Method for addressing malfunctions in automatic analysis system and automatic analysis device

Abstract

An information processing device of an automatic analysis system determines the suitability of a system parameter 31 for actual operation of an automatic analysis device from operational data related to operation of an automatic analysis device 100 stored in an operational database 32, and creates a recommended system parameter 33 to resolve any unsuitability if the system parameter is determined to be unsuitable. Through this feature, malfunctions in the automatic analysis device can thus be resolved without manual intervention by a user or requests to a service technician.

Description

Automatic analysis system and method for dealing with malfunctions in automatic analyzer

[0001] The present invention relates to an automatic analysis system and a method for dealing with a malfunction of an automatic analysis device.

[0002] Various parameters are set for automated analyzers that automatically analyze components contained in samples such as blood and urine. The parameters are set to suit the operation of the analyzer, and if parameters that are incompatible with the operation of the analyzer are set, frequent alarms may occur, throughput may decrease, and the analyzer may not be able to perform as intended.

[0003] Patent Document 1 relates to changing parameters of an automatic analyzer. There are parameters that can be changed with user administrator authority and parameters that cannot be changed without service authority held by a serviceman of the equipment manufacturer. The latter type of parameters include parameters that could have a fatal impact on the operation of the equipment depending on the changes made, and parameters that contain the know-how of the equipment manufacturer. The document discloses a technology that encodes parameters, inputs the codes, and then decodes the input codes within the equipment, thereby changing parameters that exceed the user administrator authority without the user knowing the parameters, thereby changing parameters that require service authority without waiting for the arrival of a serviceman.

[0004] Japanese Patent Application Laid-Open No. 2021-169939

[0005] Some problems that occur in automated analyzers occur because the instrument settings do not match the actual operating conditions of the instrument, and in such cases the problem can be resolved by changing the parameter settings to match the actual operation. In such cases, the user typically changes the parameters themselves while referring to the instruction manual, or contacts a service technician to request adjustments to the instrument.

[0006] However, as pointed out in Patent Document 1, not all parameters of the device are open to the user, and if a service technician is requested to make adjustments, the technician must be dispatched and perform the work, which takes time and effort to resolve the problem by changing the parameters.

[0007] An automatic analysis system that is one embodiment of the present invention has an automatic analysis device that analyzes samples and an information processing device that updates the system parameters of the automatic analysis device, in which the automatic analysis device operates in accordance with device settings based on the system parameters, and the information processing device has a processor and memory.The processor loads a system parameter update program into the memory and executes it, and the information processing device determines the suitability of the system parameters for the actual operation of the automatic analysis device from operational data related to the operation of the automatic analysis device stored in an operational database, and if the system parameters are determined to be incompatible, creates recommended system parameters that resolve the incompatibility.

[0008] The present invention provides a method for solving problems such as frequent alarms in an automatic analyzer, without requiring manual operation by the user or the calling of a service technician. Other objects and novel features will become apparent from the description of this specification and the accompanying drawings.

[0009] 1 is a diagram illustrating the overall configuration of an automatic analysis system; FIG. 2 is an example of programs and data stored in a storage device; FIG. 3 is a system parameter update flowchart; FIG. 4 is an example of the data configuration of an operation database; FIG. 5 is a flowchart of operation data analysis and creation of recommended system parameters; FIG. 6 is an example of device settings using system parameters; FIG. 7 is an example of recommended system parameters; FIG. 8 is an example of a warning screen when rewriting system parameters; FIG. 9 is an example of a system parameter rewrite confirmation screen; FIG. 10 is an example of a machine learning model; and FIG. 11 is a flowchart of algorithm creation using a machine learning model.

[0010] The embodiments of the present invention will be described in detail with reference to the drawings. In the following embodiments, it goes without saying that the components (including element steps, etc.) are not necessarily essential unless otherwise specified or considered to be obviously essential in principle.

