Medical equipment maintenance inspection method and system

Abstract

The invention discloses a medical equipment maintenance inspection method and system, and relates to the technical field of medical equipment. The system comprises a data acquisition and transmission module, a data processing and analysis module, an inspection task distribution module, a maintenance plan making module and a real-time interaction and feedback module. The method comprises the following steps: step 1, data acquisition and transmission; step 2, data processing and analysis; step 3, making a maintenance plan; step 4, distributing inspection tasks; step 5, inspection execution and feedback; and step 6, system optimization and continuous improvement. Through multi-dimensional data fusion and an advanced machine learning algorithm, faults of the medical equipment can be predicted more accurately, maintenance measures are taken in advance, the influence of the equipment faults on medical services is reduced, multiple factors are comprehensively considered by adopting an analytic hierarchy process, a personalized maintenance plan is made, the utilization efficiency of maintenance resources is improved, and the maintenance cost is reduced. And the maintenance cost is reduced.

Description

Technical Field

[0001] The present invention relates to the field of database technology, and in particular to a medical equipment maintenance and inspection method and system. Background Art

[0002] Today, as medical equipment becomes more and more numerous and complex, relying solely on manual maintenance management can easily lead to confusion and delays in maintenance work.

[0003] The invention patent with announcement number CN111816290A discloses a method and system for the maintenance and inspection of medical equipment. Through the small program of the mobile terminal, the information carrier on the medical equipment is scanned and interpreted, and then communicated with the cloud platform. The cloud platform judges the urgency of the maintenance and recommends the engineer to perform related operations such as repair, maintenance, acceptance, inspection, maintenance, and spare parts replacement of the medical equipment. The present invention is the first system to realize the maintenance and inspection of hospital medical equipment based on the small program, which meets the actual needs of hospital work, does not cause maintenance delays and maintenance sequence confusion, and ensures the normal operation of the hospital in an orderly manner. However, the patent only relies on the basic scanning code trigger process and simple rule sorting (such as department priority), lacks real-time monitoring and data analysis of the equipment operation status, and cannot realize fault prediction and dynamic maintenance strategy. For example, sensor data fusion technology is not used, and machine learning models are not built for fault prediction, resulting in passive response, and potential risks cannot be discovered in advance. Moreover, its maintenance process is limited to the management of a single hospital, and a collaborative mechanism for multiple devices and departments has not been established, and external resources (such as manufacturer technical support) have not been integrated. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides a medical equipment maintenance and inspection method and system, which realizes accurate maintenance, efficient inspection and full life cycle management of medical equipment through multi-dimensional data fusion, intelligent algorithm analysis and real-time interactive feedback, improves the reliability and availability of medical equipment, and solves the existing problems.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: A medical equipment maintenance and inspection system, comprising: Data collection and transmission module, used to collect the operation data of medical equipment in real time and transmit it to the data processing center through a secure network; Data processing and analysis module, used to clean, extract features and predict faults of collected data; The inspection task allocation module uses genetic algorithms to optimize task allocation based on the inspectors’ skills, location, and workload information; The maintenance plan formulation module uses the hierarchical analysis method to determine the weight of each factor based on the equipment's fault prediction results, service life, and performance indicators, and formulates a personalized maintenance plan; The real-time interaction and feedback module is used for inspectors to upload inspection data and equipment status information in real time. The system processes and responds to problems in real time and feeds back the processing results to inspectors and relevant managers.

[0006] Preferably, the data cleaning in the data processing and analysis module adopts a statistically based outlier detection method. , then the data point is an outlier and is processed, where Indicates the first data points, represents the mean of the data series, which is obtained by summing all data points and dividing by the number of data points. is the set threshold, usually set to 3, used to determine whether a data point is an outlier. It represents the standard deviation of the data series and reflects the degree of dispersion of the data.

[0007] Preferably, the analytic hierarchy process in the maintenance plan formulation module constructs a judgment matrix , by solving the maximum eigenvalue of the judgment matrix and the corresponding eigenvector , get the weight of each factor, where Represents the judgment matrix, which is a A matrix is ​​used to compare the relative importance of each factor. Indication factors Relative to factors The importance of is usually 1-9 and its reciprocal. Indicates the number of factors, and in the judgment matrix, indicates the order of the matrix. By solving the maximum eigenvalue of the judgment matrix and the corresponding eigenvector , and obtain the weight of each factor.

[0008] Preferably, the genetic algorithm in the inspection task allocation module defines a fitness function It is a comprehensive evaluation index for task allocation schemes, used to evaluate the pros and cons of each task allocation scheme.

[0009] A medical equipment maintenance and inspection method comprises the following steps: Step 1: Data collection and transmission: Use sensors to collect the operating data of medical equipment, generate verification codes and transmit them to the data processing center; Step 2: Data processing and analysis: verify, clean, extract features and predict faults of the transmitted data; Step 3: Formulate a maintenance plan: Comprehensively consider the equipment's fault prediction results, service life, performance indicators and other factors, use the analytic hierarchy process to determine the weight of each factor, and formulate a personalized maintenance plan; Step 4: Inspection task allocation: Based on the inspection personnel’s skills, location, workload and other information, a genetic algorithm is used to optimize task allocation; Step 5: Inspection execution and feedback: The inspectors carry mobile terminals for inspection and upload equipment information in real time. The system handles problems and provides feedback in real time. Step 6. System optimization and continuous improvement: Regularly conduct statistical analysis on system operation data and inspection and maintenance results to optimize system models and algorithms.

[0010] Preferably, the fault prediction in the data processing and analysis step adopts a support vector regression algorithm, and the optimization objective function is , the constraints are ,in, Represents a weight vector, which is used to indicate the importance of the input feature. is the bias term, in the function It plays the role of adjusting the position of the function. Represents a slack variable, which is used to deal with noise and outliers in the data and allows some data points to deviate from the prediction function within a certain range. is the penalty factor, which controls the degree of penalty for the error and balances the complexity and error of the model. is the actual value of the i-th sample, is the input feature vector of the i-th sample, Indicates that the input feature Functions mapped to high-dimensional space, For the parameters of the insensitive loss function, a range that does not consider the error is defined. Indicates the number of samples.

[0011] Preferably, the maintenance plan priority calculation formula in the maintenance plan formulation step is: ,in, The priority of the maintenance plan. The higher the value, the more urgent the maintenance needs of the equipment. Indicates The weight of each factor is determined by the hierarchical analysis method, which reflects the importance of the factor in formulating the maintenance plan. Indicates The evaluation value corresponding to each factor is the specific evaluation score of the device in this factor. It represents the number of factors, that is, the number of factors that affect the maintenance plan that are considered comprehensively.

