Medical care wireless sensor network fault diagnosis method

Abstract

The invention relates to the technical field of fault diagnosis, in particular to a medical care wireless sensor network fault diagnosis method, which comprises the following steps of: diagnosing each diagnostic instrument and network state through a preset starting check model, and determining whether the starting state of a remote medical diagnosis system is a normal state or not. Collecting detection data of each sensor, state data of a wireless sensor network state and corresponding record information in real time in a diagnosis process; performing diagnosis analysis on the detection data, the state data and the recorded information by adopting an anomaly detection model to obtain a comprehensive detection result of the remote medical diagnosis system; adopting a preset decision matching model and performing matching analysis on the comprehensive detection result to obtain a fault processing decision; and adjusting the remote medical diagnosis system and / or sending early warning information according to the fault processing decision. According to the invention, the problem that the mobile medical vehicle lacks an overall fault detection function and cannot carry out fault early warning in real time in the use process can be solved.

Description

Technical Field

[0001] The invention relates to the technical field of fault diagnosis, and in particular to a medical wireless sensor network fault diagnosis method. Background Art

[0002] As the problem of population aging becomes increasingly prominent, medical problems for the elderly are also gradually emerging. For example, the elderly cannot adapt to the bumpy ride of long-distance medical treatment, and their condition worsens due to long-distance transportation. With the rapid development of science and technology, remote medical diagnosis systems have emerged to address the above problems.

[0003] The existing telemedicine diagnosis system uses wireless sensor networks as the communication basis. It can diagnose patients in real time during remote treatment, obtain patients' diagnostic data, and make treatment plans in time according to the diagnostic data. In the wireless sensor network, doctors obtain diagnostic data collected by medical sensors with the assistance of gateway nodes, and the communication between all participants (doctors, gateways, medical sensors) is carried out in unsecure channels, which increases the risk of patient data leakage. Secondly, traditional telemedicine diagnosis systems rely on sensors of field instruments for detection and conduct remote communication through gateway nodes. For example, a mobile medical vehicle using a telemedicine diagnosis system drives to the patient's residence, and the patient is diagnosed and treated by the diagnostic equipment on the mobile medical vehicle and remote professional doctors. Since the diagnostic instruments of the mobile medical vehicle work independently and the wireless sensor network has high accuracy requirements, and the mobile medical vehicle lacks the function of overall fault detection, it is impossible to provide real-time fault warning during use. Once a diagnostic device or network fails, the entire remote diagnosis process may result in diagnostic errors, which in turn causes doctors to give wrong diagnostic plans, causing adverse effects on patients.

[0004] Therefore, the present invention provides a medical wireless sensor network fault diagnosis method to solve the above problems. Summary of the invention

[0005] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides a medical wireless sensor network fault diagnosis method to solve the problem that the above mobile medical vehicle lacks overall fault detection function and cannot provide real-time fault warning during use.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is: A medical wireless sensor network fault diagnosis method includes: when starting a diagnostic device, using a preset startup inspection model to diagnose each diagnostic instrument and network status, and determining the startup status of a remote medical diagnosis system, wherein the startup status includes a normal state and an abnormal state; When the startup state is normal, normal information is issued, and detection data of each sensor and status data of the wireless sensor network state are collected in real time during the diagnosis process, and the detection data, status data and corresponding record information are transmitted to the remote server; Anomaly detection models are used to diagnose and analyze the test data, status data and record information in the remote server to obtain comprehensive test results of the remote medical diagnosis system; Use the preset decision matching model to analyze the comprehensive detection results to obtain the fault handling decision; Adjust the remote medical diagnosis system and / or issue early warning information based on the fault handling decision.

[0007] Preferably, the processing of the preset startup check model includes: The startup data of each diagnostic instrument is obtained, the startup data is marked according to a preset tag library, and temporarily stored in a first database; the startup data is analyzed according to the preset startup inspection standard of each diagnostic instrument to determine whether the startup data contains abnormal data; if not, it is in a normal state, and if so, it is in an abnormal state, and a startup repair model is used to analyze and process the abnormal data to restore the corresponding diagnostic instrument state and / or an abnormal detection model is called to analyze the abnormal data to obtain a corresponding comprehensive detection result.

