Culture medium monitoring device, culture medium monitoring method, and program

Abstract

Provided are a culture medium monitoring device, a culture medium monitoring method, and a program, that assist efficient management of cell culture.　This culture medium monitoring device comprises: an acquisition unit that acquires, in cell culture using a culture medium, at least one of, first information on a measurement value obtained through sensing for the culture medium, second information on an environment of the cell culture, and third information on a facility of the cell culture; and an estimation unit that estimates the condition of the culture medium and / or a method for improving the condition of the culture medium on the basis of the information acquired by the acquisition unit.

Description

Culture medium monitoring device, culture medium monitoring method, and program CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application is based on Japanese Patent Application No. 2023-217052 filed on December 22, 2023, the contents of which are incorporated herein by reference.

[0002] The present disclosure relates to a culture medium monitoring device, a culture medium monitoring method, and a program.

[0003] In recent years, there has been a demand for labor-saving and automation in production management processes for purposes such as standardizing operations, improving production efficiency, and stabilizing quality and operations. In particular, the production management process for products produced using biotechnology, including cell culture, involves multiple factors such as the environment and equipment in addition to human operations, which means that the cost of achieving desirable management is high, and labor-saving and automation are therefore required. For example, Patent Document 1 discloses technology related to a cell culture management system that identifies the cause of a rejection judgment when a rejection judgment is made in a test using an intermediate product of a product manufactured by culturing cells.

[0004] Japanese Patent Application Laid-Open No. 2017-169543

[0005] However, the technology in Patent Document 1 analyzes work in units of cell culture operations extracted from instructions and manufacturing records to identify the cause of a rejection judgment, making it difficult to consider the influence of factors outside the unit of work.

[0006] There are many factors that can cause defects in cell culture, and multiple factors often interact with each other. For example, contamination by chemical components or microorganisms can occur due to multiple factors that interact with each other, such as the inconsistency of the reagent lots used, the cooling and thawing procedures and storage conditions of the cells used, the number of times and duration of opening and closing the incubator used, and particles on the hands of workers along their work routes.

[0007] Therefore, it is difficult to predict the state of the medium that may occur during a series of work processes related to cell culture. Therefore, there is a need to provide a system that supports efficient management suitable for the work of cell culture.

[0008] Therefore, an object of the present disclosure is to provide a culture medium monitoring device, a culture medium monitoring method, and a program that support efficient management of cell culture.

[0009] A culture medium monitoring device according to one aspect of the present disclosure includes an acquisition unit that acquires, in a cell culture using a culture medium, at least one of first information regarding measurement values ​​obtained by sensing the culture medium, second information regarding the environment of the cell culture, and third information regarding equipment for the cell culture, and an estimation unit that estimates the state of the culture medium and / or a method for improving the state of the culture medium based on the information acquired by the acquisition unit.

[0010] According to the present disclosure, it is possible to provide a culture medium monitoring device, a culture medium monitoring method, and a program that support efficient management of cell culture.

[0011] FIG. 1 is a diagram illustrating an example of the configuration of a culture medium monitoring system according to an embodiment of the present invention. FIG. 1 is a diagram illustrating an example of the hardware configuration of a data management server, a culture medium monitoring device, a culture medium state display device, and a recording system. FIG. 2 is a diagram illustrating an example of the functional block configuration of the culture medium monitoring device. FIG. 3 is a diagram for explaining a specific example of the processing of the culture medium monitoring device. FIG. 4 is a diagram for explaining a specific example of the processing of the culture medium monitoring device. FIG. 5 is a diagram for explaining a specific example of the processing of the culture medium monitoring device. FIG. 6 is a flowchart illustrating an example of a specific processing procedure of the culture medium monitoring device. FIG. 7 is a flowchart illustrating an example of a specific processing procedure of the culture medium monitoring device. FIG. 8 is a diagram illustrating an example of a screen display. FIG. 9 is a diagram illustrating an example of a screen display. FIG. 10 is a diagram illustrating an example of a screen display. FIG. 11 is a diagram illustrating a specific example of data stored by a data management server. FIG. 12 is a diagram illustrating a specific example of data stored by a data management server. FIG. 13 is a diagram illustrating a specific example of data stored by a data management server. FIG. 14 is a flowchart illustrating an example of the processing procedure when the culture medium monitoring device performs culture medium monitoring.

[0012] An embodiment of the present invention (hereinafter referred to as "the present embodiment") will be described with reference to the accompanying drawings, but the present invention is not limited to this and various modifications are possible without departing from the spirit of the present invention. In each drawing, parts with the same reference numerals have the same or similar configurations. The dimensional ratios of the drawings are not limited to those shown in the drawings.