[0011] FIG. 1A is a diagram illustrating the overall configuration of an automatic analysis system. The automatic analysis system includes an automatic analyzer 100 and a control device 160. The automatic analyzer 100 includes a sample input unit 120, an analysis unit 110, a sample storage unit 130, and a transport line 140. The automatic analysis system may also include an external connection terminal 170 connected to the control device 160 via a network. The automatic analysis system provides the automatic analyzer 100 with an operation monitoring function and a system parameter update function, which will be described later. However, various variations in the division of the operation monitoring function and the system parameter update function between the control device 160 and the external connection terminal 170 are conceivable. For example, these functions may be primarily performed by the control device 160, with the external connection terminal 170 being used for backup and remote monitoring. Alternatively, these functions may be primarily performed by the external connection terminal 170, with the control device 160 updating system parameters in accordance with instructions from the external connection terminal 170. Alternatively, these functions may be shared between the control device 160 and the external connection terminal 170. If the control device 160 performs all of these functions, the external connection terminal 170 may not be necessary.

[0012] First, the configuration of the automated analyzer 100 will be described. The sample input unit 120 and the sample storage unit 130 are connected by a transport line 140, and the analysis unit 110 is arranged along the transport line 140. Sample containers containing samples to be analyzed are input into the sample input unit 120, and the samples are transported to the analysis unit 110 or the sample storage unit 130 via the transport line 140. In this example, the sample containers are transported mounted on a sample rack 121 that can accommodate multiple sample containers. The samples are identified by identification information, such as a barcode or RFID tag, attached to the sample container. The identification information of the sample input into the sample input unit 120 is read by a reading device (not shown) and sent to the control device 160.

[0013] The analysis unit 110 includes a reaction disk 113 that holds reaction vessels, and a sample contained in a sample vessel transported by a transport line 140 is dispensed into the reaction vessel on the reaction disk 113 by a sample probe 114. A plurality of reagent bottles 112 containing reagents used in sample analysis are held on a reagent disk 111. A reagent probe 115 dispenses the reagent from the reagent bottle 112 into the reaction vessel into which the sample has been dispensed. The analysis unit 110 uses a measuring device (not shown) to measure the mixture (reaction solution) of the sample and reagent contained in the reaction vessel. For example, an analysis unit that performs biochemical analysis is equipped with a spectroscopic analyzer to measure the absorbance and scattered light intensity of the reaction solution. The measurements performed by the analysis unit 110 are not limited to biochemical analysis, and the analysis unit 110 is equipped with a measuring device appropriate for the measurement content.

[0014] The control device 160 has a configuration as an information processing device and includes as its main components a processor (CPU) 11, memory 12, storage device 13, input interface (I / F) 14, output interface 15, media interface 16, communication device 17, input / output interface 18, and bus 19. The processor 11 functions as a functional unit that provides a predetermined function by executing processing according to a program loaded in the memory 12. The storage device 13 stores data and programs used by the functional unit. The input I / F 14 is connected to an input device 151 such as a keyboard or pointing device, and the output I / F 15 is connected to the display device 150. The media I / F 16 exchanges data with a connected external media 180. The communication device 17 enables communication with an external connection terminal 170 and a host computer 175 via a network. The external connection terminal 170 also has a configuration as an information processing device and can exchange data with the control device 160 and remotely control the automatic analyzer 100. The host computer 175 is a computer that constitutes a higher-level system that receives test requests and manages the results for the automatic analyzer 100. The input / output I / F 18 communicates with each component included in the automatic analyzer 100. These components are connected to each other via a bus 19 so that they can communicate with each other.

[0015] 1B shows examples of programs and data stored in the storage device 13. The programs include an operation monitor program 21 and a system parameter update program 22, and the data includes system parameters 31, an operation database 32, recommended system parameters 33, and a system parameter update history 34. Details will be described later.

[0016] Here, the system parameters 31 are a group of parameters for setting the control of the automated analyzer 100 in accordance with the user's operation. The contents of the device settings based on the system parameters are diverse, as shown in Figure 5. The system parameters 31 are registered with initial values ​​when the device is shipped, and are rewritten and saved with setting values ​​that match the customer's expected device operation when the device is delivered to the customer.