[0008] Preferably, the use of a startup repair model to analyze and process the abnormal data to restore the corresponding diagnostic instrument state and / or calling an abnormal detection model to analyze the abnormal data to obtain a corresponding comprehensive detection result includes: determining the type and grade corresponding to the label of the abnormal data; if the type belongs to the first type and the grade belongs to the first preset interval, restarting the diagnostic device corresponding to the abnormal data; when the acquired second startup data contains the same abnormal data, calling the abnormal detection model to analyze the abnormal data to obtain a corresponding comprehensive detection result; and if the type does not belong to the first type, or the grade does not belong to the first preset interval, calling the abnormal detection model to analyze the abnormal data to obtain a corresponding comprehensive detection result.

[0009] Preferably, the processing process of the anomaly detection model includes: performing preliminary analysis on the detection data, status data and record information according to the preset standards corresponding to each detection data to determine whether there is initial abnormal data; if there is initial abnormal data, obtaining in real time an initial abnormal data set of a set time period corresponding to the initial abnormal data; performing time series analysis on the initial abnormal data set to determine whether the amount of abnormal data is greater than a first preset value; if it is greater than the first preset value, performing correlation analysis on the initial abnormal data set to obtain a comprehensive detection result.

[0010] Preferably, the aforementioned performing correlation analysis on the initial abnormal data set to obtain a comprehensive detection result includes: obtaining the type of each data in the initial abnormal data set; judging whether network abnormal data is included according to the type of each data; if included, using a network detection model to analyze the network abnormal data to obtain a network abnormality detection result; performing correlation analysis on the corresponding initial abnormal data set based on a preset correlation map and the type of each data to determine the abnormal location; and determining the abnormal level and abnormal cause according to the abnormal location and the initial abnormal data set; and forming a comprehensive detection result with the abnormal location, abnormal level, abnormal cause, corresponding abnormal data set and / or network abnormality detection result.

[0011] Preferably, the method of using a network detection model to analyze network anomaly data to obtain a network anomaly detection result includes: obtaining anomaly starting point data in the network anomaly data in a transmission order; determining a first network anomaly detection result including an anomaly location and an anomaly cause according to a relationship between the anomaly starting point data and a preset network anomaly matching library; classifying the network anomaly data according to the type of the network anomaly data to obtain a plurality of classified data; performing matching analysis on each classified data according to a preset network type matching library to determine a second network anomaly detection result including an anomaly type and an anomaly cause; and performing a fusion analysis on the first network anomaly detection result and the second network anomaly detection result to obtain a network anomaly detection result.

[0012] Preferably, the processing process of the preset decision matching model includes: obtaining the abnormality level in the comprehensive detection results; determining the decision type according to the abnormality level; matching and analyzing the abnormal location and abnormal cause in the comprehensive detection results according to the decision matching library corresponding to the decision type to obtain a fault handling decision, and the fault handling decision includes a conventional decision, an emergency decision and a maintenance decision.

[0013] Preferably, the anomaly detection model is constructed based on a convolutional neural network, historical record data and multiple preset rules.

[0014] Preferably, the remote medical diagnosis system is adjusted and / or a warning message is issued according to the fault handling decision, including: when the fault handling decision includes routine decisions, the routine decisions are used to adjust the diagnostic instrument status and / or network status of the remote medical diagnosis system and a prompt message is sent; when the fault handling decision includes emergency decisions, the emergency decisions are used to adjust the corresponding diagnostic instrument status and / or network status in the remote medical diagnosis system, and a warning message is sent; when the fault handling decision only includes maintenance decisions, the current medical diagnosis service is stopped and an alarm message is sent to the staff.

[0015] Preferably, a medical wireless sensor network fault diagnosis method further includes: encrypting and transmitting the detection data using an end-to-end encryption method.

[0016] The beneficial effects of the present invention are: 1. When a mobile medical vehicle using a remote medical diagnosis system is used for remote diagnosis, the present invention sets a startup check phase at startup, an abnormality detection phase during the diagnosis process, a preset decision matching phase when a fault is detected, and a final adjustment phase. In the startup phase, the present invention diagnoses each diagnostic instrument and network status through a preset startup check model to determine whether the startup state of the remote medical diagnosis system is normal. When the startup state is normal, the abnormality detection phase during diagnosis is entered, and the detection data of each sensor, the state data of the wireless sensor network state, and the corresponding record information are collected in real time during the diagnosis process; the abnormality detection model is used on the remote terminal or server to diagnose and analyze the detection data, state data, and record information to obtain a comprehensive detection result of the remote medical diagnosis system; after obtaining the comprehensive detection result, a preset decision matching model is used and the comprehensive detection result is matched and analyzed to obtain a fault handling decision; finally, the remote medical diagnosis system is adjusted and / or an early warning message is issued according to the fault handling decision, so as to provide real-time diagnostic services for the mobile medical vehicle, thereby improving the stability and robustness of the mobile medical vehicle. In addition, the present invention solves the problem that the mobile medical vehicle lacks an overall fault detection function and cannot provide a real-time fault warning during use through the above method.

