Intelligent diagnosis method and system for medical instrument failure

Abstract

The application provides a medical instrument fault intelligent diagnosis method and system, through collecting electromagnetic signals generated by current flowing at each monitoring point; decomposing each electromagnetic signal into multiple electromagnetic components, constructing a spatial gradient field of magnetic field change according to the trend characteristics of each electromagnetic component; obtaining the saturation current of the common mode choke coil in the medical instrument when the magnetic core is saturated, determining the degradation constraint of the saturation margin of the electromagnetic compatibility protection element in the medical instrument according to the time sequence change rate of the magnetic field intensity in the spatial gradient field and the saturation current; mapping the running data of the medical instrument when running into dynamic slow features of long-term aging of electromagnetic compatibility of the medical instrument when running based on the degradation constraint; performing time sequence mutation detection on the dynamic slow features to obtain the fault time of electromagnetic compatibility failure of the medical instrument. By adopting the scheme of the application, the long-term aging trend of the electromagnetic compatibility protection element of the medical instrument can be captured by excluding the influence of short-term rapid change data.

Description

Technical Field

[0001] This application relates to the field of fault diagnosis technology, and more specifically, to a method and system for intelligent fault diagnosis of medical devices. Background Technology

[0002] As core equipment for modern medical diagnosis and treatment, the operational stability of medical devices directly affects diagnostic and treatment accuracy and patient safety. Especially in scenarios with high field strength and multi-module collaborative operation (such as magnetic resonance imaging equipment and high-frequency surgical equipment), the performance degradation of electromagnetic compatibility protection components (such as common-mode chokes and filters) can lead to excessive electromagnetic interference, causing equipment malfunctions such as false triggering and signal distortion. Therefore, real-time fault diagnosis is of great significance. With the development of intelligent technology, intelligent diagnostic methods based on multi-sensor data have gradually become a research hotspot. By analyzing the physical signals during equipment operation, early warning and location of faults can be achieved.

[0003] In existing technologies, fault diagnosis of medical devices mostly relies on monitoring electrical parameters such as voltage and current. While such methods can identify sudden faults, they are difficult to effectively distinguish between short-term electromagnetic fluctuations and long-term aging trends of electromagnetic compatibility protection components. Due to the complex working environment of medical devices, electromagnetic signals are often mixed with short-term rapid interference such as equipment start-up and shutdown and load changes. These interferences can easily mask the slow performance degradation characteristics of protection components caused by magnetic core loss, insulation aging, etc., resulting in insufficient sensitivity of existing methods in capturing long-term aging trends. Therefore, how to capture the long-term aging trend of electromagnetic compatibility protection components of medical devices while eliminating the influence of short-term rapid changes in data has become a difficult problem for the industry. Summary of the Invention

[0004] This application provides a method and system for intelligent diagnosis of medical device faults, which can capture the long-term aging trend of electromagnetic compatibility protection components of medical devices without excluding the influence of short-term rapidly changing data.

[0005] In a first aspect, this application provides an intelligent diagnostic method for medical device malfunctions, comprising:

[0006] Multiple monitoring points are set up on the circuit path of the common-mode choke in the medical device to collect the electromagnetic signals generated by the current flowing through each monitoring point;

[0007] Each electromagnetic signal is decomposed into multiple electromagnetic components. The operating state space of the medical device is constructed based on the trend characteristics of each electromagnetic component. The spatial gradient field of the magnetic field change on the common mode choke current path is determined based on the correlation between each monitoring point in the operating state space.

[0008] The saturation current of the common mode choke in the medical device when the magnetic core is saturated is obtained, and the degradation constraint of the saturation margin of the electromagnetic compatibility protection element in the medical device is determined based on the temporal change rate of the magnetic field strength in the spatial gradient field and the saturation current.

[0009] Real-time monitoring of the operating data of medical devices during operation, and mapping the operating data to the dynamic slow characteristics of long-term electromagnetic compatibility aging of medical devices during operation based on the degradation constraints;

[0010] By performing time-series abrupt change detection on the dynamic slow features, the fault time of electromagnetic compatibility failure of the medical device can be obtained.

[0011] In some embodiments, decomposing each electromagnetic signal into multiple electromagnetic components specifically includes:

[0012] Select an electromagnetic signal as the selected signal, and mark all the maximum and minimum points in the selected signal;

[0013] Interpolation fitting is performed on all maxima and all minima to obtain the center envelope of the selected signal;

[0014] The selected signal is decomposed according to the central envelope to obtain multiple electromagnetic components of the selected signal;

[0015] The remaining electromagnetic signal is further decomposed into multiple electromagnetic components.

[0016] In some embodiments, constructing the operating state space of a medical device based on the trend characteristics of each electromagnetic component specifically includes:

[0017] Determine the trend characteristics of each electromagnetic component;

[0018] Based on all trend characteristics, multiple operational status components of medical devices were selected;

[0019] All operational state components are combined into the operational state space of the medical device.

