Maintenance methods for laboratory systems

Abstract

The present invention discloses a maintenance method for a laboratory system, the laboratory system comprising a first group and a second group of laboratory instruments for processing biological samples; a data acquisition component connected to the multiple groups of instruments and a remote maintenance system connected to the data acquisition component. The method comprises: the data acquisition component acquires operation data from the laboratory instruments; the first data acquisition component detects anomalies related to the laboratory instruments; when anomalies are detected, context data is transmitted to the remote maintenance system; one or more correlations between the operation data and the one or more anomalies are determined; the one or more correlations are verified; a prediction rule corresponding to the verified correlation is determined at the remote maintenance system; the one or more prediction rules are transmitted to the data acquisition component; and the data acquisition component predicts the occurrence of anomalies of the laboratory instruments based on the one or more prediction rules.

Description

Technical Field

[0001] The present application relates to a maintenance method for a laboratory system, a laboratory system configured to execute the maintenance method, and a computer program product, which, when executed by a remote maintenance system of the laboratory system, enables the laboratory system to execute the maintenance method. Background Art

[0002] As the complexity of laboratory instruments (such as pre-analytical laboratory instruments, post-analytical laboratory instruments and analytical laboratory instruments, transport systems or laboratory middleware) and laboratory systems continues to increase, the way to resolve abnormal phenomena that require some degree of intervention / maintenance becomes increasingly important. In the context of the present disclosure, anomalies relate to failures of laboratory instruments / systems or system components and / or deviations of the corresponding system or instrument from a state that is considered normal or accepted operation. Normal or accepted operation is defined by one or more of the following: operating parameters / specifications defined by the manufacturer / owner / regulatory body; statistically determined means / averages (peer comparisons) of corresponding operating parameters of the same or similar systems / instruments; and / or operations marked as normal operation by operators / technicians / experts.

[0003] One way to address anomalies is through reactive maintenance, which, as the name implies, is by reacting to anomalies that have already occurred. However, reactive maintenance is often accompanied by downtime of the affected laboratory instrument / system, which can affect the operation of one or more laboratory instruments or even the entire laboratory system. In addition, since the reaction cannot be planned, reactive maintenance often results in the need to urgently dispatch technicians to correct the anomaly.

[0004] The occurrence of downtime and the need for emergency dispatch of technicians are reduced to a certain extent through preventive maintenance. According to this approach, laboratory instruments / systems are maintained according to a set of recommendations on how to maintain the equipment based on the insights of the engineering or R&D team that created the product. Time-based maintenance requirements are usually required as part of the lease or warranty terms. However, preventive maintenance is only effective for relatively predictable repair activities. In addition, preventive maintenance brings additional costs because, as the name suggests, it is based on the concept of performing maintenance tasks before they are actually needed to prevent foreseeable (at least with a certain degree of probability) anomalies. This means that when the probability of anomalies exceeds a certain threshold (a probability based on statistical data and not dependent on the actual state of the laboratory instrument / system), replacement, maintenance, inspection, etc. of parts can be carried out. Therefore, it may happen that, for example, a spare part that is still in operation is replaced due to the expiration of a predefined time period. This leads to a lot of unnecessary costs.

[0005] Condition-based maintenance addresses the shortcomings of preventive maintenance because instead of maintaining laboratory instruments / systems based on a predefined schedule, condition-based maintenance assesses the actual condition of the asset to determine maintenance needs. With the automation of many industries and the widespread use of computers and sensors, condition-based maintenance has become increasingly automated. Sensors belonging to or connected to one or more laboratory instruments / systems provide (real-time) data to a remote maintenance system that helps the maintenance team maintain the equipment before anomalies occur.

[0006] To address the shortcomings of reactive, respectively preventive maintenance, predictive maintenance is based on the concept of proactively predicting anomalies that are specific to the respective laboratory instrument / system. Predictive maintenance takes condition-based maintenance a step further. Once operational data is available, as well as data indicative of anomalies, advanced analytics can be used to identify correlations between the operational data and one or more anomalies.

[0007] This can be done as part of an improvement process by analyzing (expert driven and / or through machine learning) the data generated by one or more laboratory instruments / systems to better understand the condition of the laboratory instrument systems. In other words, predictive maintenance uses data from one or more laboratory instruments / systems to monitor parameters of one or more laboratory instruments / systems and uses this data in conjunction with analyzed historical trends to (continuously) assess system health and predict anomalies before they occur. Additionally, data other than instruments can be used for predictions, such as environmental data, information from other systems (such as patient data), and any other data source that may be valuable.

[0008] Known implementations of predictive maintenance have the common premise of full access to data to enable advanced analytics to identify one or more conditions of one or more laboratory instruments / systems that may lead to anomalies. However, the amount of data required for predictive maintenance poses a severe limitation. Complex laboratory systems include dozens or even hundreds of instruments, each characterized by dozens of operating parameters. The sheer volume of data that needs to be transferred for analysis in order to perform predictive maintenance can cause bottlenecks and may make advanced analytics impractical.

[0009] In order to reduce the amount of data transmitted for advanced analysis, known solutions filter the data transmitted from the laboratory. However, filtering the data used to determine the cause of the anomaly carries the risk of making the data related to the anomaly unavailable.

[0010] In order to avoid the bottleneck of transmitting large amounts of data and the risk of filtering out potentially relevant data, according to another known proposed system, predictive maintenance is performed on-site.

[0011] In addition to limiting the amount of data required for advanced predictive maintenance analytics, limited data availability due to privacy issues poses a further limitation on the applicability of predictive maintenance, particularly in the area of ​​diagnostic laboratory systems that handle sensitive data related to patient health (e.g., data indicative of a disease, condition, infection, etc.). Given the sensitive nature of the data, operators of certain laboratory systems greatly restrict or even prohibit any data from leaving the laboratory system. Furthermore, even if data is allowed to leave the laboratory system, sometimes regulatory requirements restrict or even prohibit the data from leaving a specific jurisdiction (e.g., beyond the borders of a certain state). To address this issue - limiting the kind of data available for highly predictive maintenance analytics - predictive maintenance is provided within the boundaries of the region where sensitive data must be retained, according to known recommendations.

[0012] However, whether due to data volume and / or data privacy restrictions, local or regional implementations of predictive maintenance are unable to take advantage of the discoveries of advanced analytics enabled by large amounts of data. For example, in a single, locally implemented system, a rarely occurring anomaly may be detected before it has resulted in the need for reactive maintenance. At the same time, it may occur that the same anomaly appears in a different location that was not known anywhere else in the past. Since these anomalies will only occur in isolated systems, even multiple local or regional implementations of predictive maintenance, each of which cannot identify patterns between certain operational data and the occurrence of the corresponding anomaly, cannot effectively predict such rare anomalies.

[0013] Therefore, there is a need for a maintenance approach for laboratory systems that addresses the shortcomings of reactive maintenance and preventive maintenance as well as the limitations of currently known implementations of predictive maintenance. Summary of the invention

[0014] A maintenance method for a laboratory system is disclosed herein, the laboratory system comprising a first group and a second group of laboratory instruments for processing biological samples; a plurality of data acquisition components communicatively connected to the first group or the second group of laboratory instruments, and a remote maintenance system communicatively connected to the data acquisition components. The first group of laboratory instruments is connected to a first data acquisition component, and the second group of laboratory instruments is connected to a second data acquisition component. The data acquisition components are dedicated to a specific group of laboratory instruments and can directly access the operating data of the laboratory instruments. According to the embodiments disclosed herein, one data acquisition component is communicatively connected to each group of laboratory instruments.

[0015] Embodiments disclosed herein meet the above needs by a maintenance method for a laboratory system, the maintenance method comprising:

[0016] - The data acquisition component collects operating data from laboratory instruments.

[0017] The operational data collected from the laboratory instruments indicates one or more operational parameters of the respective laboratory instruments.

[0018] According to embodiments disclosed herein, operational data of laboratory instruments in the first group of laboratory instruments are collected by a first one of the plurality of data collection components, and operational data of laboratory instruments in the second group of laboratory instruments are collected by a second one of the plurality of data collection components.

[0019] - Detecting, by a first one of the plurality of data acquisition components, an anomaly associated with one or more of the plurality of laboratory instruments in the first group. Optionally, further, detecting, by a second one of the plurality of data acquisition components, an anomaly associated with one or more of the plurality of laboratory instruments in the second group.

[0020] According to embodiments disclosed herein, an anomaly indicates a faulty / suboptimal operation of a laboratory instrument / system, as reflected by a deviation of one or more operating parameters from one or more standard values ​​and / or ranges.

[0021] - When an anomaly is detected, the context data is transmitted to the remote maintenance system by a first one of the plurality of data acquisition components.

[0022] The context data sent to the remote maintenance system by the first of the plurality of data acquisition components includes the operational data and the data indicating the abnormality. Optionally, in addition, when the abnormality is detected, the context data is transmitted to the remote maintenance system by the second of the plurality of data acquisition components. The context data sent to the remote maintenance system by the second of the plurality of data acquisition components includes the operational data and the data indicating the abnormality.

[0023] One or more correlations between the operational data and the one or more anomalies are determined at the remote maintenance system.

[0024] According to embodiments disclosed herein, correlation indicates (repeated) occurrences of anomalies associated with specific operational data (such as within a certain time period).

[0025] According to embodiments disclosed herein, one or more correlations between the operational data and the one or more anomalies are determined by receiving input indicating one or more such correlations (e.g., from an expert). Alternatively or additionally, one or more correlations between the operational data and the one or more anomalies are automatically determined using a pattern recognition method, in particular based on one or more previous determinations of the expert.

[0026] According to embodiments disclosed herein, one or more correlations are determined between operational data transmitted by a first one of the plurality of data acquisition components and one or more anomalies and / or one or more correlations are determined between operational data transmitted by a second one of the plurality of data acquisition components and one or more anomalies.

[0027] -Verifies one or more dependencies and marks the corresponding dependencies as verified.

[0028] Since correlation does not always indicate causation, one or more correlations between operational data and one or more anomalies need to be verified. A verified correlation is an indication of a (possible) causal relationship between specific operational data and an anomaly.

[0029] According to embodiments disclosed herein, one or more correlations between the operational data and the one or more anomalies are verified by receiving input (e.g., from an expert) indicating the validity (possible causal relationship) of the one or more correlations between the operational data and the one or more anomalies. Alternatively or additionally, the one or more correlations between the operational data and the one or more anomalies are verified by instructing a computer-implemented verification engine to verify the one or more correlations between the operational data and the one or more anomalies, in particular based on one or more previous verifications by the expert.

[0030] According to some embodiments, the one or more correlations between the operational data transmitted by the first of the plurality of data acquisition components and the one or more anomalies are verified by comparing the one or more correlations between the operational data transmitted by the second of the plurality of data acquisition components and the one or more anomalies with verified correlations between the operational data transmitted by the second of the plurality of data acquisition components and the one or more anomalies. Thus, the verified correlations derived from the operational data transmitted by the second of the plurality of data acquisition components can be used as a reference for verifying the one or more correlations between the operational data transmitted by the first of the plurality of data acquisition components and the one or more anomalies.

[0031] - Determining, at the remote maintenance system, one or more prediction rules corresponding to the verified correlations.

[0032] According to the embodiments disclosed herein, a prediction rule includes a condition which, if satisfied by operational data, predicts the occurrence of an anomaly with a certain probability within a certain number of time frames and / or based on further prediction rules (nested or subordinate rules).

[0033] - In order to allow the data acquisition component to predict the occurrence of anomalies in a timely manner, one or more prediction rules are transmitted by the remote maintenance system to the data acquisition component.