[0017] 2. In the startup inspection phase, the present invention uses a preset label library to mark the startup data to achieve classification, so as to analyze and process the startup data later; compare and analyze the startup data according to the preset startup inspection standards of each diagnostic instrument to determine whether the startup data contains abnormal data; if it does not contain abnormal data, it will normally enter the abnormal detection phase of the diagnostic process; if there is abnormal data, the startup repair model is used to analyze and process the abnormal data to restore the corresponding diagnostic instrument status; when it cannot be repaired, the abnormal detection model is called to analyze the abnormal data to obtain the corresponding comprehensive detection results. The present invention checks the remote medical diagnosis system in the startup phase in the above manner, so that the user can clearly understand the status of the remote medical diagnosis system when using it, which provides convenience for doctors and patients.

[0018] 3. In the anomaly detection stage, the present invention sets two modules in the anomaly detection model. One module analyzes and processes the anomaly detection data of each diagnostic instrument in the initial anomaly data set; the other module uses the network detection model to perform diagnostic analysis on the network anomaly data in the initial anomaly data set; the analysis results of the two modules are integrated to obtain a comprehensive detection result. The anomaly detection model of the present invention performs division of labor diagnostic analysis in the above manner, which can improve the accuracy of the diagnostic results and the data processing speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A schematic flow chart of a medical wireless sensor network fault diagnosis method of the present invention; Figure 2 The figure is a schematic flow chart of obtaining comprehensive detection results in the anomaly detection model of the present invention. DETAILED DESCRIPTION

[0020] The following will refer to the attached Figure 1 To Attachment Figure 2 The embodiments of the present invention are described in detail. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0021] A medical wireless sensor network fault diagnosis method, as shown in the attached Figure 1 As shown, the following steps are included: Step S11: When the diagnostic device is started, a preset startup check model is used to diagnose each diagnostic instrument and network status to determine the startup status of the remote medical diagnosis system.

[0022] Preferably, the startup state includes a normal state and an abnormal state.

[0023] Step S12: When the startup state is normal, normal information is issued, and detection data of each sensor and status data of the wireless sensor network are collected in real time during the diagnosis process, and the detection data, status data and corresponding record information are transmitted to the remote server.

[0024] Preferably, the recorded information is information related to the detection data and status data, including the total amount of data, the amount of blank data, the transmission method, the acquisition time, the transmission time, etc.; the purpose of the recorded information is to compare and refer to whether the amount of data transmitted to the remote server is normal, and whether data packet loss occurs during the transmission process.

[0025] Step S13: Use an anomaly detection model to perform diagnostic analysis on the detection data, status data and record information in the remote server to obtain a comprehensive detection result of the remote medical diagnosis system.

[0026] Preferably, the detection content of the anomaly detection model includes at least network transmission failure, sensor failure of the diagnostic equipment, confidentiality and distortion rate of data transmission, stability of the conference room for expert consultation, and real-time fault diagnosis during remote diagnosis of the mobile medical vehicle. If the test result is normal, only normal information is displayed in the comprehensive test result.

[0027] Step S14: Analyze the comprehensive detection results using a preset decision matching model to obtain a fault handling decision.

[0028] Step S15: adjusting the remote medical diagnosis system and / or issuing a warning message according to the fault handling decision.

