Medical device monitoring agent

Abstract

The proposed concepts are directed to providing schemes, solutions, concepts, designs, methods and systems relating to monitoring medical devices. In particular, usage data for medical devices and other (related) medical devices are collected. The data is used to learn normal operating values of the medical device. Thus, any deviation of the operation of the medical device from the normal operating value and the standard operating strategy can be determined. The deviation is then used to generate an operational problem signal (i.e., a user alert, corrective action, etc. In this way, operational problems can be automatically detected in the early stage and may be corrected in the early stage.

Description

Technical Field

[0001] The present invention relates to the field of monitoring agents, and more particularly to the field of agents for monitoring medical devices. Background Art

[0002] Medical devices often require skilled technicians for maintenance and servicing. However, due to limited labor, owners of such devices must wait a long time for reported problems to be diagnosed and corrected. In addition, medical devices are often essential for rapid diagnosis and treatment of patients. Therefore, convenient maintenance of medical devices is key to providing effective care.

[0003] In order to understand the characteristics of erroneous / malfunctioning behavior of a medical device (and its components), it is important to understand the typical / normal usage patterns of the device. Skilled technicians are trained to know the normal usage patterns of medical devices, but technicians do not experience every deviation from normal behavior.

[0004] Therefore, automatically detecting potential future problems will solve many problems. In addition, fixing problems on the spot saves technicians’ time, which means they can allocate their time more efficiently to more complex problems. In addition, early diagnosis and correction of faulty behavior can prevent more serious failures, also saving time and costs.

[0005] Additionally, in cases where the failure is due to a specialized component, it may take a long time to replace the component, so any time that can be saved can prevent the equipment from being offline for an unmanageable amount of time.

[0006] Therefore, there is a need for a means to effectively monitor medical devices in order to detect and potentially correct malfunctioning behavior.

[0007] US2020 / 210850A1 describes a method and system for intelligent search of image modality maintenance, including an image recognition imaging device based on an imaging device. Summary of the invention

[0008] The invention is defined by the claims.

[0009] According to an example of one aspect of the present invention, there is provided an agent for monitoring a medical device, the agent comprising:

[0010] an interface configured to obtain usage data describing the operation of the medical device;

[0011] a network module configured to obtain usage data describing the operation of at least one other medical device related to the medical device;

[0012] a learning module configured to determine a normal operating value for the medical device based on the usage data for the medical device and the usage data for the at least one other medical device, and

[0013] An analysis module configured to:

[0014] comparing the usage data of the medical device to the normal operating values ​​and standard operating policies of the medical device; and

[0015] Based on the comparison, an operating problem signal is generated, which describes a difference between the detected usage data of the medical device and the normal operating value of the medical device.

[0016] Therefore, a concept related to monitoring a medical device is proposed. In particular, usage data of the medical device and other (related) medical devices are collected. This data is used to learn normal operating values ​​of the medical device. Thus, any deviations of the operation of the medical device from normal operating values ​​and standard operating strategies can be determined. The deviations are then used to generate operating problem signals (i.e. user alerts, corrective actions, etc.). In this way, operating problems can be automatically detected at an early stage and possibly corrected at an early stage.

[0017] In practice, by utilizing both data from the medical device monitored by the agent and data from other (related) medical devices, the normal / typical operation of the medical device is accurately portrayed. Determining accurate normal operation values ​​is critical to accurately detecting erroneous operation of the medical device (i.e., due to faulty components, incorrect operating conditions / parameters, etc.). In practice, in order for detected differences / deviations in the use of the medical device to accurately reflect an erroneous state, the normal operation values ​​must accurately define how the medical device should operate.

[0018] Furthermore, the same type of medical device may be used in a variety of environments, in a variety of operating modes, and for a variety of purposes. Therefore, using a medical device in a wide range of use cases is critical to understanding the (range of) normal operating values ​​for the medical device. Typically, a single instance of a medical device may only be used in one environment, in one mode, and for one purpose. Therefore, it may not be possible to determine the full range of normal operating values ​​from usage data for one device. Therefore, embodiments of the present invention overcome this problem by collecting usage data from many related medical devices in order to fully understand the various operating conditions.

[0019] Additionally, the usage data of the medical device is compared to standard operating policies and normal operating values. This may reflect the fact that the device is (or is not) intentionally configured to operate in a manner that deviates from normal / typical operation of the medical device. Thus, a more accurate assessment of the deviation can be made, thereby more appropriately signaling an operational problem.

[0020] When a discrepancy between the usage data and the normal operating values ​​is detected, an operational problem signal is generated. Such a signal may indicate to a user, operator and / or technician that the medical device is operating in an incorrect manner. Thus, this may indicate the presence of a fault, the nature of the fault, and a recommended course of action for correcting the fault (i.e., restoring the medical device to a normal operating state). In more complex embodiments, the operational problem signal may be used directly to correct the fault (i.e., by changing an operating parameter of the medical device).

[0021] In general, an agent may be a piece of hardware installed on / near a medical device. Alternatively, an agent may be software-based and installed in the operating system of a medical device. In any case, an agent may be suitably provided to monitor the medical device (i.e., access system logs, sensors, etc.) and also provide operational problem signals.

[0022] In some embodiments, the agent also includes a recommendation module, which is configured to: process the usage data of the medical device and the usage data of at least one other medical device to generate a best practice recommendation describing an operating mode of the medical device; and transmit the best practice recommendation to a user associated with the medical device.

[0023] Another benefit of obtaining information from other (related) medical devices is that device usage data can be evaluated to determine which medical devices are operating in an improved (i.e., more efficient, more accurate, safer, etc.) manner. For example, if it is found that certain operating parameters result in more accurate readings from a medical device, then such operating parameters can be recommended to the user. Alternatively, if it is determined that operating a medical device at a lower temperature means that the medical device is operating in a more efficient manner, then this approach can be recommended to the user.

