Characterization system, automated analyzer, and characterization method

Abstract

This invention provides a characterization system, an automated analyzer, and a characterization method capable of evaluating the characteristics of a sensor used for sample inspection in an automated analyzer. [Solution] In a detection process including an electrochemical reaction, parameters representing the relationship between the state of sensor 11 and the amount of change in electrical signal data due to a change in the state of sensor 11 are stored from the electrical signal data acquired by sensor 11. A state estimation index for sensor 11 is calculated using the electrical signal data and parameters, and statistical values ​​of the state estimation index are calculated. In the detection process, an evaluation target state estimation index for sensor 11a is calculated using the evaluation target electrical signal data and parameters output from the evaluation target sensor 11a. The evaluation target state estimation index and statistical values ​​are compared to evaluate the characteristics of the evaluation target sensor 11a, and the characteristic evaluation results are output.

Description

Technical Field

[0001] The present invention relates to a characteristic evaluation system, an automatic analyzer, and a characteristic evaluation method for a sample inspection sensor used in a medical automatic analyzer using an electrochemiluminescence method.

Background Art

[0002] Patent Document 1 describes a diagnostic system for diagnosing a sensor provided in an automatic analyzer that outputs an analog electrical signal, including a memory that stores data of the electrical signal output by the sensor and the sensor replacement history, and a processing device that processes the data recorded in the memory. The processing device reads out from the memory the data of the electrical signal output by a past sensor used in the automatic analyzer during a set reference period, calculates the statistical value of the data for the reference period, reads out from the memory the data recorded during a set evaluation period among the data of the electrical signal output by the diagnostic target sensor during use in the automatic analyzer, calculates the statistical value of the data for the evaluation period, and determines the abnormality of the diagnostic target sensor based on the difference obtained from the statistical value for the reference period and the statistical value for the evaluation period.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] The state of the sample inspection sensor of an automatic analyzer is not uniquely determined by the usage environment, usage method, etc., and generally does not appear in a form that can be easily confirmed visually. Therefore, when an abnormality occurs in the automatic analyzer, there has generally been a lack of means for an inspector to immediately determine whether to pick up the sample inspection sensor as an inspection item among more than a thousand types of parts.

[0005] Patent Document 1 discloses a diagnostic system that determines sensor abnormalities from electrical signal data output from a sensor of an automated analyzer.

[0006] In automated analyzers, the measurement of trace components at levels of micromoles / liter (μmol / L) or less is generally required, making it essential to improve the diagnostic accuracy of sensors.

[0007] By using the diagnostic system described in Patent Document 1, it becomes possible to detect abnormalities in sensors using electrical signals output from the sensors, thereby improving diagnostic accuracy.

[0008] However, because sensors are exposed to a wide variety of operating environments and usage methods, it has become clear that diagnostic systems aimed at distinguishing between normal and abnormal conditions still need further refinement to detect the possibility that a sensor may be in a unique state compared to a typical sensor under similar operating conditions, even if the electrical signal data appears normal at first glance.

[0009] Therefore, the present invention aims to provide a characterization system, an automated analyzer, and a characterization method capable of evaluating the characteristics of a sample inspection sensor in an automated analyzer. [Means for solving the problem]

[0010] The present invention includes multiple means for solving the above problems, but to give one example, a characteristic evaluation system for evaluating the characteristics of a sample inspection sensor provided in an automated analyzer that inspects a sample using an electrochemiluminescence method, comprising: a parameter database that stores parameters representing the relationship between the state of the sample inspection sensor and the amount of change in the electrical signal data due to a change in the state of the sample inspection sensor, from among the electrical signal data acquired by the sample inspection sensor in a detection process including an electrochemical reaction; a state estimation index calculation processor that calculates a state estimation index of the sample inspection sensor using the electrical signal data and the parameters; a statistical value calculation processor that calculates a statistical value of the state estimation index; an evaluation target state estimation index calculation processor that calculates an evaluation target state estimation index of the sample inspection sensor using the evaluation target electrical signal data output from the sample inspection sensor to be evaluated and the parameters in the detection process; a characteristic evaluation processor that evaluates the characteristics of the sample inspection sensor to be evaluated by comparing the evaluation target state estimation index and the statistical value; and a data output unit that outputs the characteristic evaluation result by the characteristic evaluation processor. [Effects of the Invention]

[0011] The present invention provides a system that can evaluate the characteristics of a sample inspection sensor for an automated analyzer. Other problems, configurations, and effects will be clarified by the following description of the embodiments. [Brief explanation of the drawing]

[0012] [Figure 1] A schematic plan view showing one example configuration of an automated analyzer to which the characterization system according to the first embodiment is applied. [Figure 2] Schematic diagram of the sensor used for sample testing in the automated analyzer shown in Figure 1. [Figure 3] A schematic diagram of a sensor installed in an automated analyzer that uses the electrochemiluminescence method shown in Figure 1 to inspect a sample. [Figure 4]A timing chart showing an example of the detection process in the automatic analyzer shown in FIG. 1. [Figure 5] A block diagram showing the processing flow in the automatic analyzer shown in FIG. 1. [Figure 6] An example of time-series data of the luminescence intensity measured during sample measurement. [Figure 7] An example of time-series data of the voltage value measured during sample measurement. [Figure 8] An example of time-series data of the current value measured during sample measurement. [Figure 9] A block diagram showing the processing flow in the preparation for characteristic evaluation in the characteristic evaluation system according to the first embodiment. [Figure 10] A diagram showing the correlation between the characteristics of the sensor and the characteristic estimation index. [Figure 11] A block diagram showing the processing flow in the execution of characteristic evaluation in the characteristic evaluation system according to the first embodiment. [Figure 12] A block diagram showing the data processing flow in the characteristic evaluation system according to the second embodiment. [Figure 13] A diagram showing an example of the screen display of the characteristic evaluation result in the characteristic evaluation system according to the third embodiment. [Figure 14] A diagram showing an example of the screen display of the characteristic evaluation result in the characteristic evaluation system according to the third embodiment. [Figure 15] A diagram showing an example of the list display of the characteristic evaluation results of a large number of sensors in the characteristic evaluation system according to the fourth embodiment. [Figure 16] A diagram showing an example of the list display of the characteristic evaluation results of a large number of sensors in the characteristic evaluation system according to the fourth embodiment. [Figure 17A] A diagram showing an example of the display method of the characteristic evaluation result in the characteristic evaluation system according to the fifth embodiment. [Figure 17B] A diagram showing an example of the display method of the characteristic evaluation result in the characteristic evaluation system according to the fifth embodiment. [Figure 18] A diagram showing an example of the status monitoring screen of each component of the automatic analyzer in the characteristic evaluation system according to the fifth embodiment. [Figure 19]A diagram showing an example of a maintenance screen in the characteristic evaluation system according to the fifth embodiment. [Figure 20] This figure shows an example of the alarm list screen in the characteristic evaluation system according to the fifth embodiment. [Figure 21] A diagram showing the real-time presentation of characteristic evaluation results in the characteristic evaluation system according to the fifth embodiment. [Modes for carrying out the invention]

[0013] Embodiments of the characterization system, automated analyzer, and characterization method of the present invention will be described below with reference to the drawings. In the drawings used herein, the same or corresponding components are denoted by the same or similar reference numerals, and repeated descriptions of these components may be omitted.

[0014] (overview) The characteristic evaluation system, automated analyzer, and characteristic evaluation method according to this embodiment are techniques for evaluating the characteristics of the sensor of an automated analyzer.

[0015] A sample testing sensor (hereinafter referred to as "sensor") equipped in an automated analyzer is, for example, a sensor used to measure analytical parameters of biological samples, such as those found in immunoassay analyzers, biochemical analyzers, blood coagulation time analyzers, and ISE analyzers.

[0016] An example of such a sensor is a flow cell type sensor for electrochemiluminescence measurement. As will be described in detail later, a flow cell type sensor is composed of, for example, a flow cell, electrodes (working electrode, counter electrode, and reference electrode) provided on the flow cell, and a photoelectric conversion sensor (such as a photomultiplier tube) positioned on the opposite side of the flow cell from the working electrode.

[0017] This type of sensor outputs not only measured values ​​of the analytical parameters of a sample indicating the concentration of the target component (for example, a signal value indicating the luminescence intensity detected by a photoelectric conversion sensor), but also analog electrical signal data (current value and voltage value) generated between the counter electrode and the working electrode when a voltage is applied between the working electrode and the reference electrode, as response values.

[0018] In automated analyzers, sensor outputs (analog electrical signal data output from the sensor) are generated at multiple timings during the measurement cycle of the same sample, such as during sample measurement, sensor cleaning, and electrode conditioning.

[0019] The liquid channel for transporting biological samples, reagents necessary for measurement, washing solutions, etc., to the above-mentioned sensor consists of a nozzle for drawing in and discharging liquid, piping for passing liquid, etc. Other components of the liquid transport system include, for example, multiple valves for opening and closing the channel, and syringes for creating a pressure difference in the channel to draw in and discharge liquid.