[0034] According to embodiments disclosed herein, one or more prediction rules are also transmitted by the remote maintenance system to the data collection component, which (for data privacy or other reasons) does not itself transmit contextual data to the remote maintenance system.

[0035] - predicting, by one or more of the plurality of data acquisition components, the occurrence of an anomaly of one or more of the plurality of laboratory instruments based on one or more prediction rules.

[0036] According to the embodiments disclosed herein, the data collection component predicts one or more anomalies with a certain probability.

[0037] The embodiments disclosed herein are advantageous because they combine the advantages associated with directly accessing an instrument's operational data to detect and predict anomalies with the ability to centrally process the operational data using a remote maintenance system.

[0038] I. By performing data collection and anomaly detection locally by the data collection component, the embodiments disclosed herein ensure that the detection of anomalies is based on all data available locally (within the corresponding set of instruments, without volume and / or privacy restrictions), and is not delayed by transmission to a remote location. This allows for more accurate detection of anomalies and ensures that the most relevant operational data is associated with the anomaly.

[0039] II. By transmitting operational data associated with anomalies to a remote maintenance system and performing data analysis (determination and verification of correlations and determination of predictive rules) at a centralized location, the embodiments disclosed herein combine data from multiple groups of laboratory instruments to jointly collect and centrally analyze as much available data as possible.

[0040] III. By transmitting one or more prediction rules to the data acquisition component and predicting anomalies by the data acquisition component, the embodiments disclosed herein allow for much more timely prediction of anomalies by avoiding delays that may be caused by transmitting data to a remote location. In addition, predicting anomalies by the data acquisition component ensures that the prediction of anomalies is based on all operational data available locally within the corresponding set of instruments.

[0041] According to further embodiments disclosed herein, the maintenance method further comprises: determining at the remote maintenance system one or more normative maintenance actions corresponding to the anomaly; transmitting the one or more normative maintenance actions from the remote maintenance system to the plurality of data acquisition components; when predicting the occurrence of anomalies associated with one or more of the laboratory instruments, one or more of the plurality of acquisition components instructing the laboratory instruments to perform the one or more normative maintenance actions. The one or more normative maintenance actions include one or more instructions that, when executed by one or more of the laboratory instruments, reduce the probability of the occurrence of the anomaly.

[0042] In order to solve the situation where an anomaly still occurs despite the use of prediction rules to predict the anomaly, further embodiments disclosed herein include: determining one or more anomaly mitigation actions corresponding to the anomaly; transmitting the one or more mitigation actions from the remote maintenance system to the multiple data acquisition components; and if the data acquisition component that commands one or more of the laboratory instruments to perform the one or more normative maintenance actions has detected the anomaly, the data acquisition component commands one or more laboratory instruments to perform one or more anomaly mitigation actions. The one or more anomaly mitigation actions include one or more instructions that, when executed by one or more of the laboratory instruments, mitigate the negative impact of the anomaly. In other words, the impact of the anomaly is reduced.

[0043] Other embodiments disclosed herein relate to a maintenance method, wherein the remote maintenance system comprises one or more regional servers each communicatively connected to a plurality of data acquisition components. The method further comprises the step of each regional server analyzing operating parameters common to all laboratory instruments and / or common to a subset of laboratory instruments connected to a particular regional server in order to detect one or more correlations between operating data and one or more anomalies specific to the respective region. Such an embodiment is advantageous because it allows a trade-off between the use of as large a data set as possible for advanced analysis and the amount / type of data transferred between different regions, thereby satisfying restrictions due to data privacy regulations.

[0044] Overall, the embodiments disclosed herein effectively combine edge computing for comprehensive and timely data capture or anomaly detection (avoiding data volume and / or privacy limitations) with cloud computing capabilities for advanced data analysis based on multiple data sources and quality assurance through verification. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Further features and advantages of the disclosed method / device / system will be described in detail below by describing and referring to the following drawings:

[0046] Figure 1 a highly schematic block diagram of an embodiment of the disclosed laboratory system;

[0047] Figure 2 a swim lane diagram illustrating the interactions between elements of a system for performing a first embodiment of the disclosed method;

[0048] Figure 3A a first page of a swim lane diagram illustrating interactions between elements of a system for performing a further embodiment of the disclosed method;

[0049] Figure 3B a second page of a swim lane diagram illustrating interactions between elements of a system for performing a further embodiment of the disclosed method;

[0050] Figure 3C a third page of a swim lane diagram illustrating interactions between elements of a system for performing a further embodiment of the disclosed method, the interactions further comprising prescriptive maintenance;

[0051] Figure 3D a fourth page of a swim lane diagram illustrating interactions between elements of a system for performing a further embodiment of the disclosed method, the interactions further comprising mitigation actions;

[0052] Figure 3E a fifth page of a swim lane diagram illustrating interactions between elements of a system for performing a further embodiment of the disclosed method, the interactions further comprising evaluating a mitigation action;

[0053] Figure 4 A highly schematic block diagram of a further embodiment of the disclosed laboratory system. DETAILED DESCRIPTION

[0054] The use of "a" or "an" is used to describe elements and components of the embodiments herein. This is done merely for convenience and to give a general sense of the inventive concept. This description should be understood to include one or at least one, and the singular also includes the plural, unless otherwise clearly indicated.

[0055] As used herein, qualifiers such as "about," "approximately," and "substantially" are intended to indicate that the qualified item or value is not limited to the exact value or amount specified, but includes some slight variations or deviations caused by, for example, measurement error or inaccuracy, manufacturing tolerances, stress exerted on various parts, wear, and combinations thereof.

[0056] The terms "sample", "patient sample" and "biological sample" refer to one or more materials that may potentially contain an analyte of interest. Patient samples can be derived from any biological source, such as physiological fluids, including blood, saliva, lens fluid, cerebrospinal fluid, sweat, urine, feces, semen, milk, ascites, mucus, synovial fluid, peritoneal fluid, amniotic fluid, tissue, cultured cells, etc. Patient samples can be pretreated before use, such as preparing plasma from blood, diluting viscous fluids, lysing, etc. The processing method can involve filtration, distillation, concentration, inactivation of interfering components and addition of reagents. Patient samples obtained from the source can be used directly, or after pretreatment to change the characteristics of the sample. In some embodiments, the initial solid or semi-solid biological material can be converted into a liquid by dissolving or suspending it in a suitable liquid medium. In some embodiments, it may be suspected that the sample contains a certain antigen or nucleic acid.

[0057] The term "analyte" is a component of a sample to be analyzed, such as molecules of various sizes, ions, proteins, metabolites, etc. The information collected on the analyte can be used to evaluate the effects of the administration of a drug on an organism or a specific tissue, or to perform a diagnosis. Therefore, "analyte" is a general term for information about the presence and / or concentration of a substance. Examples of analytes are, for example, glucose, coagulation parameters, endogenous proteins (e.g., proteins released from the myocardium), metabolites, nucleic acids, etc.

[0058] As used herein, the term "laboratory instrument" encompasses any device or device component operable to perform and / or cause one or more processing steps / workflow steps to be performed on one or more biological samples and / or one or more reagents. The expression "processing step" thus refers to a physically performed processing step, such as centrifugation, aliquoting, sample analysis, etc. The term "instrument" includes pre-analytical instruments, post-analytical instruments, analytical instruments, and laboratory middleware.

[0059] As used herein, the term "post-analytical instrument" encompasses any device or device component configured to perform one or more post-analytical processing steps / workflow steps including, but not limited to, sample unloading, transport, recapping, decapping, temporary storage / buffering, archiving (refrigerated or non-refrigerated), retrieval, and / or disposal.

[0060] As used herein, the term "pre-analytical instrument" encompasses any device or device component configured to perform one or more pre-analytical processing steps / workflow steps, including but not limited to centrifugation, resuspension (e.g., by mixing or vortexing), capping, decapping, recapping, sorting, tube type identification, sample quality determination, and / or aliquoting steps. The processing steps may also include adding chemicals or buffers to the sample, concentrating the sample, culturing the sample, etc.

[0061] As used herein, the term "analyzer" / "analytical instrument" encompasses any device or device component configured to obtain a measurement value. The analyzer is operable to determine the parameter value of a sample or its components via various chemical, biological, physical, optical or other technical procedures. The analyzer is operable to measure the parameter of a sample or at least one analyte and return the obtained measurement value. The list of possible analysis results returned by the analyzer includes but is not limited to the concentration of the analyte in the sample, the digital (yes or no) result indicating the presence of the analyte in the sample (corresponding to the concentration above the detection level), optical parameters, DNA or RNA sequences, data obtained from the mass spectrometry of proteins or metabolites, and various types of physical or chemical parameters. The analytical instrument may include units that assist in pipetting, dosing, and mixing samples and / or reagents. The analyzer may include a reagent holding unit for holding reagents for assaying. Reagents may be arranged in the form of containers or boxes, such as containing separate reagents or reagent groups, placed in a suitable receiver or position in a storage chamber or conveyor. It may include a consumable supply unit. The analyzer may include a process and a detection system, and the workflow of the process and the detection system is optimized for certain types of analysis. Examples of such analyzers are clinical chemistry analyzers, coagulation chemistry analyzers, immunochemistry analyzers, urine analyzers, nucleic acid analyzers for detecting the results of a chemical or biological reaction or monitoring the progress of a chemical or biological reaction.

[0062] As used herein, the term "laboratory middleware" refers to any physical or virtual processing device that can be configured to control a laboratory instrument / or a system including one or more laboratory instruments in a manner that the laboratory instrument / system performs one or more workflows and one or more workflow steps. The laboratory middleware can, for example, command the laboratory instrument / system to perform one or more pre-analysis, post-analysis and analysis workflows / one or more workflow steps. The laboratory middleware can receive information from the data management unit about which steps need to be performed on a certain sample. In some embodiments, the laboratory middleware is integrated with the data management unit, composed of a server computer and / or as part of a laboratory instrument, or even distributed across multiple instruments of a laboratory system. The laboratory middleware can, for example, be implemented as a programmable logic controller running a computer-readable program, and the computer-readable program is provided with instructions for performing operations.

[0063] As used herein, the term "sample transport system" encompasses any device or device component configured to transport sample carriers (each carrier holding one or more sample containers) between laboratory instruments. In particular, the sample transport system is a one-dimensional system based on a conveyor belt, a two-dimensional transport system (such as a magnetic sample carrier transport system), or a combination thereof.

[0064] As used herein, a "laboratory system" includes a system comprising one or more analytes; pre-analytical and post-analytical laboratory instruments; a sample transport system and / or laboratory middleware.

[0065] As used herein, the term "analysis" or "analytical test" encompasses laboratory procedures for characterizing a parameter of a biological sample for the qualitative assessment or quantitative measurement of the presence or amount or functional activity of an analyte.

[0066] The term "consumables" includes, but is not limited to, reagents, system fluids, quality control materials, calibration materials, microplates / microwell plates, reaction vessels, measurement cuvettes, sample tubes, pipette tips, etc.

[0067] As used herein, the term "communication network" encompasses any type of wireless network, such as WiFi TM 、GSM TM , UMTS or other wireless digital networks or cable-based networks such as Ethernet TM Etc. In particular, the communication network may implement the Internet Protocol (IP). For example, the communication network includes a combination of a cable-based network and a wireless network.

[0068] As used herein, the term "remote system" or "server" encompasses any physical or virtual machine with a physical or virtual processor that is capable of receiving, processing, and sending data. A server can run on any computer, including a special-purpose computer, which is also commonly referred to as a "server" or shared resource alone, such as a virtual server. In many cases, a computer can provide multiple services and run multiple servers. Therefore, the term server should encompass any computerized device that shares resources with one or more client processes. In addition, the term "remote system" or "server" encompasses data transmission and processing systems distributed on a data network (such as a cloud environment).

