Medical equipment full-life-cycle intelligent management method and system based on Internet of Things

Abstract

The invention provides a medical equipment full-life-cycle intelligent management method and system based on the Internet of Things, and the method comprises the steps: constructing an equipment three-dimensional model through a digital twin system in an equipment design and production stage; in the operation stage, real-time data is processed through UWB positioning and edge calculation, early warning of faults is carried out in combination with an LSTM prediction model, and a maintenance work order is generated; and in the tracing and safety stage, an intelligent auditing module is used for screening anomalies, triggering review and associating patient-equipment-diagnosis and treatment data. According to the invention, the digital twin system is used in the design and production stages to ensure accuracy and controllability; in the operation stage, the alliance chain is used for storing evidence data, the data is processed in real time in combination with the technology, and accurate prediction and work order generation are achieved; in the tracing and safety stage, a tracing system is constructed, and closed-loop management is formed; and in the scrap recovery stage, data are collected through the Internet of Things and are associated with the UDI, the recovery value is evaluated in combination with an alliance chain and machine learning, closed-loop management and resource optimization are realized, and the medical equipment management efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of device management, in particular to a medical device full life cycle intelligent management method and system based on Internet of Things. BACKGROUND

[0002] Current medical device full life cycle management mainly relies on manual recording and independent information systems, which has problems such as data fragmentation, insufficient real-time performance, and weak prediction ability. Traditional methods mostly use paper or electronic form to record device design, production, use, maintenance, and scrap data, which is difficult to realize cross-stage data penetration. Although some institutions introduce Internet of Things technology for device positioning and state monitoring, they lack deep integration with technologies such as blockchain and digital twin, resulting in low data credibility and poor traceability efficiency. For example, device usage data, maintenance records, and consumable replacement information are often scattered in multiple systems of hospitals, manufacturers, and regulatory agencies, forming "data islands" and making it difficult to support intelligent decision-making throughout the process. In addition, existing fault early warning is mostly based on single parameter threshold judgment, lacking multi-dimensional feature fusion and machine learning model support, making it difficult to capture gradual fault signs, resulting in delayed maintenance response or excessive maintenance. SUMMARY

[0003] The present application aims to at least solve the technical problems in the prior art that rely on manual recording and independent information systems, and have problems such as data fragmentation, insufficient real-time performance, and weak prediction ability. The present application provides a medical device full life cycle intelligent management method based on Internet of Things.

[0004] In order to achieve the above-mentioned purpose of the present application, the present application provides a medical device full life cycle intelligent management method based on Internet of Things, which comprises: S1, constructing a device three-dimensional model through a digital twin system in the device design and production stage, integrating CAE simulation data and clinical scene simulation in the device three-dimensional model, and synchronously assembling line sensor data to optimize process parameters; S2, in the device running stage, using a consortium chain to store device usage data, maintenance records, and consumable replacement information, performing real-time data processing through UWB positioning and edge computing, and combining an LSTM prediction model to early warn device failure and trigger an intelligent contract to automatically generate a maintenance work order; S3, in the traceability and safety stage, integrating UDI and blockchain to build a full-process traceability system, and automatically screening abnormal use patterns through an intelligent audit module based on the full-process traceability system to trigger safety review and associate patient-device-diagnosis data.

[0005] In another aspect, the present application also provides an Internet of Things-based intelligent management system for the whole life cycle of medical equipment, comprising a processor and a memory for storing processor-executable instructions; wherein the processor is configured to implement the Internet of Things-based intelligent management method for the whole life cycle of medical equipment when executing the executable instructions.