[0013] <System Configuration> FIG. 1 is a diagram showing an example of the configuration of a culture medium monitoring system 1 according to this embodiment. The culture medium monitoring system 1 shown in FIG. 1 includes a data management server 10, a culture medium monitoring device 20, a culture medium status display device 30, a recording system 40, and one or more sensors D (shown as D0, D1, D2, or Dn in the figure). The data management server 10, the culture medium monitoring device 20, the recording system 40, and the one or more sensors D are communicatively connected via a wireless or wired communication network such as the Internet, an intranet, a wireless LAN, or mobile communications. Similarly, the culture medium monitoring device 20 and the culture medium status display device 30 are communicatively connected. In the present disclosure, the culture medium monitoring system 1 is described as including the data management server 10, the culture medium status display device 30, and the recording system 40. However, as long as the culture medium monitoring system 1 includes the culture medium monitoring device 20 and the sensor D, it is not necessarily required to include the configuration of the data management server 10, the culture medium status display device 30, or the recording system 40. The functions of these components may be performed by the culture medium monitoring device 20.

[0014] In this disclosure, a "culture medium" refers to a substance for culturing cells, and may be used interchangeably with "culture medium" or "culture solution," regardless of whether it is in liquid or solid form. The type of cells may be any, such as plant cells, animal cells, or microbial cells.

[0015] In this disclosure, the term "medium state" refers to the state of a monitored medium, such as acidity, alkalinity, or overgrowth. Unless otherwise specified, it includes the current medium state and the future medium state. The medium state includes both normal and abnormal states. The medium state does not necessarily have to be one state; multiple states can exist simultaneously.

[0016] In the present disclosure, a "method for improving the condition of a medium" refers to a treatment performed on a medium to improve or further enhance the condition of the medium. For example, a method for improving the condition of a medium is the setting, change, addition, deletion, or adjustment of culture conditions, including the presentation of specific target values ​​or reference values ​​for cell culture. When the condition of the medium is normal, a method for improving the condition of the medium is a treatment for improving the quality of the medium.

[0017] In the present disclosure, "culture conditions" refer to conditions related to biological materials for cell culture, such as "cell seeding amount," conditions related to the cell culture environment, such as "temperature inside the incubator," and conditions related to cell culture equipment, such as "culture medium used." Culture conditions may also include conditions related to human factors and conditions related to non-human factors. Examples of conditions related to human factors include "time to open and close the incubator" and "hygienic conditions of the workers' hands and clothing," while examples of conditions related to non-human factors include "temperature inside the incubator" and "culture medium used."

[0018] In this disclosure, "monitoring" may be used interchangeably with "observing."

[0019] The data management server 10 is an information processing device that performs some of the functions of culture medium monitoring, such as managing sensor data (sometimes referred to as "features" in this disclosure), which are measurement values ​​obtained by sensing using the sensor D, and managing data related to estimation results. The data management server 10 may be composed of one or more information processing devices, or may be composed using a virtual server (such as a cloud server). The data management server 10 may be composed of a computer. In this disclosure, when referring to "sensor data," the number and type of sensor data are not particularly limited, and may be any number or type greater than one.

[0020] The culture medium monitoring device 20 is an information processing device that uses sensor data obtained by sensing with the sensor D to estimate the state of the culture medium to be monitored and methods for improving the state of the culture medium. For example, the culture medium monitoring device 20 acquires information about measurement values ​​obtained by sensing with one or more sensors D and information about the cell culture equipment input through operation of the recording system 40, and estimates the state of the culture medium and methods for improving the state of the culture medium. The culture medium monitoring device 20 may be composed of one or more information processing devices, or may be composed of a virtual server. The culture medium monitoring device 20 may also be composed of a computer.

[0021] The culture medium state display device 30 is an information processing device that has the function of displaying the state of the culture medium estimated by the culture medium monitoring device 20, methods for improving the state of the culture medium, various sensor data, etc.

[0022] The recording system 40 is an information processing device that performs functions such as managing information related to cell culture conditions, managing tasks and workflows related to cell culture, and analyzing process-related data. The recording system 40 may be composed of one or more information processing devices, or may be configured using a virtual server. The recording system 40 may be configured by a computer. The recording system 40 is, for example, a laboratory information management system (LIMS: Laboratory Informatics Software), a supervisory control and data acquisition system (SCADA: Supervisory Control and Data Acquisition System), a manufacturing execution system (MES: Manufacturing Execution System), or a laboratory information system (LIS: Laboratory Information System).

[0023] The one or more sensors D are devices that perform sensing of the culture medium to be monitored or the cell culture environment. The number and type of sensors D used are not particularly limited, and any number or type of sensors greater than one may be used. In FIG. 1 , "n" in "sensor Dn" represents any number. The sensor D may be, for example, an imaging device, a pH sensor, an oxygen sensor, a carbon dioxide sensor, a biological imaging sensor, a nutrient component sensor, a temperature sensor, a humidity sensor, a turbidity sensor, an illuminance sensor, a light intensity sensor, an image sensor, or a color sensor.