[0069] As used herein, the term "user interface" encompasses any suitable software and / or hardware for interaction between an operator and a machine, including but not limited to a graphical user interface for receiving commands as input from an operator and also providing feedback and conveying information to the operator. Also, a system / device may expose several user interfaces to serve different categories of users / operators.

[0070] The term "quality control" or "analytical quality control" refers to all those processes and procedures designed to ensure that the results of laboratory analyses (analytical tests) are consistent, comparable, accurate and within specified precision limits.

[0071] Embodiments of the disclosed method will now be presented with reference to the illustrative drawings.

[0072] like Figure 1 As shown in a highly schematic block diagram of , the disclosed laboratory system 100 comprises a plurality of laboratory instruments 10 grouped into a first group 1 and a second group 2 of laboratory instruments 10. The laboratory instruments 10 are communicatively connected (e.g., via a communication network) to one of a plurality of data acquisition components 20.1, 20.2. Figure 1 In the example shown, a first set 1 of laboratory instruments 10 is connected to a first data acquisition component 20.1, and a second set 2 of laboratory instruments 10 is connected to a second data acquisition component 20.2. Figure 1 In the example of , an optional laboratory middleware 15 is illustrated, which is configured to command multiple laboratory instruments 10 to process biological samples according to one or more test instructions. According to various embodiments disclosed herein, data acquisition components 20.1, 20.2 are connected to laboratory instruments 10 directly or via laboratory middleware 15. It should be noted that the data acquisition components 20.1, 20.2 are dedicated hardware units and / or part of a server computer, such as laboratory middleware 15, routers or any computing device configured to perform the functions of data acquisition components as disclosed herein (for example, in a server instance / process and / or virtual machine). The commonality of the data acquisition components 20.1, 20.2 of all embodiments is that they are dedicated to a specific set of laboratory instruments 10 and can directly access the operating data of the laboratory instruments 10.

[0073] The data acquisition components 20.1, 20.2 are communicatively connected to a remote maintenance system 50, for example via a communication network. The term remote with respect to the remote maintenance system is to be understood as a computer system located at a different location than at least one of the laboratory instruments 10 of the first or second group.

[0074] Now go to Figures 2 to 3E , embodiments of the disclosed method will be described.

[0075] In a preparatory step not explicitly shown in the figure, the laboratory instruments are divided into different groups. The grouping of laboratory instruments can be based on the common physical location of the respective laboratory instruments and / or on the logic of grouping by type; the manufacturer and / or the working area of ​​the respective laboratory instruments. According to the embodiments disclosed herein, the laboratory instruments of a particular group can even be located in different physical locations.

[0076] In a further preparatory step, each group of laboratory instruments is communicatively connected to the data acquisition component, for example via a communication network.

[0077] Figure 2 A swim lane diagram is shown illustrating the interaction between elements of a system for performing a first embodiment of the disclosed method.

[0078] In a first step 102, the data acquisition component 20.1, 20.2 acquires operating data from the first or second set 2 of laboratory instruments 10. The operating data are indicative of one or more operating parameters of the respective laboratory instrument 10.

[0079] According to the embodiments disclosed herein, the operational data includes one or more of the following:

[0080] - Environmental factors such as humidity, temperature, air, water, power quality, vibration:

[0081] o provided by sensors within the corresponding laboratory instrument, indicating the condition of one or more components of the laboratory instrument, such as temperature, humidity, voltage, current, torque (e.g., torque of a motor), noise, and / or vibration;

[0082] o provided by sensors arranged external to the laboratory instrument, providing data indicative of temperature, humidity, voltage, current, torque (e.g., torque of an electric motor), noise, and / or vibration in the vicinity of the laboratory instrument;

[0083] - Operational factors, such as throughput, QC, and calibration frequency of laboratory instruments;

[0084] - Equipment performance factors such as QC and calibration results, error messages, and log files of laboratory instruments;

[0085] - Data about consumables, including expiry date, transportation and storage conditions;

[0086] - Data about operator processing, including maintenance activities performed / not performed, customer maintenance due / overdue;

[0087] - Data indicating HW and SW performance, including availability / unavailability of storage space and / or memory, up-to-dateness of software.

[0088] According to the embodiments disclosed herein, operational factors; data on consumables; data on operator handling and / or data representing HW and SW performance are retrieved by a data acquisition component from a laboratory instrument and / or determined by the data acquisition component based on data from the laboratory instrument.

[0089] In a subsequent step 104 , an anomaly associated with one or more of the plurality of laboratory instruments 10 in the first group 1 is detected by a first one 20 . 1 of the plurality of data acquisition components.

[0090] According to the embodiments disclosed herein, the data acquisition components 20.1, 20.2 detect anomalies by applying one or more anomaly detection rules, including detecting deviations of operating parameters from set thresholds. The thresholds are set by the manufacturer of the corresponding laboratory instrument and / or by the operator of the laboratory system. In particular, the step 104 of detecting anomalies associated with one or more of the plurality of laboratory instruments 10 includes: detecting deviations of one or more operating parameters of the laboratory instrument 10 from the manufacturer's operating range; and / or detecting one or more deviations of one or more environmental parameters surrounding the laboratory instrument 10 based on data captured by one or more sensors located near the laboratory instrument 10 but external to it and communicatively connected to one of the plurality of data acquisition components 20.1, 20.2.

[0091] Additionally or alternatively, the data acquisition components 20.1, 20.2 detect anomalies by consulting one or more log files of the laboratory instrument to identify log entries that may point to anomalies associated with the respective laboratory instrument.

[0092] Thus, the data acquisition components 20.1, 20.2 can detect anomalies based on the acquired operational data (e.g., operational parameters, environmental data, and / or log file entries). For example, the data acquisition components 20.1, 20.2 can detect anomalies of one or more of the plurality of laboratory instruments 10 by detecting deviations of one or more operational parameters from set thresholds, by detecting deviations of one or more environmental parameters from set thresholds, and / or by identifying log entries that may point to anomalies.

[0093] Data indicative of an anomaly may include any data indicative of an anomaly, such as a deviation in operational data and / or an indication that an anomaly has occurred and what type of anomaly has occurred (e.g., as a flag in a log file). Data indicative of an anomaly may be created automatically and / or by input (e.g., from an expert).

[0094] The data acquisition components 20.1, 20.2 may detect anomalies in one or more of the plurality of laboratory instruments 10, for example, by detecting a malfunction of the laboratory instrument / system or system component and / or a deviation of the corresponding instrument from a state of operation that is considered normal or accepted. Normal or accepted operation may be defined, for example, by one or more of: operating parameters / specifications defined by the manufacturer / owner / regulatory body; statistically determined means / averages (peer comparisons) of corresponding operating parameters for the same or similar instruments; and / or operation marked as normal operation by an operator / technician / expert.

[0095] According to the embodiments disclosed herein, in step 104, an abnormality associated with one or more of the plurality of laboratory instruments 10 in the first group 1 is detected by a first one 20.1 of the plurality of data acquisition components and / or an abnormality associated with one or more of the plurality of laboratory instruments 10 in the second group 2 is detected by a second one 20.2 of the plurality of data acquisition components.

[0096] When an anomaly is detected, in step 106, the first 20.1 of the plurality of data acquisition components transmits context data to the remote maintenance system 50. The term "when..." may be understood to include transmission immediately after the anomaly is detected (where technically feasible) and / or transmission after the anomaly is detected based on a schedule and / or transmission when a network connection for such transmission between the first data acquisition component 20.1 and the remote maintenance system 50 becomes available. The context data includes operational data and data indicating the anomaly. According to the embodiments disclosed herein, the context data only includes operational data related to the anomaly, such as operational data collected during the detection of the anomaly and at a specified amount of time before and / or after the detection of the anomaly. In addition, the operational data related to the anomaly may include test result data generated by the corresponding laboratory instrument during the detection of the anomaly and / or at a specified amount of time before and / or after the detection of the anomaly.

[0097] According to the embodiments disclosed herein, when an anomaly is detected, in step 106 , the first of the plurality of data collection components 20 . 1 transmits context data to the remote maintenance system 50 , and / or the second of the plurality of data collection components 20 . 2 transmits context data to the remote maintenance system 50 .

[0098] According to the embodiments disclosed herein, the plurality of data collection components 20.1, 20.2 filter out sensitive data from the operational data collected in the plurality of laboratory instruments 10 before transmitting the operational data to the remote maintenance system 50. For example, sensitive data includes data identifying a patient; a result of an analytical test performed by one of the laboratory instruments and / or any data classified as sensitive according to a procedure or rule set by an operator of the laboratory system 100.

[0099] Based on the context data transmitted by the data acquisition components 20.1, 20.2, in step 108, one or more correlations between the operating data and one or more anomalies are determined at the remote maintenance system 50. According to the embodiments disclosed herein, one or more correlations between the operating data and one or more anomalies are determined by receiving an input indicating one or more such correlations (e.g. from an expert). Alternatively or additionally, a set of rules is used; a pattern recognition method, in particular one or more correlations between the operating data and one or more anomalies are automatically determined based on one or more previous determinations of an expert (expert-supervised pattern matching). According to further embodiments disclosed herein, a machine learning method is used to improve the detection of correlations, so that the correlations previously determined by the expert serve as teaching data.

[0100] However, since correlation is not necessarily associated with causality, in step 110, the correlation is verified at the remote maintenance system 50. A verified correlation is an indication of a (possible) causal relationship between specific operating data and anomalies. According to various embodiments disclosed herein, a degree / percentage of causal probability is set, and the correlation is verified to be above said degree / percentage. According to embodiments disclosed herein, the probability degree / percentage (the probability degree / percentage of causality, the correlation is verified to be above said probability degree / percentage) is set to different levels corresponding to different anomalies or types of anomalies, for example, by the severity / impact of the corresponding anomaly. According to embodiments disclosed herein, one or more correlations between operating data and one or more anomalies are verified by receiving input indicating one or more such verifications (e.g., from an expert). Alternatively or additionally, a set of rules is used; a pattern recognition method, in particular, one or more correlations between operating data and one or more anomalies are automatically verified based on one or more previous verifications by an expert (pattern matching supervised by an expert). According to further embodiments disclosed herein, a machine learning method is used to improve the verification of the correlation, so that the verification determined by the previous expert is used as teaching data.

[0101] The following example will further illustrate steps 102 to 110: The analytical instrument includes multiple components (e.g., a pipetting device, a heating device, a light source, a detection device), and each component is involved to obtain a measurement value, such as a QC result. The data acquisition components 20.1, 20.2 collect operation data including the QC result and the operation data of each component. The operation data of each component is provided by a sensor in the analytical instrument and indicates the condition of the component of the analytical instrument. The data acquisition component detects anomalies by detecting the deviation between the QC result and a predefined range or target value. When a deviation of the QC result is detected, the data acquisition component transmits context data including the deviation of the QC result (data indicating the abnormality) and the operation data of each component to the remote maintenance system 50. The transmitted context data only includes the operation data of the component related to the abnormality, for example, the operation data of each component collected during a specified amount of time before and / or after the deviation of the QC result is detected. Subsequently, the remote maintenance system 50 determines one or more correlations between the operation data of the component and the deviation of the QC result. The operation data of some components may be related to the deviation of the QC result, while the operation data of other components may not be related to the deviation of the QC result. For example, the operation data of the heating device and the light source are related to the deviation of the QC result, while the operation data of the pipetting device and the detection device are not related to the deviation of the QC result. Then, the determined correlation of the operation data of the heating device and the light source is verified to determine the causal relationship between one or more determined correlations of the deviation of the operation data of the corresponding components and the QC result. For example, only the condition of the light source is relevant and has an impact on the measurement of the QC result, while the condition of the heating device is irrelevant. Therefore, only the correlation between the operation data of the light source and the deviation of the QC result is verified. Therefore, based on the determination of the correlation and the subsequent verification of the determined correlation, the causal relationship between the condition of one or more of the multiple components and the abnormality can be determined.