[0006] The present application has the following advantages: through the deep integration of Internet of Things, blockchain, digital twin and machine learning, the present application realizes the whole-process connection and intelligent management of medical equipment whole life cycle data, effectively solves the problems of data fragmentation, insufficient real-time performance and weak prediction ability in the prior art; specifically, through the integration of CAE simulation and clinical scene simulation in the design and production stage, the digital twin system optimizes the process parameters by synchronizing the sensor data of the assembly line, ensures the accuracy of design verification and the real-time controllability of the production process; in the running stage, the device data is stored by using the alliance chain, the multi-dimensional feature data is processed in real time by combining UWB positioning and edge computing, and the accurate prediction of fault probability and the generation of intelligent maintenance work order are realized by using the LSTM model, thereby improving the timeliness of fault warning and the accuracy of maintenance response; in the tracing and safety stage, the whole-process tracing system is constructed by integrating UDI and blockchain, the abnormal use mode is automatically screened by the intelligent audit module, and the patient-equipment-diagnosis data is associated, thereby forming a closed-loop management of "data storage-abnormal screening-examination triggering-feedback optimization", and the automation of safety examination and the tracing efficiency are strengthened; finally, in the scrap recycling stage, the state data is collected by the Internet of Things sensing layer and associated with UDI, the recycling value is evaluated by combining the alliance chain storage and the machine learning model, and the closed-loop management and resource optimization of the scrap process are realized. Through the whole-process data connection, real-time intelligent analysis and closed-loop feedback mechanism, the efficiency, safety and sustainability of medical equipment management are significantly improved, and the management cost and safety risk are reduced.

[0007] Additional aspects and advantages of the application will be set forth in part in the description which follows, and in part will become apparent to those skilled in the art upon examination of the following and / or can be learned by practice of the application. BRIEF DESCRIPTION OF DRAWINGS

[0008] The above and / or additional aspects and advantages of the present application will become apparent and be readily appreciated from the description of the embodiments, which follows, including the accompanying drawings. Figure 1 is a flowchart of an Internet of Things-based intelligent management method for the whole life cycle of medical equipment. DETAILED DESCRIPTION

[0009] Embodiments of the present application are described below in detail with reference to the accompanying drawings, examples of which are shown in the drawings, wherein the same or similar notations represent the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by reference to the drawings are exemplary only, and are used only for the purpose of explaining the present application, and should not be understood as limiting the present application.

[0010] Embodiment 1 As Figure 1 shown, a medical device full life cycle intelligent management method based on Internet of Things, the method comprises: S1, constructing a three-dimensional model of the device through a digital twin system in the device design and production stage, integrating CAE simulation data and clinical scene simulation in the three-dimensional model of the device, and synchronously assembling line sensor data to optimize process parameters; S2, in the device running stage, using a consortium chain to record device usage data, maintenance records and consumable replacement information, performing real-time data processing through UWB positioning and edge computing, and combining an LSTM prediction model to early warn device failure and trigger an intelligent contract to automatically generate a maintenance work order; S3, in the traceability and safety stage, integrating UDI and block chain to build a full-process traceability system, and automatically screening abnormal use mode through an intelligent audit module based on the full-process traceability system to trigger safety review and associate patient-device-diagnosis data.

[0011] In this embodiment, the principle of a medical device full life cycle intelligent management method based on Internet of Things is: In the device design and production stage, the digital twin system integrates CAE simulation data and clinical scene simulation into the three-dimensional model of the device. By synchronously assembling line sensor data, problems existing in the process parameters can be found in time and optimized. This is like making a detailed plan and simulation in the virtual world before building a building, ensuring that every detail meets the requirements, so as to ensure the accuracy of design verification and real-time controllability of the production process.

[0012] In the device running stage, the consortium chain is like a fair recorder, accurately recording device usage data, maintenance records and consumable replacement information, preventing data from being tampered with. UWB positioning and edge computing are like sharp observers, collecting and processing multi-dimensional feature data in real time. Combined with the LSTM prediction model, the device failure probability can be accurately predicted in advance, and an intelligent maintenance work order can be automatically generated. Potential problems can be found in time and appropriate maintenance measures can be taken.

[0013] In the traceability and safety stage, the UDI is integrated with the blockchain-built full-process traceability system, which includes data of the entire life cycle of the device. The intelligent audit module automatically screens abnormal use patterns through a pre-set three-level audit rule engine. Once an abnormality is found, a safety review process is triggered, and patient-device-diagnosis data are associated. This closed-loop management mode of "data evidence-exception screening-review triggering-feedback optimization" greatly enhances the automation and traceability efficiency of safety reviews.

[0014] In the device scrapping and recycling stage, the Internet of Things sensing layer, alliance chain, and machine learning model work together. The Internet of Things sensing layer is responsible for collecting the state data of the scrapped device and associating the UDI, the alliance chain stores the scrapping full-process data, and the machine learning model evaluates the recycling value. The smart contract is triggered to allocate the recycling responsibility and generate a recycling work order, realizing closed-loop management of the scrapping process and optimized use of resources.

