A medical device supervision method, device and storage medium based on AI algorithm

Abstract

The present invention provides a medical equipment supervision method, equipment and storage medium based on an AI algorithm, which belongs to the field of data processing technology, and specifically includes: obtaining the communication protocol of the medical equipment, and determining the directly supervised medical equipment and the indirect supervised medical equipment by the number of medical equipment using the same communication protocol; at least obtaining the power consumption, usage frequency and number of the indirect supervised medical equipment to screen the core medical equipment and the non-core medical equipment, extracting the current characteristic value based on the operating current of the non-core medical equipment, using the diagnosis model based on the AI ​​algorithm to obtain the operating status of the non-core medical equipment, and determining the power quality of the power supply based on the number of non-core medical equipment with abnormal operating status. When there is a suspected abnormality, the power quality is judged based on the current waveform and voltage waveform of the power supply, thereby achieving accuracy and comprehensiveness of medical equipment supervision.

Description

Technical Field

[0001] The present invention belongs to the field of data processing technology, and in particular relates to a medical device supervision method, device and storage medium based on AI algorithm. Background Art

[0002] In order to achieve real-time supervision and dynamic management of medical equipment, in the authorized invention patent CN115361424B "A medical equipment monitoring method and system based on Internet of Things data analysis", the real-time operating status of the first medical equipment is judged by collecting and processing the multi-channel current data of the first medical equipment, thereby improving the real-time efficiency and universality of supervision. However, only using current data often cannot accurately realize the judgment of the operating status of medical equipment. For example, for large medical equipment with high power consumption, the change of power consumption of a certain component often cannot be accurately reflected in the current data. At the same time, changes in the quality of the power supply current may also lead to misjudgment of the operating status of the medical equipment. Therefore, if the medical equipment is supervised only in combination with the current data, the accuracy and reliability of the supervision will not meet the requirements.

[0003] In response to the above technical problems, the present invention provides a medical device supervision method, device and storage medium based on AI algorithm. Summary of the invention

[0004] To achieve the purpose of the present invention, the present invention adopts the following technical solutions:

[0005] According to one aspect of the present invention, a medical device supervision method based on an AI algorithm is provided.

[0006] A medical device supervision method based on AI algorithm, characterized by comprising:

[0007] S11 obtains the communication protocol of the medical device, determines the directly supervised medical device and the indirect supervised medical device by the number of medical devices using the same communication protocol, and supervises the directly supervised medical device based on the communication protocol;

[0008] S12 at least obtains the power consumption, usage frequency, and quantity of the non-directly supervised medical equipment, and uses a prediction model based on an AI algorithm to determine an evaluation value, and uses the evaluation value to screen core medical equipment and non-core medical equipment, and uses the communication protocol of the core medical equipment to supervise the core medical equipment;

[0009] S13 extracts the current characteristic value based on the operating current of the non-core medical device, and based on the current characteristic value, uses the diagnosis model based on the AI ​​algorithm to obtain the operating status of the non-core medical device, and determines the number of non-core medical devices with abnormal operating status to determine whether the power quality of the power supply is suspected to be abnormal. If so, proceed to step S14; if not, output the supervision status based on the operating status of the non-core medical device;

[0010] S14 determines the power quality based on the current waveform and voltage waveform of the power supply, and when there is no abnormality in the power quality, outputs the supervision status based on the operating status of the non-core medical equipment.

[0011] By combining the number of medical devices that use the same communication protocol to determine directly regulated medical devices and indirect regulated medical devices, the screening of directly regulated medical devices is achieved based on the number of medical devices that use the same protocol. While ensuring the efficiency and applicability of supervision, the accuracy of supervision is further improved.