[0024] 2 is a diagram showing an example of the hardware configuration of the data management server 10, the culture medium monitoring device 20, the culture medium state display device 30, and the recording system 40. The data management server 10, the culture medium monitoring device 20, the culture medium state display device 30, and the recording system 40 each include a processor 71 such as a CPU (Central Processing Unit) or GPU (Graphical Processing Unit), a communication IF (Interface) 72 for wireless or wired communication, a storage device 73 such as a memory (e.g., RAM (Random Access Memory) or ROM (Read Only Memory)), an HDD (Hard Disk Drive) and / or an SSD (Solid State Drive), an input device 74 for receiving input operations, and an output device 75 for outputting information. The input device 74 is, for example, a keyboard, a touch panel, a mouse, and / or a microphone, etc. The output device 75 is, for example, a display, a touch panel, and / or a speaker, etc. The processor 71, the communication IF 72, the storage device 73, the input device 74, and the output device 75 are connected to each other by one or more communication buses 76.

[0025] <Functional Block Configuration> (Culture Medium Monitoring Device) Fig. 3 is a diagram showing an example of the functional block configuration of the culture medium monitoring device 20. The culture medium monitoring device 20 has a memory unit 201, an acquisition unit 202, a model construction unit 203, an estimation unit 204, and an output unit 205. The memory unit 201 can be realized using the memory unit 73 included in the culture medium monitoring device 20. The acquisition unit 202, the model construction unit 203, the estimation unit 204, and the output unit 205 can be realized by the processor 71 included in the culture medium monitoring device 20 executing a program stored in the memory unit 73. The program can be stored in a storage medium. The storage medium storing the program may be a non-transitory computer-readable medium. The non-transitory storage medium is not particularly limited, but may be, for example, a storage medium such as a USB (Universal Serial Bus) memory or a CD-ROM (Compact Disc Read-Only Memory). In the present disclosure, the culture medium monitoring device 20 is described as having a model construction unit 203, but the culture medium monitoring device 20 does not necessarily have to have the model construction unit 203. The functions performed by the model construction unit 203 may be performed by an information processing device different from the culture medium monitoring device 20, and the culture medium monitoring device 20 may use a model constructed in the different information processing device.

[0026] <Storage Unit> The storage unit 201 stores various data necessary for the culture medium monitoring device to perform culture medium monitoring.

[0027] <Acquisition unit> The acquisition unit 202 has a function of acquiring at least one of first information regarding measurement values ​​obtained by sensing the culture medium, second information regarding the cell culture environment, and third information regarding the cell culture equipment.

[0028] In the present disclosure, the first information regarding the measurement values ​​obtained by sensing the culture medium is information regarding the measurement values ​​obtained by sensing the culture medium using various sensors D.

[0029] In the present disclosure, a "measured value" is a value obtained by measurement, and a "set value" is a value obtained by user input or setting without measurement. Neither value is limited to being a numerical value. Measured values ​​are, for example, pH value, temperature, or hue. Set values ​​are, for example, the cells (cultured cells) used, the type of medium used, the incubator used, or the set temperature inside the incubator. In the present disclosure, "information related to measured values" includes measured values, information based on the measured values ​​obtained by processing or analyzing the measured values, and information inferred based on the measured values.

[0030] The first information may be, for example, imaging information that acquires the hue or cell occupancy rate of the culture medium, or optical sensing information that acquires the pH, turbidity, or concentration changes of a predetermined substance in the culture medium. The first information may be, for example, a pH value measured by a pH sensor, a pH value estimated based on a hue value measured by a color sensor, etc. The acquisition unit 202 may acquire the first information from the sensor D, or may acquire the first information via communication with an information processing device different from the culture medium monitoring device 20, such as the data management server 10.

[0031] In the present disclosure, the second information related to the cell culture environment is information related to measurement values ​​obtained by sensing the cell culture environment, or information related to various setting values ​​input via the input device 74 of the recording system 40 or the culture medium monitoring device 20. The information related to measurement values ​​obtained by sensing the cell culture environment is, for example, information related to the temperature of the culture environment, such as temperature changes in an incubator, or information related to humidity, carbon dioxide concentration, or oxygen concentration. The information related to various setting values ​​is information determined by artificial settings, such as the temperature in an incubator. The acquiring unit 202 may acquire the second information from the sensor D, or may acquire the second information via communication with an information processing device different from the culture medium monitoring device 20, such as the recording system 40 or the data management server 10.

[0032] In the present disclosure, the third information related to the cell culture equipment is information related to the equipment and devices used for cell culture. For example, the third information includes information related to the equipment (including devices, equipment, and biological materials) such as the incubator (storage facility), culture medium, cell (cultured cell) species, or reagents used; information related to the usage status of the equipment, such as the cell culture method; and information related to the hygienic status of the workers' hands and clothing. The third information includes at least one of information related to measured values ​​and information related to set values.

[0033] The acquisition unit may acquire various types of information together with time-series data relating to the time series in which the information was acquired, thereby enabling monitoring of the culture medium along the time series of cell culture.

