EP4662670A1

EP4662670A1 - Systems and methods for adapting a communication output to hospital customers based on predicted bottlenecks and customer preference during case handling

Info

Publication number: EP4662670A1
Application number: EP24702730.3A
Authority: EP
Inventors: Lu Wang; Jurgen Jan RUSCH; Qi Gao; Milosh STOLIKJ; Emanuele BASTIANELLI
Original assignee: Koninklijke Philips NV
Current assignee: Koninklijke Philips NV
Priority date: 2023-02-07
Filing date: 2024-01-29
Publication date: 2025-12-17
Also published as: WO2024165353A1; CN120660146A

Abstract

A non-transitory computer readable medium (26) stores a database (30) storing a plurality of maintenance resolution workflows (32); and instructions executable by at least one electronic processor (20) to perform a method (100) of recommending one or more resolution workflows. The method includes: receiving a current service case (34) for a medical device (12); identifying one or more maintenance resolution workflows in the database for resolving the current service case; populating the identified one or more maintenance resolution workflows with case information for the current service case to produce corresponding one or more current service case resolution workflows (36); identifying at least one possible bottleneck for the current service case based on the one or more current service case resolution workflows; analyzing an impact of the identified at least one possible bottleneck; and outputting a recommendation or an explanation (38) based on the analyzed impact.

Description

SYSTEMS AND METHODS FOR ADAPTING A COMMUNICATION OUTPUT TO HOSPITAL CUSTOMERS BASED ON PREDICTED BOTTLENECKS AND CUSTOMER PREFERENCE DURING CASE HANDLING

FIELD

[0001] The following relates generally to the medical imaging arts, medical imaging device maintenance arts, medical imaging device diagnostic procedure arts, and related arts.

BACKGROUND

[0002] A medical imaging system or other medical devices may occasionally malfunction, and thus fail to work properly. The malfunction is recognized by an operator of the medical device based on observed symptoms, such as failure of the medical device to operate, unsatisfactory output of the medical device (e.g., a patient monitor unable to read a vital sign sensor, a medical imaging device producing low-quality images, or so forth), and/or by an error code produced by the medical device, or based on other symptoms. Depending on the malfunction, the problem can sometimes be resolved remotely, by a remote service engineer (RSE). During diagnosis, the RSE can interact with several tools that can aid during system troubleshooting, and identify the best course of action to resolve the problem. For example, the RSE may decide to perform several actions, such as searching through historical service records to find similar service cases, and identify how they were resolved, searching through technical documentation for possible root causes and potential solutions, inspecting the log files of the medical imaging system, to identify relevant diagnostic information such as error messages, contacting the customer to get a description of the problem, and so forth.

[0003] If the malfunction cannot be handled remotely, then the remote service engineer can issue a request for a field service engineer (FSE) to repair the medical device on-site.

[0004] When a maintenance issue case arises which will require some imaging device downtime, this can be inconvenient and costly to the customer, as imaging examinations may be abruptly canceled or rescheduled. In such a situation, the customer would like to minimize the downtime, and also expects fast and accurate information on the anticipated downtime and any measures that might be taken to mitigate the impact of the downtime on radiology laboratory operations. [0005] The following discloses certain improvements to overcome these problems and others.

SUMMARY

[0006] In some embodiments disclosed herein, a non-transitory computer readable medium stores a database storing a plurality of maintenance resolution workflows; and instructions executable by at least one electronic processor to perform a method of recommending one or more resolution workflows. The method includes: receiving a current service case for a medical device; identifying one or more maintenance resolution workflows in the database for resolving the current service case; populating the identified one or more maintenance resolution workflows with case information for the current service case to produce corresponding one or more current service case resolution workflows; identifying at least one possible bottleneck for the current service case based on the one or more current service case resolution workflows; analyzing an impact of the identified at least one possible bottleneck; and outputting a recommendation or an explanation based on the analyzed impact.

[0007] In some embodiments disclosed herein, a non-transitory computer readable medium stores a database storing a plurality of maintenance resolution workflows; and instructions executable by at least one electronic processor to perform a method of recommending one or more resolution workflows. The method includes receiving a current service case for a medical device; identifying a plurality of maintenance resolution workflows in the database for resolving the current service case; populating the identified maintenance resolution workflows with case information for the current service case to produce corresponding current service case resolution workflows; identifying at least one possible bottleneck for the current service case based on the current service case resolution workflows; analyzing an impact of the identified at least one possible bottleneck; and outputting a recommendation or an explanation based on the analyzed impact.

