Maintenance service system for medical imaging systems

Abstract

A system includes an imaging system and a malfunction module that generates a malfunction signal indicating a malfunction of the imaging system. The system further includes a maintenance service system with a data repository that stores malfunction-solution pairs, text templates, and user profiles, including repair experience of levels of user, a solutions module that determines a solution to the malfunction from the malfunction-solution pairs in response to receiving the malfunction signal, a complexity assertion module that determines a strategy for mitigating the malfunction based on the solution and a user profile of a user mitigating the malfunction, and a report generation module that automatically creates a service report for the user based on the strategy, the user profile, the solution, and information from the data repository. The maintenance service system transmits the service report to a client device of a user mitigating the malfunction.

Description

[0001] MAINTENANCE SERVICE SYSTEM FOR MEDICAL IMAGING SYSTEMS

[0002] TECHNICAL FIELD

[0003] The following generally relates to medical imaging systems, and, more particularly, to a maintenance service system for medical imaging systems.

[0004] BACKGROUND

[0005] Modem medical imaging systems, e.g., a Magnetic Resonance (MR), Computed Tomography (CT), Positron Emission Tomography (PET), Single Photon Emission Tomography (SPECT), X-ray, Ultrasound (US), and / or other medical imaging modality, are complex machines composed of many parts. Their maintenance is difficult and requires involvement of specialized personnel performing remote assistance (e.g., a Remote Service Engineer (RSE)). It is also often necessary to deploy a field service engineer (FSE) on-site when a malfunctioning part needs to be fixed or replaced. The term malfunctioning herein means a part that is not operating according to specification. RSEs and FSEs are supported in this task by predictive and proactive models automatically generating reports alert for a given malfunction from static templates about troubleshooting and service actions to be performed, helping them identify correct service actions and parts to be replaced for that malfunction.

[0006] The static templates include a template for each part and failure mode, adding up to several thousands of different templates, and the reports are often verbose and complex, mainly tailored for highly qualified staff, e.g., FSEs, rendering them unusable by other user groups, including those that could deal with the malfunction. FIG. 1 shows an example of part 102 of a model alert JSON (JavaScript Object Notation) file with a list of attributes 104, a list of values 106, and attribute-value pairs 108i , ... 108i, ... , 108N (where N is a positive integer) with information that can be included in a report. FIG. 2 shows an example of part 202 of a report generated based on the model alert file of FIG. 1. A Summary section 204 will include information summarizing the alert, and a detailed information section 206 will include detailed information about the alert. The RSE and / or FSE use such information as a guide to troubleshoot the malfunction. The whole process generally consumes time and adds to the service cost.

[0007] Hospitals often have well-trained and educated technical personnel, e.g., a biomedical engineer, etc., to perform self-help to resolve minor problems with medical devices at the moment when malfunctions occur. The term self-help herein means the owner of the imaging system handles the malfunction and not the manufacturer. However, performing maintenance in a self-help setting is often impossible with medical imaging systems. For example, template-generated reports by predictive and proactive models (and / or standard service documentation) are often too complex for even the most expert engineers, not pinpointing to the relevant section (e.g., FIGS. 1 and 2), and include proprietary and sensitive information (e.g., technical images, blueprints, intellectual property protected concepts, etc.) that is unavailable to personnel (e.g., hospital technical personnel, etc.) outside of authorized personnel of the manufacturer of the device and would have to be obscured. Such constraints may drastically reduce the possibility of self-help with medical imaging systems. As such, there is an unresolved need for an improved maintenance service approach(s) for medical imaging systems.

[0008] SUMMARY

[0009] Aspects described herein address the above-referenced problems and / or others. In one aspect, a system includes an imaging system with a malfunction module configured to generate a malfunction signal indicating a malfunction of the imaging system. The system further includes a maintenance service system with a data repository configured to store malfunctionsolution pairs, text templates, and user profiles including repair experience of levels of user, a solutions module configured to determine a solution to the malfunction from the malfunction-solution pairs in response to receiving the malfunction signal, a complexity assertion module configured to determine a strategy for mitigating the malfunction based on the solution and a user profile of a user mitigating the malfunction, and a report generation module configured to automatically create a service report for the user based on the strategy, the user profile, the solution, and information from the text templates. The maintenance service system is configured to transmit the service report to a client device of a user mitigating the malfunction.

[0010] In another aspect, a computer-implemented method includes receiving, from an imaging system, a malfunction signal indicating a malfunction with the imaging system. The method further includes determining a solution for the received malfunction signal. The method further includes determining a strategy for the solution based on the malfunction and a user profile of a user mitigating the malfunction. The method further includes generating a service report with an action for mitigating the malfunction based on the strategy, the user profile, and service information. The method further includes transmitting the service report to the user, wherein the user performs the action in the service report.

[0011] In another aspect, a computer readable medium is encoded with computer executable instructions that cause a processor to receive, from an imaging system, a malfunction signal indicating a malfunction with the imaging system. The computer executable instructions further cause the processor to determine a solution for the received malfunction signal. The computer executable instructions further cause the processor to determine a strategy for the solution based on the malfunction. The computer executable instructions further cause the processor to identify a user from a user profile to mitigate the malfunction. The computer executable instructions further cause the processor to generate service report with an action for mitigating the malfunction based on the strategy, the user profile of the user, and one or more of a parts list, images of parts of the image system, and technical documents. The computer executable instructions further cause the processor to transmit the service report to the user, wherein the user performs the action in the service report.

[0012] Those skilled in the art will recognize still other aspects of the present application upon reading and understanding the attached description. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The invention may take form in various components and arrangements of components, and in various steps and arrangements of steps. The drawings are only for the purposes of illustrating the embodiments and are not to be construed as limiting the invention.

[0014] FIG. 1 shows an example of part of a model alert fde with information that can be included in a report.

[0015] FIG. 2 shows an example of part of a report generated based on a model alert fde.

[0016] FIG. 3 diagrammatically illustrates an example system, in accordance with an embodiment(s) herein.