[0012] Preferably, the genetic algorithm in the inspection task allocation step defines a fitness function It is a comprehensive evaluation index of the task allocation scheme. The optimal task allocation scheme is found through iterative operations such as selection, crossover, and mutation.

[0013] Preferably, the system optimization and continuous improvement step optimizes and adjusts the fault prediction model, maintenance plan formulation method, inspection task allocation algorithm, etc. based on the statistical analysis results.

[0014] Beneficial Effects The present invention provides a medical equipment maintenance and inspection method and system. Compared with the prior art, it has the following beneficial effects: through multi-dimensional data fusion and advanced machine learning algorithms, it can more accurately predict the failure of medical equipment, take maintenance measures in advance, reduce the impact of equipment failure on medical services, adopt hierarchical analysis method to comprehensively consider multiple factors, formulate personalized maintenance plans, improve the utilization efficiency of maintenance resources, reduce maintenance costs, use genetic algorithms to optimize the allocation of inspection tasks, make reasonable arrangements according to the actual situation of inspection personnel, and improve inspection efficiency and quality; Through the real-time interaction and feedback module, inspectors can upload equipment information in a timely manner, and the system can respond quickly and provide solutions to ensure that problems are handled in a timely manner. The present invention can regularly evaluate and optimize the system, continuously improve the performance and adaptability of the system, and provide strong guarantees for the long-term stable operation of medical equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a schematic diagram of system connection of the present invention; Figure 2 It is the overall framework diagram of the method of the present invention; Figure 3 is a flow chart of the data processing and analysis steps of the present invention; Figure 4 Develop a flowchart of steps for the maintenance plan of the present invention; Figure 5 A flowchart of the inspection task allocation steps of the present invention; Figure 6 It is a flow chart of the inspection execution and feedback steps of the present invention. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] See also Figure 1-6 , the present invention provides three technical solutions: The first implementation method: a medical equipment maintenance and inspection system, comprising: Data collection and transmission module, which is responsible for real-time collection of various operating data of medical equipment, including the working status, performance parameters, usage time, etc. of the equipment. At the same time, the collected data is transmitted to the data processing center through a secure and stable network. In order to ensure the accuracy and integrity of the data, the data verification algorithm is adopted: ; in, It is a checksum used to verify whether errors occur during data transmission. Indicates data elements, that is, each specific value in the collected data, Indicates the number of data elements, that is, how many values ​​are there in the collected data. Represents a preset modulus, which is a fixed value used for modulo operations.

[0018] Data processing and analysis module, used to clean, extract features and predict faults of collected data; Data cleaning: Clean the collected data to remove noise and outliers. Use the statistical outlier detection method and assume the data sequence is , calculate its mean and standard deviation , for the data point ,like , then the data point is considered as an outlier and processed, where Indicates the first data points, represents the mean of the data series, which is obtained by summing all data points and dividing by the number of data points. is the set threshold, usually set to 3, used to determine whether a data point is an outlier. It represents the standard deviation of the data series and reflects the degree of dispersion of the data.

[0019] Feature extraction: Extract features that can reflect the operating status of the equipment from the cleaned data. For example, for the temperature data of the equipment, extract its mean, variance, maximum value, minimum value and other features.

[0020] Fault prediction: A fault prediction model is built based on a machine learning algorithm. The support vector regression (SVR) algorithm is used, and its goal is to find a function , so that the error between the predicted value and the actual value is minimized. The optimization objective function is: ; The constraints are: ; in, Represents a weight vector, which is used to indicate the importance of the input feature. is the bias term, in the function It plays the role of adjusting the position of the function. Represents a slack variable, which is used to deal with noise and outliers in the data and allows some data points to deviate from the prediction function within a certain range. is the penalty factor, which controls the degree of penalty for the error and balances the complexity and error of the model. For the The actual value of the samples, For the The input feature vector of samples, Indicates that the input feature Functions mapped to high-dimensional space, For the parameters of the insensitive loss function, a range that does not consider the error is defined. Indicates the number of samples.

[0021] The maintenance plan formulation module uses the analytic hierarchy process (AHP) to determine the weight of each factor based on the equipment's fault prediction results, service life, performance indicators and other factors, and formulates a personalized maintenance plan. The analytic hierarchy process in the maintenance plan formulation module constructs a judgment matrix ,in, Represents the judgment matrix, which is a A matrix is ​​used to compare the relative importance of each factor. Indication factors Relative to factors The importance of is usually 1-9 and its reciprocal. Indicates the number of factors, and in the judgment matrix, indicates the order of the matrix. By solving the maximum eigenvalue of the judgment matrix and the corresponding eigenvector , and obtain the weight of each factor. The maintenance plan priority calculation formula is: ; in, The priority of the maintenance plan. The higher the value, the more urgent the maintenance needs of the equipment. Indicates The weight of each factor is determined by the hierarchical analysis method, which reflects the importance of the factor in formulating the maintenance plan. Indicates The evaluation value corresponding to each factor is the specific evaluation score of the device in this factor. It represents the number of factors, that is, the number of factors that affect the maintenance plan that are considered comprehensively.

[0022] The inspection task allocation module uses genetic algorithm to optimize task allocation based on the skills, location, workload and other information of the inspection personnel. , the inspection task set is , the skill vector of each inspector is , the skill requirement vector of the task is , define the fitness function It is a comprehensive evaluation index for the task allocation scheme. Through the selection, crossover, mutation and other operations of the genetic algorithm, the optimal task allocation scheme is found through continuous iteration.

[0023] In the real-time interaction and feedback module, during the inspection process, the inspectors upload the inspection data and equipment status information in real time through the mobile terminal. The system processes and responds to the problems reported by the inspectors in real time, provides solutions and guidance, and promptly feeds back the processing results to the inspectors and relevant managers. The message queue mechanism is used to ensure the reliable transmission and processing of information.

[0024] Second embodiment: The present invention also discloses a medical equipment maintenance and inspection method, comprising the following steps: Step 1: Data collection and transmission 1) Sensor selection and installation: Equipment evaluation: Conduct a comprehensive evaluation of each medical device in the hospital, including the type, function, structure, working environment, etc. For example, for CT equipment, it is necessary to understand the working characteristics of key components such as the tube, detector, and scanning frame; for MRI equipment, it is necessary to consider the impact of the magnetic field environment on the sensor.

