponding table 5, in Embodiment 1, default settings are pre-defined as described above.

[0120] Figure 9 It shows that Figure 8 The distribution map is constructed by combining two factors (F1, F2) that represent the reaction characteristics of the reaction process data 8. In the example of Implementation Method 1, factor F1 is set to... Figure 8 The slope G of the asymptotic line (line 801) is given by factor F2, which is set as the mean square error (denoted by E).

[0121] exist Figure 9In the distribution plot, the horizontal axis represents the slope G of the asymptotic line as factor F1, and the vertical axis represents the mean square error (E) as factor F2. The mean square error (E) is the value obtained by squared and summing the differences between the asymptotic line and absorbance A at each time point in time T2, such as difference 802. For example, it can be expressed as follows: Let the plotted point of absorbance A at time t be A(t). Let the difference 802 be D(t). Let each time point in time T2 be multiple (n) time points such as t = t1 to t2. E = ΣD(t) 2 The operator ∑ adds t = t1 to t2. Figure 9 The distribution plot of such factor value combinations is equivalent to Figure 5 The details of the first combination of the range of factor values, which is part of the corresponding Table 5.

[0122] In this example, both factors (F1 and F2) in the distribution chart can take the value 0. The range 910 for factor F1 is from the minimum value m11 to the maximum value m12, where m11 = 0. The range 920 for factor F2 is from the minimum value m21 to the maximum value m22, where m21 = 0. Furthermore, in this example, the normal ranges for factors F1 and F2 have a predetermined relationship, and the combination of factor values ​​constitutes the normal range region 901. This normal range region 901 is not... Figure 5 Such a simple rectangular region, in this example, is a sloping strip-shaped region extending from the lower left to the upper right, containing the respective lower limit value (0). In other words, the normal range region 901 is defined by a predetermined formula as a region where the values ​​of factor F1 and factor F2 have a predetermined linear relationship. In this way, the normal range region can take any shape depending on the preconditions, etc.

[0123] Therefore, in this example, the factor value distribution map has a normal range region 901 and three abnormal range regions 902 (#1 to #3) surrounding the normal range region 901. In the range 910 of factor F1, there are values ​​f11, f12, f13, and f14, for example. In the range 920 of factor F2, there are values ​​f21, f22, f23, and f24, for example. For example, when factor F1 is m11 = 0, factor F2 is within the normal range from 0 to f21, but becomes the abnormal range region #1 above f21. When factor F1 is f11, factor F2 is within the normal range from 0 to f23, and becomes the abnormal range region #1 above f23. When factor F1 is f12, factor F2 is the abnormal range region #3 from 0 to f24, and is within the normal range from f24 to the upper limit m22. When factor F1 is valued at f14, factor F2 falls within the abnormal range #3 from 0 to f22, and within the normal range from f22 to the upper limit m22. When factor F1 is valued at f13, factor F2 falls within the abnormal range #3 from 0 to f22, and within the abnormal range #2 from f22 to the upper limit m22.

[0124] In this example, the abnormal range region 902, located to the lower right of the normal range region 901, is divided into two abnormal range regions (#2, #3) based on the value f22 (equivalent to a threshold). Furthermore, in this distribution map, the thresholds distinguishing between normal and abnormal are, for example, represented by lines 931 and 932.

[0125] In this distribution map, the plotted points represented by black dots are plotted points based on the measured data 23, consisting of pairs of factors F1 and F2 values; in other words, combinations of factor values. For example, within the normal range region 901, there is a distribution of plotted point group 903 in the lower right (0) area. Plotted point group 903 corresponds to a situation judged as normal. Additionally, in this example, there is one plotted point 904 that deviates from plotted point group 903. Plotted point 904 is located within the abnormal range region #1. Plotted point 904 corresponds to a situation judged as abnormal.

[0126] [Examples of anomalies and candidate causes]

[0127] Figure 10 Indicates inclusion Figure 9 The factor value distribution plot shows a case study of data from the reaction process data at point 904 in the outlier range region #1. Figure 10 In the reaction process data 8, the slope (G) of the asymptotic line representing factor F1 is as follows: Figure 9The result is within the normal range (0-f11), but the mean square error (E) of factor F2 becomes a larger value than the normal range (around 0 to f22).