[0024] Thus, an agent according to such embodiments provides a means for informing a user how to operate a medical device in a more efficient manner.

[0025] In some embodiments, the network module can be configured to: establish a communication channel between the agent and at least one other agent, each other agent being configured to obtain usage data of the at least one other medical device; and receive the usage data from the at least one other agent via the communication channel.

[0026] By configuring the network module in this way, the agent can collect information from a large number of networked agents. Since the agent can be networked, data can be automatically obtained from other agents, so that normal operation values ​​can be automatically generated.

[0027] Each agent may be similar to the agents of the present invention (i.e., also collects usage data of other medical devices, compares the usage data to determined normal operating values, and generates an operating problem signal), or may simply be a device capable of collecting usage data of another medical device.

[0028] In some embodiments, the analysis module may be further configured to: determine, based on the operational problem signal, a corrective action describing a change to the medical device to reduce the discrepancy; and perform the corrective action on the medical device.

[0029] After a discrepancy has been noted, corrective actions may be determined to reduce such discrepancy, thereby potentially restoring the medical device to normal operation. This may avoid the need for a skilled technician to manually inspect the medical device, and may eliminate the need for a skilled technician to perform corrections altogether.

[0030] In this case, the corrective action may include at least one of a medical device parameter change and a medical device operating sequence change.

[0031] Thus, the corrective action may change the operation of the medical device in order to attempt to restore the medical device to normal operation.

[0032] Further embodiments provide that the analysis module may be configured to determine the corrective action based on the operating problem signal and the usage data of the at least one other medical device.

[0033] One method of determining corrective action is based (at least in part) on the operational problem signal describing the discrepancy and usage data of another medical device. In practice, the other medical device may be within normal operating values. Thus, the settings of the other medical device may be copied in an attempt to eliminate the discrepancy and restore the medical device to a normal operating state.

[0034] In further embodiments, the analysis module may be configured to input the operational problem signal into a corrective machine learning module, the corrective machine learning module being adapted to identify the corrective action based on the operational problem signal.

[0035] Another method for determining corrective actions may be through a machine learning algorithm. This may result in more complex and more effective corrective actions. In practice, it may not be clear how the operating parameters of the medical device should be changed when multiple factors cause the medical device to deviate from normal operating values. A machine learning algorithm may be able to determine solutions that are not immediately obvious.

[0036] In some embodiments, the analysis module can also be configured to: determine whether the corrective action will change the manually set operating parameters of the medical device; and in response to determining that the corrective action will change the manually set operating parameters, warn a user associated with the medical device about the corrective action.

[0037] It may be the case that a medical device has a set of manually set operating parameters. This may be due, for example, to a particular use case of the medical device or the availability of certain resources (i.e., available power for the medical device, environmental factors, etc.). Therefore, once a corrective action has been determined, it is beneficial to check whether there are any manually set parameters in order to avoid making changes to the medical device that are not expected or wanted by the user. However, the user should also be warned of the impact of such manually set parameters so that they can make their own choices regarding preferred operation and normal operation trade-offs.

[0038] In some embodiments, the interface may be configured to obtain the usage data from a sensor log of the medical device and a system parameter log of the medical device.

[0039] Two specific sources of usage information may be sensor logs and system parameter logs. Therefore, for the accuracy of the generated operational problem signal, it may be beneficial for the agent to have access to this information.

[0040] In some embodiments, the usage data may include at least one of: medical device usage patterns, medical device parameters, medical device behaviors, and medical device best practices.

[0041] In an embodiment of the present invention, the usage data may further include context data, the context data including at least one of a physical parameter and an environmental parameter. In this case, the interface may be configured to obtain the context data from a sensor of the medical device.

[0042] Indeed, the context of the medical device may influence the assessment of whether the medical device is actually operating outside of the normal operating range. For example, if the room in which the device is located is particularly hot, this may be the reason for a drop in certain operating parameters of the device.

[0043] In some embodiments, the learning module can be configured to determine the normal operation of the medical device by inputting the usage data of the medical device and the usage data of at least one other medical device into a reinforcement learning model, wherein the reinforcement learning model is suitable for identifying the typical operation of the medical device based on operating parameters of multiple related medical devices.

[0044] In this way, a more accurate set of normal operating values ​​can be determined.

[0045] In some embodiments, the analysis module is further configured to: generate an operation recommendation related to an operation mode of the medical device based on the operation problem signal; and transmit the operation recommendation to a user associated with the medical device.

[0046] According to a further example of one aspect of the present invention, a system for monitoring a medical device is provided, the system comprising: a first agent for monitoring a first medical device; a second agent for monitoring a second medical device, and wherein the network module of the first agent is configured to obtain usage data describing the operation of the second medical device from the second agent.

[0047] In this way, the present invention can provide a network of agents that can determine the normal operating values ​​of the medical devices they are monitoring. This enables the automatic evaluation of a number of different medical devices, thereby reducing the amount of time a technician spends in diagnosing such devices. In some cases, this can also completely eliminate the need for a user to contact a technician to repair the device and prevent the technician from performing unnecessary work.

[0048] According to an additional example of one aspect of the present invention, a method for monitoring a medical device is provided, the method comprising: obtaining usage data describing the operation of a medical device; obtaining usage data describing the operation of at least one other medical device related to the medical device; determining a normal operating value of the medical device based on the usage data of the medical device and the usage data of the at least one other medical device; comparing the usage data of the medical device with the normal operating value and a standard operating strategy of the medical device; and generating an operation problem signal based on the comparison, the operation problem signal describing a difference between the detected usage data of the medical device and the normal operation of the medical device.

