Electronic acquisition of fault diagnostic knowledge for medical imaging scanners

Abstract

Systems or techniques are provided that facilitate electronic acquisition of fault diagnostic knowledge for a medical imaging scanner. In various embodiments, the system may access a common textual maintenance manual (e.g., 202) and a maintenance complaint (e.g., 106) associated with a medical imaging scanner (e.g., 104). In various aspects, the system may identify a semantic hierarchy of a normal text maintenance manual via named entity identification or natural language processing (e.g., 302). In various cases, the system may record, via graphical user interface tracking, how a service technician (e.g., 602) who is diagnosing a service complaint in fault navigates in sequence in a semantic hierarchy, thereby generating a fault diagnosis trace (e.g., 504) corresponding to the service complaint.

Description

Technical Field

[0001] The present disclosure relates generally to medical imaging scanners and, more particularly, to electronic acquisition of troubleshooting knowledge for medical imaging scanners. Background Art

[0002] Medical imaging scanners may be deployed in the field. During deployment, medical imaging scanners may malfunction. A service manual associated with the medical imaging scanner may contain information for troubleshooting the malfunction. However, the service manual may be voluminous, so locating such information in the service manual may not be easy. Existing techniques for locating such information require the user of the medical imaging scanner to manually look through the service manual with the help of only an electronic keyword search. Unfortunately, such prior art is not helpful.

[0003] Therefore, systems or techniques that can address one or more of these technical issues may be desirable. Summary of the invention

[0004] The following is a summary of the invention to provide a basic understanding of one or more embodiments. The summary of the invention is not intended to identify key or important elements, nor is it intended to delineate any scope of a specific embodiment or any scope of the claims. Its sole purpose is to present concepts in a simplified form as a prelude to a more detailed description presented later. In one or more embodiments described herein, an apparatus, system, computer-implemented method, device, or computer program product that facilitates electronic acquisition of troubleshooting knowledge for medical imaging scanners is described.

[0005] According to one or more embodiments, a system is provided. The system may include a non-transitory computer readable memory that may store a computer executable component. The system may also include a processor that may be operably coupled to the non-transitory computer readable memory and may execute the computer executable component stored in the non-transitory computer readable memory. In various embodiments, the computer executable component may include an access component that may access a plain text maintenance manual and a maintenance complaint associated with a medical imaging scanner. In various aspects, the computer executable component may include a parsing component that may identify a semantic hierarchy of the plain text maintenance manual via named entity recognition or natural language processing. In various cases, the computer executable component may include a tracking component that may record how a maintenance technician who is troubleshooting a maintenance complaint navigates sequentially in the semantic hierarchy via a graphical user interface tracking, thereby generating a fault diagnosis trace corresponding to the maintenance complaint. In various cases, the computer executable component may include a model component that may train a deep learning neural network for maintenance complaints and fault diagnosis traces, wherein the maintenance complaint may be considered as a training input to the deep learning neural network, and wherein the fault diagnosis trace is considered as a baseline truth annotation corresponding to the maintenance complaint.

[0006] According to one or more embodiments, a computer-implemented method is provided. In various embodiments, the computer-implemented method may include accessing, by a device operably coupled to a processor, a plain text maintenance manual and a maintenance complaint associated with a medical imaging scanner. In various aspects, the computer-implemented method may include identifying, by the device and via named entity recognition or natural language processing, a semantic hierarchy of the plain text maintenance manual. In various cases, the computer-implemented method may include recording, by the device and via a graphical user interface tracking, how a maintenance technician who is troubleshooting a maintenance complaint navigates sequentially in the semantic hierarchy, thereby generating a fault diagnosis trace corresponding to the maintenance complaint. In various cases, the computer-implemented method may include training, by the device, a deep learning neural network for maintenance complaints and fault diagnosis traces, wherein the maintenance complaints may be considered as training inputs to the deep learning neural network, and wherein the fault diagnosis traces may be considered as baseline truth annotations corresponding to the maintenance complaints.

[0007] According to one or more embodiments, a computer program product for facilitating the electronic collection of fault diagnosis knowledge is provided. In various embodiments, the computer program product may include a non-transitory computer-readable memory having program instructions embodied therein. In various aspects, the program instructions may be executable by a processor to enable the processor to access one or more ordinary text documents associated with a machine. In various cases, these program instructions may also be executable to enable the processor to identify the semantic hierarchy of one or more ordinary text documents via named entity recognition or natural language processing. In various cases, these program instructions may also be executable to enable the processor to track via a graphical user interface how a maintenance technician who is troubleshooting a maintenance complaint associated with a machine navigates sequentially in the semantic hierarchy, thereby generating a fault diagnosis trace corresponding to the maintenance complaint. In various aspects, these program instructions may also be executable to enable the processor to train a deep learning neural network for maintenance complaints and fault diagnosis traces, wherein the maintenance complaints are training inputs, and wherein the fault diagnosis traces are baseline truth annotations. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 A block diagram is shown of an exemplary, non-limiting system that facilitates electronic acquisition of troubleshooting knowledge for medical imaging scanners in accordance with one or more embodiments described herein.

[0009] Figure 2 A block diagram is shown of an exemplary, non-limiting system including a plain text maintenance manual that facilitates electronic acquisition of troubleshooting knowledge for a medical imaging scanner in accordance with one or more embodiments described herein.

[0010] Figure 3 A block diagram is shown of an exemplary, non-limiting system including a semantic hierarchy that facilitates electronic acquisition of troubleshooting knowledge for medical imaging scanners in accordance with one or more embodiments described herein.

[0011] Figure 4 An exemplary, non-limiting block diagram illustrating how a semantic hierarchy may be generated is presented in accordance with one or more embodiments described herein.

[0012] Figure 5 A block diagram is shown of an exemplary, non-limiting system including a graphical user interface and troubleshooting traceability that facilitates electronic acquisition of troubleshooting knowledge for medical imaging scanners in accordance with one or more embodiments described herein.

[0013] Figure 6 An exemplary, non-limiting block diagram illustrating how a fault diagnostic trace may be generated is presented in accordance with one or more embodiments described herein.

[0014] Figure 7 A block diagram is shown of an exemplary, non-limiting system including a deep learning neural network that facilitates electronic acquisition of troubleshooting knowledge for medical imaging scanners according to one or more embodiments described herein.

[0015] Figure 8 An exemplary, non-limiting block diagram illustrating how a deep learning neural network may be trained for fault diagnostic tracing is presented in accordance with one or more embodiments described herein.

[0016] Fig. 9 A block diagram is shown of an exemplary, non-limiting system including operational reporting that facilitates electronic acquisition of troubleshooting knowledge for a medical imaging scanner in accordance with one or more embodiments described herein.

[0017] Fig.10 A flow chart is presented of an exemplary, non-limiting computer-implemented method that facilitates electronic acquisition of troubleshooting knowledge for medical imaging scanners in accordance with one or more embodiments described herein.

[0018] Fig.11 Block diagram showing an exemplary, non-limiting operating environment in which one or more embodiments described herein may be facilitated.

[0019] Fig.12 An exemplary networking environment is presented that is operable to perform various implementations described herein. DETAILED DESCRIPTION

[0020] The following detailed description is merely illustrative and is not intended to limit the embodiments or the application / use of the embodiments. In addition, it is not intended to be bound by any express or implied information set forth in the aforementioned "background technology" or "invention content" section or "detailed description" section.

[0021] One or more embodiments are now described with reference to the accompanying drawings, wherein the same reference numerals are used to represent the same elements throughout. In the following description, for the purpose of explanation, many specific details are set forth in order to provide a more thorough understanding of one or more embodiments. However, it is apparent that in various cases, one or more embodiments may be practiced without these specific details.

[0022] Medical imaging scanners (e.g., computed tomography (CT) scanners, magnetic resonance imaging (MRI) scanners, X-ray scanners, ultrasound scanners, positron emission tomography (PET) scanners) can be deployed in the field. Thus, the medical imaging scanners can capture or generate real-world medical scan images (e.g., CT scan images, MRI scan images, X-ray scan images, ultrasound scan images, PET scan images) of real-world medical patients (e.g., humans, animals, or other objects).

[0023] During deployment, the medical imaging scanner may malfunction. For example, the medical imaging scanner may experience a hardware failure or a software failure that hinders or prevents the medical imaging scanner from correctly capturing or generating medical scan images.

[0024] A maintenance manual (e.g., an electronic copy of a technical manual) associated with a medical imaging scanner may contain information for troubleshooting a fault. However, the maintenance manual may be voluminous (e.g., containing dozens or hundreds of pages of plain text). Therefore, locating such information in the maintenance manual may not be easy or difficult.

[0025] Prior art techniques for locating such information require users of medical imaging scanners to manually peruse the service manual with the aid of only electronic keyword searches. Specifically, such prior art techniques involve users selecting various keywords that they believe describe the symptoms presented by the medical imaging scanner, and electronically searching (e.g., via a FIND operation) within the service manual for these various keywords. Unfortunately, such prior art techniques are not helpful. In practice, users tend to select keywords that are not used in the service manual, or that do not accurately or appropriately describe the symptoms of the fault. Furthermore, many different types of faults may manifest themselves via common or shared symptoms. Therefore, even if a user happens to select keywords that accurately describe the symptoms of the fault and are actually referenced in the service manual, electronically searching for such keywords may also produce many potentially irrelevant results for the user to review.

[0026] Therefore, systems or techniques that can address one or more of these technical issues may be desirable.

[0027] Various embodiments described herein can solve one or more of these technical problems. One or more embodiments described herein can include a system, a computer-implemented method, an apparatus, or a computer program product that can facilitate the electronic collection of fault diagnosis knowledge for medical imaging scanners. In particular, the inventors of the various embodiments described herein recognize that the maintenance manual of the medical imaging scanner itself does not represent a complete or complete field of information that is useful or helpful for solving the fault. In fact, the inventors recognize that quite rich information about how to solve the fault can be presented not only in the explicit wording written in the maintenance manual, but also in the order in which a qualified maintenance technician navigates in the maintenance manual when diagnosing the fault. That is, not only the written text of the maintenance manual can be considered useful for solving the fault or can provide useful information for solving the fault; the order in which a qualified maintenance technician reads or consults such written text in such written text can also be considered useful for solving the fault or can provide useful information for solving the fault. Therefore, a qualified maintenance technician can diagnose the fault by referring to an electronically displayed copy of the maintenance manual, and the various embodiments described herein can involve electronically tracking which specific parts of the maintenance manual were visited by a qualified maintenance technician in what specific order when solving the fault. This tracking can be thought of as capturing the implicit experiential knowledge of qualified repair technicians that helps solve faults but is not explicitly written down in the repair manual. When expanded across a myriad of faults, such knowledge can be exploited or aggregated through deep learning so that, given any specific fault in a medical imaging scanner, the appropriate reading or navigation path in the repair manual can be electronically predicted or inferred.

[0028] Therefore, the various embodiments described herein may be considered to improve the manner in which a service manual can be utilized to resolve a malfunction experienced by a medical imaging scanner.

[0029] The various embodiments described herein may be considered to be computerized tools (e.g., any suitable combination of computer executable hardware or computer executable software) capable of facilitating the electronic collection of troubleshooting knowledge for medical imaging scanners. In various aspects, such computerized tools may include access components, parsing components, tracking components, or model components.

[0030] In various embodiments, there can be a medical imaging scanner. In various aspects, the medical imaging scanner can be any suitable type of medical image capture equipment or modality (e.g., a CT scanner, an MRI scanner, an X-ray scanner, an ultrasound scanner, or a PET scanner, etc.). In various cases, the medical imaging scanner can be deployed in any suitable clinical or operating environment (e.g., in a hospital, in a veterinary clinic, on an emergency vehicle). In various cases, the medical imaging scanner can include any suitable human-machine interface device (e.g., a keyboard, a keypad, a touch screen, a voice command system).