[0017] FIG. 4 diagrammatically illustrates a variation of the example system, in accordance with an embodiment(s) herein.

[0018] FIG. 5 diagrammatically illustrates an example imaging system, in accordance with an embodiment(s) herein.

[0019] FIG. 6 diagrammatically illustrates a variation of the example imaging system, in accordance with an embodiment(s) herein.

[0020] FIG. 7 diagrammatically illustrates another variation of the example imaging system, in accordance with an embodiment(s) herein.

[0021] FIG. 8 diagrammatically illustrates an example of the maintenance service system, in accordance with an embodiment(s) herein.

[0022] FIG. 9 diagrammatically illustrates an example of creating a report, in accordance with an embodiment(s) herein.

[0023] FIG. 10 diagrammatically illustrates an example of a report, in accordance with an embodiment(s) herein.

[0024] FIG. 11 illustrates an example method, in accordance with an embodiment(s) herein.

[0025] FIG. 12 illustrates another example method, in accordance with an embodiment(s) herein.

[0026] DESCRIPTION OF EMBODIMENTS

[0027] The following describes a maintenance service approach for medical imaging systems that at least addresses a shortcoming of existing approaches discussed herein. In one instance, the approach includes automatically generating a maintenance report with information about required service actions that are tailored to a specific user and / or user group, such as an end user, e.g., by tailoring a report based on a profde of a specific individual and / or experience for self-help in medical equipment fixing / repair. For example, a report describing the detected malfunction and the troubleshooting and repair action can be generated according to the level of expertise of the personnel mitigating or overseeing resolution of the malfunction.

[0028] As described in greater detail below, in one instance, the approach described herein utilizes text templates and / or segments and / or machine learning techniques along with technical information stored in a database for automatic report generation, which mitigates relying on single static templates for each combination of the sub-system, part, failure mode, and / or user group. In addition, in one instance, reports can be tailored for less experienced personnel, which may enable increased self-help problem solving for the end user. Furthermore, in one instance, this approach mitigates sending a field service engineer (FSE) on-site, which may reduce service cost and repair time and improve customer satisfaction.

[0029] FIGS. 3 and 4 diagrammatically illustrate an example system 302. The example system 302 includes a first plurality of imaging systems 304, a second plurality of imaging systems 306, a medical imaging manufacturer 308, a third party service provider 310, a medical imaging maintenance service system 312, and a network 314. Medical imaging systems of the first and second plurality of imaging systems 304 and 306 include one or more types of medical imaging modalities, including an MR, a CT, a PET, a SPECT, an X-ray, an US, and / or other medical imaging modalities.

[0030] A medical imaging system 316 is a stand-alone imaging system, e.g., located in a doctor's office, part of a portable mobile scanner, etc. Medical imaging systems 318, ... , 320 are part of imaging center 322, e.g., an imaging department of a hospital, a dedicated imaging facility, etc. Medical imaging systems 324, ... , 326 are part of imaging center 328, e.g., an imaging department of a hospital, a dedicated imaging facility, etc. Medical imaging systems 330, ... , 332 are part of imaging center 334, e.g., an imaging department of a hospital, a dedicated imaging facility, etc. The imaging center 328, ... , and the imaging center 334 are part of an imaging network 336, e.g., a network of hospitals and / or a network of dedicated imaging facilities.

[0031] For clarity and brevity, the imaging systems 316, 318, ... , 320, 324, ... , 326, and 330, ... , 332 are collectively referred to herein as imaging systems, and the imaging centers 322 and 328, ... , 334 are collectively referred to herein as imaging centers.

[0032] The imaging system manufacturer 308 is a manufacturer of at least one of the imaging systems. The third-party service provider 310 includes an entity other than the medical imaging manufacturer 308 authorized to perform maintenance on one or more imaging systems. The imaging systems, the imaging system manufacturer 308, and / or the third-party service provider 310 are in electrical communication with the network 314, which may include a local area network (LAN), a wide area network (WAN), and / or other network, configured for wireless and / or wired (physical cables) technology.

[0033] In FIG. 3, the medical imaging maintenance service system 312 is included in the imaging system manufacturer 308. For example, in one instance, the medical imaging maintenance service system 312 is located at the imaging system manufacturer 308. In FIG. 4, the medical imaging maintenance service system 312 is remote from the imaging system manufacturer 308. For example, in FIG. 4, the medical imaging maintenance service system 312 is part of remote cloud-based resources, including processing, software, storage, algorithms, and / or other computing resources. In another instance, the medical imaging maintenance service system 312 is included in a server, database, etc. that is remote from the imaging system manufacturer 308. Referring to FIGS. 3 and 4, the medical imaging maintenance service system 312 is in electrical communication, via the network 314, with one or more of the imaging systems, one or more of the imaging centers, the imaging system network 336, the medical imaging manufacturer 308, and / or the third-party service provider 310. The medical imaging maintenance service system 312 is configured to receive a malfunction alert signal automatically generated by hardware and / or software monitoring one of the imaging systems without user interaction, and / or a root-cause signal generated in response to a user malfunction query at one of the imaging systems, the imaging centers, and / or the imaging system network 336.

[0034] As described in greater detail herein, the medical imaging maintenance service system 312, in response to the malfunction alert signal and / or the root-cause signal, automatically generates a service report that is tailored to personnel mitigating the malfunction, which includes personnel at the imaging systems, the imaging centers, the medical imaging manufacturer 308, and / or the third-party service provider 310, based on a profile of the personnel, which includes at least an experience of the personnel with fixing and / or repairing medical equipment, and / or a database of templates, template segments and / or machine learning, and technical information.

[0035] The medical imaging maintenance service system 312 provides the service report to the personnel. This includes providing the report to the personnel mitigating the malfunction at the imaging system (e.g., a technician, a biomedical engineer, etc.), the personnel at the imaging manufacturer 308 (e.g., an FSE, an RSE, an RME, etc.), or both the personnel mitigating the malfunction at the imaging system 320 and the personnel at the imaging manufacturer 308. As the reports are tailored to the personnel, a report transmitted to the personnel mitigating the malfunction at the imaging system and a report transmitted to the personnel at the imaging manufacturer 308 may be different.