[0049] These and other aspects of the invention will be apparent from and elucidated with reference to the embodiments described hereinafter. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] For a better understanding of the invention, and in order to more clearly show how the invention may be carried into effect, reference will now be made, by way of example only, to the accompanying drawings, in which:

[0051] Figure 1 is a simplified representation of a network of local and global agents used to monitor medical devices;

[0052] Figure 2 A simplified embodiment of using a reinforcement learning model to detect deviations of a medical device from normal operating values ​​is presented;

[0053] Figure 3A simplified block diagram of an agent for monitoring a medical device according to an embodiment of the present invention is shown;

[0054] Figure 4 A simplified block diagram of a system for monitoring a medical device using an agent according to another embodiment is presented;

[0055] Figure 5 A flow chart showing a method for monitoring a medical device according to another embodiment; and

[0056] Figure 6 is a simplified block diagram of a computer in which one or more portions of the embodiments may be employed. DETAILED DESCRIPTION

[0057] The present invention will be described with reference to the accompanying drawings.

[0058] It should be understood that the detailed description and specific examples, while indicating exemplary embodiments of the apparatus, system and method, are for illustrative purposes only and are not intended to limit the scope of the invention. These and other features, aspects and advantages of the apparatus, system and method of the invention will be better understood through the following description, claims and drawings. The fact that certain measures are cited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.

[0059] It should be understood that the drawings are merely schematic and not drawn to scale. It should also be understood that the same reference numerals are used throughout the drawings to indicate the same or similar components.

[0060] The present invention proposes a concept for monitoring a medical device. In particular, usage data of the medical device and other (related) medical devices are collected. This data is used to learn the normal operating values ​​of the medical device. Thus, any deviations of the operation of the medical device from the normal operating values ​​and the standard operating strategy can be determined. This deviation is then used to generate an operating problem signal (i.e., user alert, corrective action, etc.). In this way, operating problems can be automatically detected at an early stage and possibly corrected at an early stage.

[0061] Embodiments of the present invention provide a plurality of (intelligent) agents connected in a network for detecting incorrect usage patterns of medical devices (i.e., malfunctions of medical devices). Thus, more serious malfunctions of medical devices / machines can be avoided and, in some cases, the root problems of incorrect usage patterns can be corrected. Thus, the present invention can help provide an improved user (i.e., owner of medical devices) experience by reducing future problems with installed medical devices / machines, reducing the cost of replacement parts, and reducing the amount of service and maintenance overhead.

[0062] In other words, embodiments of the present invention detect incorrect usage patterns of medical devices, which enables mitigation measures to be taken to avoid future failures and repairs. In addition, due to the automated nature of the present invention, service and maintenance costs can be reduced, ultimately improving the user / owner experience. On a larger scale, if an agent of the present invention is provided to monitor medical devices, the workload of technicians can be reduced, allowing them to focus on more complex and / or serious problems. Similarly, the detection (and potential correction) of erroneous / incorrect / abnormal usage patterns of medical devices (or individual components / parts therein) saves the user cost, hassle and time by eliminating future failures and the need for large-scale parts replacement.

[0063] According to certain embodiments, the above benefits may be provided by a system including the following features:

[0064] (i) The ability to monitor usage of key components / parts of a medical device by an agent implemented in hardware and / or software. The agent can continuously monitor and / or mine usage pattern data (i.e., from system logs and parameters), and / or include physical and environmental changes (i.e., from sensors).

[0065] (ii) A connected network of agents for sharing usage data of medical devices, effectively providing a reference framework. The networked agents may be local (ie, within one medical center) or may be global (ie, across many medical centers).

[0066] (iii) A learning module that can determine / learn the correct / typical / normal overall usage and performance strategy of the medical device. In some cases, reinforcement learning between agents in the same system can be employed to learn the correct overall usage / performance strategy to determine the normal operating values. Therefore, any deviation in one of the agent's usage strategies may affect the convergence of the learning algorithm, resulting in instability that may trigger an operational problem signal / alarm.

[0067] (iv) Means for correcting detected faults. For example, the configuration, usage pattern, or behavior of medical device components may be changed. Such changes may be determined by reference to usage data of other medical devices provided by the networked agent.

[0068] (v) Adaptive capabilities such that the determined normal operating values ​​can be adapted to new / different use cases of the medical device.

[0069] Embodiments of the proposed invention include software and / or hardware agents for monitoring medical devices that detect deviations in usage patterns of the medical devices as well as anomalies in components / assemblies of the medical devices. Such deviations and anomalies may lead to future failures, so early detection and appropriate action can avoid larger problems.

[0070] In fact, one aspect of the present invention is to provide a set of intelligent software and / or hardware agents for monitoring medical devices. These agents can be deployed to monitor key components of medical devices and / or the entire medical device. The agent is configured to continuously survey / monitor the data of the medical device (i.e., the usage data of the medical device / its components). The agent is configured to obtain the usage data of other agents in order to verify normal behavior (i.e., check the comparison of the usage data of the device with the normal operating value). This can be achieved by establishing a network, as explained in more detail below.

[0071] In other words, each agent may be responsible for one component of the medical device. Thus, each agent has a set of goals related to the usage patterns or behavior of the component (i.e., normal operation policies). This may include a knowledge base that helps detect abnormal usage patterns of medical device components. In some cases, the agent may be able to take action to correct detected problems before large-scale damage occurs.

[0072] like Figure 1 As shown, agents can be connected together to create a network 100. For example, agents 110 monitoring different medical devices in the same hospital can create a local network 120 (i.e., in sites 1, 2, 3, and 4), and agents 110 from different hospitals can create a global network 100. Agents 110 can communicate with each other through network connections to share usage data of their respective monitored medical devices and their normal operating policies. In fact, localized configurations and best practices can be shared in local networks, while the global network provides common configurations and best practices.

[0073] Each agent is configured to monitor the usage pattern / behavior of the medical device / component, thereby generating usage data of the medical device. The agent analyzes the usage pattern data (from logs, system parameters), physical changes, environmental changes (which can come from sensors) by applying data mining methods to extract important information for detecting improper usage patterns.

[0074] Using the data mining process can involve the following steps:

[0075] (i) Data preprocessing: Data preprocessing involves noise removal, compensation for missing values, and structuring of data.