[0025] Sensor adaptation: According to the equipment evaluation results, select appropriate sensors for different equipment and key components. For example, to monitor the temperature of the equipment, a high-precision thermocouple sensor is selected, with a measurement range of -20℃ to 150℃ and an accuracy of ±0.1℃; for vibration detection of the equipment, a piezoelectric vibration sensor is used with a range of 0-5g.

[0026] Installation location determination: Determine the best installation location for the sensor to ensure that the operating data of the equipment can be accurately collected. For CT tubes, install the temperature sensor on the tube shell close to the heat source; for MRI gradient coils, install the vibration sensor on the fixed bracket of the coil.

[0027] Installation and debugging: Install the sensor on the equipment according to the determined installation position and debug it. Check whether the output signal of the sensor is normal to ensure the accuracy and stability of data collection.

[0028] 2) Data collection frequency setting Equipment classification: Medical equipment is classified according to its importance, operational stability, degree of failure impact, etc. For example, it can be divided into key equipment (such as CT, MRI), common equipment (such as ultrasound diagnostic equipment, ECG monitor) and auxiliary equipment (such as infusion pump, sphygmomanometer).

[0029] Basic frequency setting: Set the basic data collection frequency for different types of equipment. For key equipment, due to its high importance to medical diagnosis and treatment and its complex operation process, a higher collection frequency is set, such as collecting data every 5 minutes; for commonly used equipment, the collection frequency can be set to once every 10 minutes; for auxiliary equipment, the collection frequency can be appropriately reduced to once every 30 minutes.

[0030] Dynamic adjustment mechanism: Establish a dynamic adjustment mechanism for data collection frequency. Adjust the collection frequency in real time based on factors such as the equipment's operating status and historical fault records. For example, when the equipment experiences abnormal fluctuations or approaches the time of a historical fault, the collection frequency is automatically increased to capture the equipment's operating changes in a more timely manner.

[0031] 3) Data preprocessing Data cleaning: Clean the collected raw data to remove noise and outliers. Use statistical analysis methods, such as calculating the mean and standard deviation of the data, setting a reasonable threshold range, and treating data that exceeds the threshold as outliers and eliminating or correcting them.

[0032] Data normalization: Normalize the cleaned data to unify data of different types and magnitudes into the same scale. Commonly used normalization methods include minimum-maximum normalization and Z-score normalization to improve the accuracy and efficiency of subsequent data analysis.

[0033] Data encoding: For some non-numeric data, such as the operating status of the equipment (normal, faulty, under maintenance), operating mode, etc., encoding processing is performed to convert them into numerical data to facilitate computer processing and analysis.

[0034] 4) Data encryption and transmission Encryption algorithm selection: Select a suitable data encryption algorithm to ensure the security of data during transmission. Use the Advanced Encryption Standard (AES) algorithm with a key length of 256 bits to encrypt the collected data.

[0035] Encryption process: At the data collection end, the pre-processed data is encrypted using the encryption key to generate ciphertext data. During the encryption process, the initialization vector (IV) is combined to increase the randomness and security of the encryption.

[0036] Transmission network selection: According to the actual situation of the hospital, choose a stable and reliable transmission network, such as the hospital's local area network, Wi-Fi network or 4G / 5G mobile network. For data transmission of key equipment, wired network is preferred to ensure the stability and real-time performance of data transmission.

[0037] Transmission protocol configuration: Configure appropriate transmission protocols, such as TCP / IP, to ensure that data can be accurately transmitted to the data processing center. During the transmission process, use data verification mechanisms, such as cyclic redundancy check (CRC), to verify the transmitted data to ensure data integrity.

[0038] Step 2: Data processing and analysis 1) Data storage and management Database selection: Select a suitable database management system based on the characteristics of the data and analysis requirements. For the operation data of medical equipment, you can choose a relational database (such as MySQL, Oracle) or a non-relational database (such as MongoDB, Redis).

[0039] Data storage structure design: Design a reasonable data storage structure to classify and store the collected data according to dimensions such as device type, timestamp, data type, etc. For example, create an independent data table for each device to record detailed information on its operating data.

[0040] Data backup and recovery strategy: Develop a comprehensive data backup and recovery strategy, and regularly back up the data in the database to prevent data loss. The backup data is stored in different physical locations, such as remote data centers or external storage devices. At the same time, a data recovery mechanism is established to ensure rapid recovery when data is lost or damaged.

[0041] 2) Feature extraction Time domain feature extraction: Extract time domain features from the device's operating data, such as mean, variance, maximum, minimum, peak, etc. These features can reflect the device's operating status in the time dimension. For example, the device's temperature mean can reflect its overall heating situation, and the variance can reflect the degree of temperature fluctuation.

[0042] Frequency domain feature extraction: Perform Fourier transform on the data, convert it to the frequency domain, and extract frequency domain features, such as spectrum peak, center frequency, bandwidth, etc. Frequency domain features can help analyze the vibration frequency components of the equipment and determine whether the equipment has abnormal vibration.

[0043] Time-frequency domain feature extraction: Wavelet transform and other methods are used to extract the time-frequency domain features of the data, while considering the information of both time and frequency dimensions. The time-frequency domain features can more accurately capture the changes in the operating status of the equipment, which is of great significance for the early diagnosis of faults.

[0044] 3) Fault prediction model construction Model selection: Select an appropriate fault prediction model based on the characteristics of the equipment and the type of fault. For equipment data with complex nonlinear relationships, a neural network model (such as a multi-layer perceptron, long short-term memory network) can be used; for situations with small data volumes and low feature dimensions, models such as support vector machines and decision trees can be used.

[0045] Data partitioning: historical data is divided into training set, validation set and test set. The training set is used to train the model, the validation set is used to adjust the parameters of the model, and the test set is used to evaluate the performance of the model. Generally speaking, the ratio of training set, validation set and test set can be set to 7:2:1.

[0046] Model training and optimization: Use the training set to train the selected model, and optimize the model's performance by continuously adjusting the model's parameters. Use methods such as cross-validation and grid search to find the optimal combination of model parameters. At the same time, use the validation set to evaluate the model, monitor the model's training process, and prevent overfitting or underfitting.

[0047] Model evaluation: Use the test set to evaluate the trained model, and use indicators such as accuracy, recall, F1 value, and mean square error to evaluate the model's prediction performance. Based on the evaluation results, further optimize and adjust the model.

[0048] 4) Data analysis and visualization Data analysis methods: Use statistical analysis methods, machine learning algorithms, etc. to conduct in-depth analysis on the extracted feature data and fault prediction results. For example, through correlation analysis, find out the correlation between different equipment parameters; through cluster analysis, classify the operating status of the equipment.