[0020] In some embodiments, determining the spatial gradient field of the magnetic field change along the common-mode choke current path based on the correlation between monitoring points in the operating state space specifically includes:

[0021] The operating state space is standardized to obtain a standard operating state space;

[0022] Determine the correlation matrix between each monitoring point in the standard operating state space;

[0023] The correlation matrix is ​​decomposed into eigenvalues ​​to obtain an eigenvector matrix and an eigenvalue matrix.

[0024] The spatial gradient field of the magnetic field change on the common-mode choke current path is determined based on the correlation between the monitoring points in the operating state space.

[0025] In some embodiments, determining the degradation constraint of the saturation margin of the electromagnetic compatibility protection element in the medical device based on the temporal rate of change of the magnetic field strength in the spatial gradient field and the saturation current specifically includes:

[0026] The first derivative of the spatial gradient field is used to obtain the time-series rate of change matrix of the magnetic field strength;

[0027] The aging characteristics of electromagnetic compatibility protection components in medical devices are determined based on the time-series rate of change matrix.

[0028] The degradation constraint of the saturation margin of electromagnetic compatibility protection components in medical devices is determined based on the saturation current and the aging characteristics.

[0029] In some embodiments, mapping the operational data to the dynamic slow characteristics of long-term electromagnetic compatibility aging of medical devices based on the degradation constraints specifically includes:

[0030] Convert the running data into a running data matrix;

[0031] The degradation constraint maps the running data matrix into a slow feature matrix;

[0032] The slow feature matrix is ​​decomposed into dynamic slow features of electromagnetic compatibility aging during medical device operation.

[0033] In some embodiments, performing temporal abrupt change detection on the dynamic slow features to obtain the fault moment of electromagnetic compatibility failure of the medical device specifically includes:

[0034] Determine the autocorrelation sequence of the dynamic slow features;

[0035] The autocorrelation sequence is used to identify the mutation points of long-term aging;

[0036] Based on the abrupt change point, the dynamic slow feature is piecewise fitted, and then the fault moment when the electromagnetic and saturation current capacitance of the medical device fails is extracted from the piecewise fitting result.

[0037] In some embodiments, electromagnetic signals generated by the current flowing through each monitoring point are collected by a magnetoresistive sensor.

[0038] In some embodiments, the saturation current of the common-mode choke in the medical device when the magnetic core is saturated is obtained by using a Hall current sensor and a digital bridge tester.

[0039] Secondly, this application provides an intelligent diagnostic system for medical device malfunctions, comprising:

[0040] The acquisition module is used to set up multiple monitoring points on the circuit path of the common mode choke in the medical device and acquire the electromagnetic signal generated by the current flowing through each monitoring point.

[0041] The processing module is used to decompose each electromagnetic signal into multiple electromagnetic components, construct the operating state space of the medical device based on the trend characteristics of each electromagnetic component, and determine the spatial gradient field of the magnetic field change on the common mode choke current path based on the correlation between each monitoring point in the operating state space.

[0042] The processing module is also used to obtain the saturation current of the common mode choke in the medical device when the magnetic core is saturated, and to determine the degradation constraint of the saturation margin of the electromagnetic compatibility protection element in the medical device based on the temporal change rate of the magnetic field strength in the spatial gradient field and the saturation current.

[0043] The processing module is also used to monitor the operating data of the medical device in real time, and map the operating data to the dynamic slow characteristics of the long-term aging of electromagnetic compatibility of the medical device during operation based on the degradation constraint.

[0044] The execution module is used to perform temporal abrupt change detection on the dynamic slow features to obtain the fault time of electromagnetic compatibility failure of the medical device.

[0045] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects:

[0046] The intelligent diagnostic method and system for medical device faults provided in this application firstly sets up multiple monitoring points on the circuit path of the common-mode choke in the medical device, and collects the electromagnetic signals generated by the current flowing through each monitoring point; decomposes each electromagnetic signal into multiple electromagnetic components, constructs the operating state space of the medical device based on the trend characteristics of each electromagnetic component, and determines the spatial gradient field of the magnetic field change on the current path of the common-mode choke based on the correlation between each monitoring point in the operating state space; obtains the saturation current of the common-mode choke in the medical device when the magnetic core is saturated, and determines the degradation constraint of the saturation margin of the electromagnetic compatibility protection element in the medical device based on the temporal change rate of the magnetic field strength in the spatial gradient field and the saturation current; monitors the operating data of the medical device in real time, and maps the operating data to the dynamic slow characteristics of long-term aging of electromagnetic compatibility of the medical device based on the degradation constraint; performs temporal abrupt change detection on the dynamic slow characteristics to obtain the fault time of electromagnetic compatibility failure of the medical device.