[0076] (ii) Data transformation: This step involves transforming the raw usage data into a more insightful data representation. Specifically, the data can be normalized, the dimensionality of the data can be reduced, and data features can be extracted.

[0077] (iii) Modeling Data: The next step is to model the usage data and identify any patterns in the usage data. Different supervised and unsupervised techniques can be applied for this task such as classification, clustering, regression, etc.

[0078] (iv) Key insights / knowledge from the data: Key findings from the usage data that help understand the usage patterns of medical devices can then be provided.

[0079] Once the usage data of multiple medical devices are obtained, the erroneous / incorrect usage patterns of the medical devices can be detected. In certain embodiments, a reinforcement learning-based method can be used, such as Figure 2 shown.

[0080] Specifically, a reinforcement learning model 220 involving multiple agents 210 in the same medical device is developed to determine (the range of) normal operating values ​​230 for the entire medical device. Any deviation in one of the agent strategies may affect the convergence of the reinforcement learning model 220, resulting in instability (i.e., detected discrepancies). Figure 2 The example of is that of an X-ray machine, and shows agents 210 of different components of the X-ray machine inputting usage data into a reinforcement learning model 220. When the usage data provided by one of the agents deviates from the normal operating value 230, a difference is detected 240, and an operating problem signal 250 is generated.

[0081] In some embodiments, the agent may be authorized to perform certain corrective actions on the identified discrepancies / problems. In this case, usage and policy data obtained from other agents (via the agent's network) may be utilized as a reference frame. The agent may correct the fault based on the configuration of the medical device, the usage pattern of the medical device, and the local configuration learned by the agent from the medical device operating correctly.

[0082] In other words, embodiments of the proposed invention detect / identify usage patterns that deviate / deviate from normal usage processes (i.e., normal operating values) from local and / or global recommendations or practices. Based on this detection, the agent can correct such practices by changing certain parameters, operation sequences, configurations of medical devices, etc.

[0083] However, it is important to provide fail-safe techniques before correcting a medical device and to impose control points on such decisions. There may be certain usage modes / parameter sets based on the specific requirements of the user / institution. More specifically, specific settings may deviate from the recommended / normal usage mode. For example, there may be certain electrical power limitations on the KV parameters of an X-ray / CT machine, so the KV parameter may be limited to a certain predefined number (which deviates from the recommended guideline setting). This may result in images with reduced contrast, but the user may be aware of and accept this limitation. In this case, an embodiment of the present invention may not immediately correct the detected problem by changing the parameter settings, but may simply suggest that the usage mode needs to be corrected by changing the KV parameters to the user so that the user can make the decision themselves.

[0084] Therefore, embodiments of the present invention handle corrective actions (i.e., changes to usage patterns and medical device parameters) according to the following method:

[0085] (i) When an operational problem signal is generated and a corrective action is determined, the maintenance log of the medical device is checked to clarify whether the corrective action affects the set parameters. It is also checked whether the set parameters serve the intended purpose of the user.

[0086] (ii) If the corrective action may affect the setting parameters, and the setting parameters have a purpose, the agent is prevented from performing the corrective action. Instead, an alert may be generated that provides information and corrective recommendations to the user and / or technician. The user and / or technician may then decide to accept or reject the corrective action.

[0087] (iii) If the corrective action is rejected, the current usage pattern of the medical device may be marked as correct usage of that particular medical device (ie, may be accepted as normal operating values). Thus, such incorrect usage may indeed be ignored.

[0088] (iv) Finally, any accepted corrective actions will be recorded so that the settings of the medical device can be rolled back in the future.

[0089] In addition, the agent can be used to recommend best practices. In fact, since some embodiments of the present invention provide agents from different sites to be connected through a network (continuous monitoring of corresponding medical devices), this information can be effectively used to identify best practices / patterns for using medical devices (i.e., most effective, more efficient, etc.). Such best practice recommendations can be provided to users. In addition, the agent can also be used for educational purposes. Whenever a deviation from a normal operating value is determined, the user can be notified. The agent can communicate these situations to the user, including the type of deviation and the associated possible damage / impact that the deviation may cause.

[0090] Now turn to Figure 3 , provides a simplified block diagram of an agent 300 for monitoring a medical device according to an embodiment of the present invention. In particular, the agent 300 includes an interface 310, a network module 320, a learning module 330, and an analysis module 340. In some embodiments, the agent 300 may also include a recommendation module 350. The agent 300 may be implemented in hardware and / or software.

[0091] In general, agent 300 is configured to monitor a medical device, such as an X-ray machine, an MRI machine, an ECG monitor, or components thereof. In doing so, agent 300 collects usage data of the medical device, compares the usage data to a typical operating scheme, and outputs an operational problem signal based on any differences / deviations from normal operation of the device. In this way, agent 300 can alert a user / technician of existing and potential problems / malfunctions / issues with the medical device, provide recommendations for corrective actions, and potentially correct the problem on the spot.

[0092] Specifically, the interface 310 is configured to obtain usage data describing the operation of the medical device. The usage data includes information indicating how the medical device operates, for example including inputs, outputs, and parameter values. Therefore, the state / condition of the medical device can be determined based on the usage data. The usage data can be obtained in real time, or can be obtained in batches when the state of the medical device is to be determined.

[0093] The interface 310 can obtain or collect usage data directly from the medical device. In some embodiments, the usage data can be retrieved directly from a sensor log of the medical device and / or a system parameter log of the medical device. Alternatively, the interface 310 can obtain the usage data indirectly, such as from a repository of usage data information.

[0094] Additionally, interface 310 may pre-process the usage data in order to remove noise, account for missing values, and extract useful information from the usage data.

[0095] In some embodiments, the usage data includes at least one of: medical device usage patterns (i.e., number of uses, duration of use, frequency of use), medical device parameters (i.e., operating parameters), medical device behavior (i.e., response to inputs, effectiveness of outputs), and medical device best practices (i.e., recommended operating regimens). However, it should be understood that the usage data may include any information that describes how the device operates and may indicate a status of the device.