[0049] Selection of visualization tools: Choose appropriate visualization tools, such as Tableau, PowerBI, Matplotlib, etc., to display the analysis results in the form of intuitive charts, reports, etc. For example, use a line chart to display the temperature change trend of the equipment, and use a bar chart to compare the failure rates of different equipment.

[0050] Visual display design: Design a reasonable visual display interface to display different analysis results according to different user roles and needs. For equipment managers, focus on displaying the overall operating status of the equipment, fault distribution, etc.; for maintenance personnel, focus on displaying equipment fault prediction results, maintenance suggestions, etc.

[0051] Step 3: Maintenance plan formulation 1) Equipment evaluation Equipment health status assessment: Comprehensively assess the health status of the equipment based on the output of the fault prediction model, the equipment's historical fault records, operating time and other factors. The health status of the equipment is divided into three levels: good, average, and poor, providing a basis for the formulation of maintenance plans.

[0052] Equipment importance assessment: Consider the importance of equipment in hospital medical services, such as the frequency of use of equipment, the degree of impact on medical diagnosis and treatment, etc., and conduct an importance assessment on equipment. Equipment is divided into three levels: key equipment, important equipment and general equipment. Different levels of equipment are given different priorities in the maintenance plan.

[0053] 2) Maintenance strategy formulation Preventive maintenance strategy: For equipment that is in good health and of high importance, formulate a preventive maintenance strategy. Regularly clean, calibrate, lubricate and perform other maintenance work on the equipment, check the key components of the equipment, and replace aging parts in time to prevent equipment failures.

[0054] Corrective maintenance strategy: For equipment that has already failed or is in poor health, a corrective maintenance strategy is formulated. Maintenance personnel are quickly organized to repair the equipment, replace the faulty parts, and restore the normal operation of the equipment. At the same time, the cause of the failure is analyzed and corresponding improvement measures are taken to prevent similar failures from happening again.

[0055] Predictive maintenance strategy: Combine the results of the fault prediction model to perform predictive maintenance on the equipment. When the predicted probability of equipment failure exceeds a certain threshold, maintenance work is arranged in advance to avoid the impact of equipment failure on medical services.

[0056] 3) Maintenance plan scheduling Schedule: Reasonably arrange the time of maintenance plan according to the equipment operation plan, availability of maintenance resources and other factors. For key equipment, try to choose to carry out maintenance during the low period of hospital business to reduce the impact on medical services; for maintenance tasks of multiple equipment, make overall arrangements to improve the utilization efficiency of maintenance resources.

[0057] Resource allocation: According to the requirements of maintenance tasks, appropriate maintenance personnel, maintenance tools and maintenance materials are allocated. Ensure that maintenance personnel have appropriate skills and experience, and that maintenance tools and materials are sufficient and applicable.

[0058] 4) Maintenance plan approval and release Approval process: Submit the prepared maintenance plan to relevant departments and leaders for approval. During the approval process, review the rationality, feasibility and economy of the maintenance plan to ensure that the maintenance plan meets the actual situation and management requirements of the hospital.

[0059] Release and notification: After the maintenance plan is approved, it will be released to the relevant maintenance personnel, equipment management personnel and users in a timely manner. The specific content and schedule of the maintenance plan will be notified to the relevant personnel through the hospital's internal management system, email, text message, etc.

[0060] Step 4: Inspection task allocation 1) Inspection personnel information management Personnel information collection: Collect basic information of inspection personnel, including name, gender, age, contact information, etc.; professional skills information, such as maintenance skills, equipment types mastered, etc.; work experience information, such as maintenance projects participated in, fault cases handled, etc.

[0061] Personnel skill assessment: Regularly assess the skill level of patrol inspectors, using theoretical examinations and practical operation assessments to assess their maintenance and fault diagnosis capabilities for different equipment. Based on the assessment results, patrol inspectors are divided into skill levels, such as elementary, intermediate, and advanced.

[0062] Personnel training and development: Develop targeted training plans based on the inspection personnel's skill assessment results and the hospital's equipment upgrades. Regularly organize inspection personnel to participate in training courses, technical exchange activities, etc. to improve their skills and overall quality.

[0063] 2) Inspection task requirements analysis Determine equipment inspection requirements: Determine the equipment list and inspection content that needs to be inspected based on the maintenance plan and the operating status of the equipment. The inspection content includes equipment appearance inspection, operating parameter monitoring, functional testing, etc.

[0064] Task difficulty assessment: Evaluate the difficulty of each inspection task, taking into account factors such as the complexity of the equipment, the risk of failure, the skills and time required for inspection, etc. Inspection tasks are divided into three levels: simple, medium, and complex.

[0065] 3) Inspection task allocation algorithm Algorithm selection: Intelligent algorithms are used to allocate inspection tasks, such as genetic algorithms, ant colony algorithms, etc. These algorithms can comprehensively consider factors such as the skill level, location, and workload of inspection personnel, optimize task allocation plans, and improve inspection efficiency.

[0066] Objective function setting: Set the objective function of inspection task allocation, such as minimizing the total travel time of inspection personnel, maximizing the completion quality of inspection tasks, etc. By optimizing the objective function, find the optimal task allocation plan.

[0067] Consider constraints: In the process of task allocation, various constraints are considered, such as the working time limit of inspection personnel, skill matching requirements, equipment inspection time window, etc. Ensure that the task allocation plan meets the actual situation and management requirements.

[0068] 4) Adjustment and confirmation of task assignment results Result adjustment: Adjust the task allocation results generated by the algorithm according to the actual situation. For example, consider factors such as the personal wishes of the inspectors and emergencies, and make appropriate adjustments to the task allocation.

[0069] Confirmation and notification: The inspection personnel will be notified of the adjusted task allocation results and asked to confirm. After the inspection personnel confirm, the task allocation results will be officially effective.

[0070] Step 5: Inspection execution and feedback: 1) Inspection preparation Task confirmation: Before performing inspection tasks, inspection personnel should carefully confirm the task assignment results and understand the equipment list, inspection content, inspection time requirements, etc.

[0071] Tool and material preparation: According to the requirements of the inspection task, prepare the corresponding inspection tools and maintenance materials, such as multimeters, screwdrivers, calibration parts, spare parts, etc.

[0072] Safety protection measures: Wear necessary safety protection equipment, such as safety helmets, gloves, goggles, etc., to ensure personal safety during the inspection process.