[0102] According to the embodiments disclosed herein, before transmitting the operation data to the remote maintenance system 50, the plurality of data collection components 20.1, 20.2 filter out some of the operation data collected from the plurality of laboratory instruments 10. For example, the plurality of data collection components 20.1, 20.2 filter out the operation data whose correlation was not verified (i.e., rejected) in the previous verification. Therefore, by filtering out the operation data that will result in invalid correlation, the amount of data transmitted to the remote maintenance system 50 is further reduced.

[0103] In a subsequent step 112, one or more prediction rules corresponding to the verified correlation at the remote maintenance system 50 are determined. The prediction rule includes a condition that, if satisfied by the operating data, predicts the occurrence of an anomaly with a certain probability within a certain number of time frames. According to an embodiment disclosed herein, the probability of an anomaly occurrence and the time frame are set to different levels corresponding to different anomalies or types of anomalies, for example, by the severity / impact of the corresponding anomaly. According to an embodiment disclosed herein, the probability of an anomaly occurrence defined by one or more prediction rules is equal to the probability of a causal relationship. According to a further embodiment disclosed herein, the probability of an anomaly occurrence defined by one or more prediction rules is determined as a function of the probability of a causal relationship, for example, the probability of a causal relationship can be multiplied by a value specific to a particular group of laboratory instruments, such as an abnormal risk factor.

[0104] According to embodiments disclosed herein, one or more such prediction rules are determined by receiving an input indicating one or more prediction rules (e.g., from an expert). Alternatively or additionally, a set of rules is used; pattern recognition methods, in particular automatic determination of one or more prediction rules based on one or more previous prediction rules determined by an expert (expert-supervised pattern matching). According to further embodiments disclosed herein, machine learning methods are employed to improve the determination of one or more prediction rules, using previous prediction rules determined by experts as teaching data.

[0105] In a subsequent step 114, one or more prediction rules are transmitted by the remote maintenance system 50 to the data acquisition components 20.1, 20.2. The prediction rules are transmitted immediately after the prediction rules are determined and / or based on the allocation schedule and / or after a network connection between the remote maintenance system 50 and the data acquisition components 20.1, 20.2 becomes available.

[0106] According to the embodiments disclosed herein, one or more prediction rules are transmitted by the remote maintenance system 50 to the first one 20 . 1 of the plurality of data collection components and / or the second one 20 . 2 of the plurality of data collection components.

[0107] After receiving the one or more prediction rules, in step 116, one or more of the plurality of data acquisition components 20.1, 20.2 predicts the occurrence of one or more anomalies of one or more of the plurality of laboratory instruments 10 based on the one or more prediction rules. The step 116 of predicting the occurrence of one or more anomalies includes determining with a defined probability that the respective anomaly will occur within a certain amount of time.

[0108] By deploying prediction rules to data acquisition components 20.1, 20.2, the disclosed method / system allows timely prediction of the occurrence of anomalies. In addition, by deploying prediction rules to data acquisition components 20.1, 20.2, the disclosed method / system also allows prediction of anomalies regarding laboratory instruments 10, whose contextual data is not available (or limited) at the remote maintenance server 50 for privacy or technical reasons. This provides a significant advantage over known solutions by allowing anomaly prediction even for laboratory instruments 10 that may not upload data to a remote location themselves. According to specific embodiments disclosed herein, prediction rules can even be assigned to certain data acquisition components on offline data carriers (such as portable storage devices, optical or magnetic data carriers).

[0109] According to the embodiments disclosed herein, when the occurrence of an abnormality is predicted, the frequency and / or amount and / or parameter selection of the operating data from the laboratory instrument 10 captured by the data acquisition components 20.1, 20.2 is increased or changed. In other words, when the occurrence of an abnormality is predicted, the data acquisition components 20.1, 20.2 monitor the laboratory instrument 10 more closely, especially within a set close monitoring period.

[0110] Now go to FIG. 3A to FIG. 3E , further embodiments of the disclosed method will be described.

[0111] Figure 3A The first page of the swimlane diagram is shown, up to the inter-page connectors A, B1, and B2. Figure 3B Shows FIG. 3A to FIG. 3E The second page of the swimlane diagram, such as from the inter-page connectors A, B1, and B2.

[0112] According to further embodiments disclosed herein, Figure 3BAs shown, if in step 116 an anomaly is detected but not predicted, then in step 117 the respective data acquisition components 20.1, 20.2 transmit to the remote maintenance system 50 data indicating the failure of one or more prediction rules, in particular the failure of the prediction rules associated with the detected but not predicted anomaly. On the other hand, if the anomaly has been predicted by the data acquisition components 20.1, 20.2 using the prediction rules, then in step 118 the respective data acquisition components 20.1, 20.2 transmit data indicating the success of one or more prediction rules for predicting the anomaly. According to a specific embodiment, the success of the prediction rule is confirmed before the data indicating its success is transmitted in order to exclude false positives. A set of rules may be executed and / or used by receiving an input (from an expert) indicating the success of the prediction rule; a pattern recognition method, in particular an automatic determination of the confirmation of success based on a previous confirmation by an expert (pattern matching supervised by an expert). According to a further embodiment disclosed herein, a machine learning method is employed to improve the confirmation of the prediction rule, using the previous expert confirmation as teaching data.

[0113] In a subsequent step 120, the remote maintenance system 50 evaluates the data indicating the success or failure of one or more prediction rules. If the data indicating success outweighs the data indicating failure, one or more prediction rules are marked as valid. On the other hand, if the data indicating failure outweighs the data indicating success, one or more prediction rules are marked as invalid. The term outweighs should be understood to include a comparison of the number of success and failure indications. Alternatively or in addition, a weighting function can be applied, in which data indicating failure are considered with different weights compared to data indicating success. In addition, the source of such indications (specific data acquisition components) may affect the weight of success, respectively failure indications. After the evaluation of step 120, the remote maintenance system 50 commands the multiple data acquisition components 20.1, 20.2 to discard one or more prediction rules marked as invalid. Discarding is understood to include deleting and / or deactivating the prediction rule from the list of stored prediction rules and / or marking for (expert) review any predictions generated by the prediction rule marked as invalid.

[0114] Figure 3C Shows FIG. 3A to FIG. 3E The third page of the swimlane diagram, such as from Figure 3B Inter-page connectors C1, C2 and D.

[0115] According to the disclosure herein and Figure 3CIn the embodiment shown above, in step 130, one or more normative maintenance actions corresponding to the anomaly are determined at the remote maintenance system 50. The normative maintenance action includes one or more instructions that, when executed with respect to one or more of the laboratory instruments 10, reduce the probability of the anomaly occurring. According to the embodiments disclosed herein, one or more normative maintenance actions are determined by receiving input indicating such normative maintenance actions (e.g., from an expert). Alternatively or additionally, a set of rules is used; a pattern recognition method, in particular, one or more maintenance actions are automatically determined based on one or more previous normative maintenance actions determined by an expert (pattern matching supervised by an expert). According to a further embodiment disclosed herein, a machine learning method is used to improve the determination of one or more maintenance actions, so that the maintenance actions previously determined by the expert serve as teaching data.

[0116] In a subsequent step 132 , one or more normative maintenance actions are transmitted from the remote maintenance system 50 to the plurality of data acquisition components 20 . 1 , 20 . 2 .

[0117] Upon predicting the occurrence of an anomaly, in step 134, one or more of the plurality of data collection components 20.1, 20.2 triggers execution of one or more prescriptive maintenance actions.

[0118] According to embodiments disclosed herein, the prescriptive maintenance action comprises one or more computer-readable instructions, and the data acquisition components 20.1, 20.2 instruct the corresponding laboratory instrument 10 to perform the one or more prescriptive maintenance actions. According to further embodiments disclosed herein, the prescriptive maintenance action comprises one or more human-readable instructions to be executed on the laboratory instrument 10 by an operator / maintenance technician.

[0119] According to the embodiments disclosed herein, the one or more prescriptive maintenance actions include one or more of the following:

[0120] - Updating the software or software components on the respective laboratory instrument 10 .

[0121] - causing a redistribution of workload among the plurality of laboratory instruments 10. According to embodiments disclosed herein, the redistribution of workload among the plurality of laboratory instruments 10 comprises instructing the laboratory middleware 15 to adjust one or more load balancing rules to reduce the load on the affected instruments.

[0122] - redirecting the one or more biological samples from the one or more laboratory instruments 10 corresponding to the predicted anomaly to one or more laboratory instruments 10 other than the one or more laboratory instruments 10 corresponding to the predicted anomaly.

[0123] - triggering replacement and / or repair of one or more parts of one or more laboratory instruments 10 corresponding to the predicted anomaly;

[0124] - triggering a calibration and / or quality control process of one or more laboratory instruments 10 corresponding to the predicted anomaly;

[0125] - triggering the discarding of reagents and / or batches of reagents, and corresponding indications for improper handling and / or manufacturer recalls;

[0126] - Triggering the discarding of one or more laboratory instruments 10 , and a corresponding indication of improper maintenance.

[0127] - triggering automatic readjustment of one or more laboratory instruments or one or more components of one or more laboratory instruments

[0128] Triggering a prescriptive maintenance action includes the data acquisition component 20.1, 20.2 instructing the corresponding laboratory instrument 10 to perform one or more prescriptive maintenance actions and / or providing the prescriptive maintenance action as one or more human-readable instructions to the operator / maintenance technician for execution. The one or more human-readable instructions for the prescriptive maintenance action can be presented to the operator / maintenance technician on a user interface included by the corresponding laboratory instrument 10 or connected to the laboratory instrument, and / or sent to the operator / maintenance technician via an electronic communication device (such as an email, SMS, instant message alert). Additionally or alternatively, the one or more human-readable instructions are presented to the operator / maintenance technician on a wearable device, in particular an augmented reality device, which is configured to present the one or more prescriptive maintenance actions as a series of steps to be performed on the laboratory instrument 10.

[0129] In order to measure the effectiveness of the normative maintenance action, in step 136, the data acquisition components 20.1, 20.2 that trigger the execution of one or more normative maintenance actions detect the corresponding laboratory instrument 10 to determine whether the anomaly has been avoided. If no anomaly can be detected, in step 138, the data acquisition components 20.1, 20.2 transmit data indicating the success of one or more normative maintenance actions to the remote maintenance system 50. If, despite the triggering of the normative maintenance action, an anomaly has been detected, in step 139, the data acquisition components 20.1, 20.2 transmit data indicating the failure of one or more normative maintenance actions.

[0130] Figure 3D Shown from Figure 3C The fourth page of the swimlane diagram with page connectors E1, E2, and F. Figure 3DAs shown, in step 140, the remote maintenance system 50 evaluates the data indicating the success or failure of the normative maintenance action. If the data indicating success outweighs the data indicating failure, the normative maintenance action is marked as valid. On the other hand, if the data indicating failure crosses the data indicating success, the normative maintenance action is marked as invalid. The term outweighs should be understood to include a comparison of the number of successful and failed indications. Alternatively or in addition, a weighting function can be applied, in which data indicating failure are considered with different weights compared to data indicating success. In addition, the source of such indications (specific data acquisition components) may affect the weight of success, respectively failure indications. After the evaluation of step 140, in step 142, the remote maintenance system 50 commands multiple data acquisition components 20.1, 20.2 to discard the normative maintenance actions marked as invalid. Discarding is understood to include deleting and / or deactivating the normative maintenance actions from the list of stored normative maintenance actions and / or marking for (expert) review of any normative maintenance actions marked as invalid.