[0008] In some embodiments disclosed herein, a method of recommending one or more resolution workflows includes receiving a current service case for a medical device; identifying a maintenance resolution workflow stored in a database for resolving the current service case; populating the identified maintenance resolution workflow with case information for the current service case to produce a corresponding current service case resolution workflow; identifying at least one possible bottleneck for the current service case based on the current service case resolution workflow; analyzing an impact of the identified at least one possible bottleneck including estimating a resolution time for the current service case based at least on the at least one possible bottleneck; and outputting a recommendation or an explanation of the resolution time including identifying the at least one possible bottleneck.

[0009] One advantage resides in communicating potential service activities, delays, and problems to customers owning a medical device.

[0010] Another advantage resides in managing expectations of customers owning a medical device for maintenance to address an issue with the medical device.

[0011] Another advantage resides in reducing downtimes during maintenance processes.

[0012] Another advantage resides in reducing delays in delivery of parts needed to complete a maintenance process for a medical device.

[0013] Another advantage resides in accurate predictions of repair times for a medical device.

[0014] A given embodiment may provide none, one, two, more, or all of the foregoing advantages, and/or may provide other advantages as will become apparent to one of ordinary skill in the art upon reading and understanding the present disclosure.

BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The disclosure may take form in various components and arrangements of components, and in various steps and arrangements of steps. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the disclosure.

[0016] FIGURE 1 diagrammatically illustrates a medical device servicing workflow recommender apparatus in accordance with the present disclosure.

[0017] FIGURE 2 diagrammatically illustrates an embodiment of a medical device servicing workflow recommender method using the apparatus of FIGURE 1.

[0018] FIGURE 3 diagrammatically illustrates an embodiment of a medical device servicing workflow recommender method using the apparatus of FIGURE 1.

[0019] FIGURE 4 illustrates an illustrative example output of a workflow recommender apparatus. DETAILED DESCRIPTION

[0020] The following relates to a system to improve planning of maintenance resolution and communication thereof to the customer. The disclosed system includes a database of resolution workflows for various types of maintenance issues. A given resolution workflow may include various tasks with various interdependencies. For example, a workflow may include ordering/delivering a replacement part and installing the replacement part in the imaging system. The installation is dependent on the ordering/delivering since installation can only be done after the part is delivered. Some parts of the resolution workflow may also be probabilistic. For example, if the workflow includes dispatch of a field service engineer (FSE) then the portion of the workflow performed by the FSE may depend on the result of an inspection of the imaging device by the FSE. These probabilistic aspects may in some embodiments be represented as a decision tree with probabilities labeled to branches of the tree, for example based on statistics drawn from historical data. Completion times for various tasks of the workflow may also be situationally dependent, and may be estimated in real-time based on information such as the number of remote service engineers (RSEs) or FSEs on call at the moment, or at some time in the future when a task is expected to occur. Completion times that are reliant on an RSE or FSE can be estimated in real-time based on the on-call personnel data and historical information on the number of tasks a given RSE or FSE can handle per hour, for example. Similarly, part delivery may be situationally dependent and completion time for a given part to be delivered can be based on an actual delivery time estimate provided by a delivery company, or based on historical data on delivery times of that type of part given the time of day, day of week, month, year, or similar information. (For example, historical data may indicate that a part is ordinarily delivered in 8 hours but if the part is ordered over a major holiday the delivery time may be 16 hours).

[0021] Based on the workflow (retrieved from a database for the current maintenance issue case) populated with situationally specific information such as task completion times and probabilities (for uncertain branching events of the workflow), and the task interdependencies specified in the workflow, it is often the case that a specific bottleneck may be identified that is creating a substantial predicted delay in resolving the current case. A given bottleneck could also have an associated probability of occurrence. For example, it may be that an FSE inspection is required to determine whether a current maintenance issue case can be resolved by a rapid action such as a software update or calibration of installed components, or whether the current case can only be resolved by a part replacement (thus creating a bottleneck as the part is ordered/delivered). The bottleneck is assigned a probability based on the probability (e.g., labeled to the corresponding branch of the resolution workflow tree, derived from historical data) that the inspection will indicate a part replacement is required in the current case.