[0073] 2) Inspection execution On-site inspection: Inspectors arrive at the equipment site for inspection according to the inspection task requirements. They strictly follow the inspection standards and procedures and conduct a comprehensive inspection of the equipment's appearance, operating parameters, functions, etc.

[0074] Data collection and recording: During the inspection process, mobile terminal devices (such as tablets and smart phones) are used to collect equipment operation data and inspection information in real time, such as equipment temperature, pressure, vibration and other parameters, as well as problems and abnormal conditions found during the inspection process. At the same time, the inspection process is recorded by taking photos and videos to provide a basis for subsequent analysis and processing.

[0075] Problem handling: When inspection personnel find problems or abnormal conditions with equipment during the inspection process, they should handle them in a timely manner if they can be handled on the spot; if they cannot be handled on the spot, they should keep records and report to their superiors in a timely manner.

[0076] 3) Feedback and communication Real-time feedback: Inspectors use mobile terminal devices to provide real-time feedback of data collected and problems found during the inspection process to the data processing center and relevant management personnel. The data processing center analyzes and processes the feedback information in real time and provides corresponding guidance and support to inspectors.

[0077] Communication and coordination: Inspection personnel maintain close communication and coordination with equipment users, maintenance personnel, etc. They should promptly understand the use of equipment and existing problems, and provide equipment maintenance and use suggestions to equipment users; they should communicate with maintenance personnel about equipment failures and assist maintenance personnel in fault diagnosis and repair.

[0078] 4) Inspection report generation and submission Report content collation: After the inspection task is completed, the inspection personnel will organize and analyze the data and information collected during the inspection process and write an inspection report. The content of the inspection report includes basic information of the inspection task, the operating status of the equipment, problems and abnormal conditions found, treatment measures and results, etc.

[0079] Report review and submission: After the inspection report is completed, it is submitted to the superior management for review. The reviewer reviews the content of the report to ensure the accuracy and completeness of the report. After the review is passed, the inspection report is formally submitted to the data processing center for archiving and analysis.

[0080] Step 6: System optimization and continuous improvement: 1) Data evaluation and analysis Data quality assessment: Regularly assess the quality of collected data to check the accuracy, completeness, consistency, etc. Use methods such as data cleaning and outlier detection to further process and optimize the data to improve data quality.

[0081] Model performance evaluation: Evaluate the performance of fault prediction models, maintenance plan formulation models, inspection task allocation models, etc. Use new historical data and actual operation data to verify and analyze the model's prediction accuracy, decision rationality, etc. Based on the evaluation results, identify problems and deficiencies in the model.

[0082] 2) Model optimization and adjustment Parameter adjustment: Adjust the model parameters according to the model performance evaluation results. Use optimization algorithms (such as gradient descent method, genetic algorithm, etc.) to find the optimal combination of model parameters and improve the performance of the model.

[0083] Improvement of model structure: If the performance of the model still cannot meet the requirements, consider improving the structure of the model. For example, increase or decrease the number of layers or neurons in the model, or adopt a new model architecture.

[0084] Data update and retraining: Regularly update the model's training data, add newly collected data to the training set, and retrain the model, so that the model can adapt to changes in the equipment's operating status and new fault types.

[0085] 3) Process optimization and improvement Process evaluation: Evaluate the entire process of medical equipment maintenance and inspection, including data collection and transmission process, data processing and analysis process, maintenance plan formulation process, inspection task allocation process, inspection execution and feedback process, etc. Find out the bottlenecks and problems in the process and analyze their causes.

[0086] Formulate process improvement measures: formulate corresponding process improvement measures based on process evaluation results. For example, optimize data collection frequency and transmission methods to improve the efficiency of data processing and analysis; simplify maintenance plan formulation and approval process to improve work efficiency; improve inspection task allocation algorithm to improve the rationality and fairness of task allocation, etc.

[0087] Process implementation and monitoring: Implement the formulated process improvement measures and monitor the implementation process. Regularly evaluate the effect of process improvement and adjust and improve the improvement measures based on the evaluation results.

[0088] 4) Knowledge accumulation and sharing Knowledge base construction: Establish a knowledge base for medical equipment maintenance and inspection, and organize and store the equipment's technical information, failure cases, maintenance experience, inspection standards and other knowledge. The knowledge base can be managed using databases, document management systems and other methods.

[0089] Knowledge update and maintenance: Update and maintain the knowledge in the knowledge base regularly to ensure the accuracy and timeliness of the knowledge. Encourage inspection personnel and maintenance personnel to share new fault cases, maintenance experience and other knowledge in the knowledge base to continuously enrich the content of the knowledge base.

[0090] Knowledge sharing and training: Promote knowledge sharing and dissemination through internal training, technical exchange meetings, etc. Organize inspection personnel and maintenance personnel to learn knowledge in the knowledge base to improve their professional level and problem-solving ability.

[0091] The third implementation method: Example 1: Maintenance inspection of CT equipment in a hospital; Data collection and transmission Sensor deployment: Temperature sensors, current sensors, speed sensors, etc. are installed in key parts of CT equipment, such as tubes, detectors, cooling systems, etc. The temperature sensor uses a high-precision thermocouple sensor, which can monitor the temperature changes of key components of the equipment in real time. The measurement range is -20℃ to 150℃, and the accuracy is ±0.1℃. The current sensor is used to monitor the current consumption of the equipment to ensure that the equipment operates within the normal power range. The measurement range is 0-10A and the accuracy is ±0.01A. The speed sensor is installed on rotating parts, such as the rotating motor of the scanning frame, to monitor the rotation speed in real time. The measurement range is 0-1000rpm and the accuracy is ±1rpm.

[0092] Data collection frequency: In order to fully and accurately understand the operating status of the CT equipment, the data collection frequency is set to collect the equipment's operating data every 5 minutes. This can capture the slight changes in the equipment's operation process in a timely manner and provide sufficient data support for subsequent fault prediction.

[0093] Data transmission: The collected data is initially processed, and after generating a check code, it is transmitted to the data processing center through the hospital's wireless network. The wireless network uses high-speed and stable Wi-Fi6 technology to ensure the stability and timeliness of data transmission. At the same time, in order to ensure the security of the data, an encrypted transmission protocol is used to encrypt the transmitted data to prevent the data from being stolen or tampered with during the transmission process.

[0094] Data processing and analysis Data verification: After the data processing center receives the transmitted data, it first verifies it. By calculating the verification code of the received data and comparing it with the transmitted verification code, it ensures that there is no error in the data transmission process. If the verification fails, the system will automatically request to retransmit the data until the verification passes.