[0096] In order to better understand the operation of the medical device, the usage data may also include context data describing the environment / surroundings of the medical device. For example, the context data may include physical parameters (i.e., placement of the device, geographic location around the device) and / or environmental parameters (i.e., room temperature, humidity). In this case, the interface 310 may also be configured to obtain the context data from the sensors of the medical device.

[0097] Continuing, the network module 320 is configured to obtain usage data describing the operation of at least one other medical device related to the medical device. The usage data may have the same format as the usage data of the medical device, or may need to be adjusted for comparison. Thus, the usage data may be as described above.

[0098] The usage data of the relevant medical devices may be obtained directly from other medical devices or may be obtained from a repository of (historical) usage data of medical devices.

[0099] In addition, each of the at least one other medical device is related to the medical device monitored by the agent 300. That is, each other medical device may be of the same / similar model as the medical device (e.g., a device of the same / similar design), of the same / similar type as the medical device (e.g., an MRI machine), or may be a related component in the same system as the medical device (e.g., an image controller and collimator for an x-ray machine). Of course, a skilled person will understand the range of medical devices that may be considered relevant to the use of determining normal operation of a medical device.

[0100] In some cases, network module 320 is configured to establish a communication channel between agent 300 and at least one other agent. In this way, network module 320 can create a network of interconnected agents that monitor corresponding (related) medical devices. Figure 1 As explained, an agent may be local (ie, within one site / medical center) or may be global (ie, across multiple sites / medical centers).

[0101] Each other agent in the network is configured to obtain / retrieve / collect usage data of at least one other medical device. This can be implemented in a similar manner to the interface 310 of the present agent 300. In other embodiments, the agent may simply be a sensor and / or system / parameter log of the other medical device.

[0102] The communication channel may be implemented via any known wired and / or wireless communication scheme.

[0103] In the case of establishing a communication channel. The network module 320 may receive the usage data from the at least one other agent via the communication channel. In some cases, multiple other agents may send the usage data to the network module 320 in real time. In other embodiments, when the usage data is requested from other agents, the network module 320 may receive the usage data in batches. In other embodiments, the network module 320 may also send the usage data of the medical device to at least one other agent.

[0104] Then, at least a part of the usage data of both the medical device in question and the other (related) medical device is provided to the learning module 330. The learning module 330 is configured to determine (a range of) normal operating values ​​of the medical device based on the usage data of the medical device and the usage data of at least one other medical device.

[0105] In practice, it can be expected that (given usage data for a large number of medical devices) the average of the usage data indicates what can be considered normal operating values. In other words, it can be expected that most medical devices operate within normal / typical / correct limits. Such medical devices can define a range of operating values ​​that can be considered normal.

[0106] In case the (related) medical devices are related components in the same / similar medical device systems, usage data for each component may be essential to understand the interaction between the components and thus the typical overall medical device system behavior.

[0107] Furthermore, the learning module 330 may be configured to determine the normal operation of the medical device by inputting the usage data of the medical device and the usage data of at least one other medical device into a reinforcement learning model. The reinforcement learning model is adapted to identify the typical operation of the medical device based on the operating parameters of a plurality of related medical devices. In practice, the use of reinforcement learning is Figure 2 A more detailed description is given in .

[0108] Therefore, reinforcement learning can be used to determine normal operating values ​​that can be used to identify atypical / incorrect behavior of a medical device. Alternatively, a machine learning model can be used to determine normal operating values ​​from usage data of the relevant medical device.

[0109] Furthermore, the agent 300 includes an analysis module 340 configured to compare the usage data of the medical device with normal operating values ​​of the medical device (determined by the learning module 330) and standard operating policies. That is, it can be determined whether the usage data falls within (a range of) normal operating values.

[0110] The usage data is also compared to a standard operating policy of the medical device, describing the expected typical operation of the medical device. In an embodiment, the standard operating policy is determined based on technical documentation. The standard operating policy can be obtained, for example, from a designer / manufacturer of the medical device.

[0111] In any case, the comparison enables detection / identification of the presence (or lack thereof) of a difference / deviation / irregularity between the normal operating values ​​and the usage data of the medical device. An operational problem signal describing / indicative of the detected difference may then be generated. The operational problem signal may be used to alert the user / technician, to determine potential corrective procedures, or to resolve the difference itself.

[0112] In one example, the operational problem signal may be used to determine a corrective action. In this case, the analysis module 340 is further configured to determine a corrective action describing a change to the medical device to reduce the discrepancy based (at least in part) on the operational problem signal.

[0113] In other words, the corrective action indicates a change in the medical device that can restore the medical device to a normal / typical operating mode. For example, the corrective action may include at least one of a medical device parameter change and a medical device operation sequence change (i.e., a sequence change). In fact, the corrective action describes any change in the medical device that may result in a change in the behavior of the medical device.

[0114] The analysis module 340 may be configured to input the operational problem signal to the corrective machine learning module 330, which is adapted to identify the corrective action based on the operational problem signal. Thus, machine learning may be utilized to identify the most effective (or most likely effective) corrective action for restoring the medical device to normal / typical behavior.

[0115] To improve the identification of corrective actions, the corrective actions may also be based on usage data of at least one other medical device. In fact, settings / configurations of another medical device that fall within the normal operating value range may indicate how to change the medical device (through corrective actions) to restore to normal operation.

[0116] The analysis module 340 may also be configured to perform a corrective action on the medical device. Alternatively, the analysis module 340 may provide the corrective action to another component that may actuate / implement the corrective action, or may provide an indication of the corrective action to a user / technician for implementation.

[0117] Alternatively, the corrective action may not be automatically implemented / performed. In fact, the analysis module 340 may also be configured to determine whether the corrective action would change the manually set operating parameters of the medical device. This may be achieved by comparing the corrective action with the system log / history of the medical device.