[0128] exist Figure 10 In the reaction process data 8, absorbance A deviates from the asymptotic line 1001 only within the range of 1000 (time T3 near time points ta to tb). That is, within this range of 1000, the difference 1003 between the asymptotic line and absorbance A is large. Under such circumstances, the reaction is considered to occur normally except within the range of 1000. Time T3 corresponding to the range of 1000 is considered to be an abnormal state. Therefore, within time T3 within the range of 1000, as a candidate for an abnormal cause, it is speculated that the photometer ( Figure 2 The optical axis of the optical measuring mechanism 16 (a part of the optical measuring mechanism) is temporarily blocked by a foreign object.

[0129] [Output]

[0130] As an example of information output by the data processing device 1 based on the corresponding table 5, an example of screen display will be explained. The data processing device 1 provides a screen containing specific information and a graphical user interface (GUI).

[0131] Figure 11 It is in having Figure 9 In the case of the factor value distribution chart corresponding to Table 5, when an anomaly matching the anomaly range #1 is detected, the screen example shows the candidate causes and response information associated with the anomaly range #1 being notified to the operator. Figure 11 The screen includes Message 1101, Table 1102, and Button 1103. Message 1101 contains information such as the object sample and analysis items, as well as information notifying the operator (e.g., "Abnormal data has occurred, please check the following candidate causes"). Table 1102 includes "Candidate Causes of Abnormalities" item 1104 and "Response Methods" item 1105. In the "Candidate Causes of Abnormalities" item 1104, the corresponding Table 5 is displayed (e.g., ...). Figure 9 The corresponding abnormal range area (the distribution map) Figure 9 Information on one or more cause candidates associated with the anomaly range area #1. In this example, there are three cause candidates (1111, 1112, 1113) associated with the anomaly range area #1, which are displayed separately as rows in the "Cause Candidates for Anomalies" item 1104. In the "Response Methods" item 1105, information on response methods associated with the cause candidates on the left is displayed.

[0132] In this example, regarding a certain exception ( Figure 9 The drawing point 904 Figure 10The range is 1000, and three candidate causes (1111-1113) are extracted and displayed. Multiple candidate causes can also be assigned identification information for display. In this example, these multiple (3) candidate causes do not have a specific order, but are displayed vertically. For example, candidate cause 1111 is "foreign matter in the sample container," candidate cause 1112 is "foreign matter in the reaction container," and candidate cause 1113 is "foreign matter in the reagent bottle." The solution for candidate cause 1111 is "Please check the condition of the sample. If there is foreign matter, please remove it and perform the measurement again." The solution for candidate cause 1112 is "Please clean the reaction container." The solution for candidate cause 1113 is "Please check the reagent bottle. If there is foreign matter, please remove the reagent bottle in use, replace it with a new reagent, and perform the measurement again."

[0133] Button 1103 is, for example, a confirmation button. The operator observes the information on the screen to confirm, understands the possible causes and solutions, determines the cause, and tries out the solution. As a result, if the abnormal state is eliminated, the screen displays a normal state, and the operator presses button 1103 to end the screen.

[0134] Additionally, in Table 1102, for example, check items can be set for each row corresponding to each candidate cause. If the operator confirms or trials each candidate cause and its corresponding response, a check mark can be recorded in the corresponding check item. Furthermore, in Table 5, the order and priority of the likelihood of multiple candidate causes within the anomaly range can be specified. In this case, in Table 1102, multiple candidate causes can be displayed by assigning numbers according to their order and priority.

[0135] It can also be limited to Figure 11 Example of a screen display. The data processing device 1 can display on the screen. Figure 5 The image corresponding to Table 5 can also be displayed. Figure 9 Such a factor value distribution plot. In other words, a factor value distribution plot is an image that visualizes the first combination of normal / abnormal ranges of multiple factors. Data processing device 1 displays... Figure 9 In such a distribution map, for example, colors can be used to distinguish and highlight the plotted points corresponding to the combination of factor values ​​of the currently detected anomaly and the anomaly range area corresponding to that plotted point.