[0022] While the following is described for one resolution workflow, there may be two, three, or more (candidate) resolution workflows to resolve the current case. For example, one candidate workflow may send an FSE out as soon as possible to resolve the current case, another candidate workflow may use an in-house bioengineer attempt to solve the current case, another candidate workflow may utilize a contracted third party service provider to resolve the current case, while yet another candidate workflow may send out the FSE after a scheduled delay while continuing to use the imaging device on a limited basis to perform certain types of imaging examinations that can be performed in spite of the maintenance issue.

[0023] The next step is an impact analysis to analyze the impact on the customer of resolving the current case by each such candidate resolution workflow, including any identified bottlenecks. This may entail comparing the imaging examination schedule for the imaging device with the delays introduced by each candidate workflow, to determine the impact of the candidate workflow on the customer, accounting for the situationally specific completion times for the tasks of the various workflows. For a candidate workflow in which the imaging device would continue to be used on a reduced-functionality basis, the impact analysis may consider what fraction of the upcoming imaging examinations can actually be performed on the reduced-functionality basis. These analyses output, for each candidate resolution workflow, a timeline of the resolution workflow annotated with indications of times when the imaging device would be unavailable, or available on a reduced-functionality basis, along with an estimated time when the current case will be fully resolved. The timeline may also indicate the actor performing each task.

[0024] With these candidate resolution workflows analyzed as to impact, in a communication phase a recommendation or explanation is provided. The recommendation or explanation that is output depends on the identified bottlenecks. For bottlenecks with high probabilities (or confidence values), and if alternative actions could make a significant difference in the impact, the disclosed system would decide to focus on recommending the alternative actions. [0025] The communication phase is typically customer-facing. However, there may in some embodiments also be a service provider-facing communication aspect. Typically, the RSE or other service provider employee assigned to the case should be aware of the recommendation or explanation provided to the customer, and if a recommendation is provided then the RSE should also be informed of the customer’s decision. (For example, the system may recommend delaying sending out an FSE while the imaging device continues to be used on a reduced-functionality basis, but the customer may nonetheless choose a different candidate resolution workflow in which the FSE is sent out as soon as possible).

[0026] With reference to FIGURE 1, an illustrative apparatus 10 for servicing a medical device 12 is shown. The medical device 12, for example, can comprise a medical imaging device 12 (also referred to as a medical device, an imaging device, imaging scanner, and variants thereof) can be a magnetic resonance imaging (MRI) scanner, a computed tomography (CT) scanner, a positron emission tomography (PET) scanner, a gamma camera for performing single-photon emission computed tomography (SPECT), an interventional radiology (IR) device, an X-ray device, an image-guided therapy (IGT) device, an ultrasound (US) device, or so forth. Although described herein as an imaging device, the medical device 12 can also be any other suitable medical device, such as a patient monitor, a radiation therapy device, a mechanical ventilator, and so forth.

[0027] An electronic processing device 18, such as a workstation computer, or more generally a computer, a smart device (e.g., a cellular telephone (“cell phone”), a smart tablet, and so forth), is operable by a service engineer (SE). The electronic processing device 18 may also include a server computer or a plurality of server computers, e.g., interconnected to form a server cluster, cloud computing resource, or so forth, to perform more complex computational tasks. The electronic processing device 18 includes typical components, such as an electronic processor 20 (e.g., a microprocessor), at least one user input device (e.g., a mouse, a keyboard, a trackball, and/or the like) 22, and a display device 24 (e.g., an LCD display, plasma display, cathode ray tube display, and/or so forth). In some embodiments, the display device 24 can be a separate component from the electronic processing device 18, or may include two or more display devices. [0028] The electronic processor 20 is operatively connected with one or more non- transitory storage media 26. The non-transitory storage media 26 may, by way of non-limiting illustrative example, include one or more of a magnetic disk, RAID, or other magnetic storage medium; a solid-state drive, flash drive, electronically erasable read-only memory (EEROM) or other electronic memory; an optical disk or other optical storage; various combinations thereof; or so forth; and may be for example a network storage, an internal hard drive of the workstation 18, various combinations thereof, or so forth. It is to be understood that any reference to a non- transitory medium or media 26 herein is to be broadly construed as encompassing a single medium or multiple media of the same or different types. Likewise, the electronic processor 20 may be embodied as a single electronic processor or as two or more electronic processors. The non- transitory storage media 26 stores instructions executable by the at least one electronic processor 20. The instructions include instructions to generate a visualization of a graphical user interface (GUI) 28 for display on the display device 24.