[0118] Thus, in response to determining that a corrective action will change a manually set operating parameter, the analysis module 340 can alert (i.e., send a notification) to a user associated with the medical device (i.e., owner, technician, current user) of the corrective action. Thus, any unexpected changes to user preferred parameters can be avoided while still providing the user with the information needed to make a judgment about the trade-off between their preferences and the benefits of the corrective action.

[0119] In addition, the operational problem signal can also be used to generate / identify operational suggestions. To this end, the analysis module 340 will also be configured to (at least in part) generate operational suggestions related to the operating mode of the medical device based on the operational problem signal. The operational suggestions can be related to frequency, usage duration, process sequence, system parameters, etc. Basically, the operational suggestions can provide information that can keep the medical device operating within normal operating conditions / behaviors to the user / technician / owner, thereby reducing the possibility of potential damage. The operational suggestions are then communicated to the user / technician / owner of the medical device. Finally, the agent 300 can also (optionally) include a recommendation module 350, which is configured to process the usage data of the medical device and the usage data of at least one other medical device, so as to generate a best practice recommendation describing the operating mode of the medical device. In fact, by considering all usage data from (related) medical devices, it can be determined which operation scheme of the medical device leads to the most favorable medical device behavior (i.e., efficiency, effectiveness, no wear, etc.).

[0120] The best practice recommendations are then communicated to users / technicians / owners associated with the medical devices so that they can make a determination as to whether to implement the best practice recommendations.

[0121] Figure 4 A system 400 for monitoring one or more medical devices according to an exemplary embodiment is presented.

[0122] In particular, a first agent 410 for monitoring a first medical device is provided. The first agent 410 may be Figure 3 The agent 300 described above. A second agent 420 for monitoring a second medical device is also provided, which may also be Figure 3 The agent 300.

[0123] The first and second medical devices may be related devices. For example, the first and second medical devices may be of the same or similar type, the same or similar model, or may be interrelated components of the same medical device system.

[0124] Furthermore, additional agents 430 for monitoring additional (related) medical devices may be provided.

[0125] At least the network module of the first agent 410 is configured to obtain usage data describing the operation of the second medical device from the second agent 420. If an additional agent 430 is provided, the network module of the first agent 410 is configured to obtain usage data describing the operation of the additional medical device from the additional agent 430.

[0126] Such usage data may be used to determine normal operating values ​​and therefore may be used to determine when the behavior of the first medical device is erroneous or atypical.

[0127] Also in some embodiments, the second agent 420 may be configured to obtain usage data describing the operation of the first medical device from the first agent 410. The second agent 420 may then use such usage data in a similar manner as the first agent 410. The same is true for the additional agent 430, if provided.

[0128] Essentially, Figure 4 The system provides the potential for a network of agents to share usage data for determining deviations of the operating behavior of a medical device from a normal operating range.

[0129] In some embodiments, there may be two different types of systems. In a first instance, the agents may all be local agents forming a local network. In this case, the normal operating values ​​will reflect and preserve the local usage pattern. The local usage pattern may be customized according to user requirements, so it may be advantageous to retain that pattern. In a second instance, the agents may be global agents (i.e., across different sites, medical centers, etc.), so a global network may be formed. A global agent network may mean that the normal operating values ​​reflect (and preserve) a common usage pattern. The common usage pattern may correspond to the usage recommendations of the designer / manufacturer of the medical device.

[0130] Figure 5 A flow chart of a method for monitoring a medical device according to an exemplary embodiment is provided.

[0131] In step 510, usage data describing the operation of the medical device is obtained / acquired / collected. This may be obtained directly from the medical device (ie, from a system log) or from an information repository. The usage data may be provided in real time, or may be obtained in batches corresponding to (recent) history.

[0132] At step 520, usage data describing the operation of at least one other medical device related to the medical device is obtained. Similar to step 520, this can be obtained from the other medical devices, or it can be obtained from a repository. In some embodiments, the usage data of the other medical devices is obtained by establishing a communication network between agents monitoring the medical devices (i.e., forming a local and / or global network).

[0133] At step 530, a normal operating value for the medical device is determined. The normal operating value is based (at least in part) on usage data for the medical device and usage data for at least one other medical device.

[0134] In this way, the normal / typical / correct behavior of the medical device may be learned based on (a range of) normal operating values. The normal operating values ​​reflect the behavior of the medical device when operating in a state that is likely not to result in damage and / or malfunction.

[0135] At step 540, the usage data of the medical device is compared with the normal operation value and the standard operation policy of the medical device. Therefore, it can be understood whether the medical device is operating in a normal state. In other words, the state of the medical device can be interpreted based on the comparison between the usage data and the normal operation value.

[0136] Finally, in step 550, an operation problem signal is generated based on the comparison performed in step 540. The operation problem signal describes a difference between the detected usage data of the medical device and a normal operation of the medical device.

[0137] Thus, the operational problem signal can be used to generate corrective actions for reducing the variance, as well as operational recommendations for improving the operation of the medical device.Such actions and recommendations can be performed automatically, or may require user / owner / technician input to perform.

[0138] As a result of the above method, faulty behavior of medical devices can be automatically detected at an early stage. This can reduce the number and severity of faults, saving time and costs for medical device owners and technicians.

[0139] Figure 6 An example of a computer 1000 in which one or more portions of an embodiment may be employed is shown. The various operations discussed above may utilize the capabilities of the computer 1000. For example, one or more portions of a system for monitoring a medical device may be incorporated into any of the elements, modules, applications, and / or components discussed herein. In this regard, it should be understood that the system functional blocks may be run on a single computer, or may be distributed across several computers and locations (e.g., connected via the Internet).