[0136] In addition, the data processing device 1 can also be displayed on the screen. Figure 8The data processing device 1 can also display a graph of the calculated factor values ​​on the screen, for example, a graph with time on the horizontal axis and factor values ​​on the vertical axis. Furthermore, when the data processing device 1 displays the corresponding table 5 or distribution chart on the screen, it can display thresholds for dividing the normal / abnormal ranges or formulas corresponding to those thresholds, allowing for confirmation. Additionally, the thresholds or formulas can be variably set by users such as manufacturers. For example, the data processing device 1 displays... Figure 9 In the case of the distribution map, the value f22 corresponding to the threshold for dividing the abnormal range areas #2 and #3 is displayed. Users can update the settings by changing this value f22 or the corresponding line. This allows for adjustments related to multiple abnormal range areas.

[0137] [Effects, etc.]

[0138] As described above, the automatic analysis device and data processing device according to Embodiment 1 provide assistance by suggesting candidate causes of measurement anomalies based on a unique correspondence table 5, enabling operators to quickly or easily determine the cause of the measurement anomaly and improve the situation. According to Embodiment 1, even when multiple candidate causes exist for a measurement anomaly, the operator can quickly and easily determine the cause by providing information corresponding to the anomaly range region. This shortens the fault response time. In particular, even if the operator has limited experience with automatic analysis, the assistance provided by the output information from the data processing device can help with work, knowledge learning, and skill improvement. Conventionally, anomaly determination is limited to whether a single factor value is within the normal range. According to Embodiment 1, by combining multiple factor values, multiple anomaly range regions can be set outside the normal range region, and candidate causes corresponding to each anomaly range region can be suggested.

[0139] According to Implementation Method 1, reaction process data are specifically used as a combination of factors. Figure 9 The pair of slope (G) and mean square error (E) of the asymptotic line in the figure can effectively capture features related to measurement anomalies. Moreover, in this case, in the corresponding table 5 containing the factor value distribution plot, for example, as... Figure 9 As shown, multiple abnormal range regions 902 (#1 to #3) can be formed outside the normal range region 901, and each candidate cause corresponding to each abnormal range region 902 can be distinguished and captured.

[0140] As a variation of embodiment 1, the following variation can be adopted. In this variation, the data processing device 1 can be connected to... Figure 4 The process is roughly the same, but a new corresponding table (5) is generated based on the input data (especially) Figure 9(Such a factor value distribution chart). Then, the data processing device 1 displays the generated correspondence table 5 on the screen, where operators and manufacturers can input and set candidate causes and response information for each item in the correspondence table 5. For example, the data processing device 1 newly generates... Figure 5 When creating such a correspondence table, information (e.g., m11, h11, h12, m12) for specifying the normal / abnormal range of factor F1 and information (e.g., m21, h21, h22, m22) for specifying the normal / abnormal range of factor F2 are input based on user input. The data processing device 1 processes this information to generate... Figure 5 Such a distribution map includes normal range areas and multiple abnormal range areas. Then, the data processing device 1 sets the associated cause candidates and response information for each range of the distribution map based on user input.

[0141] [Variation Example 1]

[0142] As a variation of embodiment 1, the data processing device 1 has the function that allows the operator to variably set the information corresponding to Table 5. The operator can operate... Figure 3 The input device 205, while observing the screen of the display device 206, inputs new factors, candidate anomalies, and response information to set them in the correspondence table data 26. That is, the candidate anomalies and response information in the correspondence table 5 can be added to or modified by the operator.

[0143] However, it is desirable to distinguish between the information in the manufacturer's default mapping table and information added or modified by the operator (in other words, user input information) to avoid confusion. Therefore, the data processing device 1 manages the mapping table data 26 in a differentiated manner and displays the mapping table information on the screen. For example, the data processing device 1 stores / manages information from the input source (e.g., whether it is the manufacturer or the operator) for each item in the mapping table. For example, the data processing device 1 retains the default mapping table information without allowing deletion or changes. The data processing device 1 can also retain updated versions of the mapping table information if they exist.

[0144] Then, in data processing device 1, user input information, such as information added or modified by the operator, is separately maintained / managed in association with the default mapping table. When data processing device 1 displays the mapping table information on the screen, it distinguishes between the default information and the updated information provided by the operator, for example, by using different colors, displaying it in different columns, or recording the input source information together. Data processing device 1 may also display either the default information or the updated information provided by the operator on the screen, depending on the operator's selected operation. Alternatively, it may display both the default information and the updated information provided by the operator side-by-side on the screen.