[0031] In various embodiments, a medical imaging scanner may be associated with a maintenance complaint. In various aspects, a maintenance complaint may be any suitable electronic file that textually describes or otherwise indicates one or more hardware-based or software-based failure symptoms exhibited or displayed by the medical imaging scanner. In various cases, a maintenance complaint may be electronically prepared, written, or generated by or at the request of a user of the medical imaging scanner (e.g., the user may do so via a human interface device of the medical imaging scanner).

[0032] In various embodiments, an access component of a computerized tool can electronically access a medical imaging scanner or a maintenance complaint. For example, the access component can electronically communicate with a medical imaging scanner (e.g., send data to a medical imaging scanner, receive data from a medical imaging scanner). As another example, the access component can electronically retrieve or obtain a maintenance complaint from any suitable centralized or decentralized data structure (e.g., a graph data structure, a relational data structure, a hybrid data structure), whether remote from the access component or local to the access component. In any case, the access component can electronically access the medical imaging scanner or the maintenance complaint, allowing other components of the computerized tool to electronically interact with the medical imaging scanner or with the maintenance complaint (e.g., read, write, edit, copy, manipulate).

[0033] In various embodiments, the access component can also be capable of electronically accessing a plain text service manual associated with the medical imaging scanner from any suitable source. In various aspects, the plain text service manual can be an electronic document written in unstructured text (hence the term "plain text") that describes various technical or operational features, characteristics, or specifications about the medical imaging scanner. In various cases, the plain text service manual can be considered to be an electronic factory manual provided by the manufacturer of the medical imaging scanner and explaining in text form how to operate, service, maintain, or repair the medical imaging scanner. In various cases, the plain text service manual can be voluminous (e.g., can be hundreds of pages long).

[0034] In various embodiments, the parsing component of the computerized tool can generate a semantic hierarchy electronically based on a common text maintenance manual. In various aspects, the parsing component can complete this generation operation by applying any suitable natural language processing technology (e.g., named entity recognition) to a common text maintenance manual. In various cases, the semantic hierarchy can be a knowledge graph (e.g., including the relationship between nodes and nodes) representing the semantic structure or organization of a common text maintenance manual (e.g., the semantic hierarchy can be considered as a class of document object model (DOM) trees of a common text maintenance manual). Therefore, the semantic hierarchy can include a plurality of semantic nodes, each of which can be interconnected via a variety of semantic or organizational relationships (e.g., partial relationships, nested relationships). In various cases, a plurality of semantic nodes can correspond to any semantic concept (e.g., document title, document chapter, document keyword) recorded in a textual manner or otherwise present in a common text maintenance manual. In various cases, each of a plurality of semantic nodes can be marked with a corresponding position indicator that specifies where the semantic node is located in a common text maintenance manual (e.g., on which page, in which paragraph).

[0035] In various embodiments, the tracking component of the computerized tool can generate fault diagnosis tracking electronically based on the maintenance complaint and based on the semantic hierarchy. More specifically, the tracking component can include any suitable graphical user interface, or otherwise be integrated with any suitable graphical user interface electronics. In various aspects, the graphical user interface can include any suitable electronic display (e.g., computer screen, computer monitor), and any suitable human-machine interface device (e.g., keyboard, keypad, touch screen). In various cases, the tracking component can electronically present a plain text repair manual on the graphical user interface. In various cases, a maintenance technician who is qualified or certified to repair a medical imaging scanner can attempt to resolve the maintenance complaint by troubleshooting the medical imaging scanner (e.g., remotely or in person). In various aspects, the maintenance technician can interact with the graphical user interface of the tracking component during such troubleshooting to read through the plain text repair manual when searching for solutions to the maintenance complaint. In various cases, the tracking component can electronically record which semantic nodes of the plain text repair manual the maintenance technician accesses in what order during such troubleshooting. In various cases, the tracking component can facilitate such recording by implementing any suitable graphical user interface tracking technology (e.g., tracking which parts of the plain text service manual the service technician clicked on or scrolled to). In various aspects, such recorded data can be referred to as a fault diagnosis trace. In other words, the fault diagnosis trace can be considered as a directed path through the semantic hierarchy of the plain text service manual, which directed path indicates the order in which the service technician navigated or read in the plain text service manual when searching for a solution to the service complaint. In other words, the fault diagnosis trace can be considered as indicating which specific parts of the plain text service manual the service technician consulted in what specific order in order to resolve the service complaint.

[0036] In various embodiments, the model component of the computerized tool is capable of electronically training a deep learning neural network for maintenance complaint and fault diagnostic tracking.

[0037] More specifically, the model component can electronically store, maintain, control or access the deep learning neural network. In various aspects, the deep learning neural network can exhibit any suitable internal architecture. For example, the deep learning neural network may include any suitable number of layers of any suitable type (e.g., input layer, one or more hidden layers, output layer, any of which may be a convolutional layer, a dense layer, a nonlinear layer, a pooling layer, a batch normalization layer or a padding layer). For another example, the deep learning neural network may include any suitable number of neurons in various layers (e.g., different layers may have the same or different numbers of neurons). For another example, the deep learning neural network may include any suitable activation function (e.g., softmax, sigmoid, hyperbolic tangent, rectified linear unit) in various neurons (e.g., different neurons may have the same or different activation functions). For another example, the deep learning neural network may include any suitable inter-neuron connection or inter-layer connection (e.g., forward connection, skip connection, recursive connection).

[0038] In various aspects, a model component can train a deep learning neural network based on maintenance complaints and based on fault diagnosis traces. More specifically, the model component can treat maintenance complaints as training inputs for the deep learning neural network, and the model component can treat fault diagnosis traces as ground truth annotations that are known or believed to correspond to maintenance complaints.

[0039] In various cases, if the deep learning neural network has not undergone any training, the model component can randomly initialize the trainable internal parameters of the deep learning neural network (e.g., convolution kernels, weight matrices, bias vectors). In contrast, if the deep learning neural network has undergone at least some training, the model component can avoid reinitializing the trainable internal parameters of the deep learning neural network.

[0040] In various aspects, the model component can execute a deep learning neural network on the maintenance complaint, so that the deep learning neural network produces some output. Specifically, the model component can feed the maintenance complaint to an input layer of the deep learning neural network, which can complete a forward propagation through one or more hidden layers of the deep learning neural network, and such forward propagation can cause an output layer of the deep learning neural network to calculate an output based on the activations provided by the one or more hidden layers.

[0041] Note that the format, size, or dimensionality of the output may be controlled or determined by the number, arrangement, or size of neurons contained in or otherwise constituting the output layer of the deep learning neural network, or the number, arrangement, or size of other internal parameters (e.g., convolution kernels). Thus, the output may be forced to have any suitable or desired format, size, or dimensionality by adding neurons or other internal parameters to the output layer of the deep learning neural network, removing neurons or other internal parameters from the output layer of the deep learning neural network, or otherwise adjusting neurons or other internal parameters within the output layer of the deep learning neural network. Thus, the output may be forced to have the same, comparable, or commensurate format, size, or dimensionality as the fault diagnostic trace. Thus, the output may be considered to be a predicted or inferred fault diagnostic trace (e.g., a predicted or inferred reading order in a normal text repair manual), which the deep learning neural network believes should correspond to a repair complaint (e.g., believes that it will resolve or address the repair complaint). In contrast, a fault diagnostic trace recorded by a trace component may be a correct or accurate reading sequence through an ordinary text repair manual that is known or believed to resolve or handle the service complaint (e.g., after all, the fault diagnostic trace originated from a service technician who actually troubleshooted the service complaint). In various cases, if the deep learning neural network has not been trained at all or has only been trained very little to date, the output may be highly inaccurate (e.g., may differ significantly from the fault diagnostic trace).

[0042] In any case, the model component can calculate the error or loss (e.g., mean absolute error (MAE), mean square error (MSE), cross entropy error) between the output and the fault diagnosis trace. In various aspects, the model component can update the trainable internal parameters of the deep learning neural network by performing back propagation (e.g., stochastic gradient descent) driven by the calculated error or loss.

[0043] In various aspects, such a training process can be repeated for any suitable number of maintenance complaints and fault diagnosis trace pairs. Such training can ultimately cause the trainable internal parameters of the deep learning neural network to become iteratively optimized so as to accurately infer fault diagnosis traces based on the input maintenance complaints. It should be noted that the model component can implement any suitable training batch size, any suitable training termination criteria, or any suitable error, loss or objective function.

[0044] In various cases, after such training, the model component is able to electronically deploy the deep learning neural network as a third-party service. As a non-limiting example, a third party that owns or operates an instantiated medical imaging scanner can provide or create a given maintenance complaint, and the model component can execute the deep learning neural network for the given maintenance complaint. Such execution can cause the deep learning neural network to generate a predicted fault diagnosis trace, wherein such a predicted fault diagnosis trace can be considered to indicate a reading or navigation path in a plain text maintenance manual (in the eyes of the deep learning neural network) that will resolve or handle the given maintenance complaint. Therefore, the third party does not need to carefully review the plain text maintenance manual without a goal. Instead, the maintenance technician's implicit fault diagnosis knowledge may have been used to train the deep learning neural network, so that the deep learning neural network can infer or predict which parts of the plain text maintenance manual should be consulted in what order in order to perform a fault diagnosis on a given maintenance complaint.

[0045] Various embodiments described herein may be employed to solve problems that are highly technical in nature (e.g., to facilitate electronic acquisition of troubleshooting knowledge for medical imaging scanners) using hardware or software, problems that are not abstract and cannot be performed by humans as a set of mental behaviors. In addition, some of the processes performed may be performed by a dedicated computer (e.g., a graphical user interface, a text parser, a deep learning neural network) to implement defined actions associated with a medical imaging scanner. For example, such defined actions may include: accessing, by a device operably coupled to a processor, a plain text maintenance manual and maintenance complaints associated with a medical imaging scanner; identifying, by the device and via named entity recognition or natural language processing, a semantic hierarchy of the plain text maintenance manual; and recording, by the device and via a graphical user interface tracking, how a maintenance technician who is troubleshooting a maintenance complaint sequentially navigates through the semantic hierarchy to generate a troubleshooting trace corresponding to the maintenance complaint. In addition, in various cases, such defined actions may include: training, by the device, a deep learning neural network for the maintenance complaint and the troubleshooting trace, wherein the maintenance complaint is considered as a training input to the deep learning neural network, and wherein the troubleshooting trace is considered as a baseline truth annotation corresponding to the maintenance complaint.

[0046] Such defined actions are not manually performed by a human. In fact, neither the human mind nor a human with pen and paper can electronically parse a plain text maintenance manual for a medical imaging scanner (e.g., a CT scanner, an X-ray scanner) into a semantic hierarchy (e.g., a knowledge graph), electronically record (e.g., via click tracking or scroll tracking) which nodes of the semantic hierarchy were visually displayed on a graphical user interface at which moments during troubleshooting of the medical imaging scanner, and electronically train a deep learning neural network on such recorded data. In fact, the medical imaging scanner, the graphical user interface, and the deep learning neural network are inherently computerized, hardware- and software-based constructs that cannot be meaningfully implemented or trained by the human mind in any way without a computer. A computerized tool that can electronically track the order of reading or navigating in a plain text maintenance manual for a medical imaging scanner and use such order of reading or navigating to train a deep learning neural network to accurately predict the order of reading or navigating is also inherently computerized and cannot be implemented in any reasonable, practical, or appropriate manner without a computer.

[0047] In addition, various embodiments described herein can integrate various teachings related to the electronic collection of fault diagnosis knowledge for medical imaging scanners into practical applications. As described above, when a medical imaging scanner fails, the prior art involves the user or owner of the medical imaging scanner aimlessly reading through the maintenance manual of the medical imaging scanner to search for a solution to the failure. Such aimless reading is not only time-consuming, but also generally ineffective (e.g., the user or owner is likely to carefully review irrelevant parts of the maintenance manual, even with the assistance of electronic search keywords).