[0095] Data cleaning: Clean the verified data to remove noise and outliers. Take temperature data as an example to calculate the mean of the data sequence. and standard deviation , for the data point ,like , then the data point is considered to be an outlier. For example, in the data collected at a certain time, the bulb temperature data has a value that is significantly higher than the normal range. After being calculated and judged as an outlier, it is removed.

[0096] Feature extraction: Extract features that can reflect the operating status of the equipment from the cleaned data. For temperature data, extract features such as the mean, variance, maximum, and minimum. For example, by analyzing the mean and variance of the bulb temperature, you can understand the heating and stability of the bulb; by observing the maximum and minimum values, you can determine whether the bulb is overheated or overcooled. For current data, extract features such as the peak, valley, and average values ​​of the current to evaluate the power consumption of the equipment.

[0097] Fault prediction: The extracted feature data is input into the support vector regression (SVR) fault prediction model to predict the probability of CT equipment failure in the next week. The model has been trained with a large amount of historical data and can accurately capture the relationship between the equipment's operating status and failure. For example, if the mean of the tube temperature continues to increase, the variance increases, and the current consumption also fluctuates abnormally, the model will predict that the equipment has a high probability of tube failure in the next week.

[0098] Maintenance plan formulation Factors to consider: Comprehensively consider the fault prediction results, service life, performance indicators and other factors of the CT equipment. The fault prediction results directly reflect the current health of the equipment, the service life affects the aging degree of the equipment and the wear of the parts, and the performance indicators such as image quality and scanning speed are related to the working efficiency of the equipment and the quality of medical services.

[0099] Determine weights using the analytic hierarchy process: Use the analytic hierarchy process to determine the weights of each factor. Construct a judgment matrix ,in Indication factors Relative to factors The importance of the product is usually 1-9 and its reciprocal. For example, when comparing the failure prediction results and the service life, the failure prediction results are considered to be more important for the formulation of the maintenance plan. =3 (assuming that the fault prediction result is factor 1 and the service life is factor 2). By solving the maximum eigenvalue of the judgment matrix and the corresponding eigenvector , and get the weight of each factor. After calculation, the weight of the fault prediction result is 0.5, the weight of the service life is 0.3, and the weight of the performance index is 0.2.

[0100] Maintenance plan formulation: According to the weight and evaluation value of each factor, it is calculated that the maintenance priority of CT equipment is high. A detailed maintenance plan is formulated, including regular cleaning, calibration, and component replacement. For example, for the tube, due to its high predicted probability of failure, a comprehensive inspection and calibration is planned within the next week, and spare tubes are prepared so that they can be replaced in time when a failure occurs. For the detector, it is planned to be cleaned and maintained every two months to ensure image quality.

[0101] Inspection task allocation Personnel information collection: Collect information such as the skills, location, and workload of the inspectors. In terms of skills, understand whether the inspectors have the skills to repair and maintain CT equipment, such as whether they are familiar with operations such as tube replacement and detector calibration. Location information is obtained through the inspectors' mobile terminals to arrange inspection routes reasonably. Workload is evaluated based on the number and difficulty of tasks currently assigned to the inspectors.

[0102] Genetic algorithm to allocate tasks: Genetic algorithm is used to optimize task allocation. Suppose the set of patrol personnel is , the inspection task set is ,in For the tube inspection task, For the detector cleaning task, is the cooling system maintenance task. The skill vector of each inspector is , the skill requirement vector of the task is For example, patrol personnel The skill vector is =(0.8,0.6,0.7), indicating that his skill levels in tube inspection, detector cleaning, and cooling system maintenance are 0.8, 0.6, and 0.7, respectively; The skill demand vector is =(0.9, 0.2, 0.1), indicating that the tube inspection task has a high demand for tube inspection skills. The fitness function f is defined as a comprehensive evaluation index of the task allocation scheme, taking into account factors such as the skill matching degree, distance, and workload of the inspectors. Through the selection, crossover, and mutation operations of the genetic algorithm, the optimal task allocation scheme is found through continuous iteration. Finally, the tube inspection task t1 is assigned to the inspector with a higher skill level and closer to the CT equipment. .

[0103] Inspection execution and feedback Inspection execution: Inspection personnel According to the assigned inspection tasks, carry the mobile terminal to the CT equipment site for inspection. A special inspection APP is installed on the mobile terminal, which displays detailed information of the inspection task, including task content, inspection points, operation steps, etc. The inspection personnel can easily record the data and problems found during the inspection process through the APP.

[0104] Data upload: During the inspection process, the inspectors collect the actual status information of the equipment in real time, such as equipment appearance, operating sound, temperature, current, etc., and upload it to the system through the mobile terminal. For example, the inspectors found that the temperature of the bulb was slightly higher than the normal range, and immediately uploaded the temperature data and related text descriptions through the APP, and took photos of the appearance of the bulb and uploaded them to the system.

[0105] Problem handling and feedback: The system analyzes and handles the problems reported by the inspectors in real time. By comparing historical data and fault prediction models, it was found that the increase in tube temperature may be caused by poor heat dissipation in the cooling system. The system immediately provided a corresponding solution and guided the inspectors to check whether the fan of the cooling system is operating normally and whether there is enough coolant. The inspectors checked and handled the problem according to the guidance provided by the system and found that the coolant was insufficient, so they added coolant in time. The processing results are fed back to the system through the mobile terminal, and the system will promptly feedback the processing results to the inspectors and relevant managers, and update the maintenance records of the equipment at the same time.

[0106] System optimization and continuous improvement Data statistical analysis: Regularly conduct statistical analysis on the maintenance and inspection data of CT equipment to evaluate the performance and effectiveness of the system. The content of statistical analysis includes the accuracy of fault prediction, the execution of maintenance plan, the efficiency of inspection task completion, etc. For example, by comparing the fault prediction results with the actual fault situation, the accuracy of fault prediction is calculated; the completion time and quality of each task in the maintenance plan are counted to evaluate the rationality of the maintenance plan.

[0107] Model and algorithm optimization: Based on the evaluation results, optimize and adjust the fault prediction model, maintenance plan formulation method, inspection task allocation algorithm, etc. If it is found that the fault prediction model has low accuracy in some cases, analyze the reasons and collect more relevant data to retrain and optimize the model. For example, it is found that the accuracy of the tube failure prediction is low. By adding characteristic data such as the temperature and current of the tube and optimizing the parameters of the support vector regression algorithm, the accuracy of the tube failure prediction is improved. For the maintenance plan formulation method, according to the actual use of the equipment and the maintenance effect, adjust the weight and evaluation value of each factor to make the maintenance plan more reasonable. For the inspection task allocation algorithm, according to the actual work performance and task completion of the inspection personnel, optimize the parameters of the fitness function to improve the rationality and efficiency of task allocation.