[0029] The electronic processing device 18 is also in communication with a database 30 (shown in FIGURE 1 as a server computer) that stores a plurality of maintenance resolution workflows 32. For example, in the case of a database 30 of medical imaging system service cases, the plurality of maintenance resolution workflows 32 may include maintenance resolution workflows that include diagnostic procedures, repair operations, logistic data, maintenance reporting, and so forth. . It will be appreciated that the number of maintenance resolution workflows 32 in the database 30 may be large, e.g. tens of thousands of cases, hundreds of thousands of cases, or more, and a search can in many cases return a few tens of thousands of maintenance resolution workflows 32 (or more).

[0030] The apparatus 10 is configured as described above to perform a method or process 100 of recommending one or more resolution workflows 32. The non-transitory storage medium 26 stores instructions which are readable and executable by the at least one electronic processor 20 to perform disclosed operations including performing the servicing method or process 100. In some examples, the method 100 may be performed at least in part by cloud processing.

[0031] With reference to FIGURE 2, and with continuing reference to FIGURE 1, an illustrative embodiment of an instance of the method 100 is diagrammatically shown as a flowchart.

[0032] At an operation 102, a current service case for the medical device 12 is received at the electronic processing device 18. In some examples, the received current service case 34 comprises error codes generated by the medical device 12, user-provided problem descriptions, historical search results or a summary from a SE, and so forth. The operation 102 could involve, for example, an RSE or other customer call intake person receiving a call from the customer and entering such information into fields of a form; or the customer could fill out the form directly via a customer-facing web-based service request form filled out using a web browser. These are examples. A case ticket or the like may be created to track the now-open current service case.

[0033] At an operation 104, one or more maintenance resolution workflows 32 for resolving the current service case 34 are identified. The identifying operation 104 can include extracting features from the received current service case 34, and matching the extracted features with features in the maintenance resolution workflows 32 stored in the database 30. In some examples, only one maintenance resolution workflow 32 can be identified, while in other examples multiple maintenance resolution workflows 32 can be identified.

[0034] At an operation 106, the identified one or more maintenance resolution workflows 32 are populated with case information for the current service case 34 to produce corresponding one or more current service case resolution workflows 36. (If there are two or more populated current service case resolution workflows 36, these can be considered as candidate workflows). This operation 106 can include identifying future bottleneck(s) in the workflow.

[0035] At an operation 108, at least one possible bottleneck for the current service case 34 is identified based on the one or more current service case resolution workflows 36. As used herein, the term “bottleneck” (and variants thereof) refers to a timeframe issue, that is, a point in the workflow that introduces delay in resolving the service case 34. Often, although not necessarily, the bottleneck is produced at least in part by a dependency in which a later step is dependent on an earlier step, and cannot be performed until that earlier step is completed. For example, a part installation step cannot be performed until prior part ordering and part delivery steps are completed. While the part ordering is typically fast, the bottleneck is likely to be the part delivery step. A bottleneck could also be probabilistic, i.e. have a probability assigned to it. For example, if a malfunctioning system requires on-site inspection by an FSE then a bottleneck could be created depending on the findings of that inspection, e.g. if the inspection reveals a software problem, then the remediation may be fast whereas if the inspection reveals a hardware problem the remediation may be slow as it requires part order/delivery/installation. Here, a bottleneck would again be the delivery time, but now with some probability of occurrence determined based on statistics for similar historical FSE inspection cases. Each maintenance resolution workflow 32 includes tasks and interdependencies based on the bottleneck, possibly along with a probability of the bottleneck occurring. [0036] At an operation 110, an impact of the identified at least one possible bottleneck is analyzed. At an operation 112, a recommendation or an explanation 38 is output based on the analyzed impact. These operations can be performed in a variety of manners. In some examples, when the identifying operation 104 identifies a plurality of maintenance resolution workflows 32 which are populated to produce a corresponding plurality of current service case resolution workflows 36, the analyzing operation 110 includes generating alternative candidate resolution workflow timelines from the plurality of current service case resolution workflows 36. The alternative candidate resolution workflow timelines include the identified at least one possible bottleneck. The outputting operation 112 then includes presenting the alternative candidate resolution workflow timelines on the display device 24.