[0048] Various embodiments described herein can solve one or more of these technical problems. In particular, the inventors recognize that different types of rich information that are helpful or useful for solving a fault in a medical imaging scanner are not explicitly written in the maintenance manual itself. Instead, as the inventors recognize, this different type of rich information is actually the order in which an experienced, qualified or certified maintenance technician reads or navigates in the maintenance manual when diagnosing a fault. As described herein, various embodiments may involve recording such reading or navigation order via a graphical user interface tracking (e.g., tracking which parts of the maintenance manual the maintenance technician clicks on, tracking which parts of the maintenance manual the maintenance technician scrolls to). In various aspects, such reading or navigation order can then be used to train a deep learning neural network. Therefore, once trained, the deep learning neural network can respond to any given fault of the medical imaging scanner by predicting or inferring which specific parts of the maintenance manual should be consulted in what specific order in order to solve the given fault. In this way, the user or owner of the medical imaging scanner does not need to read the maintenance manual aimlessly or wait for an experienced, qualified or certified maintenance technician. Instead, a user or owner can troubleshoot a given fault themselves by relying on a deep learning neural network to indicate which parts of the repair manual should be consulted in what order (e.g., the user or owner does not need to waste time aimlessly reading through irrelevant parts of the repair manual). For at least these reasons, the various embodiments described herein are considered to be improved technologies for automatically identifying relevant troubleshooting information in a repair manual for a medical imaging scanner. Therefore, the various embodiments described herein certainly constitute tangible and concrete technical improvements or technical advantages in the field of medical imaging scanners (e.g., the prior art cannot automatically and effectively identify which parts of a plain text repair manual for a medical imaging scanner with a large amount of information are relevant to troubleshooting a medical imaging scanner). Therefore, such embodiments clearly qualify as useful and practical applications of computers.

[0049] In addition, various embodiments described herein can control real-world tangible devices based on the disclosed teachings. For example, various embodiments described herein can electronically control real-world graphical user interfaces and can electronically train or execute real-world deep learning neural networks.

[0050] It should be understood that the drawings and description herein provide non-limiting examples of various embodiments and are not necessarily drawn to scale.

[0051] Figure 1A block diagram of an exemplary, non-limiting system 100 is shown that can facilitate electronic collection of fault diagnosis knowledge for a medical imaging scanner according to one or more embodiments described herein. As shown, a fault diagnosis knowledge system 102 can be electronically integrated with a medical imaging scanner 104 or a maintenance complaint 106 via any suitable wired or wireless electronic connection.

[0052] In various embodiments, the medical imaging scanner 104 can be any suitable device, equipment or modality for capturing or generating medical images. As a non-limiting example, the medical imaging scanner 104 can be a CT scanner capable of capturing or generating a CT scan pixel array or voxel array. As another non-limiting example, the medical imaging scanner 104 can be an MRI scanner capable of capturing or generating an MRI scan pixel array or voxel array. As even another non-limiting example, the medical imaging scanner 104 can be an X-ray scanner capable of capturing or generating an X-ray scan pixel array or voxel array. As yet another non-limiting example, the medical imaging scanner 104 can be an ultrasound scanner capable of capturing or generating an ultrasound scan pixel array or voxel array. As yet another non-limiting example, the medical imaging scanner 104 can be a PET scanner capable of capturing or generating a PET scan pixel array or voxel array.

[0053] In various aspects, the medical imaging scanner 104 can be deployed, placed or otherwise implemented in any suitable clinical operating environment. As a non-limiting example, the medical imaging scanner 104 can be deployed, placed or otherwise implemented in any suitable hospital or medical center. As another non-limiting example, the medical imaging scanner 104 can be deployed, placed or otherwise implemented in any suitable scientific or medical laboratory. As another non-limiting example, the medical imaging scanner 104 can be deployed, placed or otherwise implemented in any suitable veterinary center. As even another non-limiting example, the medical imaging scanner 104 can be deployed, placed or otherwise implemented in or on any suitable means of transportation (e.g., ambulance, cruise ship, airplane).

[0054] In various cases, the medical imaging scanner 104 may include any suitable human-machine interface device through which a user or operator of the medical imaging scanner 104 can manually interact with or control the medical imaging scanner 104. As a non-limiting example, the medical imaging scanner 104 may include any suitable keyboard or keypad that can be pressed by a user or operator. As another non-limiting example, the medical imaging scanner 104 may include any suitable computer mouse that can be dragged or clicked by a user or operator. As even another non-limiting example, the medical imaging scanner 104 may include any suitable touch screen that can be manipulated by a user or operator in a tactile manner. As yet another non-limiting example, the medical imaging scanner 104 may include any suitable voice command system that can accept verbal instructions spoken by a user or operator.

[0055] In various cases, the maintenance complaint 106 may be any suitable electronic data (e.g., one or more scalars, one or more vectors, one or more matrices, one or more tensors, one or more strings, or any suitable combination of these electronic data) that indicates, specifies, or otherwise communicates one or more failure symptoms that the medical imaging scanner 104 has suffered or is experiencing. As a non-limiting example, the maintenance complaint 106 may be a plain text or structured text electronic file that describes such one or more failure symptoms. In various aspects, a user or operator of the medical imaging scanner 104 may generate the maintenance complaint 106, or otherwise cause the maintenance complaint to be created, by interacting with a human-machine interface device of the medical imaging scanner 104 (e.g., by typing the maintenance complaint 106 into a keyboard or touch screen of the medical imaging scanner 104, by speaking the maintenance complaint 106 to a voice command system of the medical imaging scanner 104). However, this is merely a non-limiting example. In other cases, a user or operator of the medical imaging scanner 104 may generate the maintenance complaint 106, or otherwise cause the maintenance complaint to be created, by interacting with any other suitable computing device.

[0056] In various embodiments, the fault diagnosis knowledge system 102 may include a processor 108 (e.g., a computer processing unit, a microprocessor), and a non-transitory computer-readable memory 110 that is operably or operatively or communicatively connected or coupled to the processor 108. The non-transitory computer-readable memory 110 may store computer-executable instructions that, when executed by the processor 108, may cause the processor 108 or other components of the fault diagnosis knowledge system 102 (e.g., an access component 112, a parsing component 114, a tracking component 116, a model component 118) to perform one or more actions. In various embodiments, the non-transitory computer-readable memory 110 may store computer-executable components (e.g., an access component 112, a parsing component 114, a tracking component 116, a model component 118), and the processor 108 may execute these computer-executable components.

[0057] In various embodiments, the fault diagnosis knowledge system 102 may include an access component 112. In various aspects, the access component 112 is capable of electronically accessing the medical imaging scanner 104 or the maintenance complaint 106. As a non-limiting example, the access component 112 is capable of electronically communicating with the medical imaging scanner 104 in any suitable manner. That is, the access component 112 is capable of electronically sending any suitable electronic data to the medical imaging scanner 104, and the medical imaging scanner 104 can likewise electronically send any suitable electronic data to the access component 112. As another non-limiting example, the access component 112 is capable of electronically retrieving or electronically obtaining the maintenance complaint 106 from any suitable centralized or decentralized data structure (not shown) or from any suitable centralized or decentralized computing device (not shown). In fact, in some cases, the access component 112 is capable of electronically receiving the maintenance complaint 106 from the medical imaging scanner 104. In any case, the access component 112 can electronically access the medical imaging scanner 104 or the service complaint 106 , allowing other components of the fault diagnostic knowledge system 102 to electronically interact with the medical imaging scanner 104 or the service complaint 106 .

[0058] In various embodiments, the access component 112 may also access a plain text service manual corresponding to the medical imaging scanner 104 , as described herein.

[0059] In various embodiments, the fault diagnosis knowledge system 102 can include a parsing component 114. In various aspects, as described herein, the parsing component 114 can generate a semantic hierarchy based on a normal text repair manual.

[0060] In various embodiments, the fault diagnosis knowledge system 102 may include a tracking component 116. In various cases, as described herein, the tracking component 116 may record the order in which a maintenance technician reads or navigates through a semantic hierarchy, and thus a normal text maintenance manual, while troubleshooting a maintenance complaint. In some examples, this recorded order may be referred to as a fault diagnosis trace.

[0061] In various embodiments, the fault diagnosis knowledge system 102 can include a model component 118. In various cases, as described herein, the model component 118 can train a deep learning neural network on maintenance complaints and fault diagnosis traces.

[0062] Figure 2 A block diagram of an exemplary, non-limiting system 200 including a plain text maintenance manual is shown, which can facilitate electronic collection of troubleshooting knowledge for medical imaging scanners according to one or more embodiments described herein. As shown, in some cases, system 200 can include the same components as system 100, and can also include a plain text maintenance manual 202.

[0063] In various embodiments, access component 112 can electronically access plain text service manual 202 from any suitable electronic source. As a non-limiting example, access component 112 can electronically receive or retrieve plain text service manual 202 from any suitable centralized or distributed computing device or database (not shown).

[0064] In any case, the plain text maintenance manual 202 may correspond to the medical imaging scanner 104 or be otherwise associated with the medical imaging scanner. Specifically, the plain text maintenance manual 202 may be any suitable electronic document written in ordinary unstructured text (e.g., written in complete human-readable sentences) and may describe or explain how to technically operate the medical imaging scanner 104, how to technically repair the medical imaging scanner 104, how to technically maintain the medical imaging scanner 104, or how to technically repair the medical imaging scanner 104. Therefore, the plain text maintenance manual 202 can be considered as a manufacturer's manual that conveys how to use and maintain the medical imaging scanner 104. In various aspects, the plain text maintenance manual 202 can be in any suitable electronic format. As a non-limiting example, the plain text maintenance manual 202 can be in a portable document format (PDF). In various cases, the plain text maintenance manual 202 can be in any suitable size or length. As a non-limiting example, the plain text maintenance manual 202 may be as long as dozens of electronic pages or hundreds of electronic pages. In various cases, plain text service manual 202 may be written in any suitable language, such as English, French, or Spanish.

[0065] In various aspects, the plain-text service manual 202 may contain or otherwise describe information that will be helpful or useful in resolving or addressing the service complaint 106. However, it may not be initially clear or known to (e.g., a user or operator of the medical imaging scanner 104) where such information is located within the plain-text service manual 202. After all, it is possible that any words documented in the service complaint 106 may not be documented verbatim in the plain-text service manual 202, or may instead be documented in multiple, potentially unrelated locations within the plain-text service manual 202.

[0066] Figure 3 A block diagram of an exemplary, non-limiting system 300 including a semantic hierarchy is shown that can facilitate electronic collection of troubleshooting knowledge for medical imaging scanners according to one or more embodiments described herein. As shown, in some cases, system 300 can include the same components as system 200, and can also include a semantic hierarchy 302.

[0067] In various embodiments, the parsing component 114 can electronically generate a semantic hierarchy 302 based on the plain text maintenance manual 202. Specifically, the semantic hierarchy 302 can be considered as a knowledge graph that represents or conveys the semantic structure or semantic organization of the plain text maintenance manual 202. Non-limiting aspects relative to Figure 4 Give a description.

[0068] Figure 4 An exemplary, non-limiting block diagram 400 illustrating how the semantic hierarchy 302 may be generated is shown according to one or more embodiments described herein.

[0069] In various embodiments, as shown, the parsing component 114 can generate the semantic hierarchy 302 electronically by applying any suitable natural language processing technique to the plain text repair manual 202. As a non-limiting example, the parsing component 114 can generate the semantic hierarchy 302 by applying any suitable named entity recognition technique to the plain text repair manual 202. As another non-limiting example, the parsing component 114 can generate the semantic hierarchy 302 by applying any suitable abstract meaning representation parsing technique to the plain text repair manual 202. As an even another non-limiting example, the parsing component 114 can generate the semantic hierarchy 302 by applying any suitable text-to-graph parsing technique to the plain text repair manual 202. As yet another non-limiting example, the parsing component 114 can generate the semantic hierarchy 302 by applying any suitable combination of any of the aforementioned natural language processing techniques to the plain text repair manual 202.