[0108] Example 2: Maintenance inspection of a large hospital's medical equipment fleet Data collection and transmission Multi-device sensor deployment: Corresponding sensors are installed for various medical equipment in the hospital, such as MRI, ultrasound diagnostic equipment, and ECG monitors. For MRI equipment, magnetic field strength sensors, gradient field stability sensors, and liquid helium level sensors are installed to monitor the key parameters of the equipment in real time. The magnetic field strength sensor can accurately measure the magnetic field strength, with a measurement range of 0-3T and an accuracy of ±0.001T; the gradient field stability sensor is used to monitor the stability of the gradient field to ensure image quality; the liquid helium level sensor monitors the liquid level of liquid helium in real time to prevent liquid helium leakage from causing equipment failure. For ultrasound diagnostic equipment, probe temperature sensors, transmission power sensors, etc. are installed to monitor the working status and transmission power of the probe. For ECG monitors, electrode connection sensors, signal strength sensors, etc. are installed to ensure accurate acquisition of ECG signals.

[0109] Different sampling frequencies: Different sampling frequencies are set according to the characteristics and operating requirements of different devices. For MRI equipment, since its operating state is relatively stable, the sampling frequency is set to collect data every 10 minutes. For ultrasound diagnostic equipment and ECG monitors, since they are used frequently and need to monitor the patient's physiological parameters in real time, the sampling frequency is set to collect data every 1 minute.

[0110] Data transmission network: After verification, the collected data is transmitted to the data processing center through the hospital's local area network. The local area network uses high-speed Gigabit Ethernet technology to ensure fast and stable data transmission. At the same time, in order to ensure data security, firewalls and intrusion detection systems are deployed at the network boundary to prevent external network attacks and data leakage.

[0111] Data processing and analysis Unified processing and analysis: The data processing center processes and analyzes a large amount of equipment operation data in a unified manner. Different feature extraction methods and fault prediction models are used for different types of equipment. For example, for MRI equipment, the focus is on extracting features such as magnetic field strength, gradient field stability, and liquid helium level; for ultrasound diagnostic equipment, the focus is on extracting features such as probe temperature and transmission power; for ECG monitors, the focus is on extracting features such as electrode connection status and signal strength.

[0112] Multi-model fault prediction: Fault prediction models are established based on the characteristic data of different devices. For MRI equipment, the support vector regression (SVR) algorithm is used for fault prediction; for ultrasound diagnostic equipment, the neural network algorithm is used for fault prediction; for ECG monitors, the decision tree algorithm is used for fault prediction. Through the training and verification of historical data, the parameters of the model are continuously optimized to improve the accuracy of fault prediction.

[0113] Maintenance plan formulation Comprehensive consideration: Comprehensively consider the failure prediction results, service life, performance indicators and other factors of various medical equipment. For key equipment, such as MRI, due to its high importance in medical diagnosis, failure will seriously affect the hospital's medical services. Therefore, when formulating maintenance plans, more attention is paid to failure prediction results and performance indicators. For non-critical equipment, such as ordinary ECG monitors, while considering the failure prediction results, appropriate consideration should be given to service life and cost factors.

[0114] Personalized maintenance plan: The analytic hierarchy process is used to determine the weight of each factor and develop a personalized maintenance plan for each device. For MRI equipment, the weight of the fault prediction result is 0.6, the weight of the service life is 0.2, and the weight of the performance index is 0.2. According to the weights and evaluation values, a detailed maintenance plan is developed, including weekly appearance inspection and cleaning of the equipment, monthly calibration of the magnetic field strength and gradient field stability, and quarterly inspection and replenishment of the liquid helium level. For ultrasound diagnostic equipment, maintenance plans are developed to clean and disinfect the probe every two weeks and calibrate the transmission power every month. For ECG monitors, maintenance plans are developed to check the electrode connection status once a day and test the signal strength once a week.

[0115] Inspection task allocation Personnel information integration: Collect and integrate the skills, location, workload and other information of inspectors. Establish an inspector skills database to record each inspector's professional skills, maintenance experience, training status and other information. Obtain the inspector's location information in real time through his / her mobile terminal to arrange the inspection route reasonably. Evaluate the inspector's workload based on the number of tasks currently assigned to him / her, the difficulty of the tasks and the remaining working time.

[0116] Genetic algorithm optimization allocation: Genetic algorithm is used to optimize task allocation. Suppose the set of inspection personnel is , the inspection task set is ,in is the number of inspectors, is the number of inspection tasks. The skill vector of each inspector is , the skill requirement vector of the task is . Define the fitness function It is a comprehensive evaluation index for the task allocation scheme, taking into account factors such as the skill matching degree, distance, and workload of the inspectors. Through the selection, crossover, and mutation of the genetic algorithm, the optimal task allocation scheme is found through continuous iteration. For example, the inspection task of MRI equipment is assigned to the inspector who has MRI equipment maintenance skills, is close to the MRI room, and has a light workload.

[0117] Inspection execution and feedback Inspection process specifications: Inspectors carry mobile terminals to inspect various medical equipment according to the assigned inspection tasks. Detailed inspection process specifications have been formulated, including preparations before inspection, key points for inspection during inspection, and report filling after inspection. For example, before inspecting MRI equipment, inspectors need to check whether personal protective equipment is complete and carry necessary testing tools; during the inspection, inspect each component of the equipment according to the specified inspection points, such as checking the sealing of the magnetic field shielding door and the valve status of the liquid helium tank; after the inspection, fill in the inspection report in time and upload it to the system.

[0118] Real-time interaction and problem handling: During the inspection process, the inspectors upload the status information of the equipment and the problems found in real time. The system processes and responds to the problems reported by the inspectors in real time, and provides detailed solutions and guidance. For example, when the inspectors were inspecting the ultrasonic diagnostic instrument, they found that the probe temperature was too high and uploaded the temperature data and related descriptions. By analyzing historical data and fault prediction models, the system determined that the cooling fan inside the probe might be at fault, and immediately provided the operating steps and precautions for replacing the cooling fan. The inspectors followed the guidance of the system and successfully solved the problem. At the same time, the system promptly fed back the processing results to the inspectors and relevant management personnel, and updated the maintenance records of the equipment.