[0047] Therefore, this application decomposes and filters electromagnetic signals, first eliminating interference noise, and then uses the remaining electromagnetic components as operational data to construct the operational state space of the medical device. Subsequently, the correlation between the data in the operational state space is analyzed to determine the changes in the internal electromagnetic field of the medical device during operation (i.e., the time-series rate of change). Then, the aging status of the common-mode choke (i.e., the degradation constraint of the saturation margin) is comprehensively evaluated by combining the time-series rate of change and the saturation current of the common-mode choke when the core is saturated. This aging status is then used to map the operational data of the medical device into slow, long-term aging characteristics, eliminating short-term changing features and focusing on the slow aging of the electromagnetic compatibility protection components over long-term operation (i.e., dynamic slow characteristics). Finally, the dynamic slow characteristics are used to identify the moment of electromagnetic compatibility failure in the medical device. In summary, the solution of this application can capture the long-term aging trend of the electromagnetic compatibility protection components of medical devices while eliminating the influence of short-term, rapidly changing data. Attached Figure Description

[0048] Figure 1 This is an exemplary flowchart of a medical device fault intelligent diagnosis method according to some embodiments of this application;

[0049] Figure 2 This is an exemplary flowchart illustrating the decomposition of electromagnetic signals according to some embodiments of this application;

[0050] Figure 3 This is an exemplary flowchart illustrating the determination of degradation constraints according to some embodiments of this application;

[0051] Figure 4 This is a schematic diagram of the structure of a medical device fault intelligent diagnosis system according to some embodiments of this application;

[0052] Figure 5 This is a schematic diagram of the structure of a computer device for implementing an intelligent diagnostic method for medical device faults, according to some embodiments of this application. Detailed Implementation

[0053] To better understand the technical solution of this application, the technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0054] refer to Figure 1 The figure is an exemplary flowchart of a medical device fault intelligent diagnosis method according to some embodiments of this application. The medical device fault intelligent diagnosis method mainly includes the following steps:

[0055] In step 101, multiple monitoring points are set on the circuit path of the common-mode choke in the medical device to collect the electromagnetic signals generated by the current flowing through each monitoring point.

[0056] In specific implementation, setting multiple monitoring points on the circuit path of the common-mode choke in the medical device can be achieved in the following way: magnetoresistive sensors are installed at multiple key locations on the circuit path of the common-mode choke as monitoring points, and each monitoring point is labeled, that is, all monitoring points are marked from 1 to n, where n is the total number of monitoring points. The key locations may include the input terminal, the output terminal, the vicinity of the magnetic core body, and the connector of the cable shield layer connected in series with the choke. In other embodiments, the key locations may also include other locations on the circuit path of the common-mode choke, which is not limited here.

[0057] It should be noted that the numbering of the monitoring points in this application is only a means to facilitate the description of the subsequent implementation. The size of the number does not limit the importance of the monitoring points or other parameters. All monitoring points can be randomly marked as 1 to n, where n is the total number of monitoring points.

[0058] In practice, the electromagnetic signals generated by the current flowing through each monitoring point can be collected in the following way: the sampling frequency of the magnetoresistive sensor at each monitoring point is set to 1kHz, the electromagnetic field strength value during the operation of the medical device is collected by the magnetoresistive sensor at each monitoring point, and all the electromagnetic field strength values ​​collected by the magnetoresistive sensor at each monitoring point are arranged in the order of collection, and the resulting sequences are used as the electromagnetic signals generated by the current flowing through each monitoring point during the operation of the medical device.

[0059] In step 102, each electromagnetic signal is decomposed into multiple electromagnetic components, and the operating state space of the medical device is constructed based on the trend characteristics of each electromagnetic component. The spatial gradient field of the magnetic field change on the common mode choke current path is determined based on the correlation between each monitoring point in the operating state space.

[0060] In some embodiments, reference Figure 2 The figure is an exemplary flowchart illustrating the decomposition of electromagnetic signals according to some embodiments of this application. In this application, the decomposition of each electromagnetic signal into multiple electromagnetic components can be achieved by the following steps:

[0061] In step 1021, an electromagnetic signal is selected as the selected signal, and all the maximum and minimum points in the selected signal are marked;

[0062] In step 1022, interpolation fitting is performed on all maxima and minima to obtain the center envelope of the selected signal;

[0063] In step 1023, the selected signal is decomposed according to the central envelope to obtain multiple electromagnetic components of the selected signal;

[0064] In step 1024, the remaining electromagnetic signal is further decomposed into multiple electromagnetic components.

[0065] In practice, the central envelope of the selected signal can be obtained by interpolating and fitting all the maxima and minima. This can be achieved as follows: First, arrange all the maxima in chronological order and fit them with the cubic spline interpolation method to obtain the upper envelope. Then, arrange all the minima in chronological order and fit them with the cubic spline interpolation method to obtain the lower envelope. Finally, calculate the average curve of the upper and lower envelopes and use this average curve as the central envelope of the selected signal.

[0066] It should be noted that the central envelope in this application is a baseline curve that reflects the overall trend of the electromagnetic signal during the fluctuation process.

[0067] In specific implementation, the selected signal is decomposed according to the central envelope to obtain multiple electromagnetic components of the selected signal. This can be achieved in the following way: using the empirical mode decomposition method in the prior art, the central envelope is used as the mean of the selected signal for elimination, and the screening condition of the intrinsic mode function is set to the number of extreme points and the number of zero crossings not exceeding 1, thereby obtaining multiple intrinsic mode functions of the selected signal, and each obtained intrinsic mode function is used as an electromagnetic component of the selected signal.