[0017] As described above, if the system parameters 31 are set to parameters that are incompatible with the actual operation of the device, the device will not be able to achieve its intended performance. In this embodiment, in order to resolve problems caused by system parameters that are incompatible with the actual operation without requiring a service technician to visit the facility, the automated analysis system monitors the actual operation of the automated analysis device 100, creates recommended system parameters tailored to the actual operation, and enables the system parameters to be updated. This eliminates the need for human intervention in parameter updates, thereby shortening the time required to resolve problems and preventing incorrect settings due to setting errors caused by human error. Furthermore, because the program is executed as a system program, it is not subject to restrictions imposed by user administrator privileges.

[0018] A system parameter update flow in this embodiment is shown in Fig. 2. To execute the flow of Fig. 2, the processor 11 loads an operation monitor program 21 and a system parameter update program 22 into the memory 12 and executes them.

[0019] When the automatic analyzer 100 is operating, the processor 11 constantly executes the operation monitor program 21 and collects operation data of the automatic analyzer 100 (S01). Figure 3 shows an example of the data structure of the operation database 32, which stores the collected operation data. The operation data is all data related to the operation of the automatic analyzer 100. The items in the operation database 32 shown in Figure 3 are exemplary and may include other items, and the illustrated items are not required. The date and time 41 indicates the date and time (time zone) when an event registered in the operation database 32 occurred. The alarm (type) 42 and alarm (number) 43 indicate the type and number of alarms that occurred on the date and time. The communication 44 indicates the average communication time between the control device 160 and the host computer 175 that occurred on the date and time. The screen display speed 45 indicates the processing time required for the control device 160 to display a GUI screen on the display device 150 on the date and time. For example, the processing time increases as the amount of data displayed increases. The retest 46 indicates the number of items that were retested on the date and time. The sample type and number of samples 47 and measurement time 48 indicate the type and number of samples measured on the relevant date and time, and the measurement time required for each sample type. This information is not limited to these, and may include, for example, measurement throughput, maintenance status, input status from the input device 151, data exchange with the external media 180, and communication status with the external connection terminal 170. The operation data collected in the operation database 32 is sorted, for example, by date and time, and is held as an indication of the status of the automated analyzer 100 within a certain period of time.

[0020] A condition is set for this operation database 32, for example, at least three times a week, and an analysis is performed periodically to determine whether the system parameters set in the automated analyzer 100 are suitable for actual operation, and if they are not suitable, the system parameters are updated (S02 to S05). To this end, the control device 160 starts and executes the system parameter update program 22 at a predetermined startup timing. Note that the system parameter update program 22 may be executed not only periodically, but also at any timing selected by the user. The specific processing of steps S02 to S04 is shown in the flowchart of FIG. 4. The following description will be given with reference to FIG. 4.

[0021] Whether the system parameter settings are suitable for actual operation is determined based on the occurrence of alarms in the automated analyzer 100. The processor 11 first uses the operation database 32 to determine whether the same alarm has occurred a predetermined number of times or more within a certain period of time (S21). If the same alarm has not occurred more than the threshold, it is determined that the system parameter settings are suitable for actual operation, and no action is required (S22). Therefore, in this case, no system parameter update is performed. On the other hand, if the same alarm has occurred more than the threshold, it is determined that the system parameter settings are not suitable for actual operation (S23), and recommended system parameters are created in the following steps to improve the situation.

[0022] As shown in Figure 5, there are a wide variety of system parameter settings, and if these settings are not suitable for actual operation, the same alarm will be issued frequently. For this reason, the flow from step S24 onwards is constructed based on the causal relationship between the system parameters and the alarms they cause. Figure 4 is a simplified flowchart that narrows down the system parameters, and the actual flow is constructed based on abnormal conditions that are expected to occur in the automated analyzer 100 when the system parameters are not suitable for actual operation. In other words, the flow is constructed so that an abnormal condition occurring in the automated analyzer 100 is identified from the operation data in the operation database 32 using some kind of algorithm, and recommended system parameters that act to resolve the abnormal condition are created.