[0036] In one instance, a service report is transmitted over VPN and / or other secure communication channel. In one instance, a notice of the service report and / or a link to the service report is transmitted to a client device. In one instance, this includes transmitting the notice of the service report and / or a link to the service report via an email client. In another instance, a notice of the service report and / or a link to the service report is transmitted via IP telephony, e.g., a voice over the internet (VoIP) client. In another instance, a notice of the service report and / or a link to the service report is transmitted via cellular and / or telephone line. In another instance, notice of the service report and / or a link to the service report is provided via an on-screen messaging application.

[0037] In one instance, the service report can be encrypted and / or otherwise converted from an original representation of the information into an alternative form requiring decrypting and / or converting back to the original representation of the information. In one instance, authorized recipients of the report have software that performs the decryption and / or converting back to the original representation of the information, wherein content of the report is unintelligible in the encrypted form. In this instance, absence the software, the report is not legible. The encryption may have different levels corresponding to the type of user, e.g., different encryption depending on whether the user is personnel at an imaging center, at the manufacturer, etc. FIGS. 5, 6 and 7 diagrammatically illustrate an example imaging system 500 of the imaging systems. In this example, the imaging system 500 is configured at least for MR imaging. Again, the imaging systems generally include imaging systems configured for different medical imaging modalities such as MR, CT, PET, SPECT, US, X-ray, etc.

[0038] The imaging system 500 includes a gantry 502. The gantry 502 includes a main magnet 504, a gradient (x, y, and z) coil(s) 506, and an RF coil 508. The main magnet 504 (which can be a superconducting, resistive, permanent, or other type of magnet) produces a substantially homogeneous, temporally constant main magnetic field B0 in an examination region 510. The gradient coil(s) 506 generates time-varying gradient magnetic fields along the x, y, and z-axes of the examination region 510.

[0039] The RF coil(s) 508 includes a transmit portion that produces radio frequency signals (at the Larmor frequency of nuclei of interest (e.g., hydrogen, etc.)) that excite the nuclei of interest in the examination region 510 and a receive portion that detects MR signals emitted by the excited nuclei. In other embodiments, the transmit portion and the receive portion of the RF coil(s) 508 are located in separate RF coils. A data acquisition system (DAS) 512 processes the MR signals, and a reconstructor 514 reconstructs the data and generates MR images.

[0040] A subject support 516 includes a tabletop 518 moveably coupled to a frame / base 520. In one instance, the tabletop 518 is slidably coupled to the frame / base 520 via a bearing or the like, and a drive system (not visible) including a controller, a motor, a lead screw, and a nut (or other drive system) translates the tabletop 518 along the frame / base 520 into and out of the examination region 510 along a longitudinal or z-axis 522. The tabletop 518 is configured to support an object or subject in the examination region 510.

[0041] A hardware / software (HW / SW)-detected malfunction module 524 is configured to automatically detect a malfunction of a component, part, module, sub-system, etc. of the imaging system 502, generate a malfunction alert signal in response thereto, and transmits the malfunction alert signal to the maintenance service system 312 of FIGS. 3 and / or 4. In one instance, the HW / SW-detected malfunction module 524 analyzes log files in search for anomalies. In another instance, the HW / SW- detected malfunction module 524 analyzes outputs to determine whether values thereof are within operating specifications. Other techniques are also contemplated herein. The malfunction alert signal, in one instance, at least includes a description of the malfunction and a unique identifier of the imaging system.

[0042] In FIG. 5, the HW / SW-detected malfunction module 524 is included in resources 526 that are remote from and external to the imaging system 500. In the illustrated embodiment, the resources 526 include cloud-based resources, including processing, software, storage, algorithms, and / or other computing resources. In another instance, the resources 526 are included in a server, database, etc., including resources at a single location and / or resources distributed across multiple servers, databases, etc. and / or located at multiple different locations. This also includes a combination of cloud-based and non-cloud-based resources. In FIG. 6, the HW / SW-detected malfunction module 524 is included in a device 602 that is local but external to the imaging system 500. In one instance, the device 602 includes a sensor, a processor, software, storage, and / or technologies that connect and exchange data with other devices and / or systems over the network 314 and / or otherwise. An example of such a device is an Internet of Things (loT) device. In the illustrated embodiment, the loT device 602 is connected to both the network 314 and directly to the imaging system 500. In another instance, the loT device 602 is not directly connected to the imaging system 500.

[0043] In FIG. 7, the HW / SW-detected malfunction module 524 is included in the imaging system 500. In the illustrated embodiment, the HW / SW-detected malfunction module 524 is shown a single module outside of any particular component of the imaging system 500. In another instance, the HW / SW-detected malfunction module 524 includes a plurality of the HW / SW-detected malfunction submodules, at least two of which are configured to detect different malfunctions. In another instance, at least one of the sub-modules is incorporated into a particular component being monitored.

[0044] Continuing with FIGS. 5, 6 and 7, in one instance, the HW / SW-detected malfunction module 524 utilizes only hardware to detect the malfunction and / or transmit the malfunction alert signal. In another instance, the HW / SW-detected malfunction module 524 includes only software to detect the malfunction and / or transmit the malfunction alert signal. In yet another instance, the HW / SW-detected malfunction module 524 includes a combination of hardware and software to detect the malfunction and / or transmit the malfunction alert signal.

[0045] The imaging system 500 further includes an operator console 528. The operator console 528 includes a computing system such as a computer, a workstation, a server, or the like, configured for operating the imaging system 500. The operator console 528 includes an input device(s) 530 such as a keyboard, mouse, touchscreen, microphone, etc., and an output device(s) 532, which includes a human readable device such as a display monitor or the like. The operator console 528 further includes input / output (I / O) 534 for transmitting and / or receiving signals and / or data.