[0029] Specifically, when a mobile medical vehicle using a remote medical diagnosis system is used for remote diagnosis, the present invention sets a startup check phase at startup, an abnormality detection phase during the diagnosis process, a preset decision matching phase when a fault is detected, and a final adjustment phase. In the startup phase, the present invention diagnoses each diagnostic instrument and network status through a preset startup check model to determine whether the startup state of the remote medical diagnosis system is normal. When the startup state is normal, the abnormality detection phase during diagnosis is entered, and the detection data of each sensor, the state data of the wireless sensor network state, and the corresponding record information are collected in real time during the diagnosis process; the abnormality detection model is used on the remote terminal or server to diagnose and analyze the detection data, state data, and record information to obtain a comprehensive detection result of the remote medical diagnosis system; after obtaining the comprehensive detection result, a preset decision matching model is used and the comprehensive detection result is matched and analyzed to obtain a fault handling decision; finally, the remote medical diagnosis system is adjusted and / or an early warning message is issued according to the fault handling decision, and a real-time diagnosis service is provided for the mobile medical vehicle, thereby improving the stability and robustness of the mobile medical vehicle. In addition, the present invention solves the problem that the mobile medical vehicle lacks an overall fault detection function and cannot provide a real-time fault warning during use through the above method.

[0030] Preferably, during the start-up inspection phase, the inspection location may be a local control terminal of the mobile medical vehicle, or a remote control terminal or server docked with the mobile medical vehicle.

[0031] In one embodiment of the present invention, the processing of the preset startup check model includes: The startup data of each diagnostic instrument is obtained, the startup data is marked according to a preset tag library, and temporarily stored in a first database; the startup data is analyzed according to the preset startup inspection standard of each diagnostic instrument to determine whether the startup data contains abnormal data; if not, it is in a normal state, and if so, it is in an abnormal state, and a startup repair model is used to analyze and process the abnormal data to restore the corresponding diagnostic instrument state and / or an abnormal detection model is called to analyze the abnormal data to obtain a corresponding comprehensive detection result.

[0032] Preferably, the preset label library is the type corresponding to the monitoring data of each sensor, and the corresponding level when the data is abnormal. The preset startup inspection standards include but are not limited to whether the detection data is in the detection range corresponding to the sensor, whether the sensor is turned on normally, whether the indicator light of the sensor is on, the diagnostic sensitivity of the relevant sensor, etc.

[0033] In this embodiment, during the startup inspection phase, the present invention uses a preset tag library to mark the startup data to achieve classification, so as to analyze and process the startup data later; the startup data is compared and analyzed according to the preset startup inspection standards of each diagnostic instrument to determine whether the startup data contains abnormal data; if it does not contain abnormal data, the abnormal detection phase of the diagnostic process is entered normally; if there is abnormal data, the startup repair model is used to analyze and process the abnormal data to restore the corresponding diagnostic instrument status; when it cannot be repaired, the abnormal detection model is called to analyze the abnormal data to obtain the corresponding comprehensive detection results. The present invention checks the remote medical diagnosis system during the startup phase in the above manner, so that the user can clearly understand the status of the remote medical diagnosis system when using it, and provides convenience for doctors and patients to use various diagnostic instruments on the mobile medical vehicle.

[0034] In one embodiment of the present invention, the use of the startup repair model to analyze and process the abnormal data to restore the corresponding diagnostic instrument state and / or calling the abnormal detection model to analyze the abnormal data to obtain the corresponding comprehensive detection result includes: Determine the type and grade corresponding to the label of the abnormal data; if the type belongs to the first type and the grade belongs to the first preset interval, restart the diagnostic device corresponding to the abnormal data; when the acquired second startup data contains the same abnormal data, call the abnormal detection model to analyze the abnormal data to obtain the corresponding comprehensive detection result; and if the type does not belong to the first type, or the grade does not belong to the first preset interval, call the abnormal detection model to analyze the abnormal data to obtain the corresponding comprehensive detection result.

[0035] In this embodiment, there is only one condition for automatically restarting the diagnostic device: the type corresponding to the label of the abnormal data belongs to the first type and the classification belongs to the first preset interval; when this condition is not met, the abnormal data is analyzed using an abnormal detection model to obtain a corresponding comprehensive detection result.

[0036] Preferably, the label includes a type and a level, wherein the type includes but is not limited to power supply, switch, line, interface, etc.; the level is divided into five levels, level one is normal, level two is ordinary startup abnormality, level three is moderate abnormality, level four is highly abnormal, and level five is severe abnormality; the first preset interval is level one and level two.

[0037] In one embodiment of the present invention, the processing process of the anomaly detection model includes: performing a preliminary analysis on the detection data, status data and record information according to the preset standards corresponding to each detection data to determine whether there is initial abnormal data; if there is initial abnormal data, obtaining in real time the initial abnormal data set of the set time period corresponding to the initial abnormal data; performing a time series analysis on the initial abnormal data set to determine whether the amount of abnormal data is greater than a first preset value; if it is greater than the first preset value, performing a correlation analysis on the initial abnormal data set to obtain a comprehensive detection result.