[0023] For example, frequent "clogging detection alarms" occur when the number of retries at the time of clogging detection is set to a normal number or a low number, even though a large amount of highly viscous reagents that are prone to clogging detection are used in actual operation. Therefore, if "clogging detection alarms" occur frequently (S24), a recommended system parameter A is created (S25) in which the number of retries at the time of clogging detection is set to a value greater than that for normal operation.

[0024] While this is an example in which recommended parameters can be created using a simple algorithm, if a "communication abnormality alarm" occurs frequently (S26), several factors may be considered. For example, the communication environment in which the device is located may be poor, resulting in slow communication speeds, or there may be many errors in the received data, or other factors may be the cause. Therefore, the cause of the communication abnormality is estimated using the operations database 32. For example, the operations database 32 may store the number of times the host computer 175 attempts to transmit data to the control device 160 and the number of times the control device 160 attempts to transmit data to the host computer 175. If an abnormality occurs in these counts, it can be assumed that a conflict occurs between the analysis request from the host computer 175 and the result transmission from the control device 160, causing the communication abnormality. Therefore, if such an abnormality occurs (Yes in S27), a recommended system parameter B-1 is created (S28) that prioritizes receiving an analysis request over transmitting results.

[0025] 4, the cause of the communication abnormality is identified only in step S27 using the operation database 32, but if other causes can be identified, they are also identified, and if it is determined that none of these apply, the cause is identified as being a poor communication environment for the device. In this case, recommended system parameter B-2 is created (S29), with a communication timeout detection time set longer than the timeout detection time in a normal communication environment.

[0026] The recommended system parameters to be created may include multiple device settings. For example, recommended system parameters B-3 may be created that prioritizes receiving analysis requests over sending results and has a long communication timeout detection time.

[0027] Figure 5 shows the device settings by system parameters, categorized by type. The possible device settings are wide-ranging, and the processor 11 can create recommended system parameters to improve the device's condition by detecting an abnormality in the device from the operation data stored in the operation database 32, in addition to using the frequent occurrence of the same alarm as a trigger. Figure 6 shows an example of recommended system parameters.

[0028] For example, if it is determined that the screen display speed is slower than normal, a recommended system parameter C is created that turns on the parameter setting for deleting unnecessary data when the device is started.

[0029] For example, if it is determined that the operation involves many analyses in which the reaction time is short and the results are obtained earlier than usual based on the sample types, number of samples, and measurement time of the measurements performed by the device, a recommended system parameter D is created that sets the number of line samples held to a large value. The recommended system parameter D has the effect of improving measurement throughput under such an operation.

[0030] For example, if quality control analysis is always performed after calibration analysis in operation, a recommended system parameter E is created that turns on automatic quality control, eliminating the need to manually request quality control. The recommended system parameter E has the effect of eliminating the need for cumbersome quality control requests.

[0031] For example, if the reexamination rate is high and the causes of reexamination are uneven, a recommended system parameter F is created that solves the causes of reexamination, such as increasing the stirring strength.

[0032] In this way, when an abnormal situation is identified by the judgment flow of the system parameter update program 22 and recommended system parameters that improve the situation are created, the system parameter settings are rewritten (FIG. 2: S05).

[0033] 4 overwriting the currently used system parameters 31 with the recommended system parameters 33 created in the flow of Figure 4 will resolve the abnormal condition that triggered the creation of the recommended system parameters, but there is a possibility that other side effects will occur. For this reason, rather than simply overwriting the system parameters, it is desirable to warn the user when updating the recommended system parameters and to keep a system parameter update history 34 so that the system parameter settings can be restored to their original state.

[0034] 7 shows an example of a warning screen displayed on the display device 150 when the system parameters are rewritten. For example, the warning screen 701 displays a message suggesting the content of a side effect and how to deal with the side effect when it occurs, thereby suggesting more appropriate operation of the device.