[0140] Computer 1000 includes but is not limited to PC, workstation, laptop computer, PDA, handheld device, server, storage, etc. In general, in terms of hardware architecture, computer 1000 may include one or more processors 1010, memory 1020 and one or more I / O devices 1030 communicatively coupled via a local interface (not shown). The local interface may be, for example, but not limited to, one or more buses or other wired or wireless connections, as known in the art. The local interface may have additional elements, such as controllers, buffers (cache), drivers, repeaters, and receivers to enable communication. In addition, the local interface may include address, control and / or data connections to enable appropriate communication between the above components.

[0141] Processor 1010 is a hardware device for executing software that may be stored in memory 1020. Processor 1010 may be virtually any custom or commercially available processor, central processing unit (CPU), digital signal processor (DSP), or auxiliary processor among several processors associated with computer 1000, and processor 1010 may be a semiconductor-based microprocessor (in the form of a microchip) or a microprocessor.

[0142] The memory 1020 may include any one or combination of volatile memory elements (e.g., random access memory (RAM), such as dynamic random access memory (DRAM), static random access memory (SRAM), etc.) and non-volatile memory elements (e.g., ROM, erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic tape, compact disk read-only memory (CD-ROM), magnetic disk, floppy disk, data tape, cassette, etc.). In addition, the memory 1020 may contain electronic, magnetic, optical and / or other types of storage media. Note that the memory 1020 may have a distributed architecture in which various components are remote from each other but can be accessed by the processor 1010.

[0143] The software in the memory 1020 may include one or more separate programs, each of which includes an ordered list of executable instructions for implementing logical functions. According to an exemplary embodiment, the software in the memory 1020 includes a suitable operating system (O / S) 1050, a compiler 1060, a source code 1070, and one or more application programs 1080. As shown, the application program 1080 includes many functional components for implementing the features and operations of the exemplary embodiment. The application program 1080 of the computer 1000 may represent various application programs, computing units, logics, functional units, processes, operations, virtual entities, and / or modules according to exemplary embodiments, but the application program 1080 is not meant to be a limitation.

[0144] The operating system 1050 controls the execution of other computer programs and provides scheduling, input-output control, file and data management, memory management, and communication control and related services. The inventors contemplate that the application program 1080 used to implement the exemplary embodiments may be applicable to all commercially available operating systems.

[0145] Application 1080 may be a source program, an executable program (object code), a script, or any other entity including a set of instructions to be executed. When it is a source program, the program is typically translated via a compiler (such as compiler 1060), an assembler, an interpreter, etc. (which may or may not be included in memory 1020) to operate normally in conjunction with O / S 1050. In addition, application 1080 may be written as an object-oriented programming language with data and method classes, or a procedural programming language with routines, subroutines, and / or functions, such as, but not limited to, C, C++, C#, Pascal, BASIC, API calls, HTML, XHTML, XML, ASP scripts, JavaScript, FORTRAN, COBOL, Perl, Java, ADA, .NET, etc.

[0146] I / O devices 1030 may include input devices such as, but not limited to, a mouse, keyboard, scanner, microphone, camera, etc. In addition, I / O devices 1030 may also include output devices such as, but not limited to, a printer, display, etc. Finally, I / O devices 1030 may also include devices that transmit both input and output such as, but not limited to, a NIC or modulator / demodulator (for accessing remote devices, other files, devices, systems, or networks), a radio frequency (RF) or other transceiver, a telephone interface, a bridge, a router, etc. I / O devices 1030 also include components for communicating over various networks such as the Internet or an intranet.

[0147] If the computer 1000 is a PC, workstation, intelligent device, etc., the software in the memory 1020 may also include a basic input and output system (BIOS) (omitted for simplicity). The BIOS is a set of basic software routines used to initialize and test the hardware at startup, start the O / S 1050, and support data transfer between hardware devices. The BIOS is stored in some type of read-only memory, such as ROM, PROM, EPROM, EEPROM, etc., so that the BIOS can be executed when the computer 800 is activated.

[0148] When the computer 1000 is working, the processor 1010 is configured to execute software stored in the memory 1020, communicate data with the memory 1020, and generally control the operation of the computer 1000 according to the software. The application 1080 and the O / S 1050 are read in whole or in part by the processor 1010, possibly buffered within the processor 1010, and then executed.

[0149] When the application 1080 is implemented in software, it should be noted that the application 1080 can be stored on virtually any computer-readable medium for use by or in conjunction with any computer-related system or method. In the context of this document, a computer-readable medium can be an electronic, magnetic, optical, or other physical device or means that can contain or store a computer program for use by or in conjunction with a computer-related system or method.

[0150] Application 1080 may be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device (e.g., a computer-based system, a system containing a processor, or other system that can obtain instructions from and execute instructions to an instruction execution system, apparatus, or device). In the context of this document, a "computer-readable medium" may be any device that can store, communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. A computer-readable medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, device, or propagation medium.

[0151] about Figure 5 The method described and about Figure 3 and 4 The described agent / system can be implemented in hardware or software or a mixture of the two (e.g., as firmware running on a hardware device). Insofar as the embodiments are partially or entirely implemented in software, the functional steps shown in the process flow chart can be performed by a suitably programmed physical computing device, such as one or more central processing units (CPUs) or graphics processing units (GPUs). Each process and its individual component steps as shown in the flow chart can be performed by the same or different computing devices. According to an embodiment, a computer-readable storage medium stores a computer program including a computer program code, which is configured to cause one or more physical computing devices to perform the encoding or decoding method as described above when the program is run on one or more physical computing devices.

[0152] Storage media may include volatile and non-volatile computer memory, such as RAM, PROM, EPROM and EEPROM, optical disks (such as CD, DVD, BD), magnetic storage media (such as hard disks and magnetic tapes). The various storage media may be fixed within the computing device or may be transportable so that one or more programs stored thereon may be loaded into a processor.