[0034] <Model Construction Unit> The model construction unit 203 has a function of constructing a mathematical model (hereinafter also referred to as a "model") based on the information acquired by the acquisition unit 202. As an example, the model construction unit 203 constructs a mathematical model used for statistical causal analysis, such as a Bayesian network model, a multiple regression analysis model, or a decision tree model, based on the information acquired by the acquisition unit 202. The information used by the model construction unit 203 for model construction may refer to information stored in the storage unit 201 as existing information related to cell culture. By inputting at least one of the first information, the second information, and the third information acquired by the acquisition unit 202 into the model constructed by the model construction unit 203, it is possible to estimate the state of the culture medium that will occur under the influence of the input information, and to calculate the probability that the estimated state of the culture medium is valid.

[0035] FIG. 4 is a diagram showing an example of the structure of a Bayesian network model constructed by the model construction unit 203. Four ellipses are nodes representing each event. In the model constructed by the model construction unit 203, for example, a model can be constructed using a learning engine from 1,000 pieces of training data created based on existing information on cell culture. Of the 1,000 pieces of training data, 800 are "normal or other state (node ​​101)" and 200 are "state 1 (e.g., overgrowth) (node ​​102)." Of the 800 pieces of training data for "normal or other state," 106 are "10.0 < sensor D1 value < 20.0 (node ​​103)" and 8 are "250.0 < sensor Dn value < 300.0 (node ​​104)." Furthermore, of the 200 training data for "State 1," 6 are "10.0<sensor D1 value<20.0 (node ​​103)," and 195 are "250.0<sensor Dn value<300.0 (node ​​104)." In the example of the model shown in FIG. 4, it is estimated that the probability of "State 1" when "sensor Dn value<300.0" is 0.96.

[0036] FIG. 5 is a diagram showing an example of the structure of a Bayesian network model constructed by the model construction unit 203. As in the case of FIG. 4, the model constructed by the model construction unit 203 can calculate the estimated probability of each event indicating the state of the culture medium (nodes 201, 202, and 203 in FIG. 5) from each event related to a predetermined parameter based on all sensor data and setting values ​​(nodes 204, 205, 206, and 207 in FIG. 5). In the example model shown in FIG. 5, for example, when the sensor data is "10.0 < sensor D1 value < 20.0 (node ​​204)," the probability that the state of the culture medium is "normal (node ​​201)" is estimated to be 0.13, the probability that the state of the culture medium is "state 1 (node ​​202)" is 0.09, and the probability that the state of the culture medium is "state 2 (node ​​203)" is 0.92.

[0037] 6 is a diagram showing an example of the structure of a Bayesian network model constructed by the model construction unit 203. The model constructed by the model construction unit 203 can also calculate an estimated probability regarding the causal relationship between events (nodes 304, 305, 306, and 307 in FIG. 6) related to predetermined parameters based on all sensor data and setting values.

[0038] FIG. 7 is a diagram illustrating an example of the structure of a Bayesian network model constructed by the model construction unit 203. As described in FIG. 6, the model constructed by the model construction unit 203 can calculate an estimated probability regarding the causal relationship between events related to predetermined parameters based on all sensor data and setting values. In the example model illustrated in FIG. 7, of 1,000 training data, 782 are "10.0 < sensor D1 value < 20.0 (node ​​404)." Of the 782 training data with "10.0 < sensor D1 value < 20.0," 203 are "250.0 < sensor D2 value < 300.0 (node ​​405)," 586 are "700.0 < sensor D3 value < 750.0 (node ​​406)," and 113 are "1,000.0 < sensor D4 value < 1,100.0 (node ​​407)." Of the 113 training data of "1000.0<sensor D4 value<1100.0", 78 are "State 2 (node ​​403)", 30 are "State 1 (node ​​402)", and 13 are "Normal (node ​​401)". In the example of the model shown in Figure 7, when the sensor data are the measured values ​​shown in table 10, the state of the culture medium is estimated to be "State 2", "State 1", and "Normal" in descending order of probability of validity.

[0039] FIG. 8 is a flowchart showing an example of a procedure for constructing a Bayesian network model by the model construction unit 203. Various sensors D are installed in an incubator or clean bench to sense the culture medium to be monitored (step S101). Various data are registered in the culture medium management DB 101a (step S102). Sensor data sensed by the various sensors D and setting values ​​input to the recording system 40 are collected and stored in various DBs (step S103). Each sensor data is labeled with an estimation result (step S104). Each continuous sensor data is discretized (step S105). A Bayesian network model is constructed from the discretized sensor data (step S106). A condition to be applied to the Bayesian network model (e.g., no arrow emanates from event A) is selected (step S107). If there is known knowhow, such as an empirical rule regarding cell culture, the knowhow is reflected as a condition to be applied to the Bayesian network model (step S108: YES). If there is no know-how, the construction of the Bayesian network model is completed (step S108: NO).

[0040] The model construction unit 203 may perform preprocessing to adjust the information acquired by the acquisition unit 202 based on time-series data. The model construction unit 203 may construct a Bayesian network model based on the information acquired by the acquisition unit 202 that has undergone preprocessing.