[0046] The operator console 528 further includes a processor(s) 536 such as a micro-processing unit (MPU), a central processing unit (CPU), a graphics processing unit (GPU), etc. The operator console 528 further includes a computer readable storage medium 538 (“MEMORY”), which includes non- transitory medium (e.g., a storage cell, device, etc.) and excludes transitory medium (i.e., signals, carrier waves, and the like). The computer readable storage medium 538 is encoded with computer executable instructions. The processor(s) 536 is configured to execute the computer executable instructions.

[0047] In the illustrated embodiment, the computer executable instructions include a user- detected malfunction module 540. The user-detected malfunction module 540, in response to receiving a user input, e.g., via the input device(s) 530, that indicates a malfunctioning behavior of the imaging system 500 that was observed by the user, infers a root cause and generates a root-cause signal for the malfunction e.g., from mappings of malfunctions to root-causes in a data structure, table, list, etc. and / or otherwise, and transmits the root-cause signal to the maintenance service system 312 of FIGS. 3 and / or 4. The signal at least includes a description of the root cause, unique identifier of the imaging system, and a unique identifier of the user.

[0048] A non-limiting example of determining a root cause is described in WO 2023 / 099412 Al, filed 11 / 28 / 22, and entitled “Case Intake System and Method with Remote Diagnostic Test Recommendation and Automatic Generation of Profiled Question,” which is incorporated in its entirety by reference herein. With this example, information describing an issue related to a functioning of a medical device is received, device log data automatically generated by the medical device is retrieved, scores for diagnostic tests for diagnosing the issue are determined based on the retrieved device log data and the information using a scoring function, and a ranked list of diagnostic tests is created based on the scores.

[0049] The HW / SW-detected malfunction module 524 and the user-detected malfunction module 540 are collectively referred to herein as a malfunction module.

[0050] In one instance, the imaging system 500 is in electrical communication with an information system 542. Examples of the information system 542 include a Radiology Information System (RIS), a Hospital Information System (HIS), etc. Communication between the imaging system 500 and the information system 542 can be via Health Level Seven (HL7) and / or other protocol for transferring data. In addition, the imaging system 500 can transfer image data to a RIS, a Picture Archiving and Communication System (PACS), an Electronic Medical Record (EMR), etc. via a Digital Imaging and Communications in Medicine (DICOM) standard for the communication and management of medical imaging information and related data.

[0051] FIG. 8 diagrammatically illustrates an example of the maintenance service system 312 of FIGS. 3 and / or 4. In this example, the maintenance service system 312 includes a data repository 802. In another instance, the data repository 802 is separate from the maintenance service system 312 and is in electrical communication therewith via wired and / or wireless technology. In either instance, the data repository 802 can be remote relative to the maintenance service system 312, e.g., resources in a cloudbased service, in a server and / or a database of the medical imaging system manufacturer 310 of FIGS. 3 and / or 4, etc., and / or local to one or more of the imaging centers, the imaging system network 336, the third party service provider 310, and / or other entity. The illustrated data repository 802 stores at least root-cause and / or alert solutions 804, parts list 806, images 808, text templates and / or segments 810, technical documents 812, and / or user profdes 814.

[0052] In the illustrated example, the root-cause solutions and the alert solutions are shown together in the data repository 802 as root-cause and / or alert solutions 804. In another instance, the root cause solutions and the alert are stored separately in the data repository 802. The root-cause and / or alert solutions 804 includes a plurality of pairs of a root-cause, solution pairs and a plurality of alert solution pairs. Such pairs can be determined by mining documented solutions to previous imaging system malfunctions, by personnel who have successfully implemented solutions to previous imaging system malfunctions, machine learning trained with successful and / or unsuccessful solutions, by manufacturer- suggested solutions, and / or otherwise. In one instance, the root cause and / or alert solutions information is added manually by personnel, e.g., by engineers as they mitigate malfunctions. For this, the root cause and / or alert solutions information can be uploaded to the date repository 802. This also includes uploading existing root-cause and / or alert solutions information.

[0053] In another instance, the root cause and / or alert solutions information is semi- automatically extracted from the historical service records, operator manuals, log files, etc. via text analysis and / or otherwise. This data may be blended into semi-structured text that requires basic text analysis to be extracted and manually checked. With automatic extraction techniques, personnel can reject, modify and / or accept a pair. In one instance, the root-cause and / or alert solutions 804 is manually and / or automatically updated to include new solution pairs and / or remove existing solution pairs, e.g., via a feedback and / or learning process. An example solution may indicate a service to perform, an identification of a part, a description of how to order the part, etc.

[0054] The parts list 806 includes a list of the different parts of each of the imaging systems. The parts list 806 can be delineated based on modality, model, serial number, age, sub-system within a modality, etc. Such pairs can be mined from manufacturer documentation and / or otherwise. In one instance, the parts list 806 is manually and / or automatically updated to include new parts and / or remove existing parts, e.g., via a feedback and / or learning process.

[0055] The images 808 include pictures, schematics, blueprints, etc. of the imaging systems and / or sub-systems thereof. Likewise, the images 808 can be mined from manufacturer documentation and / or otherwise, and delineated based on modality, model, serial number, age, sub-system within a modality, etc. One or more of the images may include textual descriptions and / or paired with a level of experience. In one instance, the images 808 are manually and / or automatically updated to include new images and / or remove existing images, e.g., via a feedback and / or learning process.

[0056] The text templates and / or segments 810 include templates or segments necessary for generating a report. Such templates would represent a larger portion of a report document to be fdled with the needed information. Segments, instead, would represent smaller parts, like a single entry in a formlike structure, or just a template sentence, to be used in some circumstances given the user-alert input pair. Both templates and segments are paired with malfunctions, and they could be associated with an experience level denoting when to use them according to a knowledge of a user. In one instance, the text templates and / or segments 810 are manually and / or automatically updated to include new text templates and / or segments and / or remove existing text templates and / or segments, e.g., via a feedback and / or learning process.