[0070] In any case, the semantic hierarchy 302 can be considered as a knowledge graph that represents, conveys, indicates, or expresses the semantic structure or semantic organization of the plain text maintenance manual 202. Therefore, the semantic hierarchy 302 may include multiple semantic nodes 402. In various aspects, for any suitable positive integer n>1, the multiple semantic nodes 402 may include n nodes: semantic node 402(1) to semantic node 402(n). In various cases, each of the multiple semantic nodes 402 may correspond to any suitable semantic concept or semantic object that is explicitly presented, written, or recorded somewhere within the plain text maintenance manual 202. For example, semantic node 402(1) may represent the first semantic concept or semantic object written or recorded somewhere within the plain text maintenance manual 202, and semantic node 402(n) may represent the nth semantic concept or semantic object written or recorded somewhere within the plain text maintenance manual 202. As some non-limiting examples, the semantic concepts or semantic objects may be: a title of the plain text maintenance manual 202; a table of contents of the plain text maintenance manual 202; an index or glossary of the plain text maintenance manual 202; a chapter or section of the plain text maintenance manual 202; a sub-chapter or sub-section of the plain text maintenance manual 202; a paragraph of the plain text maintenance manual 202; an ordered list of the plain text maintenance manual 202; an item within an ordered list of the plain text maintenance manual 202; or a keyword of the plain text maintenance manual 202 (e.g., a name of a specific hardware component of the medical imaging scanner 104; a name of a specific software component of the medical imaging scanner 104; a name of a specific hardware or software error or fault that may affect the medical imaging scanner 104; a name of a specific hardware or software symptom that may be exhibited by the medical imaging scanner 104). Note that n may be quite large (e.g., may be in the thousands or even millions) given the potentially large length of the plain text maintenance manual 202.

[0071] In various aspects, each of the plurality of semantic nodes 402 may be labeled with, or otherwise associated with, a corresponding location within the plain text maintenance manual 202, may have one or more corresponding relationships with other semantic nodes in the plurality of semantic nodes 402, or may be associated with one or more invokable electronic actions.

[0072] As a non-limiting example, a semantic node 402(1) may be associated with an in-manual location 402(1)(1), one or more inter-node relationships 402(1)(2), or one or more callable actions 402(1)(3).

[0073] In various cases, the in-manual location 402(1)(1) can be any suitable electronic data indicating the location of any semantic concept or object represented by the semantic node 402(1) within the plain text maintenance manual 202 (e.g., one or more scalars, one or more vectors, one or more matrices, one or more tensors, one or more strings, or any suitable combination of these electronic data). In various cases, the in-manual location 402(1)(1) can be specified at any suitable level of granularity (e.g., can indicate the electronic page number or electronic line number where the semantic concept or object represented by the semantic node 402(1) is located). Note that in some aspects, the in-manual location 402(1)(1) can indicate multiple locations because any semantic concept or object represented by the semantic node 402(1) may appear in multiple places within the plain text maintenance manual 202 (e.g., may appear on multiple pages or in multiple lines).

[0074] In various aspects, one or more inter-node relationships 402(1)(2) can be any suitable electronic data (e.g., one or more scalars, one or more vectors, one or more matrices, one or more tensors, one or more strings, or any suitable combination of these electronic data) that indicates which semantic relationships (e.g., directed or undirected) the semantic node 402(1) has with which other nodes in the plurality of semantic nodes 402. Non-limiting examples of such semantic relationships can include: a subject relationship; a part relationship; a hyponym relationship; a nested relationship; a containment relationship; a symptom relationship; a cause relationship; or a prerequisite relationship. Note that in various cases, one or more inter-node relationships 402(1)(2) can be considered to be a knowledge graph edge coupled to the semantic node 402(1).

[0075] In various aspects, the one or more callable actions 402(1)(3) may be any suitable electronic data (e.g., one or more scalars, one or more vectors, one or more matrices, one or more tensors, one or more strings, or any suitable combination of such electronic data) indicating any suitable electronically executable functions or operations that may be performed by, on, or otherwise relative to the medical imaging scanner 104 in response to electronically invoking (e.g., clicking) the semantic node 402(1) or any semantic concept or object represented by the semantic node 402(1). For example, assume that the semantic node 402(1) represents the name of a particular hardware component of the medical imaging scanner 104. In such a case, the one or more callable actions 402(1)(3) may include a calibration or tuning protocol associated with the particular hardware component. Thus, a call (e.g., click) to the semantic node 402(1) (e.g., a call or click to any semantic concept or object represented by the semantic node 402(1) in the plain text maintenance manual 202) can cause the medical imaging scanner 104 to automatically perform the calibration or tuning protocol for the specific hardware component. As another example, assume that the semantic node 402(1) instead represents the name of a specific software component of the medical imaging scanner 104. In such a case, one or more callable actions 402(1)(3) can include pulling an operation log associated with the specific software component. Thus, a call (e.g., click) to the semantic node 402(1) (e.g., a call or click to any semantic concept or object represented by the semantic node 402(1) in the plain text maintenance manual 202) can cause the medical imaging scanner 104 to automatically pull or generate an operation log for the specific software component. Note that in some cases, the semantic node 402(1) may not be associated with any callable action.

[0076] As another non-limiting example, a semantic node 402(n) may be associated with an in-manual location 402(n)(1), one or more inter-node relationships 402(n)(2), or one or more callable actions 402(n)(3). As above, the in-manual location 402(n)(1) may be any suitable electronic data indicating the location of any semantic concept or object represented by the semantic node 402(n) within the plain text maintenance manual 202. Also as above, one or more inter-node relationships 402(n)(2) may be any suitable electronic data indicating what semantic relationship the semantic node 402(n) has with other nodes in the plurality of semantic nodes 402. Still as above, one or more callable actions 402(n)(3) may be any suitable electronically executable function or operation that can be performed by, on, or with respect to the medical imaging scanner 104 in response to a call to the semantic node 402(n).

[0077] In any case, the semantic hierarchy 302 can be considered as a knowledge graph that represents or conveys the semantic architecture of the plain text maintenance manual 202 .

[0078] Figure 5 A block diagram of an exemplary, non-limiting system 500 including a graphical user interface and a fault diagnosis trace according to one or more embodiments described herein is shown, which can facilitate electronic collection of fault diagnosis knowledge for medical imaging scanners. As shown, in some cases, system 500 can include the same components as system 300, and can also include a graphical user interface 502 and a fault diagnosis trace 504.

[0079] In various embodiments, the tracking component 116 can electronically control or electronically access the graphical user interface 502. In various aspects, the graphical user interface 502 can include any suitable electronic display, such as a computer screen or a computer monitor. In various cases, the graphical user interface 502 can also include any suitable human-machine interface device, such as a keyboard, a keypad, or a voice command system. In some cases, the electronic display of the graphical user interface 502 can be an interactive touch screen. In various aspects, the tracking component 116 can visually present the plain text maintenance manual 202 on the graphical user interface 502, so that the maintenance technician responsible for troubleshooting the maintenance complaint 106 can read through the plain text maintenance manual 202 by viewing the graphical user interface 502 or interacting with the graphical user interface. In various cases, the graphical user interface 502 can electronically record or track the order in which the maintenance technician reads in the plain text maintenance manual 202. In various cases, such a sequence can be considered a fault diagnosis trace 504. Non-limiting aspects relative to Figure 6 Give a description.

[0080] Figure 6 An exemplary, non-limiting block diagram 600 illustrating how a fault diagnostic trace 504 may be generated is presented in accordance with one or more embodiments described herein.

[0081] In various embodiments, there may be a service technician 602. In various aspects, the service technician 602 may be any suitable licensed, certified, or otherwise qualified technician who is knowledgeable or experienced in the technical operation, technical service, technical maintenance, or technical repair of the medical imaging scanner 104. Thus, the service technician 602 may perform troubleshooting or otherwise attempt to resolve the service complaint 106.

[0082] In various cases, the graphical user interface 502 can visually present, depict, or display the text content of the plain text maintenance manual 202 on its electronic display. In various cases, the maintenance technician 602 can visually read or consult the plain text maintenance manual 202 displayed on the graphical user interface 502 while troubleshooting the maintenance complaint 106. It should be noted that in some aspects, the maintenance technician 602 can be physically present on or near the medical imaging scanner 104 (for example, the maintenance technician 602 may have personally traveled to the location of the medical imaging scanner 104 to repair it). In such a case, the graphical user interface 502 can be integrated with the medical imaging scanner 104 or otherwise become part of the medical imaging scanner. However, this is only a non-limiting example. In other aspects, the maintenance technician 602 can be physically present on or near the medical imaging scanner 104, and the graphical user interface 502 can be integrated into any other suitable computing device that the maintenance technician 602 can use or view (for example, it can be part of a mobile device or tablet that the maintenance technician 602 accompanies during the maintenance trip). In even other aspects, the maintenance technician 602 may be remote from the medical imaging scanner 104, and the graphical user interface 502 may therefore be integrated into any other suitable computing device available or viewable by the maintenance technician 602 (e.g., may be part of a desktop computer or computerized workstation of the maintenance technician 602).

[0083] In any case, the service technician 602 can interact with the graphical user interface 502 to electronically navigate through the plain text service manual 202 while troubleshooting the service complaint 106. That is, the service technician 602 can read through or electronically access pages, chapters, or paragraphs of the plain text service manual 202 while searching for a solution to the service complaint 106. In various aspects, the tracking component 116 can cause the graphical user interface 502 to electronically record or electronically track the order in which the service technician 602 reads or navigates through the plain text service manual 202. In various aspects, such recording or tracking can be facilitated by any suitable electronic tracking function of the graphical user interface 502.

[0084] As a non-limiting example, the graphical user interface 502 can record such a sequence by utilizing scroll tracking. That is, the graphical user interface 502 can record, know, or infer which parts of the plain text maintenance manual 202 are accessed or read by the maintenance technician 602 based on how the maintenance technician 602 scrolls in the plain text maintenance manual 202, and the sequence in which these parts are accessed or read. For example, assume that the maintenance technician 602 scrolls to the first page, first section, first paragraph, or first word of the plain text maintenance manual 202, and then scrolls to the second page, second section, second paragraph, or second word of the plain text maintenance manual 202. In such a case, it can be inferred that the maintenance technician 602 read the second page, second section, second paragraph, or second word after reading the first page, first section, first paragraph, or first word when troubleshooting the maintenance complaint 106. As another example, assume that the maintenance technician 602 did not scroll to the third page, third section, third paragraph, or third word of the plain text maintenance manual 202. In such a case, it can be inferred that the service technician 602 did not read the third page, the third section, the third paragraph, or the third word when troubleshooting the service complaint 106 .

[0085] As another non-limiting example, the graphical user interface 502 can record such a sequence by utilizing click tracking. That is, the graphical user interface 502 can record, know, or infer which parts of the ordinary text maintenance manual 202 are accessed or read by the maintenance technician 602 based on how the maintenance technician 602 clicks in the ordinary text maintenance manual 202, and the sequence in which these parts are accessed or read. For example, assume that the maintenance technician 602 clicks on the first page, the first section, the first paragraph, or the first word of the ordinary text maintenance manual 202, and then clicks on the second page, the second section, the second paragraph, or the second word of the ordinary text maintenance manual 202. In such a case, it can be inferred that the maintenance technician 602 read the second page, the second section, the second paragraph, or the second word after reading the first page, the first section, the first paragraph, or the first word when troubleshooting the maintenance complaint 106. As another example, assume that the maintenance technician 602 did not click on the third page, the third section, the third paragraph, or the third word of the ordinary text maintenance manual 202. In such a case, it can be inferred that the service technician 602 did not read the third page, the third section, the third paragraph, or the third word when troubleshooting the service complaint 106 .

[0086] As even another non-limiting example, the graphical user interface 502 can record such a sequence by utilizing eye tracking. That is, the graphical user interface 502 can have a camera that monitors the eye movement or orientation of the maintenance technician 602, and the graphical user interface 502 can record, know, or infer which parts of the ordinary text maintenance manual 202 are accessed or read by the maintenance technician 602 based on how the maintenance technician 602's eyes move relative to the ordinary text maintenance manual 202, and the sequence in which these parts are accessed or read. For example, assume that the maintenance technician 602 visually views the first page, first section, first paragraph, or first word of the ordinary text maintenance manual 202 for a time period exceeding any suitable threshold amount of time, and then visually views the second page, second section, second paragraph, or second word of the ordinary text maintenance manual 202 for a time period exceeding the threshold amount of time. In such a case, it can be inferred that the maintenance technician 602 reads the second page, second section, second paragraph, or second word after reading the first page, first section, first paragraph, or first word when troubleshooting the maintenance complaint 106. As another example, assume that the maintenance technician 602 does not visually review the third page, third section, third paragraph, or third word of the plain text maintenance manual 202 for more than a threshold amount of time. In such a case, it can be inferred that the maintenance technician 602 did not read the third page, third section, third paragraph, or third word when troubleshooting the maintenance complaint 106.