[0068] In some embodiments, the operational state space of a medical device can be constructed based on the trend characteristics of each electromagnetic component using the following steps:

[0069] Determine the trend characteristics of each electromagnetic component;

[0070] Based on all trend characteristics, multiple operational status components of medical devices were selected;

[0071] All operational state components are combined into the operational state space of the medical device.

[0072] In practice, the trend characteristics of each electromagnetic component can be determined in the following way: calculate the Hearst exponent of each electromagnetic component using the rescaled range analysis method in the prior art, and use the Hearst exponent of each electromagnetic component as the trend characteristics of each electromagnetic component.

[0073] It should be noted that the trend feature in this application is a parameter value that quantifies the continuous change tendency of the electromagnetic component in the time dimension.

[0074] In practice, the following method can be used to filter out the operating status components of multiple medical devices based on all trend features: according to the principle of the Hearst exponent, the trend threshold is set to 0.5. For each monitoring point, the sum of all electromagnetic components whose trend features differ from the trend threshold by less than 0.1 is taken as the operating status component of the medical device at each monitoring point.

[0075] It should be noted that the operating state component in this application is a comprehensive signal that reflects the stable change trend of electromagnetic characteristics at the corresponding monitoring point during the operation of the medical device.

[0076] It should be noted that in this application, the operating status components correspond one-to-one with the monitoring points, that is, one monitoring point corresponds to one operating status component.

[0077] In practice, combining all the operating state components into the operating state space of the medical device can be achieved in the following way: arrange all the operating state components into a matrix according to the labels of the monitoring points, that is, the operating state component of the monitoring point labeled i is the i-th column in the matrix, and use the resulting matrix as the operating state space of the medical device.

[0078] It should be noted that the operating state space in this application is a collection of electromagnetic signal components that comprehensively reflect the electromagnetic operating state of the common-mode choke and the correlation between various monitoring points.

[0079] In some embodiments, determining the spatial gradient field of the magnetic field change along the common-mode choke current path based on the correlation between monitoring points in the operating state space can be achieved using the following steps:

[0080] The operating state space is standardized to obtain a standard operating state space;

[0081] Determine the correlation matrix between each monitoring point in the standard operating state space;

[0082] The correlation matrix is ​​decomposed into eigenvalues ​​to obtain an eigenvector matrix and an eigenvalue matrix.

[0083] The spatial gradient field of the magnetic field change on the common-mode choke current path is determined based on the correlation between the monitoring points in the operating state space.

[0084] In specific implementation, the standardization of the running state space to obtain the standard running state space can be achieved in the following way: for each column of data in the running state space, calculate the mean and standard deviation of each column of data, then perform Z-score standardization on each column of data using the mean and standard deviation of each column of data, and use the matrix composed of the processed columns of data as the standard running state space.

[0085] It should be noted that the standard operating state space in this application is the operating state space after the result has been standardized.

[0086] In practice, the correlation matrix between each monitoring point in the standard operating state space can be determined in the following way: calculate the covariance matrix of the standard operating state space and use the covariance matrix as the correlation matrix between each monitoring point.

[0087] It should be noted that the correlation matrix in this application is a set of parameter values ​​that measure the degree of mutual influence between the changing trends of electromagnetic signals at various monitoring points of a medical device.

[0088] In specific implementation, the spatial gradient field of the magnetic field change on the common-mode choke current path can be determined according to the correlation between each monitoring point in the operating state space in the following way: First, the operating state space is whitened by the eigenvector matrix and the eigenvalue matrix. That is, the square root of the inverse matrix of the eigenvalue matrix is ​​taken first, and the result is multiplied with the eigenvector matrix. Then, the result of the multiplication is multiplied with the operating state space. The result is used as the whitened result of the operating state space, and the whitened result is used as the spatial gradient field of the magnetic field change on the common-mode choke current path.

[0089] It should be noted that the spatial gradient field in this application is a matrix describing the rate and direction of change of the magnetic field intensity on the common-mode choke current path between different monitoring points when the medical device is working.

[0090] In step 103, the saturation current of the common-mode choke in the medical device when the magnetic core is saturated is obtained, and the degradation constraint of the saturation margin of the electromagnetic compatibility protection element in the medical device is determined based on the temporal change rate of the magnetic field strength in the spatial gradient field and the saturation current.

[0091] In practice, the saturation current of the common-mode choke in a medical device when the magnetic core is saturated can be obtained in the following way: a Hall current sensor is connected in series at the input and output terminals of the common-mode choke, and a digital bridge tester is connected in parallel to monitor the inductance value. When the inductance drops to 80% of the nominal inductance value, the current value at this time is recorded as the saturation current. The nominal inductance value can be obtained directly from the product manual of the medical device.