[0145] Figure 12 This represents a screen example provided by the data processing device 1 in Modified Example 1. Figure 12 The screen has corresponding tables 1201 and 1202. Table 1201, for example, displays... Figure 9 In order to identify normal range areas and multiple abnormal range areas, the factor value distribution plot is displayed using identification information or color differentiation. Information such as range thresholds can also be displayed along with the corresponding table 1201. The operator can select an operating range area from the corresponding table 1201. Based on this operation, information associated with the selected range area is displayed in table 1202.

[0146] Figure 11 Table 1102 is fixed and cannot be corrected by the operator. In contrast, Figure 12 Table 1202 has a GUI that allows the operator to add and modify information. In Table 1202, the operator can add text or images to each item. For example, in the row for cause candidate 1112, the operator can add notes, URLs, or other text 1112b to the response method item. An update button 1203 is provided on the screen. The operator presses the update button 1203 (or the confirmation button) to update the information in Table 1202 to the current input state. This saves the updated version of Table 5. Additionally, an append button 1204 is provided on the screen. The operator presses the append button 1204 to add cause candidates and response information. This adds a new row to Table 1202. The operator enters cause candidates and response information in this row and presses the update button 1203 to update it.

[0147] According to Variation 1, the operator can freely set the correlation between the abnormal range of factors, candidate causes, and response information. As an advantage of this, the following benefits can be cited: The operator does not need to wait for the manufacturer to update the correspondence table; the correspondence table can be easily updated to the latest information corresponding to the usage status and history of the automatic analysis device 10. The operator can record their own thoughts, notes, etc., in the correspondence table. In cases where a specific automatic analysis device 10 has an inherent state, such as when a facility equipped with the automatic analysis device 10 exhibits unique phenomena, the operator can record information that is reflected in the correspondence table. These phenomena include, for example, foreign matter contamination from the sample due to insufficient pre-treatment of the sample (e.g., centrifugation), or abnormal reaction processes due to insufficient maintenance. Therefore, the operator can perform the operation more easily.

[0148] [Variation Example 2]

[0149] The factors used in Table 5 are not limited to the examples mentioned above. Other examples of factors include the following.

[0150] Figure 13 This example illustrates reaction process data 8 in Modified Example 2. This example is an illustration of reaction process data 8 in an endpoint method (also described in Patent Document 1) used in the determination / analysis of biological samples. In this case, the time change of absorbance A in reaction process data 8 is not... Figure 8 Instead of a straight line, the absorbance A becomes a curve, as approximated by Equation 1301 in time T4. In approximation Equation 1301, the absorbance A increases sharply from 5 minutes to time point tx (time T5), the slope gradually decreases, and after time point tx, it becomes a straight line with a slope of 0. Difference 1302 is the difference in absorbance A from the addition of the second reagent (approximately 5 minutes) to the end of the reaction (approximately 10 minutes). [The last sentence appears to be incomplete and possibly refers to a different formula.] Figure 8 Similarly, difference 1303 is the difference between absorbance A at a certain time point and approximation 1301.

[0151] For example, as described in Patent Document 1, approximation 1301 can use a logic curve. In this curve, when the reaction time has elapsed to a certain extent, the absorbance A approaches a certain value (A0 + A1). Approximation 1301 can be calculated, for example, by the least squares method, using A = A0 + A1(1 - e^(-A0 + A1)). ﹣kt ) represents the absorbance at the start of the reaction, A0 is the absorbance at the end of the reaction, A1 is the absorbance at the end of the reaction, k is the reaction rate constant, and t is time.

[0152] Figure 14 Indicates and Figure 13 An example of a factor value distribution plot corresponding to reaction process data 8. In Figure 13In the case of reaction process data 8, the factor F1 can be set as the final reaction absorbance change based on approximation formula 1301 (in other words, the approximation curve). The final reaction absorbance change is the change in absorbance for each time interval in approximation formula 1301; in other words, it is the slope of the approximation curve at each time point. Specifically, the final reaction absorbance change is close to... Figure 13 The change in a certain value (A0+A1) over time T5. Similarly, factor F2 can be set as the mean squared error (E). In the combination of these factors, it is possible to generate... Figure 14 That kind of distribution map. In Figure 14 The distribution map shows a sloping, band-shaped normal range region 1401 and three abnormal range regions 1402 (#1 to #3) surrounding the normal range region 1401. Using a threshold value of f41, the two abnormal range regions #1 and #2 located to the upper left of the normal range region 1401 have an upper abnormal range region #1 and a lower abnormal range region #2. There is one abnormal range region #3 to the lower right of the normal range region 1401. Figure 9 Similarly, the thresholds for the normal / abnormal range are represented by lines 1431 and 1432. These abnormal range regions 1402 can be associated with their respective causal candidates, etc. The plotted point 1404, as a combination of factor values, is an example of an anomaly conforming to abnormal range region #1.