[0020] The sample includes at least one (or more) of the following: patient specimens (blood, urine, etc.), calibration samples (standard samples), QC (quality control) samples, and dummy samples. Calibration samples are prepared samples measured to create a calibration curve during the calibration of the automated analyzer. QC samples are prepared samples measured during the QC of the automated analyzer. Dummy samples are predetermined samples measured as a preparatory step before measuring patient specimens.

[0021] After extensive research by the inventors, it was discovered that electrical signal data changes depending on the state of the sensor.

[0022] For example, it was found that electrical signal data changes with continued use of a sensor, and that the way it changes differs depending on how the sensor is used and the environment in which it is used. Based on this, it was discovered that by using machine learning methods such as regression models to pre-train parameters that represent the relationship between the sensor state and the amount of change in electrical signal data at that time, it is possible to calculate a state estimation index that estimates the state of the sensor from the electrical signal data obtained from the sensor being evaluated.

[0023] Furthermore, by comparing the calculated state estimation index with that of many other sensors, it becomes possible to evaluate the characteristics of the sensor being evaluated.

[0024] The sensor characterization method described above is not necessarily limited to processing by some kind of computing device (e.g., a processor), but can also be implemented as a method. For example, a user could look at electrical signal data and judge the sensor's characteristics based on changes in that data.

[0025] The present invention can be applied to automated analyzers. Examples of detection units mounted on automated analyzers include biochemical analyzers and immunoassay analyzers. However, this is merely an example, and the present invention is not limited to the embodiments described below, but can be broadly applied to automated analyzers equipped with a detection unit that analyzes a sample using an electrochemiluminescence method based on reagents and reaction results. For example, automated analyzers equipped with mass spectrometers used in clinical tests or coagulation analyzers that measure blood coagulation time may also be included as applicable.

[0026] Furthermore, the present invention is also applicable to a combined automatic analyzer equipped with multiple types of these detection units, as well as to an automatic analyzer that includes at least one automatic analyzer.

[0027] <First Embodiment> A first embodiment of the characterization system, automated analyzer, and characterization method of the present invention will be described with reference to Figures 1 to 11.

[0028] -Automatic analyzer- First, the overall configuration of the automated analyzer will be explained using Figure 1. Figure 1 is a schematic plan view showing one example of the configuration of an automated analyzer to which the component abnormality detection system according to the first embodiment is applied.

[0029] The automated analyzer 1 shown in Figure 1 comprises a transport line 2, an incubator (reaction disk) 3, a first transport mechanism 4, a sample dispensing nozzle 6, a reagent disk 7, a reagent dispensing nozzle 8, a second transport mechanism 9, a detection unit 10, a controller 21, an operating device 22, and a control device 30.

[0030] The transport line 2 is a device that transports the rack R, and transports the rack R to the sample dispensing position by the sample dispensing nozzle 6. Multiple sample containers C1 for holding samples can be installed on the rack R. In the example in Figure 1, a configuration in which samples are transported along a line is illustrated, but in some cases, a disc-shaped transport unit that rotates to transport the samples may be provided.

[0031] Incubator 3 is a rotary table-shaped device that houses reaction vessels C2, which are responsible for containing the reaction solution. Multiple reaction vessels C2 can be arranged in a ring shape. Incubator 3 is driven by a drive device (not shown) and rotates, allowing any of the reaction vessels C2 to be moved to any of several predetermined positions, such as the dispensing position by the sample dispensing nozzle 6.

[0032] The first transport mechanism 4 is a device for transporting sample dispensing tips T and reaction vessel C2. This first transport mechanism 4 is movable in the three axes XYZ along a rail and transports sample dispensing tips T and reaction vessel C2 between the incubator 3, stirring mechanism M, disposal position D, tip mounting position P, and tray 5. The stirring mechanism M is a device for stirring the sample contained in reaction vessel C2. Disposal position D is a position equipped with a disposal hole for discarding used sample dispensing tips T and reaction vessel C2. Tip mounting position P is a position for mounting sample dispensing tips T onto the sample dispensing nozzle 6.

[0033] Tray 5 is a container that holds multiple unused sample dispensing tips T and reaction vessels C2. Unused reaction vessels C2 picked up from Tray 5 are placed in a designated position in Incubator 3 by the first transport mechanism 4. Similarly, unused sample dispensing tips T picked up from Tray 5 are transported by the first transport mechanism 4 and placed in the tip mounting position P.

[0034] The sample dispensing nozzle 6 is a device for aspirating and dispensing a sample. This sample dispensing nozzle 6 is configured to rotate and move up and down. The nozzle tip is moved above the tip mounting position P and lowered, and the sample dispensing tip T prepared at the tip mounting position P is pressed into the nozzle tip and attached. Once the sample dispensing tip T is attached to the nozzle tip, the sample dispensing nozzle 6 moves its nozzle tip above the sample container C1 installed in the rack R and lowers it, aspirating a predetermined amount of sample from the sample container C1. After aspirating the sample, the sample dispensing nozzle 6 moves its nozzle tip above the incubator 3 and lowers it, dispensing the sample into an unused reaction vessel C2 installed in the incubator 3. When the dispensing of the sample is complete, the sample dispensing nozzle 6 moves its nozzle tip above the disposal position D and discards the used sample dispensing tip T into the disposal hole.

[0035] The reagent disk 7 is a rotating table-shaped device on which multiple reagent containers C3 are placed. The top of the reagent disk 7 is covered by a disk cover 7a (shown partially broken in Figure 1), and the inside is kept warm at a predetermined temperature. The disk cover 7a has an opening (omitted for illustrative purposes) at a reagent aspiration position 7b set near the incubator 3.

[0036] The reagent dispensing nozzle 8 is a device for aspirating and dispensing reagents. Like the sample dispensing nozzle 6, the reagent dispensing nozzle 8 can rotate and move up and down. It moves its nozzle tip to the reagent aspiration position 7b on the reagent disk 7 and lowers it, aspirating a predetermined amount of reagent from a predetermined reagent container C3 that has been transported to the reagent aspiration position 7b. Next, the reagent dispensing nozzle 8 lifts its nozzle tip from the reagent container C3, moves it to a predetermined position on the incubator 3 and lowers it, dispensing the reagent into the reaction vessel C2 containing the sample that has been transported to this position.

[0037] The reaction vessel C2, into which the sample and reagents have been injected, is transported to a predetermined position by the rotation of the incubator 3 and then transferred to the stirring mechanism M by the first transport mechanism 4. The stirring mechanism M mixes the sample and reagents inside the reaction vessel C2 by, for example, rotating the reaction vessel C2. After stirring is complete, the reaction vessel C2 is again transferred to a predetermined position on the incubator 3 by the first transport mechanism 4.

[0038] The second transport mechanism 9 is a device that transfers the reaction vessel C2 between the incubator 3 and the detection unit 10, and is configured to rotate and move up and down. This second transport mechanism 9 picks up the reaction vessel C2 containing the reaction solution, which has been returned to the incubator 3 after mixing the sample and reagents and after a predetermined reaction time has elapsed, and transfers it to the detection unit 10.

[0039] The detection unit 10 is a measuring instrument that measures specific biological components, chemical substances, and other measurement items contained in the reaction solution inside the reaction vessel C2.

[0040] The controller 21 is a computer attached to (forming a unit with) the mechanism 91 of the automatic analyzer 1 (the disk, transport mechanism, dispensing nozzle, detection unit 10, etc., as described above). The controller 21 controls the mechanism 91 of the automatic analyzer 1 in response to signals input from the operating device 22 and signals input from the control device 30 in response to user operations.

[0041] The control device 30 is a computer comprising a storage device 31 such as RAM, ROM, HDD, or SSD, a processor 32 such as a CPU, and is connected to the mechanical unit 91 of the automatic analyzer 1 via the controller 21. The control device 30 controls each component of the mechanical unit 91 of the automatic analyzer 1, and records and processes data input from the detection unit 10, etc.

[0042] The control device 30 may, for example, form a unit with the mechanism 91 or controller 21 of the automatic analyzer 1, or it may be installed separately from the mechanism 91 of the automatic analyzer 1 and connected directly to the controller 21 by wire or wireless connection.

[0043] In this embodiment, the control device 30 is connected to the server 40 via a communication interface 33, a network NW, and a data acquisition interface 43. The server 40 is also a computer configured with a storage device 41 such as RAM, ROM, HDD, or SSD, and a processor 42 such as a CPU. In this embodiment, the sensor 11 characteristic evaluation system is mounted on the server 40. The server 40 records the electrical signal data output by the sensor 11 of the automatic analyzer 1 in the storage device 41, and processes the data recorded in the storage device 41 with the processor 42.

[0044] The control processes for execution described below are performed by programs, but they may be combined into a single program, divided into multiple programs, or even a combination of these.

[0045] Some or all of the programs contained within each device may be implemented using dedicated hardware, or they may be modularized. Furthermore, various programs may be installed on each device via a program distribution server or external storage media, or existing devices may be updated.