[0041] The model construction unit 203 may perform preprocessing on the information acquired by the acquisition unit 202 to adjust it based on the measurement values ​​and setting values ​​of the sensor data. The model construction unit 203 may construct a Bayesian network model based on the information acquired by the acquisition unit 202 that has undergone preprocessing.

[0042] The model construction unit 203 may construct a Bayesian network model based on information on empirical rules related to cell culture input by a user. For example, even if there is no sensor data, the user may manually set the probabilities. This makes it possible to construct a model incorporating the knowledge of an expert even if there is no sensor data. In the present disclosure, empirical rules related to cell culture include, for example, empirical rules based on the number of passages of cell culture and empirical rules based on the culture phase of cell culture (adhesion phase, lag phase, logarithmic growth phase, stationary phase, etc.).

[0043] Even if the information acquired by the acquisition unit 202 does not include measured values ​​or set values, the model construction unit 203 may construct a Bayesian network model based on information about the cell culture that contributed to the state of the culture medium estimated by the estimation unit 204. This makes it possible to comprehensively utilize factors that may contribute to the state of the culture medium in model construction, even if the facility environment is not one in which all factors that may contribute to the state of the culture medium can be monitored.

[0044] <Estimation Unit> The estimation unit 204 has a function of estimating the state of the culture medium and a method for improving the state of the culture medium based on the information acquired by the acquisition unit 202. The information acquired by the acquisition unit 202 and used by the estimation unit 204 includes measured values ​​and / or set values. The information acquired by the acquisition unit 202 and used by the estimation unit 204 includes at least one of first information, second information, and third information. Furthermore, in addition to the function of estimating the state of the culture medium and a method for improving the state of the culture medium, the estimation unit 204 has a function of estimating the probability that the state of the culture medium or the method for improving the state of the culture medium is appropriate.

[0045] In the present disclosure, "estimating a method for improving the state of a culture medium" means identifying a method for improving the state of a culture medium using the result of estimation by the estimation unit 204. For example, it may mean identifying a method for improving the state of a culture medium that has a predetermined probability of being applicable from a list of methods for improving the state of a culture medium that are registered in advance according to the state of the culture medium, sensing items, or numerical values ​​of sensor data estimated by the estimation unit 204. For example, it may mean identifying a method for improving a specified state of a culture medium based on information about factors that contributed to the state of the culture medium estimated by the estimation unit 204.

[0046] The estimation unit 204 may estimate the state of the culture medium or a method for improving the state of the culture medium using a mathematical model used for statistical cause analysis, such as a Bayesian network model, constructed by the model construction unit 203.

[0047] FIG. 9 is a flowchart showing an example of the procedure for estimation by the estimation unit 204. Sensor data sensed by various sensors D and setting values ​​input to the recording system 40 are collected and stored in various DBs (step S201). The collected sensor data is input into a Bayesian network model (step S202). The state of the culture medium is estimated from the calculated estimation probability (step S203). Each data and the estimated result are stored in the estimation result data TSDB 101e (step S204). Information on estimated causes (states) is displayed in a ranking format from the culture medium state information DB 101b (step S205). Countermeasures (improvement methods) for the causes (states) ranked highest are implemented (step S206). The state of the culture medium is estimated based on the sensor data collected after the countermeasures are implemented, and it is determined whether the state of the culture medium has improved (step S207). If it is determined that the state has not improved, the process returns to step S204 (step S207: NO). If it is determined that the state has improved, the process proceeds to step S208 (step S207: YES). It is determined whether or not the estimation result data is applied to the training data of the model (step S208).

[0048] In addition to the function of estimating the state of the culture medium and methods for improving the state of the culture medium, the estimation unit 204 may also have the function of estimating information about factors that contributed to the state of the culture medium from information based on the measured values ​​or set values ​​acquired by the acquisition unit 202.

[0049] In addition to the function of estimating the state of the culture medium and methods for improving the state of the culture medium, the estimation unit 204 may also have a function of estimating information about factors that contributed to the state of the culture medium, even if the information acquired by the acquisition unit 202 does not include measured values ​​or set values. This makes it possible to comprehensively estimate the factors that caused the state of the culture medium, even if the equipment environment is not one in which all factors can be monitored.

[0050] The estimation unit 204 may estimate the state of the culture medium and a method for improving the culture medium based on information on empirical rules related to cell culture input by the user. This makes it possible to make estimations that take into account the knowledge of experts, even if sensor data is not available.

[0051] <Output Unit> The output unit 205 has a function of outputting information relating to the state of the culture medium estimated by the estimation unit 204 and a method for improving the state of the culture medium. The output method by the output unit 205 is not particularly limited, and may be a display output to a display device or the like, or an audio output to a speaker or the like. The output unit 205 has a display control unit 205a that controls the display of the output device 75.

[0052] The output of "Culture medium state" may include the name of the culture medium state and a description of the culture medium state. The description of the culture medium state may include a still image or a moving image. In the example of Fig. 10, a name field d101 showing the name of the culture medium state, a summary field d102 showing a description of the culture medium state, and a photo field d103 of the culture medium state are provided.