[0057] The technical documents 812 include specifications, service, operation, etc. information for each of the different imaging systems. The technical documents 812 can be mined from manufacturer documentation and / or otherwise, and delineated based on modality, model, serial number, age, subsystem within a modality, etc. In one instance, the technical documents 812 are manually and / or automatically updated to include new technical documents and / or remove existing technical documents, e.g., via a feedback and / or learning process. The user profiles 814 include information about users authorized to mitigate malfunctions. For example, a user profile for a particular user will identify whether the user is an imaging technician or clinician performing a scan, hospital technical personnel such as a biomedical engineer, a FSE, a RSE, a RME, a third party, and / or other who can assist with mitigating a malfunction. Additionally, or alternatively, the user profile for the particular user will include information regarding the experience level of the user with handling a malfunction of the imaging systems, including the types of malfunctions the user has encountered, and capabilities of the user.

[0058] Additionally, or alternatively, the user profile for the particular user will include information about the user’s authorization. For example, the user profile for the particular user can indicate whether the user is permitted or has access to proprietary and / or otherwise sensitive information (and to what degree / level of access where there are multiple levels, each with different access). Such information can include one or more of, but is not limited to, technical images, blueprints, intellectual property protected concepts, etc. For instance, service personnel of the manufacture of the device may have clearance to access such information for the device, whereas other users may not have access to such information.

[0059] Additionally, or alternatively, the authorization information for the particular user may indicate whether certain information the user has clearance to access may create a regulatory and / or cybersecurity issue, e.g., based on how the information is being accessed (e.g., over a secure or unsecure network, without or without data encryption technologies, locally or remotely, etc.), who, outside of the authorized personnel, might be able to view the information (e.g., a person in the room with the authorized personnel, etc.), current regulatory standards, etc. User profile information can be obtained from a user’s employer (e.g., a hospital) and include historical service record of the user, which can be updated after completing subsequent service tasks on the imaging systems.

[0060] The maintenance service system 312 further includes a solutions module 816. In one instance, the solutions module 816 receives, as input, a malfunction alert signal generated by the HW / SW-detected malfunction module 524 (FIGS. 5, 6 and / or 7). In another instance, the solutions module 816 receives, as input, a root-cause signal generated by the user-detected system -malfunction module 540 (FIGS. 5, 6 and / or 7). Again, the malfunction alert signal and / or the root-cause signal will include information at least identifying the malfunction and the imaging system, and the root-cause signal will include also include the user initiating the malfunction query and / or the user who will mitigate the malfunction.

[0061] In one instance, one or more of the root-cause and / or alert solutions 804, the parts list 806, the images 808, the text templates and / or segments 810, and / or the technical documents 812 are classified and linked together by imaging system type, imaging system model, a sub-system within the imaging system, imaging center, imaging network, etc. Additionally, or alternatively, one or more of the root-cause and / or alert solutions 804, the parts list 806, the images 808, the text templates and / or segments 810, and / or the technical documents 812 are classified and linked to a user type, e.g., a biomedical engineer, a FSE, a RSE, a RME, etc. In one instance, a model learns what data should be provided to which type of user. For this, human-prepared templates can be created and used as a training set to train the model.

[0062] The solutions module 816 searches the root-cause and / or alert solutions 804 based on the input root cause signal or alert signal to find a root cause-solution pair or an alert-solution pair for the malfunction, and outputs the solution. A solution for a given root-cause often requires some actions to be undertaken, which can be specific, e.g., replace part with provided part code, or they can be more complex, like a resolution flowchart to be followed, e.g., “check level of liquid coolant, if empty or below threshold refill, otherwise check other things... .” In one instance, each solution in the data repository 802 is associated with a unique identifier, together with all the additional information, e.g., parts to replace and / or instructions to follow.

[0063] The maintenance service system 312 further includes a complexity assertion module 818. The complexity assertion module 818 receives, as input, the solution determined by the solutions module 816. The complexity assertion module 818 either receives and / or retrieves the user profile of the personnel overseeing resolution of the malfunction from the user profiles 814. In one instance, the solutions module 816 retrieves the user profile and provides it to the complexity assertion module 818. In another instance, the solutions module 816 provides identification information of the personnel overseeing resolution and the complexity assertion module 818 retrieves the user profile.

[0064] The complexity assertion module 818, based on the solution determined by the solutions module 816 and information in the user profile 814 such as an experience level of the personnel overseeing resolution with the current malfunction and / or other malfunctions, identifies and selects a report generation strategy for the user from a collection of report generation strategies 820. In one instance, the strategies 820 include a predetermined set of report generation strategies. By way of nonlimiting example, the strategies 820 may include strategies (or more or less) identified in Table 1, which can be stored in a data structure, table, list, database, and / or otherwise. It is to be appreciated that the strategies in Table 1 below are non-limiting and are provided for explanatory purposes, and more or less strategies, including different strategies are contemplated herein.

[0065] Table 1. A collection of report generation strategies.

[0066] In Table 1, strategy 1 is “produce a report for the user.” In one instance, strategy 1 is selected, e.g., for straightforward solutions that can be carried out in self-help mode. Strategy 2 is “produce a report for the user and notify the manufacturer.” In one instance, strategy 2 is selected, e.g., for solutions that can be carried out in self-help mode, which may have consequences of some sort, so it is essential to notify the manufacturer what is being done. Strategy 3 is “produce a report only for the manufacturer and notify the user.” In one instance, strategy 3 is selected, e.g., if self-help is not possible, inform the user and send the generated report directly to the manufacturer, e.g., so that the case can be escalated.