[0131] like Figure 3D As further shown, according to the embodiments disclosed herein, in order to mitigate the impact (negative impact) of the anomaly, a mitigation action is determined and assigned. In other words, despite all measures (preventive, normative or other forms of maintenance), the anomaly mitigation action takes effect only when the anomaly occurs or is expected to occur within a specific amount of time. For example, if it is inevitable that a component of the laboratory instrument 10 will fail, the anomaly mitigation action includes unloading all unprocessed biological samples from the corresponding laboratory instrument 10 and transferring them to other laboratory instruments or storage devices until the failure is corrected. According to further embodiments disclosed herein, mitigation actions are deployed to the data acquisition components 20.1, 20.2 of the group of laboratory instruments 10, which have not yet experienced anomalies but are determined by the remote maintenance system 50 to be at risk of anomalies. Mitigation actions can, for example, take the form of triggering new or updated training activities for laboratory instrument operators.

[0132] In step 150, one or more anomaly mitigation actions corresponding to the anomaly are determined at the remote maintenance system 50. The one or more anomaly mitigation actions include one or more instructions that, when executed by one or more of the laboratory instruments 10, mitigate the negative impact of the anomaly. According to an embodiment disclosed herein, one or more anomaly mitigation actions are determined by receiving input indicating one or more such anomaly mitigation actions (e.g., from an expert). Alternatively or additionally, a set of rules is used; a pattern recognition method, in particular, one or more mitigation actions are automatically determined based on previous anomaly mitigation actions determined by an expert (pattern matching supervised by an expert). According to a further embodiment disclosed herein, a machine learning method is used to improve the determination of one or more anomaly mitigation actions, using the anomaly mitigation actions previously determined by the expert as teaching data.

[0133] According to embodiments disclosed herein, the anomaly mitigation action comprises one or more computer-readable instructions, and the data acquisition components 20.1, 20.2 instruct the corresponding laboratory instrument 10 to perform the one or more anomaly mitigation actions. According to further embodiments disclosed herein, the anomaly mitigation action comprises one or more human-readable instructions to be executed on the laboratory instrument 10 by an operator / maintenance technician.

[0134] In step 152, the remote maintenance system 50 transmits one or more mitigation actions to the plurality of data collection components 20.1, 20.2.

[0135] In a subsequent step 154, if the data acquisition component 20.1, 20.2 that instructs one or more of the laboratory instruments 10 to perform one or more prescriptive maintenance actions has detected an anomaly, the data acquisition component 20.1, 20.2 triggers the execution of one or more anomaly mitigation actions. Triggering one or more anomaly mitigation actions includes the data acquisition component 20.1, 20.2 instructing the corresponding laboratory instrument 10 to perform one or more anomaly mitigation actions and / or providing the anomaly mitigation actions as one or more human-readable instructions for the operator / maintenance technician to execute.

[0136] Figure 3E Shown from Figure 3DThe fourth page of the swimlane diagram of the page connectors G1, G2 and F. In order to measure the effectiveness of the anomaly mitigation action, in step 156, the data acquisition components 20.1, 20.2 that trigger the execution of one or more anomaly mitigation actions monitor the corresponding laboratory instruments 10 to determine whether the impact of the anomaly has been mitigated (reduced). If the anomaly can be mitigated, in step 158, the data acquisition components 20.1, 20.2 transmit data indicating the success of the anomaly mitigation action to the remote maintenance system 50. If the anomaly mitigation action is not mitigated despite the triggering of the anomaly mitigation action, in step 159, the data acquisition components 20.1, 20.2 transmit data indicating the failure of one or more anomaly mitigation actions.

[0137] like Figure 3E As shown, in step 160, the remote maintenance system 50 evaluates the data indicating the success or failure of the abnormal mitigation action. If the data indicating success outweighs the data indicating failure, the abnormal mitigation action is marked as valid. On the other hand, if the data indicating failure outweighs the data indicating success, the abnormal mitigation action is marked as invalid. The term outweighs should be understood to include a comparison of the number of successful and failed indications. Alternatively or in addition, a weighting function can be applied, wherein the data indicating failure is considered with different weights compared to the data indicating success. In addition, the source of such indications (specific data acquisition components) may affect the weight of success, corresponding failure indications. After the evaluation of step 160, in step 162, the remote maintenance system 50 commands multiple data acquisition components 20.1, 20.2 to discard the abnormal mitigation action marked as invalid. Discarding is understood to include deleting and / or deactivating the abnormal mitigation action and / or marking any abnormal mitigation action marked as invalid from the list of stored abnormal mitigation actions for expert review.

[0138] Now turn to Figure 4 , embodiments of the disclosed systems and corresponding methods are disclosed that address the need for having a remote maintenance system that provides a balance between the availability of contextual data (for predictive / prescriptive maintenance and mitigation) and the privacy and regulatory constraints of its availability at a regional or global level. Figure 4 A highly schematic block diagram of a further embodiment of the disclosed laboratory system 100 is shown, wherein the remote maintenance system 50 comprises a plurality of regional servers 150, each of the regional servers being communicatively connected to a plurality of data acquisition components 20.1, 20.2.

[0139] The disclosed method of the remote maintenance system 50 corresponding to one or more regional servers 150 that are each communicatively connected to a plurality of data acquisition components 20.1, 20.2 further includes a step in which each regional server 150 analyzes operating parameters common to all laboratory instruments 10 and / or common to a subset of laboratory instruments connected to a specific regional server 150 so as to detect one or more correlations between operating data and one or more anomalies specific to the corresponding region. This method / system is advantageous because it allows the detection of correlations between anomalies and contextual data that are specific to a certain region but common to several groups of laboratory instruments within the region. For example, it has been observed that a certain consumable occasionally causes an out-of-range analytical test result. After investigation of each data acquisition component, it has been determined that the out-of-range result occurs in several locations (multiple groups of laboratory instruments), but only in a specific region. After further investigation by the regional server, it is determined that the package insert corresponding to a certain reagent batch delivered to a specific region with a specific language has a translation error about the dilution factor. In this case, the prediction rule is determined as a condition for checking the reagent batch number used by any laboratory instrument. The normative operation corresponding to this anomaly is to provide an updated package insert with a correct translation. If an incorrectly diluted reagent has been loaded into a laboratory instrument (and therefore the anomaly can no longer be avoided), the corresponding mitigation action is to stop processing any samples with this reagent and mark any test results as invalid. As this example shows, it is advantageous to have an "abstract" level for a specific region when determining correlations, prediction rules, prescriptive actions and mitigation actions that apply to the entire region. Further regional factors include common regulatory constraints for groups of laboratory instruments in a region; common shipping and storage conditions for consumables and / or common environmental factors that affect the operation of laboratory instruments (increased humidity; high altitude, such as Mexico City).

[0140] Figure 4 Further shown is a global server 250 that is communicatively connected to a plurality of regional servers 150. The terms global and regional should be interpreted in a relative manner, i.e., the global server 250 is communicatively connected to a greater group of laboratory instruments 10 than the regional servers 150 (via the regional servers and corresponding data acquisition components). The terms global and regional include geographic and / or logical groupings. An embodiment of the disclosed method corresponding to a hierarchically organized remote maintenance system 50 further comprising a global server 250 that is communicatively connected to a plurality of regional servers 150 further comprises the following steps:

[0141] - Each regional server 150 filters out regionally sensitive data from operational data;

[0142] - Each of the plurality of regional servers 150 transmits filtered operational data to the global server 250;

[0143] The global server 250 analyses operating parameters common to all laboratory instruments 10 of the laboratory system 100 in order to detect one or more global correlations between operating data and one or more anomalies regardless of region.

[0144] Global factors regardless of region include (but are not limited to) design or manufacturing parameters that affect the operation of any laboratory instrument of a certain type; globally shared maintenance and operating procedures generated after starting up one or more corresponding laboratory instruments and / or globally shared software versions and / or insights for laboratory instruments (e.g., within a service organization of a laboratory instrument provider).

[0145] Further disclosed is a computer program product comprising instructions, which when executed by a remote maintenance system 50 and a plurality of data acquisition components 20.1, 20.2 of a laboratory system 100, causes the laboratory system 100 to perform the steps of any of the methods disclosed herein, the laboratory system comprising a first set 1 and a second set 2 of laboratory instruments 10 for processing biological samples; a plurality of data acquisition components 20.1, 20.2 communicatively connected to the first set 1 or the second set 2 of laboratory instruments 10, and a remote maintenance system 50 communicatively connected to the data acquisition components 20.1, 20.2, wherein the first set 1 of laboratory instruments 10 is connected to the first data acquisition component 20.1 and the second set 2 of laboratory instruments 10 is connected to the second data acquisition component 20.2. Thus, in particular, one, more than one or even all of the method steps disclosed herein may be performed by using a computer or a computer network (such as a cloud computing service) or any suitable data processing device. As used herein, a computer program product refers to a program as a tradable product. The product may generally be present on a computer-readable data carrier, either locally or at a remote location (cloud), in any format, such as a downloadable file. The computer program product may be stored on a non-transitory computer-readable data carrier; a server computer and a transient computer-readable data carrier such as a data carrier signal. In particular, the computer program product may be distributed via a data network. Furthermore, not only the computer program product but also the execution hardware may be located in a local deployment or a remote deployment, such as in a cloud environment.

[0146] A non-transitory computer-readable storage medium comprising instructions is further disclosed and proposed, wherein the instructions, when executed by a remote maintenance system 50 of the laboratory system 100, cause the laboratory system 100 to perform the steps of any one of the methods disclosed herein, wherein the laboratory system comprises a first group 1 and a second group 2 of laboratory instruments 10 for processing biological samples; a plurality of data acquisition components 20.1, 20.2 communicatively connected to the first group 1 or the second group 2 of laboratory instruments 10, and a remote maintenance system 50 communicatively connected to the data acquisition components 20.1, 20.2.

[0147] A modulated data signal comprising instructions is further disclosed and proposed, wherein the instructions, when executed by a remote maintenance system 50 of the laboratory system 100, cause the laboratory system 100 to perform the steps of any one of the methods disclosed herein, wherein the laboratory system comprises a first group 1 and a second group 2 of laboratory instruments 10 for processing biological samples; a plurality of data acquisition components 20.1, 20.2 communicatively connected to the first group 1 or the second group 2 of laboratory instruments 10, and a remote maintenance system 50 communicatively connected to the data acquisition components 20.1, 20.2.

[0148] According to some embodiments, operational data from a laboratory instrument 10 is collected by a data collection component 20.1, 20.2 connected to the first group 1 laboratory instrument 10 and the second group 2 laboratory instrument 10, the operational data indicating one or more operational parameters of the corresponding laboratory instrument 10. An anomaly associated with one or more of the plurality of laboratory instruments 10 in the first group 1 is detected by a first data collection component 20.1 of the plurality of data collection components connected to the first group 1 laboratory instrument 10 based on the collected operational data. When an anomaly is detected, context data is transmitted to a remote maintenance system 50 by the first of the plurality of data collection components 20.1, the context data including the operational data and data indicating the anomaly. One or more correlations between the operational data and the one or more anomalies are determined at the remote maintenance system 50. The one or more correlations are verified at the remote maintenance system 50. One or more prediction rules corresponding to the verified correlations are determined at the remote maintenance system 50. The one or more prediction rules are transmitted by the remote maintenance system 50 to the data collection components 20.1, 20.2 connected to the first group 1 laboratory instrument 10 and the second group 2 laboratory instrument 10. The second data acquisition component 20.2 of the plurality of data acquisition components connected to the second group 2 of laboratory instruments 10 predicts the occurrence of one or more abnormalities of at least one of the plurality of laboratory instruments 10 in the second group 2 of laboratory instruments based on one or more prediction rules.