[0046] Furthermore, each device may be an independent device connected by a wired or wireless network, or two or more devices may be integrated into a single unit.

[0047] -Liquid Conveying System- Figure 2 is a schematic diagram of the liquid transport system installed in the automated analyzer shown in Figure 1.

[0048] As shown in Figure 2, the detection unit 10 of the automatic analyzer 1 is equipped with a flow cell type sensor 11, a liquid transport system 12, and a turntable 13. Here, the configuration of the turntable 13 and the liquid transport system 12 will be described.

[0049] The turntable 13 is the section where the auxiliary reagent container RG for storing auxiliary reagents and the detergent container CL for storing washing solution are installed, and is equipped with a standby position SP and a reaction vessel installation position SM.

[0050] The auxiliary reagent is a chemical solution used to induce a luminescence reaction of the reaction product in the reaction solution. The washing solution is a liquid used to clean the flow path of the liquid transport system 12 and the flow cell FC of the sensor 11.

[0051] The reaction vessel C2, which is transferred from the incubator 3, is placed in the reaction vessel installation position SM. The turntable 13 is equipped with a drive device (not shown) and rotates and moves up and down when driven by a drive device controlled by a signal from the controller 21. The turntable 13 is used to transport, for example, the reaction vessel C2, detergent container CL, or auxiliary reagent container RG to the liquid suction position of the liquid transport system 12 in an appropriate time, or to align it with the standby position SP.

[0052] The liquid transport system 12 consists of flow paths F1, F2, F3, F4, F5, F6 through which the liquid passes, a plurality of valves V1, V2 for opening and closing these flow paths F1, F2, F3, F4, F5, F6, and a syringe SY that creates a pressure difference in the flow paths F1, F2, F3, F4, F5, F6 for aspirating and discharging the liquid.

[0053] Flow path F1 is a flow path that sends the aspirated liquid to the sensor 11. It consists of a nozzle for aspirating and releasing the liquid and a pipe for passing the liquid, with the other end connected to the sensor 11. Flow path F2 is connected to the sensor 11 on the opposite side of flow path F1 and connects the sensor 11 to valve V1. Flow path F3 connects valves V1 and V2. Flow path F4 connects valve V2 to a drain tank (not shown). Flow path F5 branches off from flow path F3 and connects flow path F3 to syringe SY. Flow path F6 connects syringe SY to a system water supply pump (not shown).

[0054] Valves V1 and V2 are, for example, solenoid valves. Normally open solenoid valves can be used, but in this embodiment, normally closed solenoid valves are used.

[0055] For example, if a reaction vessel C2, auxiliary reagent container RG, or detergent container CL is transported to the suction position of the liquid transport system 12, and valve V1 is opened with valve V2 closed to drive the syringe SY to suction, liquid is drawn from the container such as reaction vessel C2 via the suction nozzle. As a result, the liquid is drawn into the sensor 11 via flow path F1, and further into flow paths F3 and F5. Alternatively, if valve V1 is opened with valve V1 closed and the syringe SY is driven to discharge, the liquid drawn into flow paths F3 and F5 is discharged into the drain tank.

[0056] -Sensor for sample inspection- Figure 3 is a schematic diagram of a sensor 11 provided in an automated analyzer that uses the electrochemiluminescence method shown in Figure 1 to inspect a sample. The detection unit 10 of the automated analyzer 1 is equipped with a flow cell type sensor 11.

[0057] Sensor 11 is the part that applies a voltage to a reaction solution obtained by antigen-antibody reaction between a sample and a reagent to cause an electrochemical reaction, and includes a flow cell FC, three electrodes (working electrode E1, counter electrode E2, and reference electrode E3) located inside the flow cell FC, and a photoelectric conversion sensor PT, which measures the emission intensity of the reaction product RP produced by the electrochemical reaction, and is composed of, for example, a photomultiplier tube.

[0058] The working electrode E1, counter electrode E2, and reference electrode E3 are each controlled by a potentiostat 15 to achieve the desired voltage. When the reaction product RP of the sample and reagent in the reaction solution is collected on the working electrode E1, a specific voltage is applied between the working electrode E1 and the reference electrode E3 by the potentiostat 15, causing an electrochemical reaction to occur and the reaction product RP to emit light. A photoelectric conversion sensor PT is positioned on the opposite side of the flow cell FC from the working electrode E1 (upper side in Figure 3), and this photoelectric conversion sensor PT detects the light emission intensity of the reaction product RP.

[0059] The light emission intensity detected by the photoelectric conversion sensor PT is digitized by the A / D converter 18 and, after raw data recording processing P1, is recorded as raw data of the measured value of the measured item in the storage device 31 or the storage device of the sensor 11, along with the measurement date and time.

[0060] In this case, the current and voltage values ​​generated between the counter electrode E2 and the reference electrode E3 by applying a voltage between the working electrode E1 and the reference electrode E3 are measured by the potentiostat 15. In this embodiment, not only the output of the photoelectric conversion sensor PT, but also the applied voltage values ​​to the working electrode E1 and the reference electrode E3, and the current and voltage values ​​generated between the counter electrode E2 and the reference electrode E3 are recorded in the raw data recording process P1.

[0061] Thus, the electrical signal data and the electrical signal data to be evaluated are at least one of either current or voltage.

[0062] - Operation of automated analyzers - Calibration and QC measurements are performed in a timely manner to enable high-precision qualitative and quantitative analysis of target components contained in patient samples, which are unknown samples. For example, in the case of quantitative analysis, automated analyzer 1 is operated daily according to the following procedures (1)-(5). (1) Equipment startup First, the power is turned on to start up the automatic analyzer 1. Then, the reagent container C3 is set up and the reagents are initially filled, the internal temperature of the reagent disk 7 is adjusted, a constant voltage is applied to the electrode to continuously measure the internal standard solution, and the potential of the electrode of the sensor 11 is checked to ensure it is stable, and maintenance is performed as needed. (2) Calibration High-concentration and low-concentration standard samples of the analyte (component to be measured) with known concentrations are measured. These measurements are used to create a relationship equation (calibration curve) between the concentration of the analyte and the output of sensor 11 (photoelectric conversion sensor PT). However, the calibration frequency varies depending on the analyte; for example, calibration for each analyte is performed periodically (e.g., monthly cycle) in sequence. (3) QC measurement Multiple QC samples with different concentration levels, where the possible concentration ranges of the measurement items are known, are measured, and the concentration of the measurement item in the QC samples is calculated using a calibration curve created during calibration. The appropriateness of the calibration curve is confirmed by checking whether the calculated concentration is within the known concentration range of the QC samples. QC measurements are performed frequently because they serve as a condition check to guarantee the measurement results of patient specimens. For example, QC measurements are performed 1-3 times per day in parallel for multiple measurement items. (4) Measurement of patient specimens Patient samples with unknown concentrations of the target parameters are measured, and the concentrations of the target parameters are calculated using a calibration curve. Before measuring these patient samples, a so-called background measurement or dummy measurement may be performed to check the status of the automated analyzer 1. (5) Device shutdown If necessary, clean and inspect each part of the automatic analyzer 1, then turn off the power and shut down the automatic analyzer 1.

[0063] -Detection Process- Figure 4 is a timing chart showing an example of the detection process in the automated analyzer shown in Figure 1.

[0064] As described above, calibration, QC measurement, and patient sample measurement are performed in a timely manner from the time the device is started up until it is shut down, and the sample (patient sample, standard sample, QC sample, or dummy sample, etc.) is measured in each process. The measurement operation in each process is carried out in a series of processes as shown in Figure 4: electrode conditioning, sample introduction, measurement, and cleaning.

[0065] For example, in the electrode conditioning process, as shown in Figure 4, the auxiliary reagent container RG is transported to the aspiration position of the liquid transport system 12 by the operation of the turntable 13. When the auxiliary reagent container RG reaches the aspiration position, valve V1 opens with valve V2 closed, and syringe SY is driven to aspirate the auxiliary reagent from the auxiliary reagent container RG, introducing the auxiliary reagent into the flow cell FC. In parallel with this aspiration operation of the auxiliary reagent by syringe SY, a specific voltage pattern is applied to the electrode for a certain period of time by the potentiostat 15, preparing the electrode for measurement. Once the electrode conditioning is complete, both valves V1 and V2 are closed, and the aspiration operation by syringe SY and the application of voltage to the electrode are stopped. The sensor output at any given timing during this period is recorded as electrical signal data.

[0066] In the subsequent sample introduction step, reaction vessel C2, which is installed at reaction vessel installation position SM, is transported to the suction position of the liquid transport system 12 by the operation of the turntable 13. When reaction vessel C2 reaches the suction position, valve V1 opens while valve V2 remains closed, and syringe SY is driven to aspirate the reaction solution from reaction vessel C2, introducing the reaction solution into flow cell FC. Furthermore, with valve V1 still open, auxiliary reagent container RG is transported to the suction position of the liquid transport system 12 by the operation of the turntable 13. When auxiliary reagent container RG reaches the suction position, syringe SY is driven to aspirate the auxiliary reagent from auxiliary reagent container RG, introducing the auxiliary reagent into flow cell FC. Once the sample introduction step is complete, both valves V1 and V2 are closed, and the aspiration operation by syringe SY is stopped.