[0015] Through this full-process data connection, real-time intelligent analysis, and closed-loop feedback mechanism, the application significantly improves the efficiency, safety, and sustainability of medical device management.

[0016] As an optional embodiment of the application, optionally, in step S1, a three-dimensional model of the device is constructed through a digital twin system in the device design and production stage, CAE simulation data and clinical scenario simulation are integrated in the three-dimensional model of the device, and synchronous assembly line sensor data is used to optimize process parameters, including: S101, based on the device design parameters, a three-dimensional digital twin model of the device is constructed, multi-physics field simulation data are integrated in the three-dimensional digital twin model of the device, and the radiation dose distribution under different scanning parameters is simulated through a Monte Carlo algorithm to verify whether the device design meets the safety standards; In step S101, it needs to be specified that the Monte Carlo algorithm is a numerical calculation method based on random sampling. In this step, it simulates the radiation dose distribution under different scanning parameters through a large number of random samplings. Specifically, first, according to the physical characteristics of the device and the scanning parameter range, the parameter space for random sampling is determined. Then, a large number of random samplings are performed in this parameter space, each sampling corresponding to a specific set of scanning parameters. For each set of scanning parameters obtained by sampling, the corresponding radiation dose distribution is calculated using multi-physics field simulation data. Through statistical analysis of a large number of sampling results, the overall situation of the radiation dose distribution under different scanning parameters can be obtained.

[0017] The advantage of this is that the radiation safety of the device under different working conditions can be comprehensively evaluated before the device is actually produced. If the simulation results show that the radiation dose under certain scanning parameters exceeds the safety standard, the device design parameters can be adjusted in time to avoid safety hazards during actual production and use. At the same time, this simulation method can also provide reference for the optimization design of the device, such as determining the optimal combination of scanning parameters to minimize the radiation dose while ensuring the performance of the device.

[0018] Next, it is also necessary to synchronize the assembly line sensor data to the device three-dimensional digital twin model. The assembly line sensor can collect various data such as temperature, pressure, vibration, etc. in the production process of the device in real time. Synchronizing these data with the digital twin model can reflect the actual state of the device in the production process in real time. By comparing the theoretical state of the digital twin model with the sensor data actually collected, problems that may exist in the production process, such as process parameter deviation and equipment failure, can be found in time. Once the problem is found, the process parameters can be adjusted immediately to ensure the stability of the production process and the consistency of the product quality.

[0019] In this embodiment, the method for constructing the device three-dimensional digital twin model is: combined with point cloud data acquisition technology, a high-precision three-dimensional laser scanner is used to scan the device in all directions to obtain accurate geometric information of the device surface and generate dense point cloud data. Then the point cloud data is preprocessed, including removing noise points, smoothing and data simplification, etc. to improve data quality and processing efficiency.

[0020] Next, based on the processed point cloud data, a professional three-dimensional modeling software is used to construct the initial three-dimensional model of the device. In the modeling process, feature extraction and surface fitting methods are used to ensure that the model can accurately reflect the shape characteristics and structural details of the device. At the same time, in order to make the model more realistic and intuitive, material texture and lighting effects will be added to the model to simulate the appearance of the device in the actual environment.

[0021] After that, the integrated multi-physics simulation data is deeply fused with the constructed three-dimensional model. Through data interface and mapping relationship, the physical quantities (such as temperature, stress, strain, etc.) in the simulation data are associated with the corresponding parts in the model to realize the visualization display of physical information on the model. In this way, the physical response of the device under different working conditions can be observed intuitively in the model.

[0022] In addition, to ensure the real-time and accuracy of the three-dimensional digital twin model, a dynamic data interaction mechanism between the model and the actual equipment is established. Using Internet of Things technology, real-time data such as temperature, pressure, vibration, and other data collected by sensors during equipment operation are transmitted to the model in real time, and the model is updated and corrected in real time. At the same time, the model monitors and predicts the running state of the equipment in real time, and sends out warning signals in time when it finds that the equipment has abnormal conditions, so as to take appropriate maintenance measures.