[0119] System optimization and continuous improvement; Comprehensive evaluation and analysis: Regularly conduct statistical analysis on the maintenance and inspection data of all medical equipment in the hospital to evaluate the overall performance and effect of the system. The statistical analysis includes the accuracy of fault prediction, the execution rate of maintenance plans, the completion rate of inspection tasks, the failure rate of equipment, etc. Through the analysis of these indicators, we can understand the advantages and disadvantages of the system in terms of fault prediction, maintenance plan formulation, inspection task allocation, etc.

[0120] Comprehensive system optimization: Based on the evaluation results, the fault prediction model, maintenance plan formulation method, inspection task allocation algorithm, etc. are comprehensively optimized and adjusted. For example, if it is found that the fault prediction accuracy of a certain type of equipment is low, analyze the reasons and collect more relevant data, and retrain and optimize the fault prediction model of this type of equipment. If it is found that the execution rate of the maintenance plan is not high, analyze whether the plan is unreasonable or there are problems in the execution process, and adjust the maintenance plan formulation method. For the inspection task allocation algorithm, according to the actual work performance and task completion of the inspectors, optimize the parameters of the fitness function to improve the rationality and efficiency of task allocation. Through continuous optimization and adjustment, ensure that the system can continuously improve the efficiency and quality of maintenance and inspection of medical equipment and ensure the normal operation of medical equipment in the hospital.

[0121] Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art, and the model parameters of each electrical appliance are not specifically limited, and conventional equipment can be used.

[0122] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0123] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A medical equipment maintenance and inspection system, characterized in that: include: Data collection and transmission module, used to collect the operation data of medical equipment in real time and transmit it to the data processing center through a secure network; Data processing and analysis module, used to clean, extract features and predict faults of collected data; The inspection task allocation module uses genetic algorithms to optimize task allocation based on the inspectors’ skills, location, and workload information; The maintenance plan formulation module uses the hierarchical analysis method to determine the weight of each factor based on the equipment's fault prediction results, service life, and performance indicators, and formulates a personalized maintenance plan; The real-time interaction and feedback module is used for inspectors to upload inspection data and equipment status information in real time. The system processes and responds to problems in real time and feeds back the processing results to inspectors and relevant managers.

2. A medical equipment maintenance and inspection system according to claim 1, characterized in that: The data cleaning in the data processing and analysis module adopts a statistically based outlier detection method. , then the data point is an outlier and is processed, where Indicates the first data points, represents the mean of the data series, which is obtained by summing all data points and dividing by the number of data points. is the set threshold value, which is 3 and is used to determine whether a data point is an outlier. It represents the standard deviation of the data series and reflects the degree of dispersion of the data.

3. A medical equipment maintenance and inspection system according to claim 1, characterized in that: The AHP in the maintenance plan formulation module constructs a judgment matrix ,in, Represents the judgment matrix, which is a A matrix is ​​used to compare the relative importance of each factor. Indication factors Relative to factors The importance of is 1-9 and its reciprocal. Indicates the number of factors, and in the judgment matrix, indicates the order of the matrix. By solving the maximum eigenvalue of the judgment matrix and the corresponding eigenvector , and obtain the weight of each factor.

4. A medical equipment maintenance and inspection system according to claim 1, characterized in that: The genetic algorithm in the inspection task allocation module defines the fitness function It is a comprehensive evaluation index for task allocation schemes, used to evaluate the pros and cons of each task allocation scheme.

5. A medical equipment maintenance and inspection method, applied to a medical equipment maintenance and inspection system as described in any one of claims 1 to 4, characterized in that: The following steps are involved: Step 1: Data collection and transmission: Use sensors to collect the operating data of medical equipment, generate verification codes and transmit them to the data processing center; Step 2: Data processing and analysis: verify, clean, extract features and predict faults of the transmitted data; Step 3: Formulate a maintenance plan: Comprehensively consider the equipment's fault prediction results, service life, and performance indicators, use the analytic hierarchy process to determine the weight of each factor, and formulate a personalized maintenance plan; Step 4: Inspection task allocation: Based on the information of inspection personnel’s skills, location, and workload, a genetic algorithm is used to optimize task allocation; Step 5: Inspection execution and feedback: The inspectors carry mobile terminals for inspection and upload equipment information in real time. The system handles problems and provides feedback in real time. Step 6. System optimization and continuous improvement: Regularly conduct statistical analysis on system operation data and inspection and maintenance results to optimize system models and algorithms.

6. A medical equipment maintenance and inspection method according to claim 5, characterized in that: The data verification algorithm in the data collection and transmission step is: Ensure data accuracy, including: It is a checksum used to verify whether errors occur during data transmission. Indicates data elements, that is, each specific value in the collected data, Indicates the number of data elements, that is, how many values ​​are there in the collected data. Represents a preset modulus, which is a fixed value used for modulo operations.

7. A medical equipment maintenance and inspection method according to claim 5, characterized in that: The fault prediction in the data processing and analysis step adopts the support vector regression algorithm, and the optimization objective function is: , the constraints are ,in, Represents a weight vector, which is used to indicate the importance of the input feature. is the bias term, in the function It plays the role of adjusting the position of the function. Represents a slack variable, which is used to deal with noise and outliers in the data and allows some data points to deviate from the prediction function. is the penalty factor, which controls the degree of penalty for the error and balances the complexity and error of the model. For the The actual value of the samples, For the The input feature vector of samples, Indicates that the input feature Functions mapped to high-dimensional space, is the parameter of the insensitive loss function, Indicates the number of samples.

8. A medical equipment maintenance and inspection method according to claim 5, characterized in that: The maintenance plan priority calculation formula in the maintenance plan formulation step is: ,in, The priority of the maintenance plan. The higher the value, the more urgent the maintenance needs of the equipment. Indicates The weight of each factor is determined by the hierarchical analysis method, which reflects the importance of the factor in formulating the maintenance plan. Indicates The evaluation value corresponding to each factor is the specific evaluation score of the device in this factor. It represents the number of factors, that is, the number of factors that affect the maintenance plan that are considered comprehensively.

9. A medical equipment maintenance and inspection method according to claim 5, characterized in that: The genetic algorithm in the inspection task allocation step defines the fitness function It is a comprehensive evaluation index for the task allocation scheme. The optimal task allocation scheme is found through iterative selection, crossover and mutation operations.

10. A medical equipment maintenance and inspection method according to claim 5, characterized in that: The system optimization and continuous improvement steps optimize and adjust the fault prediction model, maintenance plan formulation method, and inspection task allocation algorithm based on the statistical analysis results.

Images