[0035] FIG. 8 shows a system parameter rewrite confirmation screen. The system parameter rewrite confirmation screen 801 includes a history information display section 802, an execute button 803, and a cancel button 804. The history information display section 802 displays system parameter update information based on the system parameter update history 34. The system parameter update history 34 stores information on the date and time of system parameter updates and backup acquisition, as well as information on the system parameters that were updated. For example, past system parameters themselves can be stored in the storage device 13, the external connection terminal 170, or the external media 180, and the system parameter update history 34 can store links to such data. The user can update the system parameters by selecting the parameter settings to be changed in the history information display section 802 and pressing the execute button 803. Since the history is displayed in the history information display section 802, the user can select a past history and press the execute button 803 to restore the system parameter settings to their previous state. To cancel the history selected in the history information display section 802, press the cancel button 804.

[0036] Returning to the explanation of FIG. 4 , even if the same alarm occurs frequently or an abnormal state of the device can be detected from the operation data stored in the operation database 32, there may be cases where an algorithm for creating recommended system parameters for such a situation has not been created. In such cases, it is determined that no action can be taken (S30), and a message such as urging a service technician to take action is displayed on the display device 150. In this embodiment, an algorithm is created using a machine learning model so that even in such cases, the situation can be improved by updating the system parameter settings in the future (S31).

[0037] An example of a machine learning model is shown in Fig. 9, and detailed processing of step S31 is shown in Fig. 10. Note that because training of a machine learning model requires a large amount of computational resources, it is also effective to perform the training by the external connection terminal 170 rather than the control device 160. In this case, in step S31, learning data to be used for training the machine learning model is created and transmitted to the external connection terminal 170 that will train the machine learning model.

[0038] The machine learning model 901 shown in FIG. 9 is a machine learning model that takes, for example, operational data 902 indicating the actual operational state of the automatic analysis device 100 and system parameters 903 as inputs, and outputs a conformance judgment 904 .

[0039] First, since the suitability of the system parameters is determined in the flow of Figure 4, the operation data and system parameters when the system parameters are determined to be suitable, and the operation data and system parameters when the system parameters are determined to be unsuitable are used as learning data and accumulated for training the machine learning model 901 (S41). Then, training of the machine learning model 901 is performed using the accumulated machine learning data (S42). Once training is complete, it becomes possible to estimate what system parameters will be judged to be suitable under specified operating conditions. Therefore, an algorithm for generating recommended system parameters corresponding to the detected abnormal state is created using the trained machine learning model (S43).

[0040] Note that the machine learning model is not limited to the model having the input and output shown in Fig. 9. For example, it is also possible to use a machine learning model that does not use the conformance determination 904, but instead takes operation data and system parameters indicating the operation environment and operation status as inputs, and operation data indicating the malfunction status of the device as output, and estimates some of the operation data from some of the operation data and system parameters.

[0041] Thereafter, by incorporating an algorithm for creating recommended parameters using the created trained machine learning model into the system parameter update program 22, it becomes possible to create recommended system parameters for abnormal situations that were previously inoperable, thereby expanding the range of abnormal situations that can be handled. Note that the algorithm for creating recommended parameters using a trained machine learning model may be incorporated into the program by an engineer, or may be automatically added by artificial intelligence.

[0042] The present invention is not limited to the above-described embodiments and includes various modifications. For example, the above-described embodiments and modifications have been described in detail to make the present invention easier to understand, and are not necessarily limited to those including all of the described configurations. Furthermore, it is possible to replace part of the configuration of one embodiment or modification with the configuration of another embodiment or modification, and it is also possible to add the configuration of another embodiment or modification to the configuration of one embodiment or modification. Furthermore, it is possible to add, delete, or replace part of the configuration of each embodiment or modification with other configurations.

[0043] 11: Processor (CPU), 12: Memory, 13: Storage device, 14: Input interface, 15: Output interface, 16: Media interface, 17: Communication device, 18: Input / output interface, 19: Bus, 21: Operation monitor program, 22: System parameter update program, 31: System parameters, 32: Operation database, 33: Recommended system parameters, 34: System parameter update history, 100: Automatic analyzer, 110: Analysis unit, 111: Reagent disk, 112: Reagent bottle, 113: Reaction disk 7: sample rack, 114: sample probe, 115: reagent probe, 120: sample input section, 121: sample rack, 130: sample storage section, 140: transport line, 150: display device, 151: input device, 160: control device, 170: external connection terminal, 175: host computer, 180: external media, 701: warning screen, 801: system parameter rewrite confirmation screen, 802: history information display section, 803: execute button, 804: cancel button, 901: machine learning model, 902: operation data, 903: system parameters, 904: conformance determination.