[0153] Insofar as the embodiments are partially or entirely implemented in hardware, the block Figure 3 and 4The boxes shown in the drawings may be separate physical components or logical subdivisions of a single physical component, or may all be implemented in an integrated manner in one physical component. The functions of a box shown in the drawings may be divided between multiple components in one implementation, or the functions of multiple boxes shown in the drawings may be combined in a single component in one implementation. Hardware components suitable for embodiments of the present invention include, but are not limited to, conventional microprocessors, application specific integrated circuits (ASICs), and field programmable gate arrays (FPGAs). One or more boxes may be implemented as a combination of dedicated hardware for performing some functions and one or more programmed microprocessors and associated circuits for performing other functions.

[0154] By studying the drawings, the disclosure and the claims, those skilled in the art can understand and implement variations to the disclosed embodiments when practicing the claimed invention. In the claims, the word "comprising" does not exclude other elements or steps, and the attributive "one" or "an" does not exclude multiple. A single processor or other unit can implement the functions of several items recorded in the claims. The fact that certain measures are cited in mutually different dependent claims does not indicate that the combination of these measures cannot be advantageous. If a computer program is discussed above, it can be stored / distributed on a suitable medium, such as an optical storage medium or solid-state medium provided together with or as part of other hardware, but can also be distributed in other forms, such as via the Internet or other wired or wireless telecommunications systems. If the term "suitable for" is used in a claim or specification, it should be noted that the term "suitable for" is intended to be equivalent to the term "configured to". Any figure marks in the claims should not be interpreted as limiting the scope.

[0155] The flow charts and block diagrams in the accompanying drawings illustrate the possible implementation architectures, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram may represent a module, segment or part of an instruction, which includes one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions annotated in the box may not occur in the order annotated in the accompanying drawings. For example, depending on the functions involved, the two boxes shown in succession may actually be executed substantially simultaneously, or these boxes may sometimes be executed in reverse order. It should also be noted that each box of the block diagram and / or flow chart and the combination of boxes in the block diagram and / or flow chart can be implemented by a dedicated hardware-based system that performs a specified function or action or performs a combination of dedicated hardware and computer instructions.

Claims

1. An agent (300) for monitoring a medical device, the agent comprising: an interface (310) configured to obtain usage data describing the operation of the medical device; a network module (320) configured to obtain usage data describing the operation of at least one other medical device related to the medical device; a learning module (330) configured to determine a normal operating value for the medical device based on the usage data for the medical device and the usage data for the at least one other medical device, and An analysis module (340) configured to: comparing the usage data of the medical device to the normal operating values ​​and a standard operating policy of the medical device; and Based on the comparison, an operating problem signal is generated, which describes a difference between the detected usage data of the medical device and the normal operating value of the medical device.

2. The agent according to claim 1, further comprising a recommendation module (350), the recommendation module being configured to: processing the usage data of the medical device and the usage data of the at least one other medical device to generate a best practice recommendation describing an operating mode of the medical device; and The best practice recommendation is communicated to a user associated with the medical device.

3. An agent according to claim 1 or 2, wherein: The network module (320) is configured to: establishing a communication channel between the agent and at least one other agent, each other agent being configured to obtain usage data of the at least one other medical device; and The usage data is received from the at least one other agent via the communication channel.

4. An agent according to any one of claims 1 to 3, wherein: The analysis module (340) is further configured to: Based on the operational problem signal, determining a corrective action describing a change to the medical device to reduce the discrepancy; and The corrective action is performed on the medical device.

5. The agent according to claim 4, wherein: The corrective action includes at least one of a medical device parameter change and a medical device operating sequence change.

6. An agent according to claim 4 or 5, wherein: The analysis module (340) is configured to determine the corrective action based on the operational problem signal and the usage data of the at least one other medical device.

7. An agent according to any one of claims 4 to 6, wherein: The analysis module (340) is configured to input the operational problem signal to a corrective machine learning module, the corrective machine learning module being adapted to identify the corrective action based on the operational problem signal.

8. An agent according to any one of claims 4 to 7, wherein: The analysis module (340) is further configured to: determining whether the corrective action will change a manually set operating parameter of the medical device; and In response to determining that the corrective action will alter the manually set operating parameter, a user associated with the medical device is alerted to the corrective action.

9. An agent according to any one of claims 1 to 8, wherein: The interface (310) is configured to obtain the usage data from a sensor log of the medical device and a system parameter log of the medical device.

10. An agent according to any one of claims 1 to 9, wherein: The usage data includes at least one of: medical device usage patterns, medical device parameters, medical device behaviors, and medical device best practices.

11. The agent according to claim 10, wherein: The usage data further comprises context data comprising at least one of a physical parameter and an environmental parameter, and wherein the interface is configured to obtain the context data from a sensor of the medical device.

12. An agent according to any one of claims 1 to 11, wherein: The learning module (330) is configured to determine normal operation of the medical device by inputting the usage data of the medical device and the usage data of the at least one other medical device into a reinforcement learning model, wherein the reinforcement learning model is suitable for identifying typical operation of the medical device based on operating parameters of a plurality of related medical devices.

13. An agent according to any one of claims 1 to 12, wherein: The analysis module (340) is further configured to: generating an operation recommendation related to an operation mode of the medical device based on the operation problem signal; and The operational recommendation is communicated to a user associated with the medical device.

14. A system (400) for monitoring a medical device, the system comprising: A first agent (410) for monitoring a first medical device according to any one of claims 1-13; A second agent (420) for monitoring a second medical device according to any one of claims 1 to 13, and Wherein the network module of the first agent is configured to obtain usage data describing the operation of the second medical device from the second agent.

15. A method (500) for monitoring a medical device, the method comprising: obtaining (510) usage data describing the operation of the medical device; obtaining (520) usage data describing the operation of at least one other medical device related to the medical device; determining (530) a normal operating value for the medical device based on the usage data for the medical device and the usage data for the at least one other medical device; comparing the usage data of the medical device to the normal operating values ​​and standard operating policies of the medical device (540); and An operation problem signal is generated (550) based on the comparison, the operation problem signal describing a difference between the detected usage data of the medical device and the normal operation of the medical device.

Images