[0067] In the measurement process, the turntable 13 moves the standby position SP to the suction position of the liquid transport system 12, and the potentiostat 15 applies the voltage necessary for the luminescence reaction of the reaction product RP captured on the working electrode E1 in the sample introduction process. The sensor output is also recorded while the voltage is applied to the electrode.

[0068] In the subsequent cleaning process, valve V2 opens while valve V1 remains closed, driving syringe SY to discharge auxiliary reagents, etc., into the drain tank. After that, both valves V1 and V2 close, and the detergent container CL is transported to the suction position of the liquid transport system 12 by the operation of the turntable 13. When the detergent container CL reaches the suction position, valve V1 opens while valve V2 remains closed, driving syringe SY to draw cleaning solution from the detergent container CL, and the cleaning solution is introduced into the flow cell FC. During this time, in order to prevent reaction product RP from remaining in the flow cell FC, a voltage with a different pattern than that of the electrode conditioning process is applied to the electrodes for a certain period of time while the cleaning solution flows through the flow cell FC. By applying voltage to the electrodes, reaction product RP, etc., attached to the electrodes are detached, and the detached reaction product RP is washed away by the cleaning solution and discharged from the flow cell FC.

[0069] During this cleaning process, the sensor output is recorded while voltage is applied to the electrodes, and both valves V1 and V2 are closed once the cleaning solution has been drawn in.

[0070] Subsequently, the turntable 13 moves the standby position SP to the suction position of the liquid transport system 12, valve V2 opens with valve V1 closed, and the cleaning fluid is discharged into the drain tank by syringe SY. Finally, valve V2 closes, returning to the state before the start of the electrode conditioning process.

[0071] Since the voltage settings applied to the electrodes differ in each step of electrode conditioning, measurement, and cleaning, the voltage applied to the electrodes at multiple timings during the same measurement cycle is complexly controlled by the potentiostat 15. In the sensor 11, a complex voltage pattern is precisely and repeatedly applied to the electrodes in accordance with the measurement of the sample (patient sample, standard sample, QC sample, dummy sample, etc.), and electrical signals such as voltage and current generated at the electrodes, as well as measured values ​​of the concentration of the analyte, are output.

[0072] Thus, the electrical signal data and the electrical signal data to be evaluated can be obtained from measurement items using at least one type of sample from among the quality control sample, the standard sample used for calibration, and the dummy sample.

[0073] In a series of detection processes, the working electrode E1 and the reference electrode E3 are controlled to a certain potential, and when the current value is approximately zero, the potential applied to the working electrode E1 and the counter electrode E2 is called the background voltage. Even when no voltage is applied by the potentiostat 15, electrochemical reactions continue to occur as long as the background voltage is present.

[0074] As described above, electrical signal data can be acquired at any time while the potential is controlled to a constant level. It is also possible to prepare a separate detection process, independent of the detection process described here, for example, one aimed at acquiring only the background voltage, and acquire electrical signal data during that process.

[0075] Furthermore, regarding electrical signal data and the electrical signal data to be evaluated, while "0" values ​​cannot be used for "current values," for "voltage values," data acquired at a time when the sensor 11 is filled with fluid can be used even if the voltage is not strictly applied.

[0076] -Data Processing Flow (Automated Analyst)- The automated analyzer 1 records alarms if abnormalities are detected during the electrode conditioning, measurement, and cleaning processes. Specifically, alarms are recorded when the measured value (concentration of the analyte) falls outside the appropriate range (too high or too low), when the current value generated when voltage is applied is higher than the appropriate value, and when the amount of light emitted by the reaction product RP is lower than the appropriate amount.

[0077] For example, the current value measured during the electrode conditioning process may increase from a normal value (e.g., around 10 mA) to (e.g., around 15 mA) if the liquid inside the flow cell FC has not been replaced from the washing solution to the auxiliary reagent.

[0078] Figure 5 is a block diagram showing the processing flow in the automated analyzer shown in Figure 1. Data is input to the control device 30 of the automated analyzer 1 from the sensor 11, the mechanism unit 91, the sample data reader 92, the reagent data reader 93, and the user interface 94.

[0079] The control device 30 receives data recorded in the raw data recording process P1 (Figure 3) as input from the sensor 11. The data input from the sensor 11 to the control device 30 includes not only the measurement cycle data for patient sample measurement, but also the measurement cycle data for each measurement type, such as QC measurement and calibration, as well as dummy measurement data performed as a preparatory operation immediately before measuring the patient sample.

[0080] As mentioned above, the mechanism unit 91 is a collective term for the various hardware components (such as the sample dispensing nozzle 6 and the incubator 3) mounted on the automatic analyzer 1. The data input from this mechanism unit 91 to the control device 30 includes, for example, the operating timing, operating amount, and current value of each motor, the signals from sensors used to control each motor, and log data such as the opening and closing timing and current value of fluid valves (valves V1, V2, etc.).

[0081] The sample data reader 92 is a device that reads registered sample data, such as a barcode or RFID reader, and is installed in the automated analyzer 1. The sample container C1 is equipped with a storage medium such as a barcode or RFID, and the sample data recorded on the storage medium is read by the sample data reader 92. The data read by the sample data reader 92 and input to the control device 30 is, for example, the sample ID.

[0082] The reagent data reader 93 is a device equipped with a barcode or RFID reader, for example, to read registered reagent data, and is installed in the automated analyzer 1. The reagent container C3 is fitted with a storage medium such as a barcode or RFID, and the reagent data recorded on the storage medium is read by the reagent data reader 93. The data read by the reagent data reader 93 and input into the control device 30 includes, for example, the reagent ID, lot number, and expiration date.

[0083] The user interface 94 consists of a monitor and input devices provided in the automated analyzer 1, and is used by the user to view data and input data into the control device 30. Various types of data can be input into the control device 30 using the user interface 94, but examples of data related to reagents and auxiliary reagents include reagent ID, lot number, expiration date, onboard expiration date, and required remaining amount. Examples of data related to samples include sample ID, measurement type for the sample ID (patient sample measurement, QC measurement, calibration, dummy measurement, etc.), and measurement items. The various data input from the sensor 11 and other devices to the control device 30 are processed in real time by the processor 32 and sent to the server 40 via the communication interface 33 as log files. The processes executed by the processor 32 include, for example, sensor output conversion processing P2 and sensor output recording processing P3 for sensor output. Other processes executed by the processor 32 include operation log recording processing P4, reagent data recording processing P5, sample data recording processing P6, alarm data recording processing P7, and log file generation processing P8. Each process will be described sequentially below.

[0084] • Sensor output conversion processing In the sensor output conversion process P2, the processor 32 converts the sensor output (raw data) input from the sensor 11 into valid values. The sensor output input from the sensor 11 consists of raw data for light emission intensity, current value, and voltage value.

[0085] Here, Figure 6 shows an example of time-series data of luminescence intensity measured during sample measurement, Figure 7 shows an example of time-series data of voltage value, and Figure 8 shows an example of time-series data of current value. The horizontal axis in each figure corresponds to time, and the time change of each value is shown. During sample measurement, the control device 30 applies voltage to the electrodes at specific timings from the start of measurement to acquire data.

[0086] In the examples shown in Figures 6 to 8, a predetermined voltage is applied to the working electrode E1 and the reference electrode E3 at the same timing in each case (10 ms after the midpoint of the sensor output cycle in this embodiment).

[0087] • Sensor output recording processing In the sensor output recording process P3, the processor 32 assigns a measurement ID for each measurement and records the raw data and valid values ​​of the measured values ​​in the storage device 31, linked to the measurement ID.

[0088] • Operation log recording process In the operation log recording process P4, the processor 32 records the operation logs input from the mechanical unit 91 and the sensor 11 in the storage device 31. The operation logs input from the mechanical unit 91, etc., include, for example, the operating timing and amount of each motor, the motor current value, the sensor signals for controlling the motor operation, and the opening and closing timing and current value of fluid valves (valves V1, V2, etc.).

[0089] • Reagent data recording and processing In reagent data recording processing P5, the processor 32 compares the reagent data input from the reagent data reader 93 with condition data pre-recorded in the storage device 31, and records it in the storage device 31 as a usable reagent if it matches the condition data. The condition data used to compare the reagent data includes the reagent ID, lot number, expiration date, onboard expiration date, required remaining amount, etc., and is input via the user interface 94 or another computer, input to the control device 30 via the communication interface 33, and recorded in the storage device 31. In addition, in reagent data recording processing P5, the processor 32 records the history of used reagents in the storage device 31 for each measurement ID. As a result, the raw data and valid values ​​of the measured values ​​and the data of the reagents used in the measurement are linked via the measurement ID.