[0092] In some embodiments, reference Figure 3 The figure is an exemplary flowchart illustrating the determination of degradation constraints according to some embodiments of this application. The determination of degradation constraints for the saturation margin of electromagnetic compatibility protection components in medical devices based on the temporal rate of change of the magnetic field strength in the spatial gradient field and the saturation current can be achieved using the following steps:

[0093] In step 1031, the first derivative of the spatial gradient field is taken to obtain the time-series rate of change matrix of the magnetic field strength;

[0094] In step 1032, the aging characteristics of the electromagnetic compatibility protection components in the medical device are determined based on the time-series rate of change matrix;

[0095] In step 1033, the degradation constraint of the saturation margin of the electromagnetic compatibility protection element in the medical device is determined based on the saturation current and the aging characteristics.

[0096] In specific implementation, the first derivative of the spatial gradient field is used to obtain the time-series rate of change matrix of the magnetic field strength. This can be achieved by the following method: For each row of data in the spatial gradient field, the derivative of each row of data is calculated. The time interval between two adjacent data in each row is the reciprocal of the sampling frequency of the magnetoresistive sensor at each monitoring point. In this application, the sampling frequency is 1 kHz, which means the time interval between two adjacent data in each row is 0.001 seconds. The derivative results of each row of data are rearranged row by row, and the resulting matrix is ​​used as the time-series rate of change matrix of the magnetic field strength.

[0097] It should be noted that the timing rate of change matrix in this application is a set of parameter values ​​describing how quickly the magnetic field strength on the circuit path of the common-mode choke changes with time at each monitoring point location during the operation of the medical device.

[0098] In specific implementation, the aging characteristics of electromagnetic compatibility protection components in medical devices can be determined based on the saturation current and the time-series rate of change matrix in the following manner: First, the covariance matrix of the time-series rate of change matrix is ​​decomposed into eigenvalues ​​to obtain the eigenvector matrix A of the time-series rate of change matrix. Then, the eigenvector matrix B and eigenvalue matrix C obtained after the eigenvalue decomposition of the operating state space are obtained. Then, based on the principle of slow feature analysis, matrix A is multiplied by the -1 / 2 power of matrix B, and the resulting matrix is ​​multiplied by C. The final result of the multiplication is used as the aging characteristics of electromagnetic compatibility protection components in medical devices.

[0099] It should be noted that the aging characteristics in this application are a feature matrix that quantifies the degree of performance degradation of the core material caused by long-term operation of the common-mode choke.

[0100] In specific implementation, the degradation constraint of the saturation margin of the electromagnetic compatibility protection element in the medical device based on the saturation current and the aging characteristics can be achieved in the following way: First, calculate the saturation margin of the electromagnetic compatibility protection element in the medical device. That is, first obtain the nominal value of the saturation current of the common mode choke in the medical device when the magnetic core is saturated. This nominal value can be directly obtained from the product manual of the medical device. Then, record the relative error between the saturation current and the nominal value as the saturation margin q. Perform eigenvalue decomposition on the matrix corresponding to the aging characteristics. Use the quantile of q for all eigenvalues ​​as the threshold. Remove the eigenvectors whose eigenvalues ​​are greater than the threshold. Reconstruct the remaining eigenvectors and eigenvalues ​​into a matrix. Use the reconstructed matrix as the degradation constraint of the saturation margin of the electromagnetic compatibility protection element in the medical device.

[0101] It should be noted that the degradation constraint in this application is a mapping matrix that quantifies the impact of the decay of the saturation margin of the electromagnetic compatibility protection element on the operating data as the aging process progresses. This matrix can remove rapidly changing indicators from the operating data and retain only the data features that can reflect the hidden nature of the long-term aging of the saturation margin of the electromagnetic compatibility protection element.

[0102] In step 104, the operating data of the medical device is monitored in real time, and the operating data is mapped to the dynamic slow characteristics of the long-term aging of electromagnetic compatibility of the medical device during operation based on the degradation constraint.

[0103] In practice, real-time monitoring of the operating data of medical devices can be achieved in the following way: the electromagnetic field strength values ​​during the operation of the medical device are collected by magnetoresistive sensors at each monitoring point on the medical device, and all the electromagnetic field strength values ​​collected by the magnetoresistive sensors at each monitoring point are arranged in the order of collection. The resulting sequences are used as the operating data at each monitoring point during the operation of the medical device. The sampling frequency of the magnetoresistive sensors at each monitoring point is set to 1kHz.

[0104] In some embodiments, mapping the operational data to the dynamic slow characteristics of long-term electromagnetic compatibility aging of medical devices based on the degradation constraints can be achieved through the following steps:

[0105] Convert the running data into a running data matrix;

[0106] The degradation constraint maps the running data matrix into a slow feature matrix;

[0107] The slow feature matrix is ​​decomposed into dynamic slow features of electromagnetic compatibility aging during medical device operation.