[0023] Finally, the constructed three-dimensional digital twin model of the equipment is comprehensively tested and verified. Through comparative analysis with the actual equipment, the accuracy and reliability of the model are checked. At the same time, experts and users in the relevant field are invited to evaluate and feedback the model, and the model is further optimized and improved according to the feedback to ensure that the model can meet the needs of actual application.

[0024] S102, embedding a clinical scenario simulation module in the three-dimensional digital twin model of the equipment, simulating the performance of the equipment in actual use based on the patient physiological parameter database, and optimizing the pressure-flow control algorithm to adapt to extreme working conditions according to the performance; S103, deploying an industrial Internet of Things sensor array to collect real-time assembly line data and building a closed-loop optimization system; In step S103, it needs to be explained in detail that the industrial Internet of Things sensor array can collect data on multiple key links of the assembly line. For example, in the assembly link of equipment parts, the sensor can monitor the installation force, angle, and other data of each part. If the installation force is too large or the angle deviation is too large, it will cause the equipment to loosen, malfunction, and other problems in subsequent use. By collecting these data in real time, potential quality problems can be found in time.

[0025] In the welding link of the equipment, the sensor can detect welding temperature, welding time and other parameters. Proper welding temperature and time are crucial to ensure welding quality. If the welding temperature is too high, the welded part will deform and become brittle; if the welding time is too short, the welding may not be firm. After the sensor collects these data in real time, it will transmit them to the closed-loop optimization system.

[0026] The closed-loop optimization system will analyze the collected assembly line data in real time. It will compare the actual collected data with the preset standard parameters. If it finds that the actual data deviates from the standard parameters, the system will immediately issue an adjustment instruction. For example, when it detects that the installation force of a part does not meet the standard, the system will remind the operator to adjust the force of the installation tool; when the welding temperature is abnormal, the system will automatically adjust the parameters of the welding equipment.

[0027] This closed-loop optimization system also has the ability of self-learning and optimization. It records and analyzes the data collected each time and the adjustment results. Over time, the system can summarize the best combination of process parameters for different devices at different production stages. When the production environment changes, such as fluctuations in raw material quality, equipment aging, etc., the closed-loop optimization system can quickly adapt and adjust the process parameters to ensure the stability of the production process and the consistency of product quality.

[0028] By deploying an array of industrial Internet of Things sensors and building a closed-loop optimization system, real-time collection, analysis, and optimization of assembly line data can be achieved. This not only improves production efficiency and reduces the rate of defective products in the production process, but also ensures that the medical devices produced meet high-quality standards.

[0029] S104、When the sensor of the closed-loop optimization system detects that the assembly error ≥ n or the temperature abnormally rises, the process parameter adjustment algorithm in the closed-loop optimization system is triggered.

[0030] In step S104, it needs to be specified that in this embodiment, the process parameter adjustment algorithm can be configured according to the actual needs of the site (such as the cutting process parameter adjustment method).

[0031] When the sensor of the closed-loop optimization system detects that the assembly error ≥ n, it means that the device has deviated beyond the acceptable range during the assembly process. At this time, the process parameter adjustment algorithm will quickly start and comprehensively analyze the parameters in the assembly process. For example, it will check whether the precision of the assembly tool meets the requirements, whether the assembly process is unreasonable, whether the size of the parts meets the requirements, etc. Through the investigation and analysis of these factors, the algorithm will determine the specific cause of the assembly error and adjust the process parameters according to the pre-set rules and models.

[0032] If it is a precision problem of the assembly tool, the algorithm will prompt to calibrate or replace the tool; if the assembly process is unreasonable, the algorithm will re-plan the assembly steps and optimize the operation sequence; if the size of the parts does not meet the requirements, the algorithm will feedback to the upstream production link in time and require adjustment or replacement of the parts. Through such adjustment, the assembly error can be corrected in time to ensure the assembly quality of the device.

[0033] When the sensor detects that the temperature abnormally rises, the process parameter adjustment algorithm will also be triggered. Abnormally high temperature may be caused by friction, excessive current, poor heat dissipation, etc. during the operation of the device. The process parameter adjustment algorithm will first diagnose the cause of the temperature rise. It will analyze the running state data of the device, such as current, voltage, speed, etc., and combine the physical characteristics and working principle of the device to determine the root cause of the temperature rise.