[0153] Figure 15 Is Figure 14 The factor value distribution plot includes an example of reaction process data 8, which corresponds to plotting point 1404 in the outlier range region #1. Figure 13 Similarly, the temporal variation of absorbance A at time T6 is represented by a curve approximating formula 1501. The range 1500 in time T7 corresponds to the portion of the final reaction approaching a certain value, representing the range of plotted points that deviate from approximation formula 1501. This range 1500 and... Figure 10 The example with a range of 1000 is similar, equivalent to an exception.

[0154] exist Figures 13-15 Given the reaction process data 8 and the corresponding table 5 for the combination of factors, it is also possible to... Figure 11 The output in that kind of image is, for example, with Figure 11 The same candidate causes and corresponding information are available. As mentioned above, the approximation related to the reaction process is not limited to a straight line, but can also be a curve, etc. By using variables such as the slope that constitute such an approximation as factors, it is possible to capture the characteristics of the reaction well and distinguish and capture a variety of abnormal causes.

[0155] [Variation Example 3]

[0156] exist Figure 16 The example of reaction process data 8, as a variation of Example 3, shows an example of data including an anomaly range 1600. The time variation of absorbance A over a period of approximately 5 to 10 minutes is represented by the asymptote 1601, which has a slope (G). At time T8, the plotted points for absorbance A have a roughly constant slope. The range 1600 at time T9 is the range of plotted points that deviate from the asymptote 1601, corresponding to an anomaly.

[0157] Figure 17 This example illustrates the factor value distribution plot in Variation Example 3. In this example, the combination of factors is that factor F1 is the measured value, and factor F2 is... Figure 8 The slope (G) of the same asymptotic line. The measured value of factor F1 is, for example, the concentration of the component calculated based on absorbance. The unit of concentration is, for example, [mg / L].

[0158] exist Figure 17 In the distribution diagram, the combination of factors F1 and F2 values ​​constitutes a normal range region 1701 and two abnormal range regions 1702 (#1, #2). Abnormal range region #1 is located to the upper left of the normal range region 1701, and abnormal range region #2 is located to the lower right. The thresholds for the normal / abnormal ranges are represented by lines 1731 and 1732. Within the normal range region 1701, Figure 16 The plotted points are distributed linearly. The plotted point 1704, representing the combination of factors F1 and F2, falls within the anomaly range #1 and has the same characteristics as... Figure 16 The relationship between the data in the range of 1600.

[0159] Based on the measured values ​​of the measurement data 23, the data processing device 1 outputs information such as candidate causes associated with the abnormal range area #1 on the screen based on the corresponding table 5, when the abnormality matches the abnormal range area #1.

[0160] Figure 18 Indicates in Figure 17 Example of the output screen under abnormal conditions of drawing point 1704. Figure 18 The structure of the picture and Figure 11 Same. In this example, two cause candidates are shown in the "Potential Causes of Anomalies" section of Table 1102. Cause candidate 1801 is "Out of the measurable range of the reagent". Cause candidate 1802 is "Calibration failed". Figure 17 This is the reaction process observed under the condition of "exceeding the measurable range of the reagent" in cause candidate 1801. Figure 17This is not equivalent to "calibration failure" for cause candidate 1802, but sometimes it is plotted in the abnormal range area #1, as in the example of plotting point 1704, in cases where a liquid with high absorbance was incorrectly measured in a calibration blank measurement. Cause candidates 1801 and 1802 can be identified based on the presence or absence of alarm outputs. The "Response Method" section displays the response method associated with each cause candidate. The response method for cause candidate 1801 is "Please verify the reaction process and re-measure under reduced conditions if necessary." The response method for cause candidate 1802 is "Please verify the calibration results. Please recalibrate if necessary."