[0067] In Table 1, strategy 4 is “request the user to perform some actions and send a report to the manufacturer with the outcomes of actions.” In one instance, strategy 4 is selected, e.g., the malfunction may not be resolved in self-help mode, but the user may carry out some actions to provide essential information to the manufacturer that may speed up maintenance, e.g., the user may be asked to run some tests, whose results would be automatically encapsulated in a report generated for the manufacturer. In one instance, the complexity assertion module 818 utilizes a static mapping 822 to identify and select the strategy. For example, a suitable mapping 822 includes a static mapping where solutions, personnel, and generation strategies are linked together.

[0068] In another instance, the complexity assertion module 818 utilizes a dynamic model 826 such as probabilistic and / or other model to identify and select the strategy. By way of non-limiting example, EQUATION 1 includes a suitable probabilistic model fg:

[0069] EQUATION 1 fa(u,s) = g, g & g, where u is a (one or multi-dimensional) vector representing a profile of a user, s is a (one or multidimensional) vector representing a solution to adopt, g is a generation strategy to follow, and g is the collection of available strategies. In one instance, u includes information including, but not limited to, a role of a user, an amount of successful and unsuccessful service actions for different severity levels, years of experience on the given modality, etc. In one instance, s includes information including, but not limited to, the part to be replaced, categorical classification of the solution, the severity of the resolution, etc.

[0070] In one instance, the dynamic model 824 and / or user profiles 814 are updated with the outcome of each service action treated following the strategy produced by the complexity assertion module 818. For example, the failure of personnel in finalizing a service action after following the information in a generated report could be used to update the dynamic model 826 so that next time a different, safest strategy would be recommended for similar profiles, and the profile of the user would be updated to indicate the failure. This can be achieved, for example, by applying a reinforcement learning strategies and / or simply by re-training the dynamic model 826. The training data may be obtained by historical data, e.g., records of hospital personnel attempting specific solutions. After the first release of the dynamic model 226, the training data could be obtained by logging all the interventions and their outcome. The report generation module 828 receives, as input, the strategy selected by the complexity assertion module 818 and the solution determined by the solutions module 816. In one instance, the report generation module 828 receives the user profile 814. In another instance, the report generation module 828 retrieves the user profile from the data repository 802. The report generation module 828 generates a report based on the strategy, the user profile and / or the root-cause and / or alert solution. For this, the report generation module 828 utilizes the report text templates and / or segments 810, the parts list 806, the images 808, and / or the technical documents 812.

[0071] The report generation module 828 employs an algorithm from a set of algorithms 830 and / or other approaches to generate a report. In the illustrated embodiment, the set of algorithms 830 includes a static algorithm 832, a dynamic algorithm 834, and a hybrid algorithm 836. In another instance, the set of algorithms 830 include a different combination, e.g., the static algorithm 832 and the dynamic algorithm 834, the dynamic algorithm 834 and the hybrid algorithm 836, the hybrid algorithm 836 and the static algorithm 832, just the static algorithm 832, just the dynamic algorithm 834, just the hybrid algorithm 836, and / or other algorithm.

[0072] An example of a suitable static algorithm includes, given the specific user profile (e.g., biomedical engineer, RSE, etc.) and a solution to implement, performing a set of queries over the data repository 802 to retrieve the information to add to a selected templated to generate a report. A specific template is selected, and the related textual segments are used to fill in different parts of a main template. These, in turn, are filled with specific terminology, e.g., part or sub-part names, where needed. Similarly, images are selected according to the user profile and attached to the template.

[0073] An example of a suitable dynamic solution includes, given users do not belong to a specific group, classifying user experience level, e.g., using a statistical profile classifier. With this example, a function fpwould output a user group according to EQUATION 2:

[0074] EQUATION 2 fp(u) = p, p ^ P' , where u is a (one or multi-dimensional) vector representing a profile of a user similar to EQUATION 1, p is a user in a set P of user group labels. Examples of P could be {inexpert, low-experience, high- experience, expert, ... }, simply {0, 1, 2, 3, ... }, etc. The information in the data repository 802 would be associated with the level of expertise, and the query would retrieve the information based on the level of expertise and incorporate the retrieved information into the argument.

[0075] With a hybrid solution, each item in the data repository 802 used for generation could be associated with a “complexity score” evaluated using different methods. This can be achieved using existing metrics or by applying trained scoring models. Metrics for textual inputs, i.e., templates, segments, and technical document sections, may include complexity reading scores, e.g., Flesch Reading Ease, Fog Scale, SMOG Index, etc. Metrics for image complexity may include a Root Mean Square Error, a Fractal Dimension, an Entropy, and a Suprathreshold detectability. An example can be found in D. Dorukalp, "Spatial Frequency and the Performance of Image-Based Visual Complexity Metrics," IEEE Access, vol. 8, pp. 100111-100119, 2020.

[0076] Dedicated probabilistic models can be trained to output complexity scores for each input type. An example can be found in M. Martinc, S. Pollak and M. Robnik-Sikonja, "Supervised and unsupervised neural approaches to text readability," Computational Linguistics Journal, vol. 47, no. 1, pp. 141-179, 2021. With this approach, the scores would be more profiled towards a specific use case, representing a level of complexity for general technical staff. Annotated data could be produced by partly relying on the aforementioned complexity metrics and employing a human-in-the-loop for corrections. The complexity scores would be integrated into tables and used together with the classified user profile from fpto select the right items to populate the template. For example, given each complexity score continuously defined between [0, 1] and p in [0, 1] (either discrete or continuous), the rows with complexity <= p would be selected (also based on solution / part, of course).

[0077] Given enough availability of text from reports, as well as template segments, a Natural Language Generation (NLG) approach could be adopted. Given a solution s, a user group or classified user experience score u (evaluated with any of the above), and a template section t (e.g., instructions), an NLG model g could generate the content for the given section with the adequate language complexity as shown in EQUATION 3:

[0078] EQUATION 3 g(u, s, t) = w±, ... ,wn, where u is the vector representing the profile of the user similar to EQUATION I . s is the vector representing the solution to adopt similar to EQUATION 1, t is defined for a given resolution written in a complex form, e.g., corresponding to a highest level of complexity, and Wj , ... , wnis a sequence of generated words composing the template segment. The model, in this case, would work more like a denoiser, simplifying pieces of text of t where needed while removing others considered too complex. Additionally, the generation could consider a retrieval approach from technical documentation, where some template sections are filled with text extracted directly from the technical documentation linked to the solution. Again, the extracted text could be processed by the same model, g, to comply with the complexity required by the user experience stated by u.