[0034] If the temperature rise is caused by friction, the algorithm will adjust the lubrication parameters of the device, increase the amount of lubricant used or replace a more suitable lubricant; if it is caused by excessive current, the algorithm will check the circuit system, adjust the size of the current or optimize the circuit layout; if it is caused by poor heat dissipation, the algorithm will adjust the parameters of the heat dissipation system, such as increasing the speed of the heat dissipation fan, improving the heat dissipation channel, etc. Through timely adjustment of process parameters, the temperature rise can be effectively controlled, avoiding damage to the equipment due to high temperature, prolonging the service life of the equipment.

[0035] Through this mechanism of triggering process parameter adjustment algorithms according to different abnormal situations, the closed-loop optimization system can respond to various problems in the production process in real time, ensuring the quality and stability of medical devices in the design and production stages.

[0036] As an optional embodiment of the present application, optionally, a clinical scene simulation module is embedded in the device three-dimensional digital twin model in step S102, the performance of the device in actual use is simulated based on the patient physiological parameter database, and the pressure-flow control algorithm is optimized to adapt to extreme working conditions, including: S1021, collect multi-dimensional clinical feature parameters to construct a patient physiological parameter database; In step S1021, it needs to be specified in detail that in the present embodiment, the multi-dimensional clinical feature parameters include the age, gender, height, weight, various physiological indicators (such as blood pressure, heart rate, blood sugar, blood lipids, etc.), disease history, allergy history, etc. of the patient. These multi-dimensional clinical feature parameters can comprehensively and meticulously reflect the physical condition and physiological characteristics of the patient.

[0037] The method for constructing the patient physiological parameter database is as follows: first, a standardized data collection process is established. Professional medical personnel are arranged to collect the multi-dimensional clinical feature parameters of the patient, ensuring the accuracy and completeness of the data. When collecting data, medical personnel need to strictly follow the established norms and standard operations, and use calibrated medical equipment to measure various physiological indicators.

[0038] For basic information such as the age, gender, height, weight, etc. of the patient, communication and exchange with the patient or his / her family members are carried out to obtain and record in detail. For various physiological indicators such as blood pressure, heart rate, blood sugar, blood lipids, etc., professional medical detection equipment is used for measurement. During the measurement process, multiple measurements are taken to obtain the average value, so as to reduce errors.

[0039] The collection of disease history and allergy history requires medical personnel to carefully inquire about the patient's past illness and allergy experience, and to verify the patient's medical records. For some special diseases and allergies, further consultation and confirmation with relevant medical institutions or experts are required.

[0040] The collected data is timely entered into a special database management system. During the entry process, the data is preliminarily audited and cleaned to remove duplicate, erroneous or incomplete data. At the same time, a unique identifier is established for each patient's data to facilitate subsequent query and management.

[0041] In order to ensure the security and privacy of the database, the patient's data is encrypted. The data is encrypted using encryption algorithms, and only authorized personnel can access and use these data. At the same time, a strict access control mechanism is established, and different access permissions are set for users of different levels to prevent data leakage and misuse.

[0042] The patient physiological parameter database is updated and maintained regularly. As the patient's condition changes and time passes, the patient's physiological parameters will also change accordingly. Therefore, the patient is regularly revisited and detected, and the data in the database is updated in a timely manner. At the same time, the database is optimized and backed up to ensure the stable operation of the database and the security of the data.

[0043] S1022, embedding a clinical scene simulation module in the device three-dimensional digital twin model based on the patient physiological parameter database, simulating the device running under different physiological parameters, and outputting the pressure-flow control parameter baseline; In step S1022, the clinical scene simulation module will construct different clinical scenes according to the various data in the patient physiological parameter database. For example, for patients of different ages, genders and physical conditions, the actual running conditions of the device when serving them are simulated. In this process, simulation technology is used, combined with the physical characteristics and working principles of the device, to accurately simulate the running state of the device under different physiological parameters.

[0044] During the simulation process, the pressure-flow control of the device will be focused on. Pressure and flow are key parameters in the operation of medical devices, and their stable and accurate control is directly related to the performance and treatment effect of the device. Through the simulation of the device running under different physiological parameters, a series of pressure-flow control parameter values can be obtained. These parameter values will vary with the changes in the patient's physiological parameters, reflecting the optimal running state of the device under various actual conditions.