[0108] In specific implementation, the operation data can be converted into an operation data matrix in the following way: First, obtain the label of each monitoring point. Then, arrange all the operation data into a matrix according to the label of the monitoring point, that is, the operation data of the monitoring point with label k is the kth column in the matrix, and the resulting matrix is ​​used as the operation data matrix.

[0109] It should be noted that the operational data matrix in this application is a collection of real-time electromagnetic field intensity change data of medical devices categorized and summarized according to monitoring points.

[0110] In a specific implementation, the mapping of the running data matrix to the slow feature matrix through the mapping matrix in the degradation constraint can be achieved in the following way: multiply the mapping matrix in the degradation constraint with the running data matrix, and use the result as the slow feature matrix.

[0111] It should be noted that the slow feature matrix in this application is a set of feature data that focuses on the long-term slow change pattern of electromagnetic compatibility after the operation data is mapped by degradation constraints. This slow feature matrix is ​​used to highlight the long-term slow change trend of electromagnetic characteristics caused by component aging.

[0112] In specific implementation, the slow feature matrix can be decomposed into dynamic slow features of long-term electromagnetic compatibility aging during medical device operation by using the screening parameters in the degradation constraint. The following method can be used: First, the slow feature matrix is ​​summed column by column, and the summed sequence is used as the dynamic slow features of long-term electromagnetic compatibility aging during medical device operation.

[0113] It should be noted that the dynamic slow feature in this application is time series data that reflects the long-term slow dynamic process of electromagnetic properties caused by the aging of electromagnetic protection components.

[0114] In step 105, a temporal abrupt change detection is performed on the dynamic slow feature to obtain the fault time of electromagnetic compatibility failure of the medical device.

[0115] In some embodiments, the following steps can be used to perform temporal abrupt change detection on the dynamic slow features to obtain the fault moment of electromagnetic compatibility failure of the medical device:

[0116] Determine the autocorrelation sequence of the dynamic slow features;

[0117] The autocorrelation sequence is used to identify the mutation points of long-term aging;

[0118] Based on the abrupt change point, the dynamic slow feature is piecewise fitted, and then the fault moment when the electromagnetic and saturation current capacitance of the medical device fails is extracted from the piecewise fitting result.

[0119] In specific implementation, the autocorrelation sequence of the dynamic slow feature can be determined in the following way: calculate the autocorrelation coefficient of the long-term aging data under different time lags using the Pearson autocorrelation coefficient method in the prior art, arrange all autocorrelation coefficients according to the size of the time lag, and use the arranged sequence as the autocorrelation sequence of the dynamic slow feature.

[0120] In specific implementation, the mutation point of long-term aging can be determined by the autocorrelation sequence in the following way: all autocorrelation coefficients in the autocorrelation sequence are compared with a preset correlation threshold in turn. The time lag corresponding to the autocorrelation coefficient that is first less than the correlation threshold is recorded as t. Then, the t-th data is found in the sequence of dynamic slow features and this data is taken as the mutation point of long-term aging.

[0121] In specific implementation, the dynamic slow feature is piecewise fitted according to the mutation point, and the fault time when the electromagnetic and saturated current capacitive failure of the medical device is extracted from the piecewise fitting result can be achieved in the following way: the two segments of data before and after the mutation point are fitted respectively using the least squares algorithm in the prior art, and the time point corresponding to the intersection of the two fitted curves is taken as the fault time when the electromagnetic and saturated current capacitive failure of the medical device is taken as the fault time.

[0122] In another aspect, in some embodiments, this application provides an intelligent diagnostic system for medical device malfunctions, with reference to... Figure 4 The figure is a schematic diagram of the structure of a medical device fault intelligent diagnosis system 400 according to some embodiments of this application. The medical device fault intelligent diagnosis system 400 includes: a data acquisition module 401, a processing module 402, and an execution module 403, which are described below:

[0123] The acquisition module 401 in this application is mainly used to set up multiple monitoring points on the circuit path of the common mode choke in the medical device and to acquire the electromagnetic signal generated by the current flowing through each monitoring point.

[0124] Processing module 402 in this application is mainly used to decompose each electromagnetic signal into multiple electromagnetic components, construct the operating state space of the medical device according to the trend characteristics of each electromagnetic component, and determine the spatial gradient field of the magnetic field change on the common mode choke current path according to the correlation between each monitoring point in the operating state space.

[0125] It should be noted that the processing module 402 in this application is also used to obtain the saturation current of the common mode choke in the medical device when the magnetic core is saturated, and to determine the degradation constraint of the saturation margin of the electromagnetic compatibility protection element in the medical device based on the temporal change rate of the magnetic field strength in the spatial gradient field and the saturation current.

[0126] It should be noted that the processing module 402 in this application is also used to monitor the operating data of the medical device in real time, and map the operating data to the dynamic slow characteristics of the long-term aging of electromagnetic compatibility of the medical device during operation based on the degradation constraint.

[0127] The execution module 403 in this application is mainly used to perform temporal abrupt change detection on the dynamic slow feature to obtain the fault time of electromagnetic compatibility failure of the medical device.

[0128] In addition, this application also provides a computer device, the computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described intelligent diagnostic method for medical device faults.