[0012] By obtaining the electricity consumption, usage frequency, and quantity of the non-directly regulated medical equipment, and using a predictive model based on an AI algorithm to determine an evaluation value, and using the evaluation value to screen core medical equipment and non-core medical equipment, it is possible to screen core medical equipment from the perspective of electricity consumption, usage, and quantity of non-directly regulated medical equipment, taking into account not only the amount of electricity consumption, but also the usage, substitutability, and importance from a quantity perspective, thereby ensuring the accuracy and comprehensiveness of the determination of core medical equipment.

[0013] By utilizing the number of non-core medical devices with abnormal operating status to determine whether there are suspected abnormalities in the power quality of the power supply, the impact of the original judgment on the accuracy of medical equipment due to problems with the power quality of the power supply is avoided, the accuracy of the judgment is improved, and unnecessary alarm information is avoided.

[0014] On the other hand, a computer device is provided in an embodiment of the present application, comprising: a memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor, characterized in that: when the processor runs the computer program, it executes the above-mentioned medical device supervision method based on the AI ​​algorithm.

[0015] On the other hand, the present invention provides a computer storage medium having a computer program stored thereon. When the computer program is executed in a computer, the computer is caused to execute the above-mentioned medical device supervision method based on an AI algorithm.

[0016] Other features and advantages will be described in the following description, and partly become apparent from the description, or understood by practicing the invention. The purpose and other advantages of the invention are realized and obtained by the structures particularly pointed out in the description and the drawings.

[0017] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The above and other features and advantages of the present invention will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings.

[0019] Figure 1 is a flow chart of a medical device supervision method based on an AI algorithm according to Example 1;

[0020] Figure 2 is a flowchart of specific steps for determining the evaluation value according to Example 1;

[0021] Figure 3 is a flowchart of specific steps for determining the operating status of a non-core medical device according to Example 1;

[0022] Figure 4 is a flowchart of specific steps of power quality assessment according to Example 1;

[0023] Figure 5 This is a framework diagram of a computer device according to Example 2. DETAILED DESCRIPTION

[0024] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that the present invention will be comprehensive and complete and fully convey the concepts of the example embodiments to those skilled in the art. The same reference numerals in the figures represent the same or similar structures, and thus their detailed description will be omitted.

[0025] The terms "a", "an", "the", and "said" are used to indicate the presence of one or more elements / components / etc.; the terms "comprising" and "having" are used to express an open-ended inclusive meaning and mean that additional elements / components / etc. may be present in addition to the listed elements / components / etc.

[0026] Example 1

[0027] To solve the above problems, according to one aspect of the present invention, Figure 1As shown, according to one aspect of the present invention, a medical device supervision method based on AI algorithm is provided.

[0028] A medical device supervision method based on AI algorithm, characterized by comprising:

[0029] S11 obtains the communication protocol of the medical device, determines the directly supervised medical device and the indirect supervised medical device by the number of medical devices using the same communication protocol, and supervises the directly supervised medical device based on the communication protocol;

[0030] It is understood that directly regulated medical devices and indirect regulated medical devices are determined by the number of medical devices that adopt the same communication protocol, including:

[0031] Medical devices using the same communication protocol are regarded as medical devices with the same protocol, and when the number of medical devices with the same protocol is greater than a preset number, the medical devices with the same protocol are regarded as directly supervised medical devices.

[0032] Generally speaking, if the number of medical devices with the same protocol is greater than 10, the medical devices with the same protocol will be treated as directly regulated medical devices.

[0033] In this embodiment, directly supervised medical devices and indirect supervised medical devices are determined by combining the number of medical devices using the same communication protocol, thereby realizing the screening of directly supervised medical devices based on the number of medical devices using the same protocol. On the basis of ensuring the efficiency and applicability of supervision, the accuracy of supervision is further improved.