[0037] In other examples, when the identifying operation 104 identifies a single maintenance resolution workflow 32 populated to produce a single current service case resolution workflow 36, the analyzing operation 110 includes estimating a resolution time for the current service case 34 based at least on the at least one possible bottleneck. The outputting operation 112 then includes presenting an explanation 38 of the resolution time including identifying the at least one possible bottleneck

[0038] In some embodiments, the analyzing operation 110 includes analyzing the impact of the at least one possible bottleneck on an examination schedule to be performed using at least the medical device 12. To do so, a reduced functionality of the medical device 12 while the current service case 34 remains unresolved is determined, and a revised examination schedule capable of being performed by the medical device 12 with the determined reduced functionality is determined. In this example, the analyzing operation 110 includes generating a timeline (comprising the recommendation or explanation 38) of the current service case resolution workflow 36 annotated with downtimes when the medical device 12 would be unavailable or available with the determined reduced functionality. In some examples, the timeline further includes an estimated time when the current case will be fully resolved. In other examples, the timeline further includes annotations indicating a medical professional performing each task of the resolution workflow.

[0039] In some embodiments, the analyzing operation includes (i) estimating a task completion time for resolving the possible bottleneck, (ii) computing a probability of resolving the possible bottleneck, or both. The operations 102-110 can be repeated for a plurality of interrelated current service cases 34.

EXAMPLE

[0040] The following describes the apparatus 10 and the method 100 in more detail. The method 100 generally relates to a method that adapts the most relevant information to communicate with hospital customers during case handling (or monitoring). The relevant information to communicate should focus on the alternative actions or explanations (when there are no better alternative actions). The method 100 proactively predicts and detects potential bottleneck, and determines the most matching communication information based on the confidence level of the prediction and the customer’s preferences. The method 100 is intended to be used in each status change during case handling, as well as when an unexpected waiting is known.

[0041] With reference to FIGURE 3, a more detailed embodiment of a medical device servicing workflow recommender used in this example is diagrammatically illustrated, and may for example be implemented using the medical device servicing workflow recommender apparatus of FIGURE 1. The electronic processing device 18 can include one or more modules implemented in the at least one electronic processor 20 to perform the method 100. A bottleneck module 120 is programmed to predict or detect a bottleneck or step that can be impactful for the customer, identify the cause, and prioritize them based on the impact of the bottleneck. Examples of elements can be logistic chain shortage, RSE will not be available etc. The bottleneck module 120 contains information of the overall maintenance process. To model the process various algorithms can be used, from process mining (i.e., comparing an event log with a constructed process model, and analyzing discrepancies), such as Business Process Management Notation (BPMN), Petri net (PN), and more. To quantify the probability of certain bottleneck steps, Petri net (PN) modeling and risk assessment or similar techniques can be used to generate a risk score. This module continuously updates while the case is progressing (e.g. with data coming from a service record management system). As an example, when an RSE inserts a service record specifying that they have completed a diagnostic, then the unit automatically updates the status of the service case to “remote diagnosis completed.” [0042] An impact analysis module 122 is configured to quantify the impact of each bottleneck step and retrieve the alternative actions, as well as their impact when using them (e.g., “how much time is needed when using an alternative company to deliver a part instead of the regular supplier?” “How many patients would be affected during the (extended) downtime?” and so forth). This module quantifies the impact of each bottleneck step. In the following examples, duration is used as an example of impact (i.e., a delivery delay of 24 hours is more impactful than getting an available FSE. The impact analysis module 122 may take the inter-dependencies among the causes into account. For example, unavailability of RSE may only cause 1 hour delay, but without an RSE, potential parts could not be determined. In this case, the RSE availability is a higher priority than the logistic chain shortage of the potential parts not being able to be determined, the duration estimation took into account the address of the hospital). Table 1 lists an example of the output of the impact analysis module 122.

Table 1. An example of the output of predicting or detecting bottleneck. The last row shows an example of predictive maintenance (instead of reactive maintenance).

[0043] For the causes with high priority, the module searches in external data sources such as third party suppliers to see if an alternative option could be found and estimates the duration. It also determines whether this alternative solution satisfies all the criteria of the customer, including regulatory requirements. If the duration needed from the alternative solution is shorter than waiting for the bottleneck to resolve, it is considered a valid alternative. For example, the RSE requests to replace the table for an MR device. Currently, such shipments can take 24 hour instead of a 4 hour (average duration). The unit could find company “New” which has clearance from the U.S. Food & Drug Administration (FDA) for the table on the MR device, and shipping will take 4 hours. This alternative option becomes valid.