[0161] As mentioned above, even when measured values ​​such as concentration or activity value are used as factors, it can capture the characteristics of the reaction well and distinguish and capture a variety of anomalies and causes.

[0162] The present invention has been specifically described above based on the embodiments, but the present invention is not limited to the above embodiments, and various modifications can be made without departing from its spirit.

[0163] Symbol Explanation

[0164] 1 Data processing device

[0165] 5 Correspondence Table

[0166] 8. Reaction process data

[0167] 10 Automatic Analysis Device

[0168] F1 and F2 factors

[0169] 901 normal range area

[0170] 902(#1~#3) Abnormal range area.

Claims

1. A data processing apparatus, characterized in that, The data processing device includes: The data receiving unit receives data related to the sample measurements performed by the automatic analysis device through communication with the automatic analysis device. The data processing unit, based on the reaction process data in the data, uses predetermined variables or measured values ​​representing the characteristics of the reaction as factors, calculates combinations of multiple factor values, makes a judgment on the normal / abnormal correlation of the multiple factors, and refers to a correspondence table to determine which abnormal range region the combination of multiple factor values ​​falls into. The correspondence table has a distribution map containing normal range regions and multiple abnormal range regions, composed of combinations of normal / abnormal ranges of the multiple factors, and information on candidate abnormal causes associated with each abnormal range region. The distribution map, at least through a combination of the normal and abnormal ranges of the first factor and the normal and abnormal ranges of the second factor, has a normal range region formed by the normal ranges of the first factor and the normal ranges of the second factor, a first abnormal range region formed by the normal ranges of the first factor and the abnormal ranges of the second factor, a second abnormal range region formed by the abnormal ranges of the first factor and the normal ranges of the second factor, and a third abnormal range region formed by the abnormal ranges of the first factor and the abnormal ranges of the second factor; and The data output department outputs information to the user about candidate anomaly causes associated with the corresponding anomaly range.

2. The data processing apparatus according to claim 1, characterized in that, The corresponding table sets corresponding response information for each of the abnormal range regions or each of the abnormal cause candidates.

3. The data processing apparatus according to claim 1, characterized in that, The correspondence table is pre-set with default settings provided by the provider of the automatic analysis device.

4. The data processing apparatus according to claim 1, characterized in that, The information in the corresponding table is displayed on the screen, and the information of the candidate abnormal causes is added or corrected based on the user's operation on the screen to update the corresponding table.

5. The data processing apparatus according to claim 1, characterized in that, The reaction process data represents the change in absorbance over time. The variable that is set as one of the factors is a variable that constitutes an approximation in the reaction process data.

6. The data processing apparatus according to claim 1, characterized in that, The reaction process data represents the change in absorbance over time. As one of the factors, the variable is the mean square error, which is related to the difference between the approximation and absorbance in the reaction process data.

7. The data processing apparatus according to claim 1, characterized in that, The reaction process data represents the change in absorbance over time. The measured value, which is set as one of the factors, uses the concentration or activity value of the component of the sample.

8. The data processing apparatus according to claim 1, characterized in that, The distribution map of the corresponding table is displayed on the screen.

9. An automatic analysis device, characterized in that, The automatic analysis device includes: The interface circuit, through communication with the data processing device, sends data related to the determination of the sample to the data processing device. The control device, based on the reaction process data in the data, uses predetermined variables or measured values ​​representing the characteristics of the reaction as factors, calculates combinations of multiple factor values, makes judgments regarding the normal / abnormal correlation of the multiple factors, and refers to a correspondence table to determine which abnormal range region the combination of multiple factor values ​​falls into. The correspondence table has a distribution map containing normal range regions and multiple abnormal range regions, composed of combinations of normal / abnormal ranges of the multiple factors, and information on candidate abnormal causes associated with each abnormal range region. The distribution map, at least through a combination of the normal and abnormal ranges of the first factor and the normal and abnormal ranges of the second factor, has a normal range region formed by the normal ranges of the first factor and the normal ranges of the second factor, a first abnormal range region formed by the normal ranges of the first factor and the abnormal ranges of the second factor, a second abnormal range region formed by the abnormal ranges of the first factor and the normal ranges of the second factor, and a third abnormal range region formed by the abnormal ranges of the first factor and the abnormal ranges of the second factor; and An output device that outputs to the user information on candidate anomaly causes associated with a defined range of anomalies.

Images