[0149] According to the example, the operation data from the laboratory instrument 10 is collected by a first data collection component 20.1 of the plurality of data collection components connected to the first group 1 of laboratory instruments, and is used to detect anomalies, transmit context data, determine one or more correlations between the operation data and one or more anomalies, verify the one or more correlations, and determine one or more prediction rules corresponding to the verified correlations. The operation data from the laboratory instrument 10 is collected by a second data collection component 20.2 of the plurality of data collection components connected to the second group 2 of laboratory instruments, and is used to predict the occurrence of anomalies of at least one or more of the plurality of laboratory instruments 10 in the second group 2 of laboratory instruments based on one or more prediction rules.

[0150] According to an example, operation data from a laboratory instrument 10 is collected by a first data collection component 20.1 of a plurality of data collection components connected to a first group 1 of laboratory instruments, the operation data indicating one or more operation parameters of the corresponding laboratory instrument 10. An anomaly associated with one or more of the plurality of laboratory instruments 10 in the first group 1 is detected by the first data collection component 20.1 of the plurality of data collection components connected to the first group 1 of laboratory instruments 10 based on the collected operation data. When the anomaly is detected, context data is transmitted by the first of the plurality of data collection components 20.1 to a remote maintenance system 50, the context data including the operation data and data indicating the anomaly. One or more correlations between the operation data and the one or more anomalies are determined at the remote maintenance system 50. The one or more correlations are verified at the remote maintenance system 50. One or more prediction rules corresponding to the verified correlations are determined at the remote maintenance system 50. The one or more prediction rules are transmitted by the remote maintenance system 50 to the data collection components 20.1, 20.2 connected to the first group 1 of laboratory instruments 10 and the second group 2 of laboratory instruments 10. The operation data from the laboratory instrument 10 is collected by a second data collection component 20.2 among the multiple data collection components connected to the second group 2 of laboratory instruments, the operation data indicating one or more operation parameters of the corresponding laboratory instrument 10, and is used by the second data collection component 20.2 among the multiple data collection components connected to the second group 2 of laboratory instruments 10 to predict the occurrence of one or more abnormalities of at least one of the multiple laboratory instruments 10 in the second group 2 of laboratory instruments based on one or more prediction rules.

[0151] According to some embodiments, operation data from the laboratory instrument 10 is collected by a data collection component connected to the first group 1 laboratory instrument 10, to the second group 2 laboratory instrument 10, and to the third group laboratory instrument, the operation data indicating one or more operation parameters of the corresponding laboratory instrument 10. An anomaly associated with one or more of the plurality of laboratory instruments 10 in the first group 1 is detected by a first data collection component 20.1 of the plurality of data collection components connected to the first group 1 laboratory instrument 10 based on the collected operation data and / or an anomaly associated with one or more of the plurality of second group 2 laboratory instruments 10 is detected by a second data collection component 20.2 of the plurality of data collection components connected to the second group 2 laboratory instruments 10 based on the collected operation data. When an anomaly is detected, context data is transmitted to the remote maintenance system 50 by the first of the plurality of data collection components 20.1 and / or by the second of the plurality of data collection components 20.2, the context data including the operation data and data indicating the anomaly. One or more correlations between the operation data and the one or more anomalies are determined at the remote maintenance system 50. One or more correlations are verified at the remote maintenance system 50. One or more prediction rules corresponding to the verified correlations are determined at the remote maintenance system 50. The one or more prediction rules are transmitted by the remote maintenance system 50 to the data acquisition components connected to the first group 1 of laboratory instruments 10, to the second group 2 of laboratory instruments 10, and to the third group of laboratory instruments. The third data acquisition component connected to the third group of laboratory instruments 10 among the plurality of data acquisition components predicts the occurrence of an abnormality of at least one or more of the plurality of laboratory instruments in at least the third group of laboratory instruments based on the one or more prediction rules.

[0152] According to the example, the operation data from the laboratory instrument 10 is collected by the data collection components 20.1, 20.2 connected to the first group 1 laboratory instruments 10 and the second group 2 laboratory instruments 10 for detecting one or more anomalies, transmitting context data, determining one or more correlations between the operation data and the one or more anomalies, verifying the one or more correlations and determining one or more prediction rules corresponding to the verified correlations. The operation data from the laboratory instrument is collected by a third data collection component connected to the third group of laboratory instruments among the plurality of data collection components for predicting the occurrence of anomalies of one or more of the plurality of laboratory instruments in at least the third group of laboratory instruments based on one or more prediction rules.

[0153] According to some embodiments, if an anomaly has been detected but not predicted, one or more of the plurality of data acquisition components 20.1, 20.2 connected to the first set 1 laboratory instrument 10 and the second set 2 laboratory instrument 10 transmits data indicating failure of one or more prediction rules to the remote maintenance system 50. And / or in particular if the prediction of an anomaly has been confirmed, one or more of the plurality of data acquisition components 20.1, 20.2 connected to the first set 1 laboratory instrument 10 and the second set 2 laboratory instrument 10 transmits data indicating success of one or more prediction rules predicting the anomaly to the remote maintenance system 50. The remote maintenance system 50 evaluates the data indicating success or failure of the one or more prediction rules, marks the one or more prediction rules as valid if the data indicating success outweighs the data indicating failure, and marks the one or more prediction rules as invalid if the data indicating failure outweighs the data indicating success. The remote maintenance system 50 instructs the plurality of data acquisition components 20.1, 20.2 connected to the first set 1 laboratory instrument 10 and the second set 2 laboratory instrument 10 to discard the one or more prediction rules marked as invalid.

[0154] According to some embodiments, one or more normative maintenance actions corresponding to the anomaly are determined at the remote maintenance system 50. The one or more normative maintenance actions include one or more instructions that, when executed with respect to one or more of the laboratory instruments 10, reduce the probability of the anomaly occurring. The one or more normative maintenance actions are transmitted from the remote maintenance system 50 to a plurality of data acquisition components 20.1, 20.2 connected to the first group 1 laboratory instruments 10 and the second group 2 laboratory instruments 10. When the occurrence of the anomaly is predicted, one or more of the plurality of acquisition components 20.1, 20.2 connected to the first group 1 laboratory instruments 10 and the second group 2 laboratory instruments 10 trigger the execution of the one or more normative maintenance actions, including commanding one or more of the laboratory instruments 10 to perform the one or more normative maintenance actions and / or providing one or more normative maintenance actions to the operator.

[0155] According to some embodiments, if the data collection components 20.1, 20.2 that trigger the execution of one or more normative maintenance actions cannot detect an anomaly, data indicating the success of the normative maintenance action is transmitted to the remote maintenance system 50. If the data collection components 20.1, 20.2 that trigger the execution of one or more normative maintenance actions have detected an anomaly, data indicating the failure of the normative maintenance action is transmitted to the remote maintenance system 50. The remote maintenance system 50 evaluates the data indicating the success or failure of the normative maintenance action, marks the one or more normative maintenance actions as valid if the data indicating success outweighs the data indicating failure, and marks the one or more normative maintenance actions as invalid if the data indicating failure outweighs the data indicating success. The remote maintenance system 50 commands the multiple data collection components 20.1, 20.2 connected to the first group 1 laboratory instruments 10 and the second group 2 laboratory instruments 10 to discard the one or more normative maintenance actions marked as invalid.

[0156] According to some embodiments, one or more anomaly mitigation actions corresponding to the anomaly are determined at the remote maintenance system 50. The one or more anomaly mitigation actions include one or more instructions that, when executed by one or more of the laboratory instruments 10, mitigate the negative impact of the anomaly. The one or more mitigation actions are transmitted from the remote maintenance system 50 to a plurality of data acquisition components 20.1, 20.2 connected to the first set 1 laboratory instruments 10 and the second set 2 laboratory instruments 10. If the data acquisition component 20.1, 20.2 that instructs one or more of the laboratory instruments 10 to perform one or more prescriptive maintenance actions has detected an anomaly, the data acquisition component 20.1, 20.2 triggers execution of the one or more anomaly mitigation actions.

[0157] According to some embodiments, if the negative impact of the abnormality has been mitigated, data indicating the success of one or more mitigation actions are transmitted from the plurality of data acquisition components 20.1, 20.2 connected to the first group 1 laboratory instrument 10 and the second group 2 laboratory instrument 10 to the remote maintenance system 50. If the negative impact of the abnormality cannot be mitigated, data indicating the failure of one or more mitigation actions are transmitted to the remote maintenance system 50. The remote maintenance system 50 integrates the data indicating the success or failure of the one or more mitigation actions, marks the one or more mitigation actions as valid if the data indicating success outweighs the data indicating failure, and marks the one or more mitigation actions as invalid if the data indicating failure outweighs the data indicating success. The remote maintenance system 50 commands the plurality of data acquisition components 20.1, 20.2 connected to the first group 1 laboratory instrument 10 and the second group 2 laboratory instrument 10 to discard the one or more mitigation actions marked as invalid.

[0158] According to some embodiments, the step of detecting an anomaly associated with one or more of the plurality of laboratory instruments 10 includes:

[0159] - detecting one or more deviations of one or more operating parameters of the laboratory instrument 10 from the manufacturer's operating range;

[0160] - Detecting one or more deviations of one or more environmental parameters surrounding the laboratory instrument 10 based on data captured by one or more sensors 12, wherein the one or more sensors are located near the laboratory instrument 10 but external to it and are communicatively connected to one of a plurality of data acquisition components 20.1, 20.2, which are connected to the first group 1 laboratory instruments 10 and the second group 2 laboratory instruments 10.

[0161] According to some embodiments, the maintenance method further includes increasing the frequency and / or amount and / or selection of parameters of operating data from the laboratory instrument 10 captured by the data acquisition components 20.1, 20.2 connected to the first group 1 laboratory instruments 10 and the second group 2 laboratory instruments 10 when the occurrence of an abnormality is predicted.

[0162] According to some embodiments, the remote maintenance system 50 comprises one or more regional servers 150 each communicatively connected to the plurality of data acquisition components 20.1, 20.2. The method further comprises the step of each regional server 150 analyzing operating parameters common to all laboratory instruments 10 and / or common to a subset of laboratory instruments connected to a particular regional server 150 in order to detect one or more correlations between operating data and one or more anomalies specific to the respective region. According to some embodiments, each of the plurality of data acquisition components 20.1, 20.2 communicatively connected to the regional server 150 is dedicated to a particular set of laboratory instruments.

[0163] According to a first aspect, a maintenance method for a laboratory system is provided, the laboratory system comprising a first group and a second group of laboratory instruments for processing biological samples; a plurality of data acquisition components communicatively connected to the first group or the second group of laboratory instruments and a remote maintenance system communicatively connected to the data acquisition components, the method comprising:

[0164] - collecting, by the data collection component, operating data from the laboratory instrument, the operating data being indicative of one or more operating parameters of the respective laboratory instrument;

[0165] - detecting, by a first of a plurality of data acquisition components, an anomaly associated with one or more of a plurality of laboratory instruments in a first group;

[0166] - when an anomaly is detected, transmitting context data to a remote maintenance system by a first one of the plurality of data acquisition components, the context data comprising the operational data and data indicating the anomaly;

[0167] - determining, at the remote maintenance system, one or more correlations between the operational data and the one or more anomalies;

[0168] - verifying one or more dependencies at the remote maintenance system;

[0169] - determining, at the remote maintenance system, one or more prediction rules corresponding to the verified correlations;

[0170] - Transmitting one or more prediction rules from the remote maintenance system to the data acquisition component;

[0171] - predicting, by one or more of the plurality of data acquisition components, the occurrence of an anomaly of one or more of the plurality of laboratory instruments based on one or more prediction rules.