[0053] The output of the "method for improving the condition of the culture medium" may include the content of the improvement method and information on the sensed items and sensor data, such as "the measured value by sensor D1 is below the recommended value." The content of the improvement method may be output based on the information on the sensed items and sensor data. In the example of FIG. 10 , the content d201 of the improvement method and advice d202 on a culture operation for improving the condition of the culture medium based on the feature amount (sensor data) are output.

[0054] The output unit 205 may output the medium condition or a method for improving the medium condition estimated by the estimation unit 204 according to the probability that the medium condition or the method for improving the medium condition is appropriate. For example, the medium condition or the method for improving the medium condition may be output in a ranking format according to the probability that the medium condition or the method for improving the medium condition is appropriate. For example, the medium condition or the method for improving the medium condition may be output together with a numerical value or a symbol indicating the probability that the medium condition or the method for improving the medium condition is appropriate. In the example of FIG. 10 , a list screen d301 is output, ranking the medium conditions according to the probability of occurrence. Also, in the example of FIG. 10 , when a name field d101 indicating the name of the medium condition whose details are to be confirmed is selected from the list of medium conditions output in ranking format, a screen d302 showing details of the selected medium condition is output. In the present disclosure, the term "appropriateness probability" may be used interchangeably with the term "occurrence probability."

[0055] The output unit 205 may output factor information, which is information that contributed to the state of the culture medium estimated by the estimation unit 204, from among the information acquired by the acquisition unit 202. The factor information may be output as information based on sensor data values ​​such as "sensor D1 value: 15" or "10.0 < sensor D1 value < 20.0", or may be output as the item name of a factor such as "cell culture temperature" or "temperature inside the incubator". In the example of FIG. 10 , a graph d401 showing data on feature 1 is output as factor information that contributed to the state of the culture medium from among the information acquired by the acquisition unit 202.

[0056] The output unit 205 may output the medium state estimated by the estimation unit 204 and the method for improving the medium state, linked to factor information, which is information that contributed to the medium state and the method for improving the medium state, among the information based on the measured values ​​or set values ​​acquired by the acquisition unit. For example, if the medium state estimated by the estimation unit 204 is "overgrowth," the medium state "overgrowth" may be output linked to information, such as "cell culture temperature exceeds the recommended value," which is factor information that contributed to the occurrence of the medium state "overgrowth," among the information acquired by the acquisition unit 202. This enables the user to efficiently understand the factors that caused the medium state and the method for improving the medium state, even in the operation of cell culture, in which multiple factors are intricately intertwined.

[0057] The output unit 205 may output the medium state estimated by the estimation unit 204 and a method for improving the medium, linked to information about the cell culture that does not have a measured value or a set value but that the estimation unit 204 estimates contributes to the medium state. For example, when the estimation unit 204 estimates a medium state called "overgrowth" and the estimation unit 204 estimates via a model that "cell culture temperature exceeding the recommended value" contributed to the medium state "overgrowth," even if the estimation unit 204 does not have a measured value or a set value for "cell culture temperature," the output unit 205 may output the medium state "overgrowth" linked to the information that "cell culture temperature exceeding the recommended value." In this case, the information may be output in a manner that allows the user to recognize that the factor is estimated by the estimation unit 204 but does not have a measured value or a set value. This allows the user to comprehensively understand the factors that caused the medium state and methods for improving the medium state, even if the user is not in an equipment environment where all factors can be monitored in a cell culture operation where multiple factors are intricately intertwined.

[0058] The factor information may be output according to the probability that it contributes to the state of the culture medium or the method for improving the culture medium. For example, the factor information contributing to the state of the culture medium may be output in a ranking format according to the probability that it contributes to the state of the culture medium or the method for improving the state of the culture medium. For example, the factor information contributing to the state of the culture medium or the method for improving the state of the culture medium may be output together with a numerical value or a symbol indicating the probability that it contributes to the state of the culture medium or the method for improving the state of the culture medium. For example, the output may be a Bayesian network model diagram including nodes indicating factor information contributing to the state of the culture medium and nodes indicating the state of the culture medium, with arrows indicating the causality between the nodes. Figure 12 is a diagram showing an example of the output of a Bayesian network model diagram.

[0059] In a Bayesian network model diagram, the direction and strength of causality may be expressed using the direction of the arrow, the type of line, the color of the line, the thickness of the line, etc. This makes it possible to visualize the strength of the causal relationship and identify factors that should be improved with priority.

[0060] In the Bayesian network model diagram, only objects showing a probability equal to or greater than a predetermined threshold may be output, which makes it possible to efficiently grasp the underlying factors.