Claims

1. An automatic analysis system having an automatic analyzer that analyzes samples and an information processing device that updates system parameters of the automatic analyzer, wherein the automatic analyzer operates according to device settings based on the system parameters, and the information processing device has a processor and a memory, and the processor loads a system parameter update program into the memory and executes it, whereby the information processing device determines the suitability of the system parameters for the actual operation of the automatic analyzer from operational data related to the operation of the automatic analyzer stored in an operation database, and if the system parameters are determined to be incompatible, creates recommended system parameters to eliminate the incompatibility.

2. An automatic analysis system as claimed in claim 1, wherein the processor loads and executes an operation monitor program into the memory while the automatic analysis device is in operation, so that the information processing device collects the operation data from the automatic analysis device and stores it in the operation database.

3. The automatic analysis system according to claim 2, wherein the processor periodically initiates the system parameter update program.

4. The automatic analysis system according to claim 1, wherein the system parameter update program creates recommended system parameters that resolve the incompatibility using a trained machine learning model when the system parameters are determined to be incompatible.

5. An automatic analysis system according to claim 4, wherein the machine learning model is a machine learning model trained using at least the operational data stored in the operational database and the system parameters at the time the operational data was collected as teacher data.

6. An automatic analysis system according to claim 1, wherein the information processing device, when rewriting the settings of the system parameters to the recommended system parameters, issues a warning about side effects on the operation of the automatic analysis device caused by the rewriting.

7. An automatic analysis system according to claim 1, wherein the information processing device comprises a storage device, and the storage device stores a system parameter update history that records the update date and time of the system parameters and information on the updated system parameters, and the information processing device displays the system parameter update history on a display device based on the system parameter update history, and rewrites the setting of the system parameter to a past system parameter selected on the update history.

8. An automatic analysis system as described in claim 1, wherein the information processing device determines that the system parameters are unsuitable for the actual operation of the automatic analysis device when the automatic analysis device generates the same alarm a number of times equal to or greater than a threshold within a specified period of time.

9. A method for dealing with malfunctions of an automatic analyzer in an automatic analysis system having an automatic analyzer that analyzes samples and an information processing device that updates system parameters of the automatic analyzer, wherein the automatic analyzer operates according to device settings based on the system parameters, the information processing device has a processor and a memory, and the processor loads a system parameter update program into the memory and executes it, whereby the information processing device determines the suitability of the system parameters for actual operation of the automatic analyzer from operational data related to the operation of the automatic analyzer stored in an operation database, and if the system parameters are determined to be incompatible, creates recommended system parameters to eliminate the incompatibility, and the information processing device determines that the system parameters are incompatible with the actual operation of the automatic analyzer if the automatic analyzer generates the same alarm a number of times equal to or greater than a threshold within a specified period.

10. A method for dealing with malfunctions as claimed in claim 9, wherein the processor loads and executes an operation monitor program into the memory while the automatic analysis device is in operation, so that the information processing device collects the operation data from the automatic analysis device and stores it in the operation database.

11. The method for dealing with a malfunction according to claim 10, wherein the processor periodically starts the system parameter update program.

12. A method for dealing with a problem as described in claim 9, wherein the system parameter update program is a method for creating recommended system parameters that resolve the incompatibility using a trained machine learning model when the system parameters are determined to be incompatible.

13. A method for dealing with malfunctions according to claim 12, wherein the machine learning model is a machine learning model trained using at least the operational data stored in the operational database and the system parameters at the time the operational data was collected as teacher data.

14. A method for dealing with a malfunction as described in claim 9, wherein the information processing device, when rewriting the settings of the system parameters to the recommended system parameters, issues a warning about side effects on the operation of the automatic analysis device caused by the rewriting.

Images