[0044] A partial functionality module 124 is configured to determine the impacted functions of the medical device 12 and match them with appointments planned during the predicted downtime. The result is the number of appointments that can stay or should be postponed during the timeframe. Alternatively, the result can be a list of exams that can still be performed with a partially functioning system. Once the diagnosis is completed, and the impact of the issue is known (e.g. by specifying the malfunction area of the medical device 12), then the partial functionality of the device 12 can be determined. This module generates a list of possible operations/exams of the device 12 when it has a known malfunction area. Then it matches these with the scheduled exams and determines the list of unimpacted appointments and the ones that need to be shifted till after the estimated downtime period. When the malfunction is not known, an operational user (e.g., biomed engineers at hospital) could provide input on the type of operations that could still be performed. To ensure patient safety, various checks and balances can be performed. For example, the output to the customer in FIGURE 3 of a recommendations for continued use generated by the partial functionality module 124 may be transmitted to a radiology department manager or other suitably qualified person for review and approval before proceeding with performing the listed exams with the partially functioning system. Such review can be performed, for example, by sending the output from the block 126 of FIGURE 3 to an instance of the electronic processing device 18 of FIGURE 1 operated by the radiology department manager or other qualified reviewer. If the reviewer approves the proposal to continue operation with the partial functionality, additional measures could optionally be taken depending on the functionality limitations of the imaging device, such as having an on-call radiologist review and approve the images acquired by these exams prior to discharging the patient, and/or assigning a more senior imaging technologist to the imaging device with reduced functionality to conduct these exams. On the other hand, if the reviewer rejects the proposal to proceed with the listed examinations using the reduced-functionality imaging device, then this decision could be fed back to the impact analysis 122 which then updates the alternative actions accordingly. For example, if the initial run of the impact analysis 122 outputs a list of exams proposed to be performed with a partially functioning system, but this proposal is rejected by the human reviewer, then a subsequent run of the impact analysis 122 is performed with this option excluded, thereby outputting another alternative action or actions such as recommending to call an RSE on an expedited basis.

[0045] In a variant approach, the output to the customer may be a list of exams proposed to be performed with the partially functioning system, and the user (e.g. radiology department manager) can accept or reject proceeding with each imaging examination on the list individually. Hence, only a sub-set of the listed imaging examinations may be approved to proceed with the reduced-functionality imaging device. This updated information can again be fed back to the impact analysis 122 which may update its recommended alternative action(s) based on this additional information. In some examples, the selected portion of the output recommendation or explanation 40 can be automatically implemented in the medical device 12 in response to the received input.

[0046] A communication focus module 126 is configured to, based on the predicted/detected bottleneck(s), determine the focus of the communication output, that is, whether to focus on an alternative action or on explanations to manage expectations. If the bottleneck is predicted with high probability (or confidence) and alternative actions are possible, the communication output would focus on such actions. If the bottleneck is predicted with low probability (or confidence) and/or alternative actions are not possible, then the communication output would focus on providing explanations that provide transparency to the user about what is happening in the process. The explanation topics can be selected based on customer preference.

[0047] This module synthesizes the probability of each predicted bottleneck, available alternative actions and the impact of each of them. For bottlenecks with high probability (or confidence) and if alternative actions could make a significant difference in the impact, the unit would decide to focus on recommending the alternative actions.

[0048] With reference to FIGURE 4, a nonlimiting illustrative example of a suitable output of the communication focus module 126 is shown for a case in which there are multiple (illustrative three) different candidate resolution workflows identified. The output of the communication focus module 126 can be presented on the GUI 28 of the electronic processing device 18 of FIGURE 1. As shown in FIGURE 4, a first candidate resolution workflow is illustrated in a first row on the GUI 28 and labeled ‘with bottleneck’. A highlighted background or other graphical representation can indicate the estimated downtime with the identified bottleneck. The bottleneck is highlighted with, for example, a red icon. This first workflow is problematic due to the bottleneck introduced by the delivery time (labeled “Waiting for parts, delay” in FIGURE 4). Additional (e.g., two) rows shown in FIGURE 4 depict two other candidate resolution workflows, labeled “Recommendation 1” and “Recommendation 2”, one of which is to get a third party visit and the other to reschedule some of the appointments. Both of them are considered making significant change to the impact compared with the “With bottleneck” candidate resolution workflow, and hence are recommended. The customer can choose which option to take (i.e. selected between the “With bottleneck,” “Recommendation 1”, or “Recommendation 2” candidate workflows), or can request more information. Moreover, the user can provide one or more inputs via the user input device 22 in order to select an icon, a resolution workflow, or any other item displayed on the GUI 28 in order to view additional information about the selected item. In cases in which a recommendation includes a list of examinations proposed to be performed with the limited-functionality imaging device, the user interface may list these examinations and allow the user to approve or reject the list in its entirety, and/or approve or reject proceeding with each imaging examination on the list individually.