[0061] The output unit 205 may output the relationship between the factor information that contributed to the state of the culture medium estimated by the estimation unit 204. For example, the output may include a causal relationship between multiple pieces of factor information that contributed to the state of the culture medium, among information based on the measured values ​​or set values ​​acquired by the acquisition unit 202. This allows the user to understand the relationship between factors, which can lead to improved culture efficiency and culture medium quality. In the example of FIG. 11 , a factor information list screen d311 showing multiple factors affecting the state of the culture medium is output. When the user selects a factor (feature A: cell density value) from the list of factor information, a screen d312 showing details of the selected factor (feature A: cell density value) is output. When the user selects a label d411 of the discretized sensor data corresponding to the selected factor, a ranking d412 of factor information showing other factors that influenced the value range of the sensor data is output. For example, if the value of the sensor data called feature A (cell density value) is 685.0, when the user selects the label d411, a ranking d412 of factor information indicating other factors that contributed to the numerical value called feature A (cell density value) is displayed. In the present disclosure, "contributes to the state of the culture medium" and "affects the state of the culture medium" may be used interchangeably.

[0062] The output unit 205 may output information regarding a target cell culture indicator. The information regarding the target cell culture indicator may be information indicating the cell culture environment when the culture is successful or unsuccessful, information comparing the cell culture environment of the monitored cell with the cell culture environment when the culture is successful or unsuccessful, or a target numerical range of the sensor data. The information regarding the target cell culture indicator may include still images or moving images. This allows the user to more objectively grasp the state of the culture medium, which can lead to improvements in culture efficiency and culture medium quality. In the example of FIG. 10, a reference normal value range d501 of the sensor data is displayed. In the example of FIG. 11, a reference value range d511 of the sensor data is displayed.

[0063] The output unit 205 may output the sensor data constituting the factor information as time-series data that reflects the time series in which the data was acquired. This allows the user to recognize and analyze the time when the sensor data constituting the factor information contributed as a factor. In the example of Fig. 10, the sensor data, such as feature 1 and feature 2, are displayed as a graph d401 that reflects a time series with the horizontal axis as the time axis.

[0064] The output unit 205 may output the number of cell culture samples estimated to have a predetermined medium state relative to the total number of cell culture samples in a manner that allows the user to recognize the number. This allows the user to review and improve the cell culture operation while also taking into account the incidence rate of the predetermined medium state relative to the total number of samples, thereby providing an efficient improvement to the cell culture environment.

[0065] The output unit 205 may output information on system errors (errors due to non-human factors) and information on human errors (errors due to human factors) from among the cause information in a manner that allows the user to distinguish between them. For example, among the cause information whose probability of contributing to the culture medium state is equal to or greater than a predetermined value, information on system errors such as "temperature inside the incubator" and information on human errors such as "forgetting to close the lid of the culture vessel" may be displayed in different colors, thereby differentiating the output mode. This allows the user to efficiently grasp the cause information that contributed to the occurrence of the culture medium state.

[0066] The output unit 205 may output information regarding the state of the culture medium, a method for improving the state of the culture medium, and the timing of culture medium replacement or collection, thereby enabling the user to understand the state of the culture medium and whether the state of the culture medium requires replacement or collection of the culture medium.

[0067] (Data Management Server) The data management server 10 has at least a storage unit 101. The storage unit 101 stores various types of sensor data obtained by sensing using one or more sensors D and various types of information related to the cell culture equipment input through the operation of the recording system 40. The data stored by the data management server 10 in the storage unit 101 is stored in a culture medium management DB (Data Base) 101a, a culture medium state information DB 101b, a sensor data TSDB (Time Series Data Base) 101c, a pH value data TSDB 101d, and an estimation result data TSDB 101e.

[0068] FIG. 13 is a diagram showing a specific example of data stored in the data management server 10. The culture medium management DB 101a manages the product ID, lot number, name, and ID of the sensor D of the culture medium. The product ID of the culture medium is, for example, an identifier for identifying the culture medium on a product-by-product basis, and is stored in association with the type of culture medium, product name, etc. The lot number is, for example, an identifier for identifying the culture medium on a lot-by-lot basis, and is stored in association with the manufacturing date, materials, etc. of the culture medium. The ID of the sensor D is, for example, used for the purpose of managing the system configuration, and is an identifier for identifying the sensor used when sensing the culture medium with various sensors. The ID of the sensor D may also be an identifier for identifying information about the sensor being used (such as name, model, resolution, etc.), and may be stored in association with sensor data obtained when sensing the culture medium.

[0069] The culture medium state information DB101b manages the name of the state of the culture medium, the symptoms occurring in the culture medium (for example, acidification or alkalinity, etc.), the color change of the state of the culture medium using a pH indicator (for example, color change using phenol red), a description of the state of the culture medium (for example, cell metabolites, mainly lactic acid, have accumulated in the culture medium, etc.), and how to deal with the state of the culture medium (for example, centrifugation at 150 × g for about 5 minutes, plating on new culture medium at a cell density higher than the recommended concentration, adding about 10 to 20% conditioned medium, etc.).

[0070] In this disclosure, the term "measures" may be used interchangeably with the term "improvement methods."