[0034] S12 at least obtains the power consumption, usage frequency, and quantity of the non-directly supervised medical equipment, and uses a prediction model based on an AI algorithm to determine an evaluation value, and uses the evaluation value to screen core medical equipment and non-core medical equipment, and uses the communication protocol of the core medical equipment to supervise the core medical equipment;

[0035] Specific examples, such as Figure 2 As shown, the specific steps for determining the evaluation value are:

[0036] S21 obtains the power consumption of the indirect supervision medical device, and determines whether the indirect supervision medical device is a core medical device based on the power consumption. If so, the indirect supervision medical device is regarded as a core medical device. If not, proceed to step S22;

[0037] It should be noted that when the power consumption is high or low, it is not accurate to judge the fault based solely on the power consumption. Therefore, this part of non-directly regulated medical equipment is regarded as core medical equipment.

[0038] S22 determines whether the indirect supervision medical device is a frequently used medical device based on the usage frequency of the indirect supervision medical device, if so, proceeds to step S23, if not, proceeds to step S25;

[0039] It should be noted that when the frequency of use of non-directly regulated medical devices is greater than a certain frequency, they will be treated as high-frequency medical devices. Only when the number of high-frequency medical devices is small or they cannot be replaced, can they be treated as core medical devices.

[0040] S23 determines whether the frequently used medical devices are core medical devices based on the number of the frequently used medical devices. If yes, the frequently used medical devices are core medical devices. If not, proceed to step S24.

[0041] It should be noted that when the number of high-frequency used medical devices is small, it means that they are core medical devices.

[0042] S24: Determine whether the frequently used medical device is a core medical device based on the alternative evaluation result of the frequently used medical device. If yes, take the frequently used medical device as the core medical device. If no, proceed to step S25.

[0043] S25 determines the evaluation value based on the power consumption, usage frequency, quantity, and alternative evaluation results of the non-directly regulated medical equipment and a prediction model based on an AI algorithm.

[0044] To give a specific example, the prediction model based on the AI ​​algorithm adopts the prediction model of the GA-ACO-GRU algorithm, wherein the specific construction steps are as follows:

[0045] Step 1: Initialize the GRU model. Determine the topological structure of the GRU model, use real number encoding to compose the initial weights and thresholds of the GRU model into real number strings, and set the GA algorithm parameters, namely: population size P, crossover probability ρc, mutation probability ρm, etc.

[0046] Step 2: Select the fitness function. The fitness function is the Euclidean distance error between the actual output and the expected output of the training sample. The roulette rule is used to perform fitness proportional selection, and individuals with low fitness are selected to form a new population.

[0047] Step 3: Crossover operation. Randomly select two individuals in the population and perform the operation according to the set crossover probability. The two parent individuals xi and xj perform linear crossover at the dth chromosome point;

[0048] Step 4: Mutation operation. Using Gaussian mutation operator, Gaussian distribution can improve GA algorithm to avoid falling into local optimum, and at the same time has local search capability;

[0049] Step 5: Start GA algorithm iteration. When the set number of iterations is reached, the ACO algorithm pheromone is initialized with the obtained optimization solution, and the initial parameters of the ACO algorithm are set. The independent variables are set to the total number of weight thresholds and the maximum number of iterations N. Amax , pheromone volatility coefficient rou, transfer probability constant ρ, expected value heuristic factor β, etc., and arrange the ant pheromones in ascending order and randomly set the initial position of the ant; if the maximum number of iterations is not reached, proceed to step 3.

[0050] Step 6: Ants search for paths based on the state transition function. When a single ant completes the path search, the local pheromone is updated; when all ants complete the path search, the global pheromone is updated and the evolution information is stored. After reaching the maximum number of evolutions, the optimal solution is used as the initial parameter of the GRU model to determine whether it meets the operating requirements. If so, the algorithm ends and the predicted value is output; otherwise, the GRU model continues to be trained to obtain the optimal solution.

[0051] It should be noted that in the ant colony algorithm, ants search for food and calculate the state transition probability based on the remaining information and heuristic information on the path, and the transition probability determines the next direction to move to ensure the randomness of the ants. The state transition probability is mainly composed of the amount of information and the heuristic factor. The heuristic factor includes the information heuristic factor and the expected heuristic factor. Among them, α is the information heuristic factor, which indicates the degree of influence of pheromones on the path when the ants choose the next path to take; β is the expected heuristic factor, which indicates the degree of influence of heuristic information on path selection. If the α value is too large, the random search ability of the ant will be reduced, because it will choose the path with high pheromone concentration; if the β value is too large, it will easily fall into the local optimum, because it will choose the path closest to the current node.