[0038] Preferably, the set time period is adjusted according to actual needs, and can be selected and determined between 10s and 60s. The preset standard is the standard interval corresponding to each data, including but not limited to the upper and lower limits of the value, the upper and lower limits of the precision, etc.

[0039] Specifically, the initial abnormal data set is a data set formed by the detection data of the preset time period corresponding to the abnormal data; the initial abnormal data set of the set time period is obtained to avoid accidental failures and improve the accuracy of data diagnosis. The initial abnormal data set will mark the abnormalities caused by network transmission, and the marking method includes name marking for known abnormalities and unknown marking for unknown abnormalities.

[0040] Through the configuration of this embodiment, the present invention starts subsequent diagnostic detection steps when abnormal data is detected to reduce the amount of data processing and improve diagnostic efficiency. When abnormal data is detected, the corresponding data of the set time period is detected in real time to avoid occasional failures and improve the accuracy of abnormal detection model detection.

[0041] In one embodiment of the present invention, as shown in the attached Figure 2 As shown, the correlation analysis of the initial abnormal data set to obtain a comprehensive detection result includes the following steps: Step S21: Obtain the type of each data in the initial abnormal data set; determine whether network abnormal data is included according to the type of each data; if so, execute step S22, if not, execute step S23.

[0042] Step S22: using a network detection model to analyze the network anomaly data to obtain a network anomaly detection result.

[0043] Step S23: performing correlation analysis on the corresponding initial abnormal data set based on the preset correlation map and the type of each data to determine the abnormal location.

[0044] Step S24: Determine the abnormality level and abnormality cause according to the abnormality location and the initial abnormality data set; form a comprehensive detection result with the abnormality location, abnormality level, abnormality cause, corresponding abnormal data set and / or network abnormality detection result.

[0045] In step S24, when the abnormal location, abnormal level, abnormal cause, corresponding abnormal data set and / or network abnormality detection results form a comprehensive detection result, redundant and repeated results will be removed to simplify the comprehensive detection result, so as to streamline the data and reduce the data matching amount of the preset decision matching model.

[0046] The network detection model is used to detect the causes of network anomalies, including but not limited to network congestion, data packet loss, network transmission speed, network attacks, etc. Based on the preset association graph and the type of each data, the corresponding initial abnormal data set is analyzed for correlation to detect whether the sensor is faulty, the cause and location of the fault, etc.

[0047] Through the configuration of this embodiment, the present invention sets two modules in the anomaly detection model, one module analyzes and processes the anomaly detection data of each diagnostic instrument in the initial anomaly data set; the other module uses the network detection model to perform diagnostic analysis on the network anomaly data in the initial anomaly data set; the analysis results of the two modules are merged to obtain a comprehensive detection result. The anomaly detection model of the present invention performs division of labor diagnostic analysis in the above manner, which can improve the accuracy of the diagnostic results and the data processing speed.

[0048] In one embodiment of the present invention, the adopting of a network detection model to analyze network anomaly data to obtain a network anomaly detection result includes: The abnormal starting point data in the network abnormal data is obtained in the transmission order; a first network abnormality detection result including the abnormal position and the abnormal cause is determined according to the relationship between the abnormal starting point data and a preset network abnormality matching library; the network abnormality data is classified according to the type of the network abnormality data to obtain a plurality of classified data; matching analysis is performed on each classified data according to the preset network type matching library to determine a second network abnormality detection result including the abnormal type and the abnormal cause; and the first network abnormality detection result and the second network abnormality detection result are fused and analyzed to obtain a network abnormality detection result.

[0049] Preferably, the preset network anomaly matching library includes abnormal data, abnormal positions corresponding to the abnormal data, abnormal causes, etc.; the preset network type matching library is a mapping library constructed according to the abnormal data type, including multiple classification data, abnormal causes corresponding to each classification data, abnormal core positions, etc.