[0172] According to a second aspect, a maintenance method according to the first aspect is provided, the method further comprising:

[0173] - if an anomaly has been detected but not predicted, one or more of the plurality of data acquisition components transmits data indicating failure of one or more prediction rules to the remote maintenance system; and / or

[0174] - in particular if the prediction of the anomaly has been confirmed, one or more of the plurality of data acquisition components transmits to the remote maintenance system data indicating the success of the one or more prediction rules predicting the anomaly;

[0175] - The remote maintenance system evaluates the data indicating the success or failure of one or more predictive rules:

[0176] - marking one or more prediction rules as valid if the data indicating success outweighs the data indicating failure; and

[0177] - marking one or more prediction rules as invalid in the event that data indicating failure outweighs data indicating success;

[0178] The remote maintenance system instructs the plurality of data collection components to discard one or more prediction rules marked as invalid.

[0179] According to a third aspect, a maintenance method according to the first aspect or the second aspect is provided, wherein the method further comprises the following steps:

[0180] - determining, at the remote maintenance system, one or more prescriptive maintenance actions corresponding to the anomaly, the one or more prescriptive maintenance actions comprising one or more instructions that, when executed with respect to one or more of the laboratory instruments, reduce the probability of the anomaly occurring;

[0181] -Transmitting one or more prescriptive maintenance actions from the remote maintenance system to multiple data acquisition components;

[0182] -When an abnormality is predicted to occur, one or more of the plurality of acquisition components triggers execution of one or more prescriptive maintenance actions, including commanding one or more of the laboratory instruments to execute one or more prescriptive maintenance actions and / or providing one or more prescriptive maintenance actions to an operator.

[0183] According to a fourth aspect, a maintenance method according to the third aspect is provided, the method further comprising the following steps:

[0184] - if the data collection component that triggers the execution of one or more prescriptive maintenance actions fails to detect the anomaly, transmitting data indicating the success of the prescriptive maintenance action to the remote maintenance system;

[0185] - if the data acquisition component that triggered the execution of one or more prescriptive maintenance actions has detected an anomaly, transmitting data indicating the failure of the prescriptive maintenance action to the remote maintenance system;

[0186] - The remote maintenance system evaluates the data indicating the success or failure of the prescriptive maintenance action:

[0187] - marking one or more normative maintenance actions as valid where the data indicating success outweighs the data indicating failure; and

[0188] - marking one or more normative maintenance actions as invalid in the event that data indicating failure outweighs data indicating success;

[0189] The remote maintenance system instructs the plurality of data collection components to discard one or more normative maintenance actions marked as invalid.

[0190] According to a fifth aspect, a maintenance method according to the third aspect or the fourth aspect is provided, wherein the one or more normative maintenance actions include one or more of the following:

[0191] - Update SW on corresponding laboratory instruments;

[0192] - Causes a redistribution of workload between multiple laboratory instruments;

[0193] - redirecting one or more biological samples from the one or more laboratory instruments corresponding to the predicted abnormality to one or more laboratory instruments other than the one or more laboratory instruments corresponding to the predicted abnormality;

[0194] - triggering replacement and / or repair of one or more parts of one or more laboratory instruments corresponding to the predicted anomaly;

[0195] - triggering a calibration and / or quality control process of one or more laboratory instruments corresponding to the predicted anomaly;

[0196] - triggering the discarding of reagents and / or batches of reagents, and corresponding indications for improper handling and / or manufacturer recalls;

[0197] - Triggering the discarding of one or more laboratory instruments, and corresponding indication of improper maintenance.

[0198] According to a sixth aspect, a maintenance method according to any one of the third to fifth aspects is provided, wherein the method further comprises the following steps:

[0199] - determining, at the remote maintenance system, one or more anomaly mitigation actions corresponding to the anomaly, the one or more anomaly mitigation actions comprising one or more instructions that, when executed by one or more of the laboratory instruments, mitigate a negative impact of the anomaly;

[0200] - transmitting one or more mitigation actions from the remote maintenance system to the plurality of data acquisition components;

[0201] - If a data collection component that commands one or more of the laboratory instruments to perform one or more prescriptive maintenance actions has detected an anomaly, the data collection component triggers the execution of one or more anomaly mitigation actions.

[0202] According to a seventh aspect, a maintenance method according to the sixth aspect is provided, the method further comprising the following steps:

[0203] - if the negative impact of the anomaly has been mitigated, transmitting data indicating the success of one or more mitigation actions from the plurality of data acquisition components to the remote maintenance system;

[0204] - if the negative impact of the anomaly cannot be mitigated, transmitting data indicating the failure of one or more mitigation actions to the remote maintenance system;

[0205] - The remote maintenance system integrates the data indicating the success or failure of one or more mitigation actions:

[0206] - marking one or more mitigation actions as valid where the data indicating success outweighs the data indicating failure; and

[0207] - marking one or more mitigation actions as invalid in the event that data indicating failure outweighs data indicating success;

[0208] The remote maintenance system instructs the plurality of data collection components to discard one or more mitigation actions marked as invalid.

[0209] According to an eighth aspect, a maintenance method according to any of the preceding aspects is provided, wherein the step of detecting an anomaly associated with one or more of the plurality of laboratory instruments comprises:

[0210] - Detecting one or more deviations of one or more operating parameters of a laboratory instrument from the manufacturer's operating range;

[0211] - Detecting one or more deviations in one or more environmental parameters surrounding the laboratory instrument based on data captured by one or more sensors located near but external to the laboratory instrument and communicatively connected to one of the plurality of data acquisition components.

[0212] According to a ninth aspect, a maintenance method according to any one of the preceding aspects is proposed, the method further comprising increasing the frequency and / or quantity and / or parameter selection of the operating data from the laboratory instrument captured by the data acquisition component when an abnormality is predicted to occur.

[0213] According to a tenth aspect, a maintenance method according to any one of the preceding aspects is provided, wherein the step of determining, at the remote maintenance system, one or more correlations between the operating data and the one or more anomalies comprises:

[0214] - receiving input indicating one or more correlations between said operational data and said one or more anomalies; and / or

[0215] - A command pattern recognition system identifies correlations between the operational data and one or more anomalies of the same type based on a database of the operational data and anomalies associated with the same or other laboratory instruments in the plurality of laboratory instruments.

[0216] According to an eleventh aspect, a maintenance method according to any one of the preceding aspects is provided, wherein the step of verifying one or more dependencies and marking the corresponding dependencies as verified comprises:

[0217] - receiving input indicating the validity of one or more correlations between the operational data and the one or more anomalies; and / or

[0218] - Commanding a validation engine to validate said one or more correlations between said operational data and said one or more anomalies.

[0219] According to the twelfth aspect, a maintenance method according to any one of the preceding aspects is proposed, wherein the remote maintenance system includes one or more regional servers each communicatively connected to a plurality of data acquisition components, and the method further includes a step in which each regional server analyzes operating parameters common to all laboratory instruments and / or common to a subset of laboratory instruments connected to a specific regional server in order to detect one or more correlations between operating data and one or more anomalies specific to the corresponding region.

[0220] According to a thirteenth aspect, a maintenance method according to the twelfth aspect, wherein the remote maintenance system comprises a global server communicatively connected to a plurality of regional servers; the method further comprises:

[0221] - Each regional server filters out regionally sensitive data from operational data;

[0222] -each of the plurality of regional servers transmits filtered operational data to the global server; and

[0223] - The global server analyses operating parameters common to all laboratory instruments of the laboratory system in order to detect one or more global correlations between the operating data and one or more anomalies regardless of region.

[0224] According to a fourteenth aspect, a laboratory system is provided, comprising:

[0225] - a first plurality of laboratory instruments for processing biological samples, said first plurality of laboratory instruments being communicatively connected to a first data acquisition component;

[0226] - a second plurality of laboratory instruments for processing biological samples, said second plurality of laboratory instruments being communicatively connected to a second data acquisition component;

[0227] - a remote maintenance system, the remote maintenance system being communicatively connected to a plurality of data acquisition components;

[0228] The laboratory system is configured to perform the method according to any one of the first to thirteenth aspects.

[0229] According to a fifteenth aspect, a computer program product comprising instructions is provided, wherein when the instructions are executed by a remote maintenance system of a laboratory system, the laboratory system performs the steps of any one of the methods described in any one of the first to thirteenth aspects, wherein the laboratory system comprises a first group and a second group of laboratory instruments for processing biological samples; a plurality of data acquisition components communicatively connected to the first group or the second group of laboratory instruments, and a remote maintenance system communicatively connected to the data acquisition components.

[0230] List of reference numerals:

[0231] Group 1 (Laboratory Instruments) 1

[0232] Group 2 (Laboratory Instruments) 2

[0233] Laboratory Instruments10

[0234] Laboratory Middleware 15

[0235] Laboratory System 100

[0236] First data acquisition component 20.1

[0237] Second data acquisition component 20.2

[0238] Remote maintenance system50

[0239] Collecting operation data step 102

[0240] Detect one or more anomalies step 104

[0241] Transmit context data step 106

[0242] Determine one or more correlations (between context data and anomalies) step 108

[0243] Verify the correlation step 110

[0244] Determine one or more prediction rules Step 112

[0245] Transmit one or more prediction rules to the data collection component step 114

[0246] Predict one or more abnormal steps 116

[0247] Transmit data indicating that one or more prediction rules have failed Step 117

[0248] Transmitting data indicating success of one or more prediction rules Step 118

[0249] Evaluate the effectiveness of one or more prediction rules step 120

[0250] Discard one or more invalid prediction rules step 122

[0251] Determine the steps for prescriptive maintenance actions 130

[0252] Transmission specification maintenance action step 132

[0253] Triggering normative maintenance action step 134

[0254] Determine if the exception step 136 has been avoided

[0255] Transmit data indicating the success of the prescriptive maintenance operation Step 138

[0256] Transfer data indicating failure of prescriptive maintenance action Step 139

[0257] Evaluating the effectiveness of prescriptive maintenance actions Step 140

[0258] Discard invalid prescriptive maintenance action steps 142

[0259] Determine abnormal mitigation action step 150

[0260] Transmission anomaly mitigation action steps 152

[0261] Trigger abnormal mitigation action step 154

[0262] Determine whether the anomaly has been mitigated Step 156

[0263] Transmit data indicating success of anomaly mitigation action Step 158

[0264] Transmit data indicating failure of anomaly mitigation action Step 158

[0265] Evaluate the effectiveness of anomaly mitigation actions Step 160

[0266] Discard invalid exception mitigation actions step 162

Claims

1. A maintenance method for a laboratory system (100), The laboratory system comprises a first set (1) and a second set (2) of laboratory instruments (10) for processing biological samples; a plurality of data acquisition components; and a remote maintenance system (50) communicatively connected to the data acquisition components, wherein the first group (1) of laboratory instruments (10) is connected to a first data acquisition component (20.1), The second group (2) of laboratory instruments (10) is connected to a second data acquisition component (20.2), the method comprising: - collecting, by the data collection component, operating data from the laboratory instrument (10), the operating data being indicative of one or more operating parameters of the respective laboratory instrument (10); - detecting, by the first data acquisition component (20.1) of the plurality of data acquisition components, an anomaly associated with one or more of the plurality of laboratory instruments (10) in the first group (1) based on the acquired operating data, wherein the first data acquisition component (20.1) detects the anomaly by detecting a deviation of the one or more operating parameters from a set threshold value; - when an abnormality is detected, the first data acquisition component (20.1) of the plurality of data acquisition components transmits context data to the remote maintenance system (50), the context data comprising operating data and data indicating the abnormality, the operating data being acquired during the detection of the abnormality and before and / or after the detection of the abnormality, and the data indicating the abnormality comprising test result data of the laboratory instrument (10) generated during the detection of the abnormality and / or before and / or after the detection of the abnormality, wherein sensitive data is filtered out; - determining, at the remote maintenance system (50), one or more correlations between the operational data and one or more of the anomalies based on the contextual data, wherein the one or more correlations between the operational data and one or more of the anomalies are automatically determined using a pattern recognition method; - verifying the one or more correlations at the remote maintenance system (50), wherein the one or more correlations between the operational data and the one or more anomalies are verified by instructing a computer-implemented verification engine to verify the one or more correlations between the operational data and the one or more anomalies; - determining, at the remote maintenance system (50), one or more prediction rules corresponding to the verified correlations; - transmitting the one or more prediction rules to the data acquisition component by the remote maintenance system (50); - predicting, by one or more of the plurality of data acquisition components, the occurrence of an anomaly in one or more of the plurality of laboratory instruments (10) based on the one or more prediction rules.