[0129] In some embodiments, reference Figure 5 The figure is a schematic diagram of the structure of a computer device for implementing an intelligent diagnostic method for medical device faults according to some embodiments of this application. The intelligent diagnostic method for medical device faults in the above embodiments can... Figure 5 The computer device shown is used to implement this, and the computer device 500 includes at least one processor 501, a communication bus 502, a memory 503, and at least one communication interface 504.

[0130] Processor 501 can be a general-purpose central processing unit (CPU) or an application-specific integrated circuit (ASIC).

[0131] The communication bus 502 can be used to transmit information between the aforementioned components.

[0132] Memory 503 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CDROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital versatile optical discs, Blu-ray discs, etc.), magnetic disks or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. Memory 503 may exist independently and be connected to processor 501 via communication bus 502. Memory 503 may also be integrated with processor 501.

[0133] The memory 503 stores program code for executing the solution of this application, and its execution is controlled by the processor 501. The processor 501 executes the program code stored in the memory 503. The program code may include one or more software modules. In the above embodiments, the intelligent diagnosis method for medical device faults can be implemented by the processor 501 and one or more software modules in the program code in the memory 503.

[0134] Communication interface 504 uses any transceiver-like device to communicate with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0135] In a specific implementation, as one example, a computer device may include multiple processors, each of which may be a single-core (single CPU) processor or a multi-core (multi CPU) processor. Here, a processor may refer to one or more devices, circuits, and / or processing cores used to process data (e.g., computer program instructions).

[0136] The aforementioned computer device can be a general-purpose computer device or a special-purpose computer device. In specific implementations, the computer device can be a desktop computer, a portable computer, a network server, a handheld digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This application does not limit the type of computer device.

[0137] In addition, this application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described intelligent diagnostic method for medical device faults.

[0138] In summary, the intelligent diagnostic method and system for medical device faults disclosed in this application firstly sets up multiple monitoring points on the circuit path of the common-mode choke in the medical device, and collects electromagnetic signals generated by the current flowing through each monitoring point; decomposes each electromagnetic signal into multiple electromagnetic components, constructs the operating state space of the medical device based on the trend characteristics of each electromagnetic component, and determines the spatial gradient field of the magnetic field change on the current path of the common-mode choke based on the correlation between each monitoring point in the operating state space; obtains the saturation current of the common-mode choke in the medical device when the magnetic core is saturated, and determines the degradation constraint of the saturation margin of the electromagnetic compatibility protection element in the medical device based on the temporal change rate of the magnetic field strength in the spatial gradient field and the saturation current; monitors the operating data of the medical device in real time, and maps the operating data to the dynamic slow characteristics of long-term aging of electromagnetic compatibility of the medical device based on the degradation constraint; performs temporal abrupt change detection on the dynamic slow characteristics to obtain the fault time of electromagnetic compatibility failure of the medical device.

[0139] Therefore, this application decomposes and filters electromagnetic signals, first eliminating interference noise, and then uses the remaining electromagnetic components as operational data to construct the operational state space of the medical device. Subsequently, the correlation between the data in the operational state space is analyzed to determine the changes in the internal electromagnetic field of the medical device during operation (i.e., the time-series rate of change). Then, the aging status of the common-mode choke (i.e., the degradation constraint of the saturation margin) is comprehensively evaluated by combining the time-series rate of change and the saturation current of the common-mode choke when the core is saturated. This aging status is then used to map the operational data of the medical device into slow, long-term aging characteristics, eliminating short-term changing features and focusing on the slow aging of the electromagnetic compatibility protection components over long-term operation (i.e., dynamic slow characteristics). Finally, the dynamic slow characteristics are used to identify the moment of electromagnetic compatibility failure in the medical device. In summary, the solution of this application can capture the long-term aging trend of the electromagnetic compatibility protection components of medical devices while eliminating the influence of short-term, rapidly changing data.