[0079] K validation module 838 is employed where the report generation module 828 automatically generates a report based on a probabilistic algorithm without human interaction. In such instance, the validation module 838 determines whether a generated service report is in-line with service guidelines of the imaging system manufacturer 308. In one instance, this includes computing a similarity metric / value based on service standards provided by the imaging system manufacturer 308 and comparing the similarity metric / value with a predetermined threshold. Where the similarity metric / value satisfies the threshold, the report is validated as acceptable and transmitted. However, where the similarity metric / value fails to satisfy the threshold, the report is not validated as acceptable. In one instance, the report generation module 828 receives feedback indicating the report did not satisfy the threshold and re-generates a report based on the dynamic algorithm 834. In another instance, the report generation module 828 receives feedback indicating the report did not satisfy the threshold and generates a report based on the static algorithm 832 and / or the hybrid algorithm 836. In either instance, the maintenance service system 312 provides information regarding the failure to the imaging system manufacturer 308 of FIGS. 3 and / or 4.

[0080] FIG. 9 graphically illustrates an example of generating a report. The report generation module 828 receives / retrieves a user profde 902 (e.g., “Biomed”) from the user profdes 814 and a solution 904 (“Fix keyboard part: keyboard xyz) from the complexity assertion module 818. Based on this information, the report generation module 828 selects, from the data repository 802, a template 906 (“External device fix) linked to a user profile 902 and the solution 904, a segment 908 (including a section) linked to a user profile 902 and the solution 904, a document 910 (including a section and a paragraph) linked to a user profile 902 and the solution 904, a first image identifier 912 linked to a user profile 902 and the solution 904, and a second image identifier 914 linked to a user profile 902 and the solution 904.

[0081] The report generation module 828 combines the selected information, creating a template report 916. The illustrated template report 916 includes a section 918 of the segment 908 in a segment region 920, a section and a paragraph 922 of the segment 908 in a segment region 924, and a first image 926 corresponding to the first image identifier 912 and a second image 928 corresponding to the second image identifier 914, both in an image region 930. It is to be understood that the report 916 is an example and non-limiting, and in other instances, more, less, and / or different information in included in a report.

[0082] FIG. 10 diagrammatically illustrates an example of a report 1000 for where it is determined the manufacturer should mitigate the malfunction. The report 1000 includes a first field 1002 with instructions for the user of the imaging system who is not personnel of the manufacturer of the imaging system. The report 1000 includes a first field 1002 with instructions for the user. The report 1000 includes a second field 1004 with a picture 1006 of the malfunctioning part and identification of a region 1008 with the part where the malfunction is located. The report 1000 includes a schematic 1010 of the region 1008. By way of non-limiting example, for a malfunctioning a coil, the picture 1006 may include a picture of the coil, the region 1008 specifies the region of the coil where the malfunction occurs, and the schematic 1010 includes a schematic of the circuitry and components in the region.

[0083] In the above, the service report is generated based on the user mitigating the malfunction. In a variation, a solution to be applied is determined and then a user is automatically identified, based on their profile, to perform the repair action. For this, the complexity assertion module 818 and / or the report generation module 828 communicate with the hospital information system (HIS) to identify available personnel, e.g., based on their work schedules, and an experience of the available personnel, e.g., success rates, failure rates, previous experience with a similar solution, etc. With this variation, the personnel initiating the malfunction query may not end up being the personnel mitigating the malfunction. In another variation, a solution may be associated with a complexity score for a given root cause solution. In one instance, the computation of the score is performed statically, e.g., by humans at solution database creation time and / or update time. Such a solution can be considered a “universal complexity value” at least because the complexity score will be the same for each install-base of the database among different customers, regardless of which profde is going to apply the solution. For example, changing the magnet of an MR imaging system will always have a score of ‘very difficult,’ regardless of who is going to change the magnet. However, this value could be used in the selection of hospital staff from the HIS. For example, such a value can be used in the selection of the right profile to fix a problem, as a filter over the personnel profiles (e.g. who is able to perform ‘very difficult’ solutions, or who has performed a ‘very difficult’ solution in the past), as a feature in a probabilistic model for staff selection, etc. In one instance, the universal complexity value is stored in the data repository 802.

[0084] FIG. 11 discloses a computer-implemented method in accordance with an embodiment herein. It is to be appreciated that the ordering of the acts of the method is not limiting. As such, other orderings are contemplated herein. In addition, one or more acts may be omitted, and / or one or more additional acts may be included. At 1102, a root-cause signal is received by the maintenance service system 312, as described herein and / or otherwise. At 1104, a solution is determined for the root-cause, as described herein and / or otherwise. At 1106, a strategy is determined based on the root-cause, the solution, and a user mitigating the malfunction, as described herein and / or otherwise. At 1108, a report is generated for the strategy based on a profde of the user and one or more of a text template, a template segment, a part list, images, and technical documents, as described herein and / or otherwise. The report is transmitted to the user mitigating the malfunction, as described herein and / or otherwise. Where the report is generated based on Al, the report is validated before being transmitted to the user mitigating the malfunction, as described herein and / or otherwise. After the service is performed, at least the user profde of the user mitigating the malfunction is updated with the outcome of the service, e.g., successful or unsuccessful.