[0050] Among them, matching analysis is performed on each classification data according to the preset network type matching library to determine the second network anomaly detection result containing the anomaly type and the anomaly cause, including: judging the type corresponding to each classification data according to the data in the preset network type matching library to determine the anomaly type, and when the classification data contains unrecorded data, marking it with an unknown method to facilitate targeted repairs during later maintenance. The first network anomaly detection result and the second network anomaly detection result are fused and analyzed to remove duplicate data, so that the obtained network anomaly detection result is streamlined.

[0051] In one embodiment of the present invention, the processing process of the preset decision matching model includes: obtaining the abnormality level in the comprehensive detection results; determining the decision type according to the abnormality level; matching and analyzing the abnormal location and abnormal cause in the comprehensive detection results according to the decision matching library corresponding to the decision type to obtain a fault handling decision, wherein the fault handling decision includes a conventional decision, an emergency decision and a maintenance decision.

[0052] In this embodiment, the routine decision is the decision corresponding to an occasional failure or an abnormal startup failure, and the failure can be eliminated by restarting. The emergency decision is the decision corresponding to when the network is in a blocked state or the diagnostic instrument is in a set range of diagnostic accuracy, which does not affect the final result state, and an emergency time is set; the maintenance failure is the decision corresponding to encountering an unknown network attack or the diagnostic instrument fails and cannot be used.

[0053] In one embodiment of the present invention, the anomaly detection model is constructed based on a convolutional neural network, historical record data and multiple preset rules.

[0054] In this embodiment, a plurality of preset rules include, but are not limited to, preset startup inspection standards for detection data in the above-mentioned embodiments, preset label libraries, preset standards corresponding to detection data in the diagnosis phase, preset association maps, preset network anomaly matching libraries, and preset network type matching libraries. Historical record data includes, but is not limited to, network attack data, network attack type, and fault data known to the diagnostic instrument. Based on the historical record data and a plurality of preset rules, the initial anomaly detection model based on the convolutional neural network is trained and tested to complete the construction of the anomaly detection model.

[0055] In one embodiment of the present invention, the remote medical diagnosis system is adjusted and / or a warning message is issued according to the fault handling decision, including: when the fault handling decision includes a routine decision, the routine decision is used to adjust the diagnostic instrument status and / or network status of the remote medical diagnosis system and a prompt message is sent; when the fault handling decision includes an emergency decision, the emergency decision is used to adjust the corresponding diagnostic instrument status and / or network status in the remote medical diagnosis system, and a warning message is sent; when the fault handling decision only includes a maintenance decision, the current medical diagnosis service is stopped and an alarm message is sent to the staff.

[0056] Preferably, the warning information includes but is not limited to emergency time, accuracy of emergency collection data, etc.

[0057] Through the configuration of this embodiment, the present invention divides the fault handling decisions so as to provide fault decision services for the mobile medical vehicle in a targeted manner, so as to facilitate the reference of the on-site staff and provide convenience for the staff.

[0058] In one embodiment of the present invention, a medical wireless sensor network fault diagnosis method further includes: using an end-to-end encryption method to encrypt and transmit the detection data.

[0059] Specifically, when using end-to-end encryption, encryption can be controlled by the user and not all information needs to be encrypted; data is protected from source to destination across the network; encryption is transparent to network nodes and can also be used during network reorganization, ensuring the security of patient diagnostic data.

[0060] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0061] It should be noted that in the description of the present invention, the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer" and the like indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings, which are only for the convenience of description, and do not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0062] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0063] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0064] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0065] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0066] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0067] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0068] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. A medical wireless sensor network fault diagnosis method, characterized in that: include: When the diagnostic equipment is started, the preset startup check model is used to diagnose each diagnostic instrument and the network status to determine the startup status of the remote medical diagnosis system, which includes a normal state and an abnormal state; When the startup state is normal, normal information is issued, and detection data of each sensor and status data of the wireless sensor network state are collected in real time during the diagnosis process, and the detection data, status data and corresponding record information are transmitted to the remote server; Anomaly detection models are used to diagnose and analyze the test data, status data and record information in the remote server to obtain comprehensive test results of the remote medical diagnosis system; Use the preset decision matching model to analyze the comprehensive detection results to obtain the fault handling decision; Adjust the remote medical diagnosis system and / or issue early warning information based on the fault handling decision.