[0052] Therefore, in order to make full use of the two heuristic factors of pheromone and path, and enable ants to select the next node when the two are balanced, the pheromone heuristic factor and the expected heuristic factor are linearly combined in the state transition probability to obtain the improved ant colony algorithm (Improved Ant Colony Optimization, IACO), which will neither reduce its random search ability nor easily fall into the local optimum. Therefore, the following improvements are made to the state transition probability:

[0053]

[0054] in: represents the probability that ant k moves from city node i to city node j at time t; τ ij(t) represents the pheromone concentration on the path connecting city node i and city node j at time t; ηij(t) is called the heuristic function representing the path connecting city node i and city node j at time t; allowed k (k=1, 2, ..., n) is the set of city nodes that the ants have not passed through; α is the pheromone importance factor; β is the heuristic function importance factor, and N(0,1) is a random number between 0 and 1 that conforms to the normal distribution.

[0055] To give a specific example, the alternative evaluation result is determined based on the types and quantities of medical devices with the same function as the medical device, wherein the more types of medical devices with the same function as the medical device and the more quantities of medical devices with the same function as the medical device, the greater the alternative evaluation result.

[0056] It should be noted that when the evaluation value is greater than a preset value, the non-directly supervised medical device will be regarded as a core medical device.

[0057] In this embodiment, the power consumption, usage frequency, and quantity of the non-directly regulated medical equipment are obtained, and an evaluation value is determined using a predictive model based on an AI algorithm. The evaluation value is used to screen core medical equipment and non-core medical equipment, thereby achieving screening of core medical equipment from the perspective of power consumption, usage, and quantity of non-directly regulated medical equipment. Not only the amount of power consumption is taken into account, but also the usage, substitutability, and importance from a quantity perspective, thereby ensuring the accuracy and comprehensiveness of the determination of core medical equipment.

[0058] S13 extracts the current characteristic value based on the operating current of the non-core medical device, and based on the current characteristic value, uses the diagnosis model based on the AI ​​algorithm to obtain the operating status of the non-core medical device, and determines the number of non-core medical devices with abnormal operating status to determine whether the power quality of the power supply is suspected to be abnormal. If so, proceed to step S14; if not, output the supervision status based on the operating status of the non-core medical device;

[0059] Specific examples, such as Figure 3 As shown, the specific steps for determining the operating status of the non-core medical equipment are:

[0060] S31 extracts current characteristics based on the operating current of the non-core medical device to obtain kurtosis, skewness, specified harmonic level, and frequency, and based on the kurtosis, skewness, specified harmonic level, and frequency, adopts a diagnostic model based on an AI algorithm to obtain the operating status value of the non-core medical device, and determines whether there is any abnormality in the operating status of the non-core medical device based on the operating status value. If so, proceed to step S32, if not, determine the operating status based on the operating status value of the non-core medical device.

[0061] S32: obtaining the operating current waveform of the non-core medical device within a preset time, and converting the operating current waveform within the preset time into a current waveform image, extracting image features based on the current waveform image, and obtaining the image operating state value of the non-core medical device based on the image features by using an image diagnosis model based on a CNN algorithm, and determining whether the operating state of the non-core medical device is abnormal based on the image operating state value, and if so, determining that the operating state of the non-core medical device is abnormal, and if not, proceeding to step S33;

[0062] S33: obtaining a comprehensive operating status value based on the image operating status value and the operating status value, and determining whether the operating status of the non-core medical device is abnormal based on the comprehensive operating status value; if so, determining that the operating status of the non-core medical device is abnormal; if not, proceeding to step S34;

[0063] S34 determines the operating status of the non-core medical device based on the failure rate of the non-core medical device of the same type and the comprehensive operating status value.