[0049] If the bottleneck is of low probability (or confidence), or no better alternative actions could be identified, then the unit would decide to focus on the explanation to the hospital customer. An example of communication output could be “Based on the estimated schedule of the first available RSE, there is a low chance to get an RSE to diagnose the problem in time.” The explanation also adapts to customers’ preferences. It could consist of multiple aspects, such as the contact person, the source of the estimation, the investigation result of alternative actions etc. Which aspects to focus on could be guided by the customer input (see the optional unit Customer Input) or an analysis from NPS or customer complaints records. From an NLP analysis, one can cluster the factors that satisfy or dissatisfy a customer regarding communication. E.g., Hospital ABC complained in the past that the vendor did not tell the estimation is based on which data source. Therefore, from this analysis, it is guided that the explanation should expose the data sources used in the estimation.

[0050] In some examples, a customer input module can allow users to specify on which aspects and to what level of granularity they want for an explanation, and to react if they have further questions on the explanations and recommendations. Some customers want to know why certain changes happened; some customers may want to know how the determination was made (intermediate steps of applying artificial intelligence). Table 2 shows an example of various communication outputs based on the predicted bottleneck and availability of the alternative actions.

Table 2. An example of determining communication output based on the predicted bottleneck and availability of the alternative actions.

[0051] The disclosed system 10 and method 100 can be implemented for a fleet of medical device 12. For example, the system 10 can analyze current service cases 34 for a plurality of devices 12 in the fleet of devices, and analyze a nature of the scheduled examinations to help determine what examinations can take place, thereby providing some different alternatives to weigh against the determined bottlenecks.

[0052] The disclosure has been described with reference to the preferred embodiments. Modifications and alterations may occur to others upon reading and understanding the preceding detailed description. It is intended that the exemplary embodiment be constructed as including all such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.

Claims

CLAIMS:

1. A non-transitory computer readable medium (26) storing: a database (30) storing a plurality of maintenance resolution workflows (32); and instructions executable by at least one electronic processor (20) to perform a method (100) of recommending one or more resolution workflows, the method including: receiving a current service case (34) for a medical device (12); identifying one or more maintenance resolution workflows in the database for resolving the current service case; populating the identified one or more maintenance resolution workflows with case information for the current service case to produce corresponding one or more current service case resolution workflows (36); identifying at least one possible bottleneck for the current service case based on the one or more current service case resolution workflows; analyzing an impact of the identified at least one possible bottleneck; and outputting a recommendation or an explanation (38) based on the analyzed impact.

2. The non-transitory computer readable medium (26) of claim 1 , wherein when a plurality of maintenance resolution workflows (32) are identified which are populated to produce a corresponding plurality of current service case resolution workflows (36): the analyzing includes generating alternative candidate resolution workflow timelines from the plurality of current service case resolution workflows wherein the alternative candidate resolution workflow timelines include the identified at least one possible bottleneck; and the outputting includes presenting the alternative candidate resolution workflow timelines on a display device (24).

3. The non-transitory computer readable medium (26) of claim 1, wherein when a single maintenance resolution workflow (32) is identified: the analyzing includes estimating a resolution time for the current service case (34) based at least on the at least one possible bottleneck; and the outputting includes presenting an explanation (38) of the resolution time including identifying the at least one possible bottleneck.

4. The non-transitory computer readable medium (26) of any one of claims 1-3, wherein analyzing an impact of the identified at least one possible bottleneck includes: analyzing the impact of the at least one possible bottleneck on an examination schedule to be performed using at least the medical device (12).

5. The non-transitory computer readable medium (26) of claim 4, wherein the analyzing of the impact on the examination schedule includes: determining a reduced functionality of the medical device (12) while the current service case (34) remains unresolved; and determining a revised examination schedule capable of being performed by the medical device with the determined reduced functionality.