[0140] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0141] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for intelligent diagnosis of medical device faults, characterized in that, include: Multiple monitoring points are set up on the circuit path of the common-mode choke in the medical device to collect the electromagnetic signals generated by the current flowing through each monitoring point; Each electromagnetic signal is decomposed into multiple electromagnetic components. The operating state space of the medical device is constructed based on the trend characteristics of each electromagnetic component. The spatial gradient field of the magnetic field change on the common mode choke current path is determined based on the correlation between each monitoring point in the operating state space. The saturation current of the common mode choke in the medical device when the magnetic core is saturated is obtained, and the degradation constraint of the saturation margin of the electromagnetic compatibility protection element in the medical device is determined based on the temporal change rate of the magnetic field strength in the spatial gradient field and the saturation current. Real-time monitoring of the operating data of medical devices during operation, and mapping the operating data to the dynamic slow characteristics of long-term electromagnetic compatibility aging of medical devices during operation based on the degradation constraints; By performing temporal abrupt change detection on the dynamic slow features, the fault time of electromagnetic compatibility failure of the medical device can be obtained; Specifically, determining the spatial gradient field of the magnetic field change along the common-mode choke current path based on the correlation between monitoring points in the operating state space includes: The operating state space is standardized to obtain a standard operating state space; Determine the correlation matrix between each monitoring point in the standard operating state space; The correlation matrix is ​​decomposed into eigenvalues ​​to obtain an eigenvector matrix and an eigenvalue matrix. The spatial gradient field of the magnetic field change on the common-mode choke current path is determined based on the correlation between each monitoring point in the operating state space. The degradation constraint for the saturation margin of electromagnetic compatibility protection components in medical devices, determined based on the temporal rate of change of magnetic field strength in the spatial gradient field and the saturation current, specifically includes: The first derivative of the spatial gradient field is used to obtain the time-series rate of change matrix of the magnetic field strength; The aging characteristics of electromagnetic compatibility protection components in medical devices are determined based on the time-series rate of change matrix. The degradation constraint of the saturation margin of electromagnetic compatibility protection components in medical devices is determined based on the saturation current and the aging characteristics. The determination of the aging characteristics of electromagnetic compatibility protection components in medical devices based on the time-series rate of change matrix is ​​achieved in the following manner: First, the covariance matrix of the time-series rate of change matrix is ​​decomposed into eigenvalues ​​to obtain the eigenvector matrix A of the time-series rate of change matrix. Then, the eigenvector matrix B and eigenvalue matrix C obtained by eigenvalue decomposition after standardization of the operating state space are obtained. Then, according to the principle of slow feature analysis, matrix A is multiplied by the -1 / 2 power of matrix B, and the resulting matrix is ​​multiplied by C. The final result of the multiplication is used as the aging characteristics of electromagnetic compatibility protection components in medical devices.

2. The method as described in claim 1, characterized in that, Decomposing each electromagnetic signal into multiple electromagnetic components specifically includes: Select an electromagnetic signal as the selected signal, and mark all the maximum and minimum points in the selected signal; Interpolation fitting is performed on all maxima and all minima to obtain the center envelope of the selected signal; The selected signal is decomposed according to the central envelope to obtain multiple electromagnetic components of the selected signal; The remaining electromagnetic signal is further decomposed into multiple electromagnetic components.

3. The method as described in claim 1, characterized in that, The operational state space of the medical device is constructed based on the trend characteristics of each electromagnetic component, specifically including: Determine the trend characteristics of each electromagnetic component; Based on all trend characteristics, multiple operational status components of medical devices were selected; All operational state components are combined into the operational state space of the medical device.

4. The method as described in claim 1, characterized in that, Mapping the operational data to the dynamic slow characteristics of long-term electromagnetic compatibility aging of medical devices based on the aforementioned degradation constraints specifically includes: Convert the running data into a running data matrix; The degradation constraint maps the running data matrix into a slow feature matrix; The slow feature matrix is ​​decomposed into dynamic slow features of electromagnetic compatibility aging during medical device operation.

5. The method as described in claim 1, characterized in that, By performing temporal abrupt change detection on the aforementioned dynamic slow features, the specific fault moments of electromagnetic compatibility failure in medical devices are obtained, including: Determine the autocorrelation sequence of the dynamic slow features; The autocorrelation sequence is used to identify the mutation points of long-term aging; Based on the abrupt change point, the dynamic slow feature is piecewise fitted, and then the fault moment when the electromagnetic and saturation current capacitance of the medical device fails is extracted from the piecewise fitting result.

6. The method as described in claim 1, characterized in that, Electromagnetic signals generated by the current flowing through each monitoring point are collected using a magnetoresistive sensor.

7. The method as described in claim 1, characterized in that, The saturation current of the common-mode choke in a medical device when the magnetic core is saturated is obtained by using a Hall current sensor and a digital bridge tester.

8. A medical device fault intelligent diagnosis system, which performs intelligent fault diagnosis of medical devices using the method described in any one of claims 1 to 7, characterized in that, The system includes: The acquisition module is used to set up multiple monitoring points on the circuit path of the common mode choke in the medical device and acquire the electromagnetic signal generated by the current flowing through each monitoring point. The processing module is used to decompose each electromagnetic signal into multiple electromagnetic components, construct the operating state space of the medical device based on the trend characteristics of each electromagnetic component, and determine the spatial gradient field of the magnetic field change on the common mode choke current path based on the correlation between each monitoring point in the operating state space. The processing module is also used to obtain the saturation current of the common mode choke in the medical device when the magnetic core is saturated, and to determine the degradation constraint of the saturation margin of the electromagnetic compatibility protection element in the medical device based on the temporal change rate of the magnetic field strength in the spatial gradient field and the saturation current. The processing module is also used to monitor the operating data of the medical device in real time, and map the operating data to the dynamic slow characteristics of the long-term aging of electromagnetic compatibility of the medical device during operation based on the degradation constraint. The execution module is used to perform temporal abrupt change detection on the dynamic slow features to obtain the fault time of electromagnetic compatibility failure of the medical device.

Images