2. The fault diagnosis method according to claim 1, characterized in that: The processing process of the preset startup check model includes: The startup data of each diagnostic instrument is obtained, the startup data is marked according to a preset tag library, and temporarily stored in a first database; the startup data is analyzed according to the preset startup inspection standard of each diagnostic instrument to determine whether the startup data contains abnormal data; if not, it is in a normal state, and if so, it is in an abnormal state, and a startup repair model is used to analyze and process the abnormal data to restore the corresponding diagnostic instrument state and / or an abnormal detection model is called to analyze the abnormal data to obtain a corresponding comprehensive detection result.

3. The fault diagnosis method according to claim 2, characterized in that: The use of the startup repair model to analyze and process the abnormal data to restore the corresponding diagnostic instrument state and / or calling the abnormal detection model to analyze the abnormal data to obtain the corresponding comprehensive detection result includes: Determine the type and grade corresponding to the label of the abnormal data; if the type belongs to the first type and the grade belongs to the first preset interval, restart the diagnostic device corresponding to the abnormal data; when the acquired second startup data contains the same abnormal data, call the abnormal detection model to analyze the abnormal data to obtain the corresponding comprehensive detection result; and if the type does not belong to the first type, or the grade does not belong to the first preset interval, call the abnormal detection model to analyze the abnormal data to obtain the corresponding comprehensive detection result.

4. The fault diagnosis method according to claim 3, characterized in that: The processing of the anomaly detection model includes: A preliminary analysis is performed on the detection data, status data and record information according to the preset standards corresponding to each detection data to determine whether there is initial abnormal data; if there is initial abnormal data, an initial abnormal data set of a set time period corresponding to the initial abnormal data is obtained in real time; a time series analysis is performed on the initial abnormal data set to determine whether the amount of abnormal data is greater than a first preset value; if it is greater than the first preset value, a correlation analysis is performed on the initial abnormal data set to obtain a comprehensive detection result.

5. The fault diagnosis method according to claim 4, characterized in that: The aforementioned correlation analysis of the initial abnormal data set to obtain a comprehensive detection result includes: Obtain the type of each data in the initial abnormal data set; determine whether network abnormal data is included according to the type of each data; if included, use the network detection model to analyze the network abnormal data to obtain the network anomaly detection result; perform correlation analysis on the corresponding initial abnormal data set based on the preset correlation map and the type of each data to determine the abnormal location; and determine the abnormal level and abnormal cause according to the abnormal location and the initial abnormal data set; form a comprehensive detection result based on the abnormal location, abnormal level, abnormal cause, corresponding abnormal data set and / or network anomaly detection result.

6. The fault diagnosis method according to claim 5, characterized in that: The network anomaly data is analyzed using a network detection model to obtain a network anomaly detection result, including: The abnormal starting point data in the network abnormal data is obtained in the transmission order; a first network abnormality detection result including the abnormal position and the abnormal cause is determined according to the relationship between the abnormal starting point data and a preset network abnormality matching library; the network abnormality data is classified according to the type of the network abnormality data to obtain a plurality of classified data; matching analysis is performed on each classified data according to the preset network type matching library to determine a second network abnormality detection result including the abnormal type and the abnormal cause; and the first network abnormality detection result and the second network abnormality detection result are fused and analyzed to obtain a network abnormality detection result.

7. The fault diagnosis method according to claim 5, characterized in that: The processing process of the preset decision matching model includes: Obtain the abnormality level in the comprehensive detection result; determine the decision type according to the abnormality level; match and analyze the abnormal location and abnormal cause in the comprehensive detection result according to the decision matching library corresponding to the decision type to obtain a fault handling decision, which includes a conventional decision, an emergency decision and a maintenance decision.

8. The fault diagnosis method according to claim 1, characterized in that: The anomaly detection model is built based on a convolutional neural network, historical record data and multiple preset rules.

9. The fault diagnosis method according to claim 7, characterized in that: Adjust the remote medical diagnosis system and / or issue early warning information based on the fault handling decision, including: When the fault handling decision includes routine decisions, the routine decisions are used to adjust the diagnostic instrument status and / or network status of the remote medical diagnosis system and send prompt information; when the fault handling decision includes emergency decisions, the emergency decisions are used to adjust the corresponding diagnostic instrument status and / or network status in the remote medical diagnosis system, and send early warning information; when the fault handling decision only includes maintenance decisions, the current medical diagnosis service is stopped and an alarm message is sent to the staff.

10. The fault diagnosis method according to claim 1, characterized in that: Also includes: End-to-end encryption is used to encrypt and transmit detection data.

Images