[0090] • Sample data recording and processing In the sample data recording process P6, the processor 32 compares the sample data input from the sample data reader 92 with condition data previously recorded in the storage device 31. If the sample matches the condition data, it is recorded in the storage device 31 as a sample that can be measured, and measurement is performed in a timely manner. The condition data used to compare the sample data includes the sample ID, measurement type (QC measurement, patient sample measurement, etc.), measurement items, etc., and is input via the user interface 94 or another computer, then input to the control device 30 via the communication interface 33 and recorded in the storage device 31.

[0091] • Alarm data recording process In alarm data recording process P7, the processor 32 determines whether there is an abnormality in the measurement each time a measurement is taken (each time a control voltage is applied between the working electrode E1 and the reference electrode E3 of the sensor 11). If there is an abnormality in the measurement, alarm data is added to the dataset related to the measurement that was determined to be abnormal and recorded in the storage device 31. Whether there is an abnormality in the measurement is determined by comparing the measured value (concentration of the analyte), the current value generated when the control voltage is applied to the electrode, and the amount of light emitted by the reaction product with preset values, for example.

[0092] More specifically, if the measured value is higher than the set value (upper limit of the appropriate range), or if the measured value is lower than the set value (lower limit of the appropriate range), it is determined that there is an abnormality in the measurement, and an alarm is added to the data set of sensor 11 output acquired in this measurement. Also, if the current value generated between the counter electrode E2 and the reference electrode E3 is higher than the set value, or if the amount of light emitted by the reaction product RP is lower than the set value, it is determined that there is an abnormality in the measurement, and an alarm is added to the data set of sensor output acquired in this measurement. The alarm data includes information about the nature of the abnormality (measured value is too high / measured value is too low / current value is too high / light emission is too low).

[0093] • Log file generation process In the log file generation process P8, the processor 32 generates a log file containing electrical signal data aggregated from the data stored in the storage device 31 for each measurement (each measurement timing in the measurement cycle), including data necessary for characterizing the sensor 11, such as the measurement type. The processor 32 also transmits the log file to the server 40 via the communication interface 33 and the network NW. A log file is created for each measurement, uploaded sequentially, and stored on the server 40.

[0094] -Sensor Characterization Flow (Automated Analyzer)- The characterization flow of the present invention is divided into two parts: preparation for characterization and execution of characterization. Since both parts are performed on server 40, server 40 constitutes a characterization system that evaluates the characteristics of the sensor 11 provided in the automated analyzer 1 that inspects a sample using an electrochemiluminescence method.

[0095] • Preparation for characterization First, we will explain the preparations for characterization using Figure 9.

[0096] When a log file is received from one or more automated analysis devices 1 via the network NW, the processor 42 records the log file in the storage device 41 during the log file storage process P21.

[0097] Next, in the process P22 for extracting electrical signal data to be evaluated from the log file, the processor 42 extracts from the log file stored in the storage device 41 the electrical signal data necessary for characterizing the sensor 11, such as time-series data of luminescence intensity, voltage value, and current value at any given time when an electrochemical reaction occurs during the detection process.

[0098] At this time, it is not necessarily required to extract all time-series data for luminescence intensity, voltage, and current values; it is acceptable to extract either the voltage or current value, or one or any combination of both. Furthermore, as long as the electrochemical reaction is occurring, there may be one or more intervals in the detection process for extracting time-series data.

[0099] If the detection process for extracting time-series data has multiple intervals, it can flexibly handle combinations such as extracting only voltage time-series data in one interval and only current time-series data in another.

[0100] In the subsequent parameter acquisition step P24, parameters representing the relationship between the state of sensor 11 and the amount of change in electrical signal data due to the change in this state are acquired.

[0101] The state of the sensor 11 can be expressed by indicators such as the elapsed time since the sensor 11 was attached to the detection unit 10, the number of times the detection process has been executed, the number of times voltage has been applied, the number of QC measurements and patient sample measurements, and the cumulative amount of reaction solution, auxiliary reagents, and washing solution introduced into the flow cell FC. The indicators are not limited to those listed here, as they relate to changes in the quality, performance, and reliability of the sensor 11.

[0102] The state of sensor 11 and the amount of change in electrical signal data due to the change in this state are combined as training data, and the relationship between them is identified through machine learning. The parameters obtained from this process represent the relationship between the state of sensor 11 and the amount of change in electrical signal data due to the change in this state. For example, let's explain using the simple linear regression model shown in equation (1) below.

[0103]

number

[0104] At this time, in equation (1), y represents the state of sensor 11, and x represents the state of sensor 11. n If we consider this as electrical signal data, we set it to adjust the behavior of the machine learning model. n This corresponds to a parameter.

[0105] The formulas shown here are merely examples; they could also be parameters for nonlinear regression models such as neural networks or random forests.

[0106] Furthermore, the model is not limited to regression; it may also be a clustering model that clusters the state of sensor 11 based on the similarity of electrical signal data. In this case, the method for calculating the similarity of electrical signal data and the parameters related to the clustering model are relevant.

[0107] Alternatively, the classification model may be one that divides the state of sensor 11 into arbitrary ranges and learns based on training data labeled for each divided state; in this case, the parameters of the classification model are relevant.

[0108] This also applies to machine learning methods that recognize graphs plotted based on electrical signal data as images.

[0109] The parameters obtained here are recorded in the storage device 41 during parameter storage processing P25.

[0110] Thus, in the detection process including an electrochemical reaction, the memory device 41 stores parameters that represent the relationship between the state of the sensor 11 and the amount of change in the electrical signal data due to the change in the state of the sensor 11, among the electrical signal data acquired by the sensor 11.

[0111] Subsequently, in the state estimation index calculation process P26, the processor 42 applies parameters to the machine learning model to calculate a state estimation index that estimates the state of the sensor 11. In this way, the processor 42 (state estimation index calculation process P26) calculates the state estimation index of the sensor 11 using electrical signal data and parameters.

[0112] The state estimation index calculated using the regression model shown as an example has a correlation with the state of sensor 11 as shown in Figure 10. The horizontal dashed line in Figure 10 indicates identical values ​​for a given state estimation index. Points a and b on the dashed line in Figure 10 represent the same state estimation index, even though the states of sensor 11 are different. This means that the electrical signal data acquired at points a and b were identical.

[0113] In the subsequent state estimation index statistical calculation process P27, statistical values ​​for multiple state estimation indexes are calculated. In this way, processor 42 (statistical value calculation process P27) calculates statistical values ​​for state estimation indexes.

[0114] Here, a statistical value refers to a value obtained based on any of the following values: mean, median, mode, standard deviation, variance, maximum value, minimum value, first quartile, third quartile, sum, or frequency, or at least one of these values. Then, in the statistical value storage process P28, the statistical values ​​of the characteristic estimation index are recorded in the storage device 41, completing the preparation part for characteristic evaluation.

[0115] • Performing characteristic evaluations Next, we will explain the execution of the characteristic evaluation using Figure 11. Note that the process from the automated analyzer 1, which is equipped with the sensor 11a to be evaluated, to the electrical signal data extraction process P22 is the same as the preparation for the characteristic evaluation, so we will omit the explanation.

[0116] In the evaluation target state estimation index calculation process P31, parameters recorded in the storage device 41 are read in the parameter reading process P32, and the evaluation target state estimation index of the evaluation target sensor 11a is calculated from the extracted evaluation target electrical signal data using a machine learning model to which the parameters have been applied. In this way, the processor 42 (evaluation target state estimation index calculation process P31) calculates the evaluation target state estimation index of the evaluation target sensor 11a using the evaluation target electrical signal data and parameters output from the evaluation target sensor 11a in the detection process.

[0117] Next, in the characteristic evaluation process P33, the statistical values ​​of the state estimation index recorded in the memory device 41 are read in the statistical value reading process P34, and the state estimation index to be evaluated is compared with the statistical values. In this way, the processor 42 (characteristic evaluation process P33) evaluates the characteristics of the sensor 11a to be evaluated by comparing the state estimation index to be evaluated with the statistical values.

[0118] For example, if we use a moving average of the sensor 11's state as a statistical value, the moving average will be the solid line lave shown in Figure 10. Points c and d, which lie on the vertical dashed line representing the same state, will have significantly different state estimation indices even though their states are identical.

[0119] For example, if the state of sensor 11 is represented by the number of voltage applications, point c is larger than the moving average lave, so it can be evaluated that this number of voltage applications corresponds to a state equivalent to an even larger number of voltage applications.

[0120] On the other hand, since point d is smaller than the moving average lave, it can be evaluated that this number of voltage applications corresponds to a state equivalent to a smaller number of voltage applications.

[0121] Finally, the characteristic evaluation results are displayed on the server user interface 45 via the data output interface 44. In this way, the data output interface 44 outputs the characteristic evaluation results from the processor 42 (characteristic evaluation process P33).

[0122] The server user interface 45 does not necessarily need to be attached to the server 40; it can also be accessed remotely via a network (NW) from an information terminal held by a user or inspector. Alternatively, it can be displayed on the user interface 94 of the automatic analyzer 1 via the data acquisition interface 43, network (NW), and communication interface 33.