[0087] As yet another non-limiting example, graphical user interface 502 may implement any suitable combination of scroll tracking, click tracking, or eye tracking.

[0088] In any case, the graphical user interface 502 can electronically record or track the order in which the maintenance technician 602 reads or navigates in the plain text maintenance manual 202 when troubleshooting the maintenance complaint 106, and such order can be considered or regarded as a troubleshooting trace 504. Since each position in the plain text maintenance manual 202 can correspond to multiple semantic nodes 402 of the semantic hierarchy 302 (for example, each semantic node can be marked with one or more corresponding manual position indicators), the troubleshooting trace 504 can be regarded as a directed path through the semantic hierarchy 302.

[0089] Specifically, the fault diagnosis trace 504 can be any suitable electronic data (e.g., one or more scalars, one or more vectors, one or more matrices, one or more tensors, one or more strings, or any suitable combination of these electronic data) capable of indicating a sequence of visited nodes 604. In various aspects, for any suitable positive integer p>1, the sequence of visited nodes 604 can include p nodes: visited node 604(1) to visited node 604(p). In various aspects, the visited node 604(1) can be any one of the multiple semantic nodes 402 that the maintenance technician 602 initially or first visits or navigates to (e.g., clicks, scrolls to, views) when troubleshooting the maintenance complaint 106. In contrast, the visited node 604(p) can be any one of the multiple semantic nodes 402 that the maintenance technician 602 last visits or navigates to (e.g., clicks, scrolls to, views) when troubleshooting the maintenance complaint 106. In other words, the visited node 604 ( p ) may be the last or final semantic node visited or navigated to by the service technician 602 before resolving, overcoming, or closing the service complaint 106 .

[0090] Note that in some cases, p may be less than n because the maintenance technician 602 may read or navigate through only a portion of the plurality of semantic nodes 402 when troubleshooting the maintenance complaint 106 (e.g., the semantic nodes read or navigated through may only constitute a portion of the normal text maintenance manual 202). However, note that in other cases, p may be greater than n because the maintenance technician 602 may read or navigate to one or more of the plurality of semantic nodes 402 multiple times when troubleshooting the maintenance complaint 106 (e.g., some portions of the normal text maintenance manual 202 may be visited more than once).

[0091] In some aspects, the fault diagnostic trace 504 can be viewed as a type of segmentation mask that corresponds to multiple nodes of the semantic hierarchy 302, respectively (e.g., for each node, the fault diagnostic trace 504 can indicate whether the node has been visited, and if so, what ordinal level or position the node has in the reading or navigation order taken by the maintenance technician 602).

[0092] Note that in some cases, the troubleshooting trace 504 may also indicate whether any semantic nodes visited by the maintenance technician 602 were invoked. As a non-limiting example, and as mentioned above, the visited node 604(1) may be any of the plurality of semantic nodes 402 that the maintenance technician 602 first visited or navigated to. In various cases, the troubleshooting trace 504 may not only indicate that such a semantic node was first visited or navigated to during troubleshooting of the maintenance complaint 106, but may also indicate whether the maintenance technician 602 invoked or activated one or more of any electronically-invokable actions (if any) associated with the semantic node.

[0093] In any case, the semantic hierarchy 302 can be a knowledge graph representing the semantic structure of the plain text maintenance manual 202, and the fault diagnosis trace 504 can represent or indicate a directed path through the knowledge graph, wherein such directed path indicates or represents the order in which the maintenance technician 602 reads or navigates in the plain text maintenance manual 202 to resolve the maintenance complaint 106 or perform fault diagnosis on the maintenance complaint.

[0094] In various embodiments, there can be any suitable number of different service complaints (e.g., multiple instances of 106) corresponding to the medical imaging scanner 104 (e.g., corresponding to any instantiation of the medical imaging scanner 104, where different instantiations can be deployed in different operating environments). Thus, as described above, the tracking component 116 can generate a corresponding fault diagnosis trace (e.g., multiple instances of 504) for each of those different service complaints. This can ultimately produce any suitable number of complaint trace tuples or complaint trace pairs, where each fault diagnosis trace represents any reading or navigation order in the plain text service manual 202 to resolve or handle the corresponding service complaint. Note that in such cases, the parsing component 114 need not repeatedly or iteratively generate the semantic hierarchy 302.

[0095] Figure 7A block diagram of an exemplary, non-limiting system 700 including a deep learning neural network is shown that can facilitate electronic acquisition of fault diagnosis knowledge for medical imaging scanners according to one or more embodiments described herein. As shown, in some cases, system 700 can include the same components as system 500, and can also include a deep learning neural network 702.

[0096] In various embodiments, the model component 118 can electronically store, electronically maintain, electronically control, or electronically access the deep learning neural network 702. In various cases, the deep learning neural network 702 can have or otherwise exhibit any suitable deep learning internal architecture. For example, the deep learning neural network 702 can have an input layer, one or more hidden layers, and an output layer. In various instances, any of such layers can be coupled together by any suitable inter-neuron connection or inter-layer connection (such as a forward connection, a skip connection, or a recursive connection). In addition, in various cases, any of such layers can be a neural network layer of any suitable type with any suitable learnable or trainable internal parameters. For example, any of such input layer, one or more hidden layers, or output layer can be a convolutional layer, and the learnable or trainable parameters of the convolutional layer can be a convolution kernel. As another example, any of such input layer, one or more hidden layers, or output layer can be a dense layer, and the learnable or trainable parameters of the dense layer can be a weight matrix or a bias value. As yet another example, any of such input layer, one or more hidden layers, or output layer may be a batch normalization layer, whose learnable or trainable parameters may be shift factors or scale factors. Still further, in various cases, any of such layers may be any suitable type of neural network layer with any suitable fixed or non-trainable internal parameters. For example, any of such input layer, one or more hidden layers, or output layer may be a non-linear layer, a padding layer, a pooling layer, or a cascade layer.

[0097] In various aspects, the model component 118 can electronically train the deep learning neural network 702 on the complaint trace tuples obtained by the trace component 116. As a non-limiting example, the model component 118 can train the deep learning neural network 702 on the maintenance complaint 106 and the fault diagnosis trace 504, wherein the model component 118 can treat the maintenance complaint 106 as a training input, and wherein the model component 118 can treat the fault diagnosis trace 504 as a ground truth annotation corresponding to the training input. Non-limiting aspects relative to Figure 8 Give a description.

[0098] Figure 8An exemplary, non-limiting block diagram 800 illustrating how a deep learning neural network 702 may be trained for a fault diagnostic trace 504 is presented in accordance with one or more embodiments described herein.

[0099] Prior to commencing such training, model component 118 can initialize the trainable internal parameters (e.g., weight matrices, bias vectors, convolution kernels) of deep learning neural network 702 in any suitable manner (e.g., random initialization).

[0100] In various aspects, the model component 118 can electronically execute the deep learning neural network 702 for the maintenance complaint 106, and such execution can cause the deep learning neural network 702 to generate an output 802. More specifically, as mentioned above, the maintenance complaint 106 can be one or more scalars, one or more vectors, one or more matrices, one or more tensors, or one or more strings. Therefore, the model component 118 can feed the maintenance complaint 106 to the input layer of the deep learning neural network 702 (e.g., any scalar, vector, matrix, tensor, or string defining the maintenance complaint 106 can be fed to the input layer of the deep learning neural network 702). In various cases, the maintenance complaint 106 (e.g., any scalar, vector, matrix, tensor, or string defining the maintenance complaint 106) can complete a forward propagation through one or more hidden layers of the deep learning neural network 702. In various cases, the output layer of the deep learning neural network 702 can calculate the output 802 based on the activation map generated by the one or more hidden layers.

[0101] Note that the format, size, or dimensionality of the output 802 may be dictated by the number, arrangement, size, or other characteristics of the neurons, convolution kernels, or other internal parameters of the output layer (or any other layer) of the deep learning neural network 702. Thus, the output 802 may be forced to have the same, comparable, or commensurate format, size, or dimensionality as the fault diagnostic trace 504. Thus, the output 802 may be viewed as a predicted or inferred fault diagnostic trace that the deep learning neural network 702 believes should correspond to the maintenance complaint 106. In other words, the output 802 may be viewed as a predicted or inferred reading or navigation order in the ordinary text maintenance manual 202 that the deep learning neural network 702 believes will resolve or handle the maintenance complaint 106. In contrast, the fault diagnostic trace 504 may be viewed as a ground truth fault diagnostic trace that is known or believed to correspond to the maintenance complaint 106. That is, the fault diagnostic trace 504 may be a correct or accurate reading order in the ordinary text maintenance manual 202 that is known or believed to be able to resolve or handle the maintenance complaint 106. Note that if deep learning neural network 702 has not been trained at all or has only been trained very little to date, output 802 may be highly inaccurate (e.g., may differ significantly from fault diagnostic trace 504).

[0102] In various aspects, the model component 118 can calculate any suitable error or loss (e.g., MAE, MSE, cross entropy) between the output 802 and the fault diagnostic trace 504. In various cases, the model component 118 can update the trainable internal parameters of the deep learning neural network 702 by performing back propagation (e.g., stochastic gradient descent) driven by the calculated error or loss.

[0103] In various aspects, the above training process can be repeated for any suitable number of complaint trace tuples (e.g., for each complaint trace tuple obtained by the tracking component 116). Such training can ultimately cause the trainable internal parameters of the deep learning neural network 702 to become iteratively optimized so as to accurately or correctly infer the fault diagnosis trace in response to the input maintenance complaint (e.g., accurately or correctly infer the reading or navigation order in the plain text maintenance manual 202). In various cases, when training the deep learning neural network 702, the model component 118 can implement any suitable training termination criteria, any suitable training batch size, or any suitable error, loss, or objective function.

[0104] In any case, the deep learning neural network 702 can thus be trained or configured to receive as input any maintenance complaint associated with an instantiation of the medical imaging scanner 104, and determine as output which specific portions of the plain text maintenance manual 202 should be read in what specific order in order to resolve or handle the maintenance complaint. Thus, in various embodiments, the model component 118 can electronically deploy the deep learning neural network 702 as a third-party service. Therefore, the deep learning neural network 702 can be viewed as helping any third party that owns or operates an instantiation of the medical imaging scanner 104 to troubleshoot maintenance complaints more quickly or more efficiently (e.g., without having to wait for a maintenance technician, and without having to aimlessly peruse the plain text maintenance manual 202). In other words, a third-party's computing device can provide a given maintenance complaint, the model component 118 can execute the deep learning neural network 702 for the given maintenance complaint, thereby generating a predicted fault diagnosis trace, and the model component 118 can send the predicted fault diagnosis trace back to the third-party's computing device (e.g., the predicted fault diagnosis trace can indicate the order in which the third party should read in the plain text maintenance manual 202 in order to resolve or handle their given maintenance complaint).

[0105] Fig. 9 A block diagram of an exemplary, non-limiting system 900 including an operation report is shown that can facilitate electronic collection of troubleshooting knowledge for a medical imaging scanner according to one or more embodiments described herein. As shown, in some cases, system 900 can include the same components as system 700, and can also include an operation report 902.

[0106] In various embodiments, the access component 112 can electronically receive, retrieve, or otherwise access the operation report 902. In various aspects, the operation report 902 can be any suitable structured electronic text document that can be generated by the medical imaging scanner 104 and can represent, indicate, or otherwise convey current or current time status information about the medical imaging scanner 104. As some non-limiting examples, the operation report 902 can be any suitable error log, calibration log, or tuning log that can be output by the medical imaging scanner 104.