[0085] FIG. 12 discloses another computer-implemented method in accordance with an embodiment herein. It is to be appreciated that the ordering of the acts of the method is not limiting. As such, other orderings are contemplated herein. In addition, one or more acts may be omitted, and / or one or more additional acts may be included. At 1202, a malfunction alert signal is received by the maintenance service system 312, as described herein and / or otherwise. At 1204, a solution is determined for the malfunction, as described herein and / or otherwise. At 1206, a strategy is determined based on the malfunction, the solution, and a user mitigating the malfunction, as described herein and / or otherwise. At 1208, a report is generated for the strategy based on a profile of the user and one or more of a text template, a template segment, a part list, images, and technical documents, as described herein and / or otherwise. The report is transmitted to the user mitigating the malfunction, as described herein and / or otherwise. Where the report is generated based on Al, the report is validated before being transmitted to the user mitigating the malfunction, as described herein and / or otherwise. After the service is performed, at least the user profile of the user mitigating the malfunction is updated with the outcome of the service, e.g., successful or unsuccessful. The above methods can be implemented by way of computer readable instructions, encoded, or embedded on the computer readable storage medium, which, when executed by a computer processor, cause the processor to carry out the described acts or functions. Additionally, or alternatively, at least one of the computer readable instructions is carried out by a signal, carrier wave or other transitory medium, which is not computer readable storage medium.

[0086] While the invention has been illustrated and described in detail in the drawings and foregoing description, such illustration and description are to be considered illustrative or exemplary and not restrictive; the invention is not limited to the disclosed embodiments. Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims.

[0087] The word “comprising” does not exclude other elements or steps, and the indefinite article “a” or “an” does not exclude a plurality. A single processor or other unit may fulfill the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. A computer program may be stored / distributed on a suitable medium, such as an optical storage medium or a solid-state medium supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the Internet or other wired or wireless telecommunication systems. Any reference signs in the claims should not be construed as limiting the scope.

Claims

CLAIMSClaim 1. A system, comprising: an imaging system; a malfunction module configured to generate a malfunction signal indicating a malfunction of the imaging system; and a maintenance service system, including: a data repository configured to store malfunction-solution pairs, text templates, and user profiles including repair experience of levels of user; a solutions module configured to determine a solution to the malfunction from the malfunction-solution pairs in response to receiving the malfunction signal; a complexity assertion module configured to determine a strategy for mitigating the malfunction based on the solution and a user profde of a user mitigating the malfunction; and a report generation module configured to automatically create a service report for the user based on the strategy, the user profile, the solution, and information from the text templates, wherein the maintenance service system is configured to transmit the service report to a client device of a user mitigating the malfunction.Claim 2. The system of claim 1, wherein the user profile further includes an indication of the information the user is permitted to access, and the report generation module is configured to automatically create the service report further based on the information the user is permitted to access.Claim 3. The system of claim 1, wherein the data repository is further configured to store at least one of text segments, a parts list and technical documents, and the report generation module is configured to automatically create the service report based further on one or more of the text segments, a parts list, and technical documents.Claim 4. The system of claim 1, wherein the malfunction module is a cloud-based resource.Claim 5. The system of claim 1, wherein the malfunction module includes a user-detected malfunction module configured to detect the malfunction based on a user input.Claim 6. The system of claim 1, wherein the complexity assertion module is configured to determine the strategy based on a static mapping between the solution, the user experience, and the strategy.Claim 7. The system of claim 1, wherein the complexity assertion module is configured to determine the strategy based on a dynamic module.Claim 8. The system of claim 1, wherein the report generation module is configured to create the service report based on a static algorithm.Claim 9. The system of claim 1, wherein the report generation module is configured to create the service report based on a dynamic algorithm.Claim 10. A computer-implemented method, comprising: receiving, from an imaging system, a malfunction signal indicating a malfunction with the imaging system; determining a solution for the received malfunction signal; determining a strategy for the solution based on the malfunction and a user profde of a user mitigating the malfunction; generating a service report with an action for mitigating the malfunction based on the strategy, the user profde, and service information; and transmitting the service report to the user, wherein the user performs the action in the service report.Claim 11. The computer-implemented method of claim 10, wherein the malfunction signal includes a malfunction alert generated by hardware, software, or hardware and software monitoring the imaging system for malfunctions without user interaction.Claim 12. The computer-implemented method of claim 10, wherein the malfunction signal includes a root-cause alert generated in response to a user initiate malfunction query at the imaging system.Claim 13. The computer-implemented method of claim 10, further comprising: retrieving a template and automatically populating the template with the service information.Claim 14. The computer-implemented method of claim 13, wherein the service information includes one or more of a parts list, images of parts of the imaging system, and technical documents.Claim 15. The computer-implemented method of claim 10, where the strategy is based on a probabilistic model: fa(u,s) = g, g & g, where u is a vector representing a profile of a user, s is a vector representing a solution to adopt, g is a generation strategy to follow, and g is a collection of available strategies.Claim 16. The computer-implemented method of claim 10, where the information in the report is retrieved based on a dynamic model: fp(u) = p, p E?, where u is a vector representing a profde of a user, p is a user in a set J3of user group labels.Claim 17. The computer-implemented method of claim 10, where the information in the report is determined based on a dynamic model: g u,s, t) = w , ... , wn, where u is a vector representing a profde of a user, s is a vector representing a solution to adopt, t corresponds to a highest level of complexity, and W] , ... , wnis a sequence of generated words composing a template segment.Claim 18. A computer readable medium encoded with computer executable instructions, which, when executed by a processor, causes the processor to: receive, from an imaging system, a malfunction signal indicating a malfunction with the imaging system; determine a solution for the received malfunction signal; determine a strategy for the solution based on the malfunction; identify a user from a user profde to mitigate the malfunction; generate service report with an action for mitigating the malfunction based on the strategy, the user profde of the user, and one or more of a parts list, images of parts of the image system, and technical documents; and transmit the service report to the user, wherein the user performs the action in the service report.Claim 19. The computer readable medium of claim 18, wherein the instructions further cause the processor to update the user profile with an outcome of the action after the action in the service report is performed.Claim 20. The computer readable medium of claim 19, wherein the instructions further cause the processor to update at least one of the parts list, the images of parts of the image system, and the technical documents to add new information or remove existing information.

Images