[0123] While maintaining a certain level of quality, performance, and reliability, the sensor 11 tends to gradually decline as related indicators, such as elapsed time and the number of detection process executions, increase. Furthermore, this trend varies depending on the usage method and environment, ranging from almost no decline to a significant decline.

[0124] Conventional anomaly detection methods often rely on binary classifications of normal and abnormal, such as whether a certain arbitrary threshold has been exceeded, and therefore cannot detect changes in trends. In contrast, by using the configuration and processing of this embodiment, it becomes possible to evaluate the trends as characteristics by comparing them with statistical values ​​of estimated state indicators based on general usage methods and usage environments obtained from a large number of sensors 11.

[0125] Next, the effects of this embodiment will be described.

[0126] The first embodiment of the present invention described above evaluates the characteristics of a sensor 11 provided in an automatic analyzer 1 that inspects a sample using an electrochemiluminescence method, and comprises: a storage device 41 (parameter storage process P25) that stores parameters representing the relationship between the state of the sensor 11 and the amount of change in the electrical signal data due to a change in the state of the sensor 11 from among the electrical signal data acquired by the sensor 11 in a detection process including an electrochemical reaction; a processor 42 (state estimation index calculation process P26) that calculates a state estimation index of the sensor 11 using the electrical signal data and parameters; a processor 42 (statistical value calculation process P27) that calculates statistical values ​​of the state estimation index; a processor 42 (evaluated state estimation index calculation process P31) that calculates an evaluation target state estimation index of the evaluation target sensor 11a using the evaluation target electrical signal data and parameters output from the evaluation target sensor 11a in the detection process; a processor 42 (characteristic evaluation process P33) that evaluates the characteristics of the evaluation target sensor 11a by comparing the evaluation target state estimation index and statistical values; and a data output interface 44 that outputs the characteristic evaluation results by the processor 42 (characteristic evaluation process P33).

[0127] This makes it possible to detect the state or signs of an unusual sensor condition, allowing for the evaluation of sensor characteristics exposed to a wide variety of usage environments and methods.

[0128] Furthermore, since the electrical signal data and the electrical signal data to be evaluated are obtained from measurement items using at least one type of sample from among the quality control sample, the standard sample used for calibration, and the dummy sample, a large amount of electrical signal data and the electrical signal data to be evaluated can be acquired, enabling more accurate evaluation of the characteristics.

[0129] Furthermore, by ensuring that the parameters are those of a machine learning model, and that the statistical values ​​are calculated based on at least one of the following: mean, median, mode, standard deviation, variance, maximum, minimum, first quartile, third quartile, sum, or frequency, or at least one of these values, we can steadily improve the accuracy of the evaluation.

[0130] Furthermore, because the sensor 11 and the sensor under evaluation 11a are flow cell type, diagnostics can be performed even for sensors whose operating environment is extremely unstable.

[0131] Furthermore, because the electrical signal data and the electrical signal data to be evaluated are either current or voltage, the characteristics can be evaluated using data acquired during the evaluation of the sample. This allows for more opportunities to perform characteristic evaluations, thereby improving accuracy.

[0132] <Second Embodiment> A second embodiment of the present invention, comprising a characterization system, an automated analyzer, and a characterization method, will be described with reference to Figure 12.

[0133] Figure 12 is a block diagram showing the data processing flow in the characterization system according to the second embodiment of the present invention.

[0134] The difference between this embodiment and the first embodiment is that the characterization system for the sensor 11a to be evaluated is provided within the automated analyzer 1A itself, which inspects the sample using an electrochemiluminescence method. The processing shown in the characterization system F in Figure 12 is the series of processing related to the characterization system that was handled by the server 40 in the first embodiment.

[0135] Regarding the characteristic evaluation system F, the data and processing that were allocated to the storage device 41 and processor 42 in the first embodiment are allocated to the storage device 31 and processor 32 of the control device 30 as an example in this embodiment. The characteristic evaluation system for the state of the sensor 11a under evaluation is the same as in the first embodiment. The characteristic evaluation results are notified to the user, etc., through the user interface 94.

[0136] Although omitted for illustrative purposes, the automatic analyzer 1A of this embodiment can also be configured to communicate with the server 40 and other automatic analyzers via a network NW. In that case, data from other sensors can be incorporated into the characterization, similar to the first embodiment.

[0137] Furthermore, if the automated analyzer 1A is a standalone type, characterization can be performed by inputting parameters and statistical values ​​into the automated analyzer 1A, which is equipped with a characterization function, using a USB memory stick or the like. Alternatively, by calculating parameters and statistical values ​​based on the data obtained by the automated analyzer 1A, it is possible to configure a characterization system for the sensor 11a to be evaluated without considering other automated analyzers.

[0138] Other configurations and operations are substantially the same as those of the characterization system, automated analyzer, and characterization method of the first embodiment described above, and details are omitted.

[0139] In the second embodiment of the present invention, the characterization system, automatic analyzer, and characterization method also provide substantially the same effects as those of the first embodiment described above.

[0140] <Third Embodiment> A third embodiment of the present invention, comprising a characterization system, an automated analyzer, and a characterization method, will be described with reference to Figures 13 and 14.

[0141] In the first embodiment, a method for evaluating the characteristics of the sensor 11a under evaluation was explained using Figure 10, based on a comparison between lave and a state estimation index.

[0142] In this embodiment, the state estimation index is standardized as shown in equation (2) below, using the mean and standard deviation of the statistical values ​​of the state estimation index recorded in the storage device 41, which are read in the statistical value reading process P34.

[0143]

number

[0144] At this time, y in equation (2) ^ The values ​​represent the state estimation index of the sensor 11a being evaluated, μ is the mean value, σ is the standard deviation, and Z is the standardized state estimation index.

[0145] Figure 13 shows an example of the display of the characteristic evaluation results of the sensor 11a under evaluation according to the third embodiment of the present invention, and also outputs the past history of the characteristic evaluation results. Here, the characteristic evaluation results of three sensors 11a under evaluation, "C0", "C1", and "C3", will be explained as an example.

[0146] Assuming the reference value Z=0 as shown in Figure 13, it can be seen that "C0" indicates that Z is fluctuating around the reference value. This shows that the evaluation sensor 11a, which is "C0", exhibits characteristics in an average state.

[0147] Next, for the evaluation sensor 11a, which is "C1," it can be seen that Z increases as the state index of the evaluation sensor 11a increases. This corresponds to a case where the state estimation index is larger compared to the moving average lave, such as points a and c in Figure 10.

[0148] Finally, it can be seen that the evaluation sensor 11a, designated "C2," shows a low value for Z. This corresponds to a case where the state estimation index is smaller compared to the average lave, such as points b and d in Figure 10.

[0149] By displaying the changes in the state estimation index corresponding to the state index of the sensor 11a being evaluated as a progression, it becomes possible to visually represent the characteristics evaluation results of the sensor 11a being evaluated.

[0150] If the form for outputting past history of characteristic evaluation results, as in this embodiment, is incorporated into the first embodiment, it is displayed on the server user interface 45 of the server 40; if incorporated into the second embodiment, it is displayed on the user interface 94 of the automatic analyzer 1. This allows users and inspectors to easily grasp the characteristics. There are no particular limitations on the display method; it may be displayed on the entire screen, or displayed in a pop-up format or as a replacement in part of the screen, and there are no particular limitations. The following display screen examples are similar.

[0151] In this case, the results are displayed on the server user interface 45 or user interface 94, as shown in the example screen display of the characteristics evaluation results of the sensor 11a to be evaluated in Figure 14. In this example, when detection unit Ch1 of the automatic analyzer 1 is selected, the characteristics evaluation results of the sensor 11a to be evaluated attached to it are displayed. By selecting detection unit Ch2 at the same time, it is also possible to display the characteristics evaluation results of multiple sensors 11a to be evaluated simultaneously.

[0152] Other configurations and operations are substantially the same as those of the characterization system, automated analyzer, and characterization method of the first or second embodiment described above, and details are omitted.

[0153] In the third embodiment of the present invention, the characterization system, automatic analyzer, and characterization method also provide substantially the same effects as those of the characterization system, automatic analyzer, and characterization method of the first or second embodiment described above.

[0154] Furthermore, the data output interface 44 outputs a history of the characteristic evaluation results, allowing users to understand the trends in the characteristic evaluation results and thus determine if an anomaly is approaching.

[0155] <Fourth Embodiment> A fourth embodiment of the present invention, comprising a characterization system, an automated analyzer, and a characterization method, will be described with reference to Figures 15 and 16.

[0156] In the third embodiment, the change in the state estimation index was displayed as a line graph. In this case, if the number of sensors 11a to be evaluated is small, multiple graphs may be displayed on top of each other. However, as the number of sensors 11a to be evaluated increases, if the graphs overlap, it may be desirable to improve the display to easily grasp the change in the state estimation index. In Figure 13 and similar figures, the generally important information is that the state estimation index Z becomes larger than the average, that is, the state estimation index Z increases, and it is important to make this easily understandable to maintenance personnel and users.