[0107] In various cases, the parsing component 114 can generate the semantic hierarchical structure 302 not only based on the normal text maintenance manual 202, but also based on the operation report 902. That is, the semantic hierarchical structure 302 can be a result obtained by applying any suitable natural language processing technology (e.g., named entity recognition, abstract meaning representation parsing, or text-to-graph parsing) to the normal text maintenance manual 202 and the operation report 902. Therefore, the semantic hierarchical structure 302 can be a knowledge graph representing the semantic structure or organization of both the normal text maintenance manual 202 and the operation report 902 (e.g., some of the plurality of semantic nodes 402 can represent semantic concepts or objects recorded in the normal text maintenance manual 202, while other semantic nodes of the plurality of semantic nodes 402 can represent semantic concepts or objects recorded in the operation report 902). In such cases, the fault diagnosis trace 504 can be considered to represent the order in which the maintenance technician 602 reads or navigates not only the plain text maintenance manual 202, but also the operation report 902 (e.g., when troubleshooting the maintenance complaint 106, the maintenance technician 602 can jump back and forth between reading the plain text maintenance manual 202 and reading the operation report 902). In addition, in such cases, the model component 118 can train the deep learning neural network 702 as described above, except that the training input to be fed to the deep learning neural network 702 can be both the maintenance complaint 106 and the operation report 902 (e.g., the concatenation of the two). In such cases, the deep learning neural network 702 can be configured to receive a given maintenance complaint and a corresponding operation report as input, and then generate a fault diagnosis trace as output that indicates the order in which both the plain text maintenance manual 202 and the corresponding operation report should be read in order to resolve or handle the given maintenance complaint.

[0108] It should be noted that the operation report 902 can have an unchanged, standardized or constant structure or format, regardless of its specific substantive content. Therefore, the knowledge graph representation of each possible version of the operation report 902 can also be unchanged, standardized or constant. Therefore, as described above, the parsing component 114 may not need to repeatedly or repeatedly generate the semantic hierarchy 302.

[0109] Fig.10 A flow chart showing an exemplary, non-limiting computer-implemented method 1000 that can facilitate electronic collection of troubleshooting knowledge for a medical imaging scanner according to one or more embodiments described herein. In various cases, the troubleshooting knowledge system 102 can facilitate the computer-implemented method 1000 .

[0110] In various embodiments, act 1002 can include accessing, by a device operably coupled to a processor (e.g., 108) (e.g., via 112), a plain text maintenance manual (e.g., 202) and maintenance complaints (e.g., 106) associated with a medical imaging scanner (e.g., 104).

[0111] In various aspects, act 1004 can include identifying, by a device (eg, via 114 ) and via named entity recognition or natural language processing, a semantic hierarchy of a plain text repair manual (eg, 302 ).

[0112] In various cases, action 1006 can include recording, by a device (e.g., via 116) and via a graphical user interface tracking, how a maintenance technician (e.g., 602) who is troubleshooting a maintenance complaint sequentially navigates through the semantic hierarchy to generate a troubleshooting trace (e.g., 504) corresponding to the maintenance complaint.

[0113] In various cases, action 1008 may include training, by the device (e.g., via 118), a deep learning neural network (e.g., 702) on maintenance complaints and fault diagnostic traces, where the maintenance complaints may be considered training inputs for the deep learning neural network and where the fault diagnostic traces may be considered ground truth annotations corresponding to the maintenance complaints.

[0114] Although not in Fig.10 Although not explicitly shown in , the computer-implemented method 1000 may include: deploying the deep learning neural network as a third-party service after training it by the device (e.g., via 118).

[0115] Although not in Fig.10 , but the semantic hierarchy can be a knowledge graph representation of a plain text repair manual, and the fault diagnosis trace can be a directed path represented by the knowledge graph, which indicates the reading order adopted by the maintenance technician in the plain text repair manual to resolve the maintenance complaint.

[0116] Although not in Fig.10 As explicitly shown in , the computer-implemented method 1000 may include: accessing, by a device (e.g., via 112), an operation report (e.g., 902) generated by a medical imaging scanner, wherein the semantic hierarchy may be a knowledge graph representation of both a plain text maintenance manual and the operation report, and wherein the fault diagnosis trace may be a directed path represented by the knowledge graph indicating a reading order adopted by a maintenance technician in both the plain text maintenance manual and the operation report to resolve a maintenance complaint.

[0117] Although not in Fig.10, but graphical user interface tracking can include click tracking, scroll tracking, or eye movement tracking.

[0118] Although not in Fig.10 , but a node of the semantic hierarchy (e.g., 402(1)) can be based on a keyword located in a normal text maintenance manual that can be associated with an electronic action (e.g., 402(1)(3)) that can be performed by the medical imaging scanner, and clicking on or invoking the node can cause the medical imaging scanner to automatically perform the electronic action.

[0119] So far, various embodiments have been described with respect to a plain text maintenance manual for a medical imaging scanner. However, this is merely a non-limiting example. In various aspects, the various embodiments described may be applied to or extrapolated to any suitable text document for any suitable machine (e.g., even a machine that is not a medical imaging scanner).

[0120] For example, various embodiments may include a computer program product for facilitating the electronic collection of fault diagnosis knowledge. In various aspects, the computer program product may include a non-transitory computer-readable memory (e.g., 110) having program instructions embodied therein. In various cases, these program instructions may be executable by a processor (e.g., 108) to cause the processor to: access one or more ordinary text documents (e.g., 202, 902) associated with a machine (e.g., 104); identify the semantic hierarchy of one or more ordinary text documents via named entity recognition or natural language processing (e.g., 302); record how a maintenance technician (e.g., 602) who is troubleshooting a maintenance complaint (e.g., 106) associated with the machine sequentially navigates in the semantic hierarchy via a graphical user interface tracking, thereby generating a fault diagnosis trace corresponding to the maintenance complaint (e.g., 504); and train a deep learning neural network (e.g., 702) for the maintenance complaint and the fault diagnosis trace, wherein the maintenance complaint is a training input, and wherein the fault diagnosis trace is a baseline truth annotation. In various cases, the semantic hierarchy may be a knowledge graph representation of one or more ordinary text documents, and the fault diagnosis trace may be a directed path represented by the knowledge graph, which indicates a reading order adopted by a maintenance technician in one or more ordinary text documents to resolve a maintenance complaint. In various aspects, the one or more ordinary text documents may include a maintenance manual of a machine (e.g., 202), or an operation report generated by a machine (e.g., 902). In various cases, the graphical user interface tracking may include click tracking, scroll tracking, or eye movement tracking. In various cases, these program instructions may also be executable to enable the processor to: deploy the deep learning neural network as a third-party service after training it. In various aspects, such deployment may include: receiving a third-party maintenance complaint associated with the machine from a third-party computing device; executing a deep learning neural network for the third-party maintenance complaint, thereby generating an inferred fault diagnosis trace, wherein the inferred fault diagnosis trace may represent a sequential reading path through one or more ordinary text documents that is predicted to resolve or handle the third-party maintenance complaint; and sending the inferred fault diagnosis trace to a third-party computing device.

[0121] In various instances, the machine learning algorithm or model may be implemented in any suitable manner to facilitate any suitable aspect described herein. In order to facilitate some of the above machine learning aspects of various embodiments, consider the following discussion of artificial intelligence (AI). Various embodiments described herein may employ artificial intelligence to facilitate the automation of one or more features or functionalities. These components may employ various AI-based schemes to perform various embodiments / examples disclosed herein. In order to provide or contribute to the numerous determinations described herein (e.g., determination, detection, inference, calculation, prediction, prognosis, estimation, derivation, forecast, detection, estimation), the components described herein may examine the entirety or subset of the data to which they are granted access rights, and may provide reasoning about or determine the state of a system or environment from a set of observations captured via events and / or data. For example, determination may be used to identify a specific context or action, or a probability distribution of a state may be generated. These determinations may be probabilistic; that is, the calculation of the probability distribution of the state of interest is based on consideration of data and events. Determination may also refer to a technique for composing a higher-level event from a set of events or data.

[0122] Such determinations may result in the construction of new events or actions from a set of observed events or stored event data, whether or not the events are closely related in time, and whether or not the events and data come from one or more event and data sources. The components disclosed herein may employ various classification (explicitly trained (e.g., via training data) and implicitly trained (e.g., via observed behavior, preferences, historical information, receiving external information, etc.)) schemes or systems (e.g., support vector machines, neural networks, expert systems, Bayesian belief networks, fuzzy logic, data fusion engines, etc.) in conjunction with performing automatic or determined actions related to the claimed subject matter. Thus, the classification schemes or systems may be used to automatically learn and perform a variety of functions, actions, or determinations.

[0123] The classifier can take the input attribute vector z=(z1,z2,z3,z4,z n ) is mapped to the confidence that the input belongs to a certain category, such as according to f(z) = confidence(category). Such classification can use probability-based or statistical analysis (e.g., analyzing utility and cost considerations) to determine the actions to be automatically performed. Support vector machines (SVMs) may be examples of classifiers that can be used. SVMs operate by finding a hypersurface in the space of possible inputs, where the hypersurface attempts to separate triggering criteria from non-triggering events. Intuitively, this makes the classification correct for testing data that is close to but different from the training data. Other directed and undirected model classification methods include, for example, naive Bayes, Bayesian networks, decision trees, neural networks, fuzzy logic models, or any of the probabilistic classification models that provide different independent patterns. Classification as used herein also includes statistical regression for developing priority models.

[0124] In order to provide additional context for the various embodiments described herein, Fig.11 The following discussion is intended to provide a brief general description of a suitable computing environment 1100 in which various embodiments described herein may be implemented. Although the embodiments have been described above in the general context of computer-executable instructions that may be executed on one or more computers, those skilled in the art will recognize that these embodiments may also be implemented in conjunction with other program modules or as a combination of hardware and software.

[0125] Generally, program modules include routines, programs, components, data structures, etc. that perform specific tasks or implement specific abstract data types. In addition, those skilled in the art will appreciate that the methods of the present invention can be practiced with other computer system configurations, including single-processor or multi-processor computer systems, minicomputers, mainframe computers, Internet of Things (IoT) devices, distributed computing systems, as well as personal computers, handheld computing devices, microprocessor-based or programmable consumer electronics, etc., each of which is operatively coupled to one or more associated devices.

[0126] The illustrated embodiments of the embodiments herein may also be practiced in distributed computing environments where certain tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote memory storage devices.

[0127] Computing devices typically include various media, which may include computer-readable storage media, machine-readable storage media, or communication media, where the two terms are used differently in this article, as described below. Computer-readable storage media or machine-readable storage media can be any available storage media that can be accessed by a computer, and include volatile and non-volatile media, removable and non-removable media. By way of example and not limitation, computer-readable storage media or machine-readable storage media can be implemented in conjunction with any method or technology for storing information such as computer-readable or machine-readable instructions, program modules, structured data, or unstructured data.

[0128] Computer-readable storage media may include, but are not limited to, random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD), Blu-ray disk (BD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, solid-state drives or other solid-state storage devices, or other tangible or non-transitory media that can be used to store the desired information. In this regard, the terms "tangible" or "non-transitory" as applied to storage, memory, or computer-readable media herein should be understood to exclude only propagating transient signals themselves as a modifier, and not to disclaim all standard storage, memory, or computer-readable media that are not merely propagating transient signals themselves.

[0129] Computer-readable storage media may be accessed by one or more local or remote computing devices, eg, via access requests, queries, or other data retrieval protocols, to perform various operations relative to the information stored by the media.

[0130] Communication media typically embodies computer readable instructions, data structures, program modules, or other structured or unstructured data in a data signal, which may be, for example, a modulated data signal, such as a carrier wave or other transport mechanism, and includes any information delivery or transmission media. The term "modulated data signal" or "signal" refers to a signal that has one or more of its characteristics set or changed to encode information in one or more signals. By way of example, and not limitation, communication media include wired media (such as a wired network or direct-wired connection) and wireless media (such as acoustic, RF, infrared and other wireless media).