[0071] The sensor data TSDB 101c manages various types of sensor data obtained by sensing using the sensor D together with the acquisition date and time of the data.

[0072] The pH value data TSDB 101d manages the pH value data of the culture medium obtained by sensing with the sensor D, together with the date and time of acquisition of the data. The pH value data TSDB 101d may also manage the product ID of the culture medium.

[0073] 14 is a diagram showing a specific example of data stored in the data management server 10. The estimation result data TSDB 101e manages the estimation results of the culture medium state together with the estimation date and time, the culture medium lot number, the product ID, the sensor data from the sensor D, the pH value, etc. The estimated culture medium state is stored as the estimation result.

[0074] Fig. 15 is a diagram showing a detailed example of data stored in the culture medium status information DB 101b. In the example of Fig. 15, the culture medium status information DB 101b manages the presumed causes of pH changes, such as acidification or alkalinization, and the methods for identifying them.

[0075] In this disclosure, the terms "cause" and "factor" may be used interchangeably.

[0076] 3. Operation Figure 16 is a flowchart showing an example of the processing procedure when the culture medium monitoring device performs culture medium monitoring. The culture medium monitoring device 20 acquires at least one of first information, second information, and third information using the acquisition unit 202 (step S301). The culture medium monitoring device 20 constructs a model for estimating the state of the culture medium using the model construction unit 203 based on the information acquired by the acquisition unit 202 (step S302). The culture medium monitoring device 20 inputs the information acquired by the acquisition unit 202 into the model constructed by the model construction unit 203, and estimates the state of the culture medium and a method for improving the state of the culture medium using the estimation unit 204 (step S303). The culture medium monitoring device 20 outputs information related to the state of the culture medium and a method for improving the state of the culture medium estimated by the estimation unit 204 using the output unit 205 (step S304).

[0077] 4. Program A program according to one aspect of the present disclosure causes a computer to execute, in cell culture using a culture medium, the steps of acquiring at least one of first information including a measurement value obtained by sensing the culture medium, second information related to the environment of the cell culture, and third information related to equipment for the cell culture, and estimating a state of the culture medium and / or a method for improving the state of the culture medium based on the information acquired by the acquisition unit.

[0078] 1...culture medium monitoring system, 10...data management server, 20...culture medium monitoring device, 201...storage unit, 202...acquisition unit, 203...model construction unit, 204...estimation unit, 205...output unit, 205a...display control unit, 30...culture medium state display device, 40...recording system, 71...processor, 72...communication IF, 73...storage device, 74...input device, 75...output device, 76...communication bus, D...sensor

Claims

1. A culture medium monitoring device for culturing cells using a culture medium, comprising: an acquisition unit that acquires at least one of first information on measurement values ​​obtained by sensing the culture medium, second information on the environment of the cell culture, and third information on equipment for the cell culture; and an estimation unit that estimates a state of the culture medium and / or a method for improving the state of the culture medium based on the information acquired by the acquisition unit.

2. The culture medium monitoring device according to claim 1, wherein the estimation unit estimates the state of the culture medium and / or a method for improving the state of the culture medium through statistical cause analysis based on the information acquired by the acquisition unit.

3. The culture medium monitoring device according to claim 1, further comprising an output section which outputs the state of the culture medium estimated by the estimation section and / or a method for improving the state of the culture medium.

4. The culture medium monitoring device of claim 1, wherein the first information includes at least one of imaging information of the culture medium and optical sensing information of the culture medium, the second information includes at least one of information on the temperature of the culture environment, information on humidity, information on carbon dioxide concentration, and information on the oxygen concentration, and the third information includes at least one of information on the type of culture medium, information on the cultured cells, and information on the culture method.

5. The culture medium monitoring device according to claim 1, wherein the method for improving the state of the culture medium includes culture conditions including a target value for the cell culture.

6. The culture medium monitoring device of claim 3, wherein the output unit outputs the state of the culture medium and / or a method for improving the state of the culture medium depending on the probability of the occurrence of the state of the culture medium and / or the probability that the improvement method is appropriate, as estimated by the estimation unit.

7. The culture medium monitoring device according to claim 3, wherein the output unit outputs information that contributes to the state of the culture medium and / or a method of improving the culture medium estimated by the estimation unit, from among the information acquired by the acquisition unit.

8. A culture medium monitoring method, in which a culture medium monitoring device executes the steps of: acquiring at least one of first information including a measurement value obtained by sensing the culture medium during cell culture using the culture medium, second information relating to the environment of the cell culture, and third information relating to equipment for the cell culture; and estimating a state of the culture medium and / or a method for improving the state of the culture medium based on the information acquired in the acquiring step.

9. A program that causes a culture medium monitoring device to execute the steps of: acquiring at least one of first information including a measurement value obtained by sensing the culture medium during cell culture using the culture medium, second information regarding the environment of the cell culture, and third information regarding equipment for the cell culture; and estimating the state of the culture medium and / or a method for improving the state of the culture medium based on the information acquired in the acquiring step.

Images