[0064] Specifically, the comprehensive operation status value is determined based on the image operation status value and the operation status value by using a mathematical model based on the hierarchical analysis method.

[0065] S14 determines the power quality based on the current waveform and voltage waveform of the power supply, and when there is no abnormality in the power quality, outputs the supervision status based on the operating status of the non-core medical equipment.

[0066] It should be noted that if Figure 4 As shown, the specific steps of the power quality assessment are:

[0067] S41 extracts current characteristics based on the current waveform of the power supply to obtain current kurtosis, current skewness, current specified harmonic level, and current frequency;

[0068] S42 extracts voltage characteristics based on the voltage waveform of the power supply to obtain voltage kurtosis, voltage skewness, voltage specified harmonic level, and voltage frequency;

[0069] S43 constructs an input set based on the current kurtosis, current skewness, current specified harmonic level, current frequency, voltage kurtosis, voltage skewness, voltage specified harmonic level, and voltage frequency, and uses a power quality diagnosis model based on an AI algorithm to determine whether the power quality is abnormal.

[0070] In this embodiment, the number of non-core medical devices with abnormal operating status is used to determine whether there is any suspected abnormality in the power quality of the power supply, thereby avoiding the impact of the accuracy of the judgment on the medical equipment due to problems with the power quality of the power supply, improving the accuracy of the judgment, and avoiding unnecessary alarm information.

[0071] Example 2

[0072] On the other hand, Figure 5 As shown, a computer device is provided in an embodiment of the present application, comprising: a memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor, characterized in that: when the processor runs the computer program, the above-mentioned medical device supervision method based on the AI ​​algorithm is executed.

[0073] Example 3

[0074] On the other hand, the present invention provides a computer storage medium having a computer program stored thereon. When the computer program is executed in a computer, the computer is caused to execute the above-mentioned medical device supervision method based on an AI algorithm.

[0075] In several embodiments provided in the present application, it should be understood that the disclosed system and method can also be implemented in other ways. The system embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, methods and computer program products according to multiple embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and a part of the module, program segment or code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0076] In addition, the functional modules in the various embodiments of the present invention may be integrated together to form an independent part, or each module may exist independently, or two or more modules may be integrated to form an independent part.

[0077] If the function is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0078] Based on the above ideal embodiments of the present invention, the relevant staff can make various changes and modifications without departing from the technical concept of the present invention through the above description. The technical scope of the present invention is not limited to the contents of the specification, and its technical scope must be determined according to the scope of the claims.

Claims

1. A medical device supervision method based on AI algorithm, characterized in that: Specifically include: Obtaining a communication protocol of the medical device, and determining directly supervised medical devices and indirect supervised medical devices by the number of medical devices using the same communication protocol, and supervising the directly supervised medical devices based on the communication protocol; At least the power consumption, usage frequency, and quantity of the non-directly supervised medical equipment are obtained, and an evaluation value is determined using a prediction model based on an AI algorithm, and the evaluation value is used to screen core medical equipment and non-core medical equipment, and the core medical equipment is supervised using the communication protocol of the core medical equipment; Extracting current characteristic values ​​based on the operating current of the non-core medical device, and obtaining the operating status of the non-core medical device using a diagnostic model based on an AI algorithm based on the current characteristic values, and determining the number of non-core medical devices with abnormal operating status to determine whether the power quality of the power supply is suspected to be abnormal, if so, proceeding to the next step, if not, outputting the supervision status based on the operating status of the non-core medical device; The power quality is judged based on the current waveform and the voltage waveform of the power supply, and when there is no abnormality in the power quality, the supervision status is output based on the operating status of the non-core medical device; The specific steps for determining the evaluation value are: S21 obtains the power consumption of the indirect supervision medical device, and determines whether the indirect supervision medical device is a core medical device based on the power consumption. If so, the indirect supervision medical device is regarded as a core medical device. If not, proceed to step S22; S22 determines whether the indirect supervision medical device is a frequently used medical device based on the usage frequency of the indirect supervision medical device, if so, proceeds to step S23, if not, proceeds to step S25; S23 determines whether the frequently used medical devices are core medical devices based on the number of the frequently used medical devices. If yes, the frequently used medical devices are core medical devices. If not, proceed to step S24. S24: Determine whether the frequently used medical device is a core medical device based on the alternative evaluation result of the frequently used medical device. If yes, take the frequently used medical device as the core medical device. If no, proceed to step S25. S25 determines the evaluation value based on the power consumption, usage frequency, quantity, and alternative evaluation results of the non-directly regulated medical equipment and a prediction model based on an AI algorithm.