[0131] Reference again Fig.11 , an exemplary environment 1100 for implementing various embodiments of various aspects described herein includes a computer 1102, which includes a processing unit 1104, a system memory 1106, and a system bus 1108. The system bus 1108 couples system components including, but not limited to, the system memory 1106 to the processing unit 1104. The processing unit 1104 can be any of a variety of commercially available processors. Dual microprocessors and other multi-processor architectures can also be used as the processing unit 1104.

[0132] The system bus 1108 may be any of several types of bus structures capable of further interconnecting to a memory bus (with or without a memory controller), a peripheral bus, and a local bus using any of a variety of commercially available bus architectures. The system memory 1106 includes ROM 1110 and RAM 1112. A basic input / output system (BIOS) may be stored in a nonvolatile memory such as ROM, erasable programmable read-only memory (EPROM), EEPROM, where the BIOS contains basic routines that help to transfer information between elements within the computer 1102, such as during startup. The RAM 1112 may also include high-speed RAM, such as static RAM for caching data.

[0133] The computer 1102 also includes an internal hard disk drive (HDD) 1114 (e.g., EIDE, SATA), one or more external storage devices 1116 (e.g., a magnetic floppy disk drive (FDD) 1116, a memory stick or flash drive reader, a memory card reader, etc.), and a drive 1120, such as a solid-state drive, an optical drive, which can read or write from a disk 1122 (such as a CD-ROM disk, a DVD, a BD, etc.). Alternatively, in the case of a solid-state drive, the disk 1122 will not be included unless separated. Although the internal HDD 1114 is illustrated as being located within the computer 1102, the internal HDD 1114 can also be configured to be used externally in a suitable infrastructure (not shown). In addition, although not shown in the environment 1100, a solid-state drive (SSD) can be used in addition to or instead of the HDD 1114. The HDD 1114, external storage device 1116, and drive 1120 can be connected to the system bus 1108 via a HDD interface 1124, an external storage interface 1126, and a drive interface 1128, respectively. The interface 1124 for an external drive specific implementation can include at least one or both of a Universal Serial Bus (USB) and an Institute of Electrical and Electronics Engineers (IEEE) 1394 interface technology. Other external drive connection technologies are within the contemplation of the embodiments described herein.

[0134] The drives and their associated computer-readable storage media provide non-volatile storage of data, data structures, computer-executable instructions, etc. For the computer 1102, the drives and storage media accommodate the storage of any data in a suitable digital format. Although the above description of computer-readable storage media refers to corresponding types of storage devices, it should be understood by those skilled in the art that other types of computer-readable storage media (whether currently existing or developed in the future) can also be used in the exemplary operating environment, and further, any such storage media can contain computer-executable instructions for performing the methods described herein.

[0135] A number of program modules may be stored in the drives and RAM 1112, including an operating system 1130, one or more application programs 1132, other program modules 1134, and program data 1136. All or portions of the operating system, application programs, modules, or data may also be cached in RAM 1112. The systems and methods described herein may be implemented using various commercially available operating systems or combinations of operating systems.

[0136] Computer 1102 may optionally include emulation technology. For example, a hypervisor (not shown) or other intermediary may emulate the hardware environment for operating system 1130, and the emulated hardware may optionally be different from the hardware of the operating system 1130. Fig.11 1102. In such an embodiment, operating system 1130 may include one of a plurality of virtual machines (VMs) hosted at computer 1102. In addition, operating system 1130 may provide a runtime environment, such as a Java runtime environment or a .NET framework, to application 1132. A runtime environment is a consistent execution environment that allows application 1132 to run on any operating system that includes the runtime environment. Similarly, operating system 1130 may support containers, and application 1132 may be in the form of containers that are lightweight, stand-alone, executable software packages that include, for example, the application's code, runtime, system tools, system libraries, and settings.

[0137] In addition, the computer 1102 can be enabled with a security module, such as a trusted processing module (TPM). For example, in the case of a TPM, the boot component is hashed in the next boot component, and the result is waited for to match the security value before loading the next boot component. This process can occur at any layer in the code execution stack of the computer 1102, such as applied to the application execution level or the operating system (OS) kernel level, thereby achieving security at any code execution level.

[0138] A user may enter commands and information into the computer 1102 through one or more wired / wireless input devices, such as a keyboard 1138, a touch screen 1140, and a pointing device such as a mouse 1142. Other input devices (not shown) may include a microphone, an infrared (IR) remote control, a radio frequency (RF) remote control or other remote control, a joystick, a virtual reality controller or virtual reality headset, a game pad, a stylus, an image input device (e.g., a camera), a gesture sensor input device, a visual movement sensor input device, an emotion or facial detection device, a biometric input device (e.g., a fingerprint or iris scanner), and the like. These input devices and other input devices are often connected to the processing unit 1104 through an input device interface 1144, which may be coupled to the system bus 1108, but these input devices and other input devices may be connected through other interfaces, such as a parallel port, an IEEE 1394 serial port, a game port, a USB port, an IR port, a fingerprint or iris scanner, or the like. interface, etc.) connection.

[0139] A monitor 1146 or other type of display device may also be connected to the system bus 1108 via an interface, such as a video adapter 1148. In addition to the monitor 1146, computers typically include other peripheral output devices (not shown), such as speakers, printers, and the like.

[0140] The computer 1102 can operate in a networked environment using logical connections to one or more remote computers, such as remote computer 1150, via wired or wireless communications. The remote computer 1150 can be a workstation, server computer, router, personal computer, portable computer, microprocessor-based entertainment appliance, peer device, or other public network node, and typically includes many or all of the elements described with respect to the computer 1102, although only a memory / storage device 1152 is illustrated for simplicity. The depicted logical connections include wired / wireless connections to a local area network (LAN) 1154 or a larger network, such as a wide area network (WAN) 1156. Such LAN and WAN networking environments are common in offices and companies, and facilitate enterprise-wide computer networks, such as intranets, all of which can be connected to a global communication network, such as the Internet.

[0141] When used in a LAN networking environment, the computer 1102 can be connected to a local network 1154 through a wired or wireless communication network interface or adapter 1158. The adapter 1158 can facilitate wired or wireless communication with the LAN 1154, which can also include a wireless access point (AP) disposed thereon to communicate with the adapter 1158 in a wireless mode.

[0142] When used in a WAN networking environment, the computer 1102 may include a modem 1160, or may be connected to a communications server on the WAN 1156 via other means for establishing communications over the WAN 1156, such as through the Internet. The modem 1160, which may be an internal or external device and a wired or wireless device, may be connected to the system bus 1108 via the input device interface 1144. In a networked environment, program modules depicted relative to the computer 1102, or portions thereof, may be stored in the remote memory / storage device 1152. It should be appreciated that the network connections shown are examples and other means of establishing a communications link between the computers may be used.

[0143] When used in a LAN or WAN networking environment, in addition to or as an alternative to the external storage devices 1116 described above, the computer 1102 can access a cloud storage system or other network-based storage system, such as, but not limited to, a network virtual machine that provides one or more aspects of information storage or processing. Generally speaking, the connection between the computer 1102 and the cloud storage system can be established through the LAN 1154 or WAN 1156, for example, by an adapter 1158 or a modem 1160, respectively. When the computer 1102 is connected to the associated cloud storage system, the external storage interface 1126 can manage the storage provided by the cloud storage system with the help of the adapter 1158 or the modem 1160, just like other types of external storage devices. For example, the external storage interface 1126 can be configured to provide access to cloud storage sources as if those storage sources were physically connected to the computer 1102.

[0144] The computer 1102 may be configured to communicate with any wireless device or entity that is operatively configured to communicate wirelessly, such as a printer, scanner, desktop or portable computer, portable data assistant, communication satellite, any piece of equipment or location associated with a wirelessly detectable tag (e.g., a self-service machine, a newsstand, a store shelf, etc.), and a telephone. This may include Wireless Fidelity (Wi-Fi) and Wireless technology. Therefore, the communication can be a predefined structure like a conventional network, or just an ad hoc communication between at least two devices.

[0145] Fig.121 is a schematic block diagram of a sample computing environment 1200 with which the disclosed subject matter can interact. The sample computing environment 1200 includes one or more clients 1210. The client 1210 can be hardware or software (e.g., a thread, a process, a computing device). The sample computing environment 1200 also includes one or more servers 1230. The server 1230 can also be hardware or software (e.g., a thread, a process, a computing device). For example, the server 1230 can accommodate threads to perform transformations by adopting one or more embodiments as described herein. One possible communication between the client 1210 and the server 1230 can be in the form of data packets suitable for transmission between two or more computer processes. The sample computing environment 1200 includes a communication framework 1250 that can be used to facilitate communication between the client 1210 and the server 1230. The client 1210 is operably connected to one or more client data repositories 1220, which can be used to store information local to the client 1210. Similarly, server 1230 is operably connected to one or more server data repositories 1240 , which may be used to store information local to server 1230 .

[0146] Various embodiments can be systems, methods, devices or computer program products at any possible level of technical detail of integration. A computer program product may include a computer-readable storage medium (or multiple media) having computer-readable program instructions thereon for various aspects of various embodiments implemented by a processor. A computer-readable storage medium may be a tangible device that can hold and store instructions for use by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device or any suitable combination of the foregoing. An incomplete list of more specific examples of computer-readable storage media may also include the following: a portable computer floppy disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device (such as a punch card or a raised structure in a groove with instructions recorded thereon), and any suitable combination of the above items. As used herein, computer-readable storage media should not be understood as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses passing through fiber optic cables), or electrical signals transmitted through wires.

[0147] Computer-readable program instructions as described herein can be downloaded to corresponding computing / processing equipment from computer-readable storage media, or downloaded to external computers or external storage devices via networks (e.g., the Internet, local area networks, wide area networks, or wireless networks). Networks can include copper transmission cables, optical transmission optical fibers, wireless transmissions, routers, firewalls, switches, gateway computers, or edge servers. Network adapter cards or network interfaces in each computing / processing equipment receive computer-readable program instructions from the network, and forward computer-readable program instructions for storage in computer-readable storage media in corresponding computing / processing equipment. Computer-readable program instructions for the operation of various embodiments for implementation can be assembler instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, configuration data of integrated circuits, or source code or object code written in any combination of one or more programming languages ​​(including object-oriented programming languages, such as Smalltalk, C++, etc.) and process programming languages ​​(such as "C" programming languages ​​or similar programming languages). Computer-readable program instructions can be executed completely on a user's computer, partially on a user's computer, executed as an independent software package, partially on a user's computer and partially on a remote computer, or completely on a remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network (including a local area network (LAN) or a wide area network (WAN)), or can be connected to an external computer (for example, by using the Internet of an Internet service provider). In some embodiments, an electronic circuit including, for example, a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA) can execute computer-readable program instructions to personalize the electronic circuit by utilizing the state information of the computer-readable program instructions, so as to perform various aspects.

[0148] Flowchart illustrations and / or block diagrams of methods, devices (systems) and computer program products according to various embodiments are used to describe various aspects described herein. It should be understood that each frame of a flowchart illustration or a block diagram, and a combination of frames in a flowchart illustration or a block diagram can be implemented by computer-readable program instructions. These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device create a component for implementing the function / action specified in one or more frames of a flowchart or a block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium, which can instruct a computer, a programmable data processing device or other equipment to act in a particular manner, so that a computer-readable storage medium with instructions stored therein includes an article, which includes instructions for implementing various aspects of the function / action specified in one or more frames of a flowchart or a block diagram. Computer-readable program instructions can also be loaded onto a computer, other programmable data processing devices or other devices, so that a series of operating actions are performed on a computer, other programmable devices or other devices to produce a computer-implemented process, so that the instructions executed on a computer, other programmable devices or other devices implement the function / action specified in one or more frames of a flowchart or a block diagram.

[0149] The flow chart and block diagram in the accompanying drawings illustrate the possible specific implementation architecture, functionality and operation of the system, method and computer program product according to various embodiments. In this regard, each frame in the flow chart or block diagram can represent a module, fragment or part of an instruction, which includes one or more executable instructions for realizing a specified logical function. In some alternative specific implementations, the function indicated in the frame may not occur in the order indicated in the figure. For example, in fact, the two frames shown in succession can be executed substantially at the same time, or sometimes these frames may be executed in reverse order, depending on the functionality involved. It will also be noted that the combination of each frame of the block diagram or flow chart illustration and the frame in the block diagram or flow chart illustration can be realized by a system based on dedicated hardware that performs a specified function or action or implements a combination of dedicated hardware and computer instructions.