[0157] Figure 15 is a diagram showing an example of a list display of the characteristic evaluation results of a number of sensors according to the fourth embodiment.

[0158] Therefore, as shown in Figure 15, there is a method of color-coding the values ​​in stages from the reference value shown in Figure 13 up to an arbitrary threshold. Here, the reference value is shown as white and the threshold as black, and for convenience, it is displayed in 4 levels, indicating that Z increases as the color approaches black. Note that there is no need to limit the number of levels, and it is also acceptable to display it as a color gradient.

[0159] This method allows for a clear view at a glance of how the state estimation index of the sensor 11a being evaluated is changing, even when displaying the evaluation results of multiple sensors 11 side by side, thereby improving the visibility of the characteristic evaluation results.

[0160] For example, if we set the acceptable range to be up to level 2, and consider replacement for levels 3 and above, then we can see that sensors "A", "D", "E", and "H" have characteristics such that their state estimation index remains close to or below the standard value, indicating that they are remaining within the acceptable range.

[0161] On the other hand, sensors "B", "C", "F", "G", and "I" exhibit characteristics that tend to remain at level 3 or higher, indicating that replacement should be considered. In particular, sensor "F" was found to remain at level 4, where the state estimation index was near or above the threshold, for most of the interval, indicating that sensor 11 should be considered for immediate replacement.

[0162] By improving visibility in this way, it becomes possible to immediately identify which sensors 11 require attention, making management easier. Furthermore, by limiting the displayed sensors to specific regions, rearranging their order, etc., the system can be used to help with maintenance planning.

[0163] In Figure 15, the state estimation index is represented by color coding, but it is also possible to represent it by, for example, the size of the area. This is merely one example of a representation method, and it is not limited to the method given here, as long as it changes according to the value of the state estimation index and improves visibility.

[0164] When combined with the first embodiment, the characteristics can be easily understood by users and inspectors by displaying them on the server user interface 45 of the server 40, or when combined with the second embodiment, by displaying them on the user interface 94 of the automatic analyzer 1.

[0165] In this case, the results of the characterization of multiple sensors are displayed on the server user interface 45 or user interface 94, as shown in the example screen display of the multiple sensor characterization results in Figure 16. In this example, nine channels are selected from the channels of the five automatic analyzers 1, and the characterization results of the sensors 11 attached to each channel are displayed simultaneously.

[0166] In this way, the characteristic evaluation results of any number of sensors 11 can be easily displayed and compared. Note that the number of sensors 11 displayed does not necessarily have to be multiple; it is also possible to display only one.

[0167] Other configurations and operations are substantially the same as those of any of the characterization systems, automated analyzers, and characterization methods described in the first to third embodiments above, and details are omitted.

[0168] In the fourth embodiment of the present invention, the characterization system, automatic analyzer, and characterization method also provide substantially the same effects as those of any of the first to third embodiments described above.

[0169] <Fifth Embodiment> A fifth embodiment of the present invention, comprising a characterization system, an automated analyzer, and a characterization method, will be described with reference to Figures 17A to 21.

[0170] The third and fourth embodiments aimed to visualize changes in state estimation indices as a method for displaying the characteristic evaluation results of the sensor 11a under evaluation. If the only purpose is to show what actions are needed for the sensor 11a under evaluation during maintenance, then such visualization is unnecessary.

[0171] In this embodiment, the data output interface 44 outputs whether or not the sensor 11a to be evaluated needs to be replaced and when, based on the characteristic evaluation results, and a method for displaying the judgment result from the trend of the state estimation index will be described. Examples of methods for displaying the trend of the state estimation index include the slope of the line graph shown in Figure 13 and threshold exceedance.

[0172] If the slope exceeds a certain value, it means that the state estimation index of the sensor 11a being evaluated is increasing rapidly, indicating that the deviation from the average value is large.

[0173] Furthermore, when using threshold exceedance, there is a possibility that the state estimation index may become large as an outlier due to factors other than the sensor 11a being evaluated. Therefore, by referring to indicators such as the period and number of times the threshold is exceeded, or the mean and median of the state estimation index within a certain period, it becomes possible to express the trend of the state estimation index.

[0174] The trends of the state estimation indicators listed here are represented, and if they exceed a certain threshold, they are judged as red; otherwise, they are judged as green.

[0175] As shown in Figure 17A, in the case of a red judgment, for example, the four levels shown in Figure 15 can be used to indicate that the state estimation index level is high, and the replacement of sensor 11 at the next maintenance can be prompted, thereby improving the efficiency of maintenance.

[0176] The reason why the replacement of sensor 11 is scheduled for the next maintenance is that the purpose of the present invention is not to detect abnormalities in sensor 11, but rather to detect whether it may be in an unusual state by evaluating its characteristics.

[0177] For example, if an anomaly in the sensor 11a under evaluation is detected by another method and determined to be normal, then the sensor 11a can continue to be used, and there is no need to immediately replace it, incurring extra costs and time. However, because it is in an unusual state, the possibility of failure increases, and from the perspective of preventing failure, we recommend replacing it at the next maintenance. However, the timing of replacement shown in this embodiment is merely an example, and it is also possible to specify a more concrete time, such as one month from now.

[0178] Furthermore, a green rating indicates that the sensor 11a being evaluated has characteristics within the acceptable range for state estimation indicators. Therefore, as shown in Figure 17B, clearly stating that replacement is unnecessary makes it possible to reliably prevent unnecessary replacements and contribute to reducing maintenance costs.

[0179] When combined with the first embodiment, the characteristics can be easily understood by users and inspectors by displaying them on the server user interface 45 of the server 40, or when combined with the second embodiment, by displaying them on the user interface 94 of the automatic analyzer 1.

[0180] As one example of the display, Figure 18 shows an example of a status monitoring screen for each component of the automatic analyzer, where each component of the automatic analyzer 1 is displayed in a list. If a sensor is judged red by the characteristic evaluation system of the present invention, the color of the sensor item changes as shown in the figure. Similarly, the alarm button on the right side of the screen also changes color, indicating that an alarm has occurred.

[0181] Pressing the alarm button will take you to a maintenance screen, such as the one shown in Figure 19. Selecting the item where a problem is occurring will display the characteristic evaluation results of the sensor 11a being evaluated, and the system will then suggest appropriate actions to take to address the issue.

[0182] On the other hand, for users who cannot replace the sensors, the alarm list screen shown in Figure 20 clearly indicates items that are marked in red, and selecting them displays a message prompting them to contact an inspector. At the same time, an identification number is displayed, allowing inspectors to quickly understand the current situation and determine the appropriate course of action.

[0183] As a variation of Figure 20, Figure 21 shows the real-time presentation of characteristic evaluation results. The sensor 11a under evaluation can perform characteristic evaluation in real time from the electrical signal data obtained while inspecting the sample. When it is detected that the characteristics of the sensor 11a under evaluation are in an unusual state, it is possible to see which measurement item caused the detection, as shown in Figure 21. In the example shown here, "Warning" is displayed when the measurement items are Test A, Test B, and Test C. The "Result" item displays the numerical values ​​obtained from the sample inspection. By selecting the corresponding measurement item, a message is displayed at the bottom of the screen, indicating that it is necessary to contact the inspector.

[0184] Other configurations and operations are substantially the same as those of the characterization system, automated analyzer, and characterization method of the first embodiment described above, and details are omitted.

[0185] In the fifth embodiment of the present invention, the characterization system, automatic analyzer, and characterization method also provide substantially the same effects as those of any of the first to fourth embodiments described above.

[0186] Furthermore, the data output interface 44 outputs whether or not the sensor 11a under evaluation needs to be replaced and when, based on the characteristic evaluation results, making it possible to take measures such as procuring replacement parts in advance, thus enabling smoother maintenance.

[0187] <Sixth Embodiment> A characterization system, an automated analyzer, and a characterization method according to a sixth embodiment of the present invention will be described.

[0188] In this embodiment, the electrical signal data used for characterization and the electrical signal data to be evaluated are limited to those obtained when the reaction solution, which is the reaction between the sample and the reagent, is not contained within the sensor 11 or the sensor 11a to be evaluated. Specifically, the electrical signal data used to calculate the state estimation index of the sensor 11 and the sensor 11a to be evaluated, as described in the first to fifth embodiments, is data within the range obtained by calibration or QC measurement.

[0189] In patient sample measurements, the contents of the reaction solution vary significantly from patient to patient, raising concerns about the potential for unstable electrical signal data. On the other hand, calibration and QC measurements, by their very nature, use a stable reaction solution composition, thus minimizing the impact on electrical signal data. Therefore, by consistently using electrical signal data obtained through calibration or QC measurements from preparation to execution of characterization, more accurate characterization becomes possible.

[0190] For similar reasons, the same effect can be obtained by using electrical signal data obtained during electrode conditioning or washing processes that do not contain the reaction solution, or by using electrical signal data obtained when only auxiliary reagents are introduced into the flow cell FC and voltage is applied.