2. The medical device supervision method according to claim 1, characterized in that: Directly regulated medical devices and indirect regulated medical devices are determined by the number of medical devices that use the same communication protocol, including: Medical devices using the same communication protocol are regarded as medical devices with the same protocol, and when the number of medical devices with the same protocol is greater than a preset number, the medical devices with the same protocol are regarded as directly supervised medical devices.

3. The medical device supervision method according to claim 1, characterized in that: The alternative evaluation result is determined based on the types and quantities of medical devices with the same function as the medical device, wherein the more types of medical devices with the same function as the medical device and the more quantities of medical devices with the same function as the medical device, the larger the alternative evaluation result.

4. The medical device supervision method according to claim 1, characterized in that: When the evaluation value is greater than a preset value, the non-directly supervised medical device is regarded as a core medical device.

5. The medical device supervision method according to claim 1, characterized in that: The specific steps for determining the operating status of the non-core medical equipment are: Extracting current characteristics based on the operating current of the non-core medical device to obtain kurtosis, skewness, specified harmonic level, and frequency, and based on the kurtosis, skewness, specified harmonic level, and frequency, using a diagnostic model based on an AI algorithm to obtain an operating status value of the non-core medical device, and determining whether the operating status of the non-core medical device is abnormal based on the operating status value, if so, proceeding to the next step, if not, determining the operating status based on the operating status value of the non-core medical device; Obtaining the operating current waveform of the non-core medical device within a preset time, and converting the operating current waveform within the preset time into a current waveform image, extracting image features based on the current waveform image, and obtaining the image operating state value of the non-core medical device based on the image features using an image diagnosis model based on a CNN algorithm, and determining whether the operating state of the non-core medical device is abnormal based on the image operating state value, and if so, determining that the operating state of the non-core medical device is abnormal, and if not, proceeding to the next step; Obtaining a comprehensive operating status value based on the image operating status value and the operating status value, and determining whether the operating status of the non-core medical device is abnormal based on the comprehensive operating status value, and if so, determining that the operating status of the non-core medical device is abnormal, and if not, proceeding to the next step; The operating status of the non-core medical device is determined based on the failure rate of the non-core medical device of the same type and the comprehensive operating status value.

6. The medical device supervision method according to claim 1, characterized in that: The specific steps of the power quality assessment are: Extracting current characteristics based on the current waveform of the power supply to obtain current kurtosis, current skewness, current specified harmonic level, and current frequency; Extracting voltage characteristics based on the voltage waveform of the power supply to obtain voltage kurtosis, voltage skewness, voltage specified harmonic level, and voltage frequency; An input set is constructed based on the current kurtosis, current skewness, current specified harmonic level, current frequency, voltage kurtosis, voltage skewness, voltage specified harmonic level, and voltage frequency, and a power quality diagnosis model based on an AI algorithm is used to determine whether the power quality is abnormal.

7. A computer device comprising: A memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor, characterized in that the processor executes the medical device supervision method described in any one of claims 1-6 when running the computer program.

8. A computer storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to execute the medical device supervision method according to any one of claims 1 to 6.

Images