6. The non-transitory computer readable medium (26) of claim 5, wherein analyzing an impact of the identified at least one possible timeframe issue includes: generating a timeline (38) of the current service case resolution workflow (36) annotated with downtimes when the medical device (12) would be unavailable or available with the determined reduced functionality, the timeline being output on a display device (24).

7. The non-transitory computer readable medium (26) of claim 6, wherein the timeline (38) further includes an estimated time when the current case will be fully resolved.

8. The non-transitory computer readable medium (26) of either one of claims 6 and 7, wherein the timeline (38) further includes annotations indicating a medical professional performing each task of the current service case resolution workflow (36).

9. The non-transitory computer readable medium (26) of any one of claims 1-8, wherein the analyzing includes: estimating a task completion time for resolving the possible bottleneck.

10. The non-transitory computer readable medium (26) of any one of claims 1-9, wherein the analyzing includes: computing a probability of resolving the possible bottleneck.

11. The non-transitory computer readable medium (26) of any one of claims 1-10, wherein the method (100) further includes receiving an input indicative of an approval of the output recommendation or an explanation (38).

12. The non-transitory computer readable medium (26) of any one of claims 1-10, wherein the method (100) further includes receiving an input indicative of a selection of a portion of the output recommendation or an explanation (38); and at least one of:

(i) displaying additional information related to the selected portion of the output recommendation or an explanation; and

(ii) automatically implementing the selected portion of the output recommendation or explanation in response to the received input.

13. The non-transitory computer readable medium (26) of any one of claims 1-12, wherein the method (100) further includes: repeating the receiving, identifying, identifying, populating, analyzing, and outputting for a plurality of current service cases (34).

14. The non-transitory computer readable medium (26) of any one of claims 1-13, wherein each maintenance resolution workflow (32) includes tasks and interdependencies based on the at least one bottleneck.

15. The non-transitory computer readable medium (26) of any one of claims 1-14, wherein the device (12) comprises a fleet of devices, and the method (100) further includes: repeating the receiving, identifying, identifying, populating, analyzing, and outputting for a plurality of current service cases (34) for the devices of the fleet of devices.

16. A non-transitory computer readable medium (26) storing: a database (30) storing a plurality of maintenance resolution workflows (32); and instructions executable by at least one electronic processor (20) to perform a method (100) of recommending one or more resolution workflows, the method including: receiving a current service case (34) for a medical device (12); identifying a plurality of maintenance resolution workflows in the database for resolving the current service case; populating the identified maintenance resolution workflows with case information for the current service case to produce corresponding current service case resolution workflows (36); identifying at least one possible bottleneck for the current service case based on the current service case resolution workflows; analyzing an impact of the identified at least one possible bottleneck; and outputting a recommendation or an explanation (38) based on the analyzed impact.

17. The non-transitory computer readable medium (26) of claim 16, wherein: the analyzing includes generating alternative candidate resolution workflow timelines from the plurality of current service case resolution workflows wherein the alternative candidate resolution workflow timelines include the identified at least one possible bottleneck; and the outputting includes presenting the alternative candidate resolution workflow timelines on a display device (24).

18. The non-transitory computer readable medium (26) of claim 16, wherein the analyzing of the impact on the examination schedule includes: determining a reduced functionality of the medical device (12) while the current service case (34) remains unresolved; and determining a revised examination schedule capable of being performed by the medical device with the determined reduced functionality.

19. The non-transitory computer readable medium (26) of claim 18, wherein analyzing an impact of the identified at least one possible timeframe issue includes: generating a timeline (38) of the current service case resolution workflow (36) annotated with downtimes when the medical device (12) would be unavailable or available with the determined reduced functionality.

20. A method (100) of recommending one or more resolution workflows, the method including: receiving a current service case (34) for a medical device (12); identifying a maintenance resolution workflow (32) stored in a database (30) for resolving the current service case; populating the identified maintenance resolution workflow with case information for the current service case to produce a corresponding current service case resolution workflow (36); identifying at least one possible bottleneck for the current service case based on the current service case resolution workflow; analyzing an impact of the identified at least one possible bottleneck including estimating a resolution time for the current service case (34) based at least on the at least one possible bottleneck; and outputting a recommendation or an explanation (38) of the resolution time including identifying the at least one possible bottleneck.