2. The maintenance method according to claim 1, further comprising: - if an anomaly has been detected but not predicted, one or more of the plurality of data acquisition components transmits data indicating failure of the one or more prediction rules to the remote maintenance system (50); and / or - in particular if the prediction of said anomaly has been confirmed, one or more of said plurality of data acquisition means transmits to said remote maintenance system (50) data indicating the success of said one or more prediction rules predicting said anomaly; - the remote maintenance system (50) evaluates the data indicating success or failure of one or more predictive rules: - marking one or more prediction rules as valid where data indicating success outweighs data indicating failure; as well as - marking one or more prediction rules as invalid in the event that data indicating failure outweighs data indicating success; The remote maintenance system (50) instructs the plurality of data collection components to discard one or more prediction rules marked as invalid.

3. The maintenance method according to claim 1 further comprises the following steps: - determining, at the remote maintenance system (50), one or more prescriptive maintenance actions corresponding to the anomaly, the one or more prescriptive maintenance actions comprising one or more instructions that, when executed with respect to one or more of the laboratory instruments (10), reduce the probability of the anomaly occurring; - transmitting the one or more normative maintenance actions from the remote maintenance system (50) to the plurality of data acquisition components; -When the occurrence of an abnormality is predicted, one or more of the multiple data acquisition components trigger the execution of the one or more normative maintenance actions, including commanding one or more of the laboratory instruments (10) to perform the one or more normative maintenance actions and / or providing the one or more normative maintenance actions to an operator.

4. The maintenance method according to claim 3 further comprises the following steps: - if the data acquisition component that triggers the execution of the one or more normative maintenance actions is unable to detect the anomaly, transmitting data indicating the success of the normative maintenance action to the remote maintenance system (50); - if the data acquisition component that triggers the execution of the one or more normative maintenance actions has detected the anomaly, transmitting data indicating the failure of the normative maintenance action to the remote maintenance system (50); - said remote maintenance system (50) evaluates said data indicating success or failure of a prescriptive maintenance action: - marking one or more normative maintenance actions as valid where data indicating success outweighs data indicating failure; as well as - marking one or more normative maintenance actions as invalid in the event that data indicating failure outweighs data indicating success; The remote maintenance system (50) instructs the plurality of data collection components to discard one or more normative maintenance actions marked as invalid.

5. The maintenance method according to claim 3, wherein: The one or more prescriptive maintenance actions include one or more of the following: - Updating the software on the corresponding laboratory instrument (10); - causing a redistribution of the workload between a plurality of laboratory instruments (10); - redirecting one or more biological samples from the one or more laboratory instruments (10) corresponding to the predicted abnormality to one or more laboratory instruments (10) other than the one or more laboratory instruments (10) corresponding to the predicted abnormality; - triggering replacement and / or repair of one or more parts of one or more laboratory instruments (10) corresponding to the predicted anomaly; - triggering a calibration and / or quality control process of one or more laboratory instruments (10) corresponding to the predicted anomaly; - triggering the discarding of reagents and / or batches of reagents, and corresponding indications for improper handling and / or manufacturer recalls; - triggering the discarding of one or more laboratory instruments (10) and a corresponding indication of improper maintenance; - triggering automatic readjustment of one or more laboratory instruments or one or more components of one or more laboratory instruments.

6. The maintenance method according to claim 3 further comprises the following steps: - determining, at the remote maintenance system (50), one or more anomaly mitigation actions corresponding to the anomaly, the one or more anomaly mitigation actions comprising one or more instructions that, when executed by one or more of the laboratory instruments (10), mitigate a negative impact of the anomaly; - transmitting the one or more mitigation actions from the remote maintenance system (50) to the plurality of data acquisition components; - If a data acquisition component commanding one or more of the laboratory instruments (10) to perform the one or more prescriptive maintenance actions has detected the anomaly, the data acquisition component triggers the execution of one or more anomaly mitigation actions.

7. The maintenance method according to claim 6 further comprises the following steps: - if the negative impact of the anomaly has been mitigated, transmitting data indicating the success of the one or more mitigation actions from the plurality of data acquisition components to the remote maintenance system (50); - if the negative impact of the anomaly cannot be mitigated, transmitting data indicating the failure of the one or more mitigation actions to the remote maintenance system (50); - said remote maintenance system (50) integrates said data indicating success or failure of one or more mitigation actions: - marking one or more mitigation actions as valid where data indicating success outweighs data indicating failure; as well as - marking one or more mitigation actions as invalid in the event that data indicating failure outweighs data indicating success; The remote maintenance system (50) instructs the plurality of data collection components to discard one or more mitigation actions marked as invalid.

8. The maintenance method according to claim 1, wherein: The step of detecting an anomaly associated with one or more of the plurality of laboratory instruments (10) comprises: - detecting one or more deviations of one or more operating parameters of the laboratory instrument (10) from the manufacturer's operating range; - Detecting one or more deviations of one or more environmental parameters surrounding the laboratory instrument (10) based on data captured by one or more sensors (12), the one or more sensors being located near but external to the laboratory instrument (10) and being communicatively connected to one of the plurality of data acquisition components.

9. The maintenance method according to claim 1 further comprises the following steps: In predicting the occurrence of an anomaly, the frequency and / or amount and / or selection of parameters of the operational data from the laboratory instrument (10) captured by the data acquisition component is increased.

10. The maintenance method according to claim 1, wherein: The step of determining, at the remote maintenance system (50), one or more correlations between the operating data and the one or more anomalies comprises: - A command pattern recognition system identifies, based on a database of operational data and anomalies associated with the same or other laboratory instruments of the plurality of laboratory instruments (10), a correlation between the operational data and one or more anomalies of the same type.

11. The maintenance method according to claim 1, wherein: The remote maintenance system (50) includes one or more regional servers (150) each communicatively connected to a plurality of data acquisition components, the method further comprising the step of each regional server (150) analyzing operating parameters common to all laboratory instruments (10) and / or common to a subset of the laboratory instruments (10) connected to the corresponding regional server (150) to detect one or more correlations between the operating data and one or more anomalies specific to the corresponding region.

12. The maintenance method according to claim 11, wherein: The remote maintenance system (50) includes a global server (250) communicatively connected to a plurality of regional servers (150); the method further includes: - Each regional server (150) filters out regionally sensitive data from the operational data; - each of the plurality of regional servers (150) transmits the filtered operational data to the global server (250); and - the global server (250) analyses operating parameters common to all laboratory instruments (10) of the laboratory system (100) in order to detect one or more global correlations between operating data and one or more anomalies regardless of region.

13. A laboratory system (100), comprising: - a first set (1) of a plurality of laboratory instruments (10) for processing biological samples, said first set of a plurality of laboratory instruments being communicatively connected to a first data acquisition component (20.1); - a second set (2) of laboratory instruments (10) for processing biological samples, said second set of laboratory instruments being communicatively connected to a second data acquisition component (20.2); - a remote maintenance system (50), the remote maintenance system being communicatively connected to a plurality of data acquisition components; The first data acquisition component (20.1) and the second data acquisition component (20.2) are included in the laboratory system (100); wherein: The first data acquisition component (20.1) and the second data acquisition component (20.2) are configured to acquire operating data from the laboratory instrument (10), the operating data indicating one or more operating parameters of the corresponding laboratory instrument (10); The first data acquisition component (20.1) is configured to: detecting anomalies associated with one or more of the plurality of laboratory instruments (10) of the first group (1) based on the acquired operational data by detecting deviations of the one or more operational parameters from set thresholds; and When an anomaly is detected, context data is transmitted to the remote maintenance system (50), the context data comprising operating data and data indicating the anomaly, the operating data being collected during the detection of the anomaly and before and / or after the detection of the anomaly, and the data indicating the anomaly comprising test result data of the laboratory instrument (10) generated during the detection of the anomaly and / or before and / or after the detection of the anomaly, wherein sensitive data is filtered out; and The remote maintenance system (50) is configured as follows: determining one or more correlations between the operational data and the one or more anomalies based on the contextual data, wherein the one or more correlations between the operational data and the one or more anomalies are automatically determined using a pattern recognition method; verifying the one or more correlations, wherein the one or more correlations between the operational data and the one or more anomalies are verified by instructing a computer-implemented verification engine to verify the one or more correlations between the operational data and the one or more anomalies; determining one or more prediction rules corresponding to the validated correlations; and transmitting the one or more prediction rules to the data collection component; One or more of the plurality of data acquisition components is configured to predict the occurrence of an abnormality of one or more of the plurality of laboratory instruments (10) based on the one or more prediction rules.

14. A computer program product comprising instructions for execution by a remote maintenance system (50) of a laboratory system (100), The laboratory system includes: A remote maintenance system (50), a plurality of data acquisition components, and a first group (1) and a second group (2) of laboratory instruments (10) for processing biological samples; wherein the remote maintenance system (50) is communicatively connected to the data acquisition components, and wherein the first group (1) of laboratory instruments (10) is connected to a first data acquisition component (20.1), and the second group (2) of laboratory instruments (10) is connected to a second data acquisition component (20.2), wherein: The first data acquisition component (20.1) and the second data acquisition component (20.2) are configured to acquire operating data from the laboratory instrument (10), the operating data indicating one or more operating parameters of the corresponding laboratory instrument (10); and The first data acquisition component (20.1) is configured to: detecting anomalies associated with one or more of the plurality of laboratory instruments (10) of the first group (1) based on the acquired operational data by detecting deviations of the one or more operational parameters from set thresholds; and When an anomaly is detected, transmitting context data to the remote maintenance system (50), the context data comprising operating data and data indicating the anomaly, the operating data being acquired during the detection of the anomaly and before and / or after the detection of the anomaly, and the data indicating the anomaly comprising test result data of the laboratory instrument (10) generated during the detection of the anomaly and / or before and / or after the detection of the anomaly, wherein sensitive data is filtered out; The computer program product, when executed by the remote maintenance system (50), enables the remote maintenance system (50): determining one or more correlations between the operational data and the one or more anomalies based on the contextual data, wherein the one or more correlations between the operational data and the one or more anomalies are automatically determined using a pattern recognition method; verifying the one or more correlations, wherein the one or more correlations between the operational data and the one or more anomalies are verified by instructing a computer-implemented verification engine to verify the one or more correlations between the operational data and the one or more anomalies; determining one or more prediction rules corresponding to the validated correlations; and The one or more prediction rules are transmitted to the data acquisition component for predicting the occurrence of anomalies of one or more of the plurality of laboratory instruments (10) based on the one or more prediction rules.

Images