[0150] Although the present subject matter has been described above in the general context of computer executable instructions of a computer program product running on one or more computers, it will be appreciated by those skilled in the art that the present disclosure may also or may be implemented in combination with other program modules. Typically, a program module includes routines, programs, components, data structures, etc. that perform specific tasks or implement specific abstract data types. In addition, it will be appreciated by those skilled in the art that various aspects may be practiced with other computer system configurations, including single-processor computer systems or multi-processor computer systems, small computing devices, large computers, and computers, handheld computing devices (e.g., PDAs, phones), microprocessor-based or programmable consumer or industrial electronics, etc. The illustrated aspects may also be practiced in a distributed computing environment in which tasks are performed by a remote processing device linked by a communication network. However, some (if not all) aspects of the present disclosure may be practiced on a stand-alone computer. In a distributed computing environment, program modules may be located in local and remote memory storage devices.

[0151] As used in this application, the terms "component", "system", "platform", "interface", etc. may refer to or include computer-related entities or entities related to an operating machine having one or more specific functionalities. The entities disclosed herein may be hardware, a combination of hardware and software, software, or software in execution. For example, a component may be, but is not limited to, a program, a processor, an object, an executable file, an execution thread, a program, or a computer running on a processor. By way of illustration, both an application running on a server and a server may be components. One or more components may reside within a process or a thread of execution, and a component may be located on a computer or distributed between two or more computers. As another example, the corresponding component may be executed according to various computer-readable media having various data structures stored thereon. The components may communicate via a local or remote process, such as according to a signal having one or more data packets (e.g., data from a component that interacts with another component in a local system, a distributed system, or a network (such as the Internet with other systems) via a signal). As another example, a component may be a device having a specific functionality provided by a mechanical part operated by an electrical or electronic circuit, which is operated by a software or firmware application executed by a processor. In such cases, the processor may be internal or external to the device and may execute at least a portion of a software or firmware application. As yet another example, a component may be a device that provides specific functionality through electronic components rather than mechanical parts, where the electronic components may include a processor or other means for executing software or firmware that at least partially imparts the functionality to the electronic components. In one aspect, the component may emulate an electronic component, such as via a virtual machine within a cloud computing system.

[0152] In addition, the term "or" is intended to mean an inclusive "or" rather than an exclusive "or". That is, unless otherwise specified or clear from the context, "X employs A or B" is intended to mean any natural inclusive replacement. That is, if X employs A; X employs B; or X employs both A and B, then "X employs A or B" is satisfied in any of the foregoing cases. As used herein, the term "and / or" is intended to have the same meaning as "or". In addition, unless otherwise specified or clear from the context that it is for a singular form, the articles "a" and "an" used in this specification and the drawings should generally be understood to mean "one or more". As used herein, the term "example" or "exemplary" is used to indicate use as an example, instance, or illustration. For the avoidance of doubt, the subject matter disclosed herein is not limited to such examples. In addition, any aspect or design described herein as an "example" or "exemplary" is not necessarily to be understood as being preferred or advantageous over other aspects or designs, nor is it intended to exclude equivalent exemplary structures and techniques known to those of ordinary skill in the art.

[0153] The disclosure herein describes non-limiting examples. For ease of description or explanation, when discussing various examples, the various parts of the disclosure herein utilize the terms "each," "each," or "all." Such usage of the terms "each," "each," or "all" is non-limiting. In other words, when the disclosure herein provides a description of "each," "each," or "all" specific objects or parts that are applied to some specific objects or parts, it should be understood that this is a non-limiting example, and it should also be understood that in various examples, it may be that such description is applicable to less than "each," "each," or "all" specific objects or parts in the specific object or part.

[0154] As used in this specification, the term "processor" may refer to substantially any computing processing unit or device, including but not limited to a single-core processor; a single processor with software multithreaded execution capability; a multi-core processor; a multi-core processor with software multithreaded execution capability; a multi-core processor with hardware multithreading technology; a parallel platform; and a parallel platform with distributed shared memory. In addition, a processor may refer to an integrated circuit, an application-specific integrated circuit (ASIC), a digital signal processor (DSP), a field programmable gate array (FPGA), a programmable logic controller (PLC), a complex programmable logic device (CPLD), discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. In addition, the processor may utilize nanoscale architectures (such as, but not limited to, transistors, switches, and gates based on molecules and quantum dots) in order to optimize space usage or enhance the performance of user equipment. The processor may also be implemented as a combination of computing processing units. In the present disclosure, terms such as "repository", "storage device", "data repository", "data storage device", "database", and substantially any other information storage component related to the operation and functionality of the component are used to refer to a "memory component", an entity embodied in a "memory", or a component including a memory. It should be understood that the memory or memory components described herein may be volatile memory or non-volatile memory, or may include both volatile memory and non-volatile memory. By way of illustration and not limitation, non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM), flash memory, or non-volatile random access memory (RAM) (e.g., ferroelectric RAM (FeRAM)). For example, volatile memory may include RAM that can act as an external cache memory. By way of illustration and not limitation, RAM can be provided in a variety of forms, such as synchronous RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), Synchlink DRAM (SLDRAM), direct Rambus RAM (DRRAM), direct Rambus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM). In addition, the disclosed memory components of the system or computer-implemented method herein are intended to include, but are not limited to, including these and any other suitable types of memory.

[0155] What has been described above includes only examples of systems and computer-implemented methods. Of course, it is not possible to describe every conceivable combination of components or computer-implemented methods for the purposes of describing the present disclosure, but many other combinations and permutations of the present disclosure are possible. In addition, to the extent that the terms "including," "having," "having," and the like are used in the detailed description, claims, appendices, and drawings, such terms are intended to be inclusive in a manner similar to the term "comprising," as interpreted when "including" is used as a transitional word in a claim.

[0156] Descriptions of various embodiments have been given for purposes of illustration, but these descriptions are not intended to be exhaustive or limited to the disclosed embodiments. Many modifications and variations will be apparent without departing from the scope and spirit of the described embodiments. The terminology used herein is selected to best illustrate the principles of the embodiments, practical applications or technical improvements over technologies found on the market, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A system, comprising: A processor (e.g., 108) that executes computer-executable components stored in a non-transitory computer-readable memory (e.g., 110), wherein the computer-executable components include: accessing a component (e.g., 112) that accesses a plain text maintenance manual (e.g., 202) and a maintenance complaint (e.g., 106) associated with a medical imaging scanner (e.g., 104); a parsing component (e.g., 114) that identifies a semantic hierarchy of the plain text maintenance manual via named entity recognition or natural language processing (e.g., 302); and A tracking component (e.g., 116) that records, via a graphical user interface, how a maintenance technician (e.g., 602) who is troubleshooting the maintenance complaint navigates sequentially in the semantic hierarchy to generate a troubleshooting trace (e.g., 504) corresponding to the maintenance complaint.

2. A system according to claim 1, wherein the semantic hierarchical structure is a knowledge graph representation of the plain text maintenance manual, and wherein the fault diagnosis trace is a directed path through the knowledge graph representation, which indicates the reading order adopted by the maintenance technician in the plain text maintenance manual to resolve the maintenance complaint.

3. A system according to claim 1, wherein the access component accesses an operation report (e.g., 902) generated by the medical imaging scanner, wherein the semantic hierarchical structure is a knowledge graph representation of both the plain text maintenance manual and the operation report, and wherein the fault diagnosis trace is a directed path represented by the knowledge graph, which indicates a reading order adopted by the maintenance technician in both the plain text maintenance manual and the operation report to resolve the maintenance complaint.

4. The system of claim 1, wherein the graphical user interface tracking comprises click tracking, scroll tracking, or eye movement tracking.

5. The system of claim 1 , wherein the computer executable component further comprises: A model component (e.g., 118) that trains a deep learning neural network (e.g., 702) on the maintenance complaint and the fault diagnostic trace, wherein the maintenance complaint is considered as a training input to the deep learning neural network, and wherein the fault diagnostic trace is considered as a ground truth annotation corresponding to the maintenance complaint.

6. The system of claim 5, wherein the model component deploys the deep learning neural network as a third-party service after training it.

7. The system according to claim 1, wherein the nodes of the semantic hierarchy (e.g., 402) is based on a keyword located in the plain text maintenance manual, wherein the keyword is associated with an electronic action capable of being performed by the medical imaging scanner, and wherein clicking or invoking the node or the keyword causes the medical imaging scanner to automatically perform the electronic action.

8. A computer-implemented method, the computer-implemented method comprising: accessing, by a device operably coupled to a processor (e.g., 108) (e.g., via 112), a plain text maintenance manual (e.g., 202) and a maintenance complaint (e.g., 106) associated with a medical imaging scanner (e.g., 104); identifying, by the device (e.g., via 114) and via named entity recognition or natural language processing, a semantic hierarchy of the plain text maintenance manual (e.g., 302); and How a maintenance technician (e.g., 602) who is troubleshooting the maintenance complaint sequentially navigates through the semantic hierarchy is recorded by the device (e.g., via 116) and tracked via a graphical user interface to generate a troubleshooting trace (e.g., 504) corresponding to the maintenance complaint.

9. A computer-implemented method according to claim 8, wherein the semantic hierarchical structure is a knowledge graph representation of the plain text maintenance manual, and wherein the fault diagnosis trace is a directed path through the knowledge graph representation, which indicates the reading order adopted by the maintenance technician in the plain text maintenance manual to resolve the maintenance complaint.

10. The computer-implemented method of claim 8, further comprising: An operation report (e.g., 902) generated by the medical imaging scanner is accessed by the device (e.g., via 112), wherein the semantic hierarchy is a knowledge graph representation of both the plain text maintenance manual and the operation report, and wherein the fault diagnosis trace is a directed path through the knowledge graph representation, which indicates a reading order adopted by the maintenance technician in both the plain text maintenance manual and the operation report to resolve the maintenance complaint.

11. The computer-implemented method of claim 8, wherein the graphical user interface tracking comprises click tracking, scroll tracking, or eye movement tracking.

12. The computer-implemented method of claim 8, further comprising: A deep learning neural network is trained (e.g., 702) by the device (e.g., via 118) for the maintenance complaint and the fault diagnostic trace, wherein the maintenance complaint is considered as a training input for the deep learning neural network, and wherein the fault diagnostic trace is considered as a ground truth annotation corresponding to the maintenance complaint.

13. The computer-implemented method of claim 12, further comprising: The deep learning neural network is deployed as a third-party service by the device (e.g., via 118) after training.

14. A computer-implemented method according to claim 8, wherein the nodes of the semantic hierarchy (e.g., one of 402) are based on keywords located in the plain text maintenance manual, wherein the keywords are associated with electronic actions capable of being performed by the medical imaging scanner, and wherein clicking on or invoking the node or the keyword causes the medical imaging scanner to automatically perform the electronic action.

15. A computer program product for facilitating electronic collection of fault diagnosis knowledge, the computer program product comprising a non-transitory computer readable memory (e.g., 110), the non-transitory computer readable memory having program instructions embodied therein, the program instructions being executable by a processor (e.g., 108) to cause the processor to: Accessing one or more plain text documents (e.g., 202 or 902) associated with a machine (e.g., 104); Identifying a semantic hierarchy of the one or more plain text documents via named entity recognition or natural language processing (e.g., 302); Recording, via a graphical user interface tracking, how a maintenance technician (e.g., 602) who is troubleshooting a maintenance complaint (e.g., 106) associated with the machine sequentially navigates through the semantic hierarchy to generate a troubleshooting trace (e.g., 504) corresponding to the maintenance complaint; and A deep learning neural network is trained on the maintenance complaint and the fault diagnostic trace (e.g., 702), wherein the maintenance complaint is a training input and wherein the fault diagnostic trace is a ground truth annotation.

Images