[0191] Other configurations and operations are substantially the same as those of any of the characterization systems, automated analyzers, and characterization methods described in the first to fifth embodiments above, and details are omitted.

[0192] In the sixth embodiment of the present invention, the characterization system, automatic analyzer, and characterization method also provide substantially the same effects as those of any of the first to fifth embodiments described above.

[0193] Furthermore, stable characterization can be achieved because the electrical signal data and the electrical signal data to be evaluated are obtained when the reaction solution, which is the reaction between the sample and the reagent, is not contained within the sensor 11 or the sensor 11a to be evaluated.

[0194] <Seventh Embodiment> A seventh embodiment of the present invention, comprising a characterization system, an automated analyzer, and a characterization method, will be described.

[0195] In this embodiment, as described in the sixth embodiment, the electrical signal data used for characterization and the electrical signal data to be evaluated are obtained when the reaction solution, which is obtained by reacting the sample and the reagent, is not contained in the sensor 11 or the sensor to be evaluated 11a, and are obtained after a pretreatment step has been performed to remove impurities unnecessary for inspection from the reaction solution. Specifically, this embodiment uses electrical signal data obtained from measurement items in calibration or QC measurement where the reaction solution is pretreated before being introduced into the flow cell FC.

[0196] The reaction solution pretreatment step is performed before the second transport mechanism 9 transfers the reaction vessel C2 to the detection unit 10. Specifically, this step aims to remove impurities other than the reaction product RP that are unnecessary for measurement. After this, the solution is transferred to the detection unit 10, and the subsequent steps are the same as for other measurement items.

[0197] The other configurations and operations are substantially the same as those of any of the characterization systems, automated analyzers, and characterization methods described in the first to sixth embodiments above, and details are omitted.

[0198] In the seventh embodiment of the present invention, the characterization system, automatic analyzer, and characterization method also yield substantially the same effects as those obtained in any of the first to sixth embodiments described above.

[0199] Furthermore, since the electrical signal data and the electrical signal data to be evaluated are obtained after a pretreatment process has been carried out to remove impurities unnecessary for the test from the reaction solution, this pretreatment process makes the contents of the reaction solution more homogenized, further reducing the impact on the electrical signal data and enabling more accurate characterization.

[0200] <Other> It should be noted that the present invention is not limited to the embodiments described above, and various modifications are included. The embodiments described above are explained in detail for the purpose of clearly illustrating the present invention, and are not necessarily limited to those having all the configurations described.

[0201] Furthermore, it is possible to replace parts of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add configurations from other embodiments to the configuration of one embodiment. In addition, it is possible to add, delete, or replace parts of the configuration of each embodiment with other configurations. [Explanation of Symbols]

[0202] 1,1A: Automatic analyzer 2: Conveyor line 3: Incubator (reaction disk) 4: First conveying mechanism 5: Tray 6: Sample dispensing nozzle 7: Reagent disk 7a: Disc cover 7b: Reagent aspiration position 8: Reagent dispensing nozzle 9: Second transport mechanism 10: Detection Unit 11: Sensor 11a: Sensor to be evaluated 12: Liquid Conveying Systems 13: Turntable 15: Potentiostat 18: A / D converter 21: Controller 22: Operating device 30: Control device 31: Storage device 32: Processor 33: Communication Interface 40: Server 41: Memory device (parameter database) 42: Processors (State estimation index calculation processor, statistical value calculation processor, evaluation target state estimation index calculation processor, characteristic evaluation processor) 43: Data Acquisition Interface 44: Data output interface (data output section) 45: Server User Interface 91: Mechanism department 92: Sample data reading device 93: Reagent data reader 94: User Interface C1: Sample container C2: Reaction vessel C3: Reagent container RG: Auxiliary reagent container CL: Detergent container E1: Working electrode (electrode) E2: Counter electrode E3: Reference electrode (electrode) F: Characterization System F1: Flow channel (flow channel, nozzle, piping) F2, F3, F4, F5, F6: Flow channels FC: Flow Cell PT: Photoelectric conversion sensor SY: Syringe V1, V2: Valve NW: Network P25: Parameter storage process (parameter database) P26: State Estimation Index Calculation Process (State Estimation Index Calculation Processor) P27: Statistical value calculation process (Statistical value calculation processor) P31: Processing for calculating the evaluation target state estimation index (evaluation target state estimation index calculation processor) P33: Characterization process (characterization processor)

Claims

1. A characterization system for evaluating the characteristics of a sample inspection sensor provided in an automated analyzer that inspects a sample using an electrochemiluminescence method, In a detection process including an electrochemical reaction, a parameter database is used to store parameters representing the relationship between the state of the sample inspection sensor and the amount of change in the electrical signal data due to a change in the state of the sample inspection sensor, among the electrical signal data acquired by the sample inspection sensor. A state estimation index calculation processor that calculates a state estimation index for the sample inspection sensor using the aforementioned electrical signal data and parameters, A statistical value calculation processor that calculates statistical values ​​of the state estimation index, In the detection process, an evaluation target state estimation index calculation processor calculates an evaluation target state estimation index for the evaluation target sensor using the evaluation target electrical signal data output from the evaluation target sensor and the parameters, A characteristic evaluation processor that evaluates the characteristics of the sensor for inspecting the sample to be evaluated by comparing the evaluation target state estimation index with the statistical value, The system includes a data output unit that outputs the characteristic evaluation results from the characteristic evaluation processor. Characterization system.

2. In the characteristic evaluation system according to claim 1, The aforementioned electrical signal data and the aforementioned electrical signal data to be evaluated were obtained using measurement items with at least one type of sample from among quality control samples, standard samples used for calibration, and dummy samples. Characterization system.

3. In the characteristic evaluation system according to claim 1, The aforementioned electrical signal data and the aforementioned electrical signal data to be evaluated were obtained when the reaction solution, obtained by reacting the sample and the reagent, was not contained within the sample inspection sensor or the sample inspection sensor to be evaluated. Characterization system.

4. In the characteristic evaluation system described in claim 3, The aforementioned electrical signal data and the electrical signal data to be evaluated were obtained after a pretreatment process was carried out to remove impurities unnecessary for the inspection from the reaction solution. Characterization system.

5. In the characteristic evaluation system according to claim 1, The aforementioned parameters are parameters of a machine learning model. Characterization system.

6. In the characteristic evaluation system according to claim 1, The aforementioned statistical values ​​are values ​​obtained based on at least one of the following: mean, median, mode, standard deviation, variance, maximum value, minimum value, first quartile, third quartile, sum, or frequency. Characterization system.

7. In the characteristic evaluation system according to claim 1, The data output unit outputs the past history of the characteristic evaluation results. Characterization system.

8. In the characteristic evaluation system according to claim 1, The data output unit outputs whether or not the sensor for inspecting the sample under evaluation needs to be replaced and when, based on the characteristic evaluation results. Characterization system.

9. In the characteristic evaluation system according to claim 1, The aforementioned sample inspection sensor and the aforementioned sensor for inspecting the sample to be evaluated are of the flow cell type. Characterization system.

10. In the characteristic evaluation system according to claim 1, The electrical signal data and the electrical signal data to be evaluated are at least one of either current or voltage. Characterization system.

11. An automated analyzer that uses an electrochemiluminescence method to examine a sample, Sensor for sample inspection, In a detection process including an electrochemical reaction, a parameter database is used to store parameters representing the relationship between the state of the sample inspection sensor and the amount of change in the electrical signal data due to a change in the state of the sample inspection sensor, among the electrical signal data acquired by the sample inspection sensor. A state estimation index calculation processor that calculates a state estimation index for the sample inspection sensor using the aforementioned electrical signal data and parameters, A statistical value calculation processor that calculates statistical values ​​of the state estimation index, In the detection process, an evaluation target state estimation index calculation processor calculates an evaluation target state estimation index for the evaluation target sensor using the evaluation target electrical signal data output from the evaluation target sensor and the parameters, A characteristic evaluation processor that evaluates the characteristics of the sensor for inspecting the sample to be evaluated by comparing the evaluation target state estimation index with the statistical value, The system includes a data output unit that outputs the characteristic evaluation results from the characteristic evaluation processor. Automatic analyzer.

12. A characteristic evaluation method for evaluating the characteristics of a sample inspection sensor provided in an automated analyzer that inspects a sample using an electrochemiluminescence method, In a detection process involving an electrochemical reaction, a parameter is determined from the electrical signal data acquired by the sample inspection sensor that represents the relationship between the state of the sample inspection sensor and the amount of change in the electrical signal data due to a change in the state of the sample inspection sensor. Using the aforementioned electrical signal data and parameters, the state estimation index of the sample inspection sensor is calculated. The statistical values ​​of the aforementioned state estimation index are calculated, In the detection process, the evaluation target state estimation index of the evaluation target sensor is calculated using the evaluation target electrical signal data output from the evaluation target sensor and the parameters, The characteristics of the sensor for inspecting the sample to be evaluated are evaluated by comparing the evaluation target state estimation index with the statistical value. Output characteristic evaluation results Characterization method.

Images