[0045] After statistical analysis of these simulated pressure-flow control parameter values, removing outliers and unreasonable data, a representative parameter range is determined. This parameter range is the pressure-flow control parameter baseline, which provides a reference standard for the pressure-flow control of the device in actual operation. Based on this baseline, the device can automatically adjust the pressure-flow control strategy according to the patient's specific physiological parameters to adapt to the needs of different patients.

[0046] In this way, the pressure-flow control parameter baseline can enable the device to operate more accurately in actual use, improving the performance and therapeutic effect of the device.

[0047] S1023, based on the pressure-flow control parameter baseline, the performance of the device under extreme conditions is evaluated through Monte Carlo simulation and dynamic response analysis, and a performance evaluation report is obtained; In step S1023, it needs to be explained in detail that Monte Carlo simulation is a statistical method based on random sampling, which can simulate various possible situations of the device under extreme conditions through a large number of random experiments. In this embodiment, a series of boundary conditions of extreme conditions will be set according to the pressure-flow control parameter baseline, such as extreme pressure value, flow value, temperature value, etc. Then a large number of random samples conforming to these boundary conditions are generated by Monte Carlo simulation, each sample represents the running state of the device under a possible extreme condition.

[0048] For each random sample, the performance of the device under this condition is simulated by combining the physical model of the device and the dynamic response analysis method. The dynamic response analysis will consider the inertia, elasticity, damping and other physical characteristics of the device, as well as the interaction between the various components inside the device, so as to more accurately predict the dynamic behavior of the device under extreme conditions. For example, the deformation, stress distribution and whether the flow control of the device can respond and adjust in time when the device is subjected to a great pressure impact in an instant, etc.

[0049] During the simulation process, various performance indicators of the device are recorded, such as pressure fluctuation range, flow regulation accuracy, fatigue life of the device, whether a fault occurs, etc. Through statistical analysis of the simulation results of a large number of random samples, the probability distribution of each performance indicator of the device under extreme conditions can be obtained. For example, the probability of device failure under extreme conditions, the probability of pressure fluctuation exceeding the allowable range, etc.

[0050] According to these statistical analysis results, a performance evaluation report is generated. The report will describe in detail the performance of the device under extreme conditions, including statistical parameters such as the average, standard deviation, maximum, minimum, etc. of each performance indicator, as well as the risk level of device failure under different extreme conditions. At the same time, the report will also analyze and evaluate the weak links of the device, pointing out which aspects of the device need to be further improved and optimized.

[0051] S1024, based on the performance evaluation report, the pressure-flow control algorithm is dynamically optimized by using adaptive PID control strategy; The expression for optimizing the pressure-flow control algorithm is: wherein, denotes the pressure control value at time , denotes the proportional gain, denotes the integral gain, denotes the derivative gain, denotes the pressure error function, and both denote adaptive adjustment coefficients, denotes the body mass index, denotes the respiratory rate; In step S1024, it needs to be explained in detail that the adaptive PID control strategy is an advanced control method that can automatically adjust control parameters according to the running state of the system. In this embodiment, based on the performance data of the equipment in extreme working conditions provided by the performance evaluation report, the adaptive PID control strategy is used to dynamically optimize the pressure-flow control algorithm.

[0052] The performance evaluation report presents in detail the statistical parameters of various performance indicators of the equipment in extreme working conditions and the fault risk level and other information. For example, if the report shows that the probability of pressure fluctuation exceeding the allowed range of the equipment in some extreme working conditions is relatively high, it indicates that the current pressure-flow control algorithm has deficiencies in dealing with such working conditions. At this time, the adaptive PID control strategy will automatically adjust the proportional gain, integral gain and derivative gain and other parameters in the pressure-flow control algorithm according to these feedback information.

[0053] The proportional gain is mainly used to quickly respond to the pressure error. When the pressure error is large, increasing the proportional gain can make the control quantity change rapidly, and speed up the response speed of the system. However, too large proportional gain may cause the system to have a large overshoot, and even cause the system to be unstable. The integral gain is used to eliminate the steady-state error of the system, which integrates the pressure error and gradually adjusts the control quantity, so that the output of the system tends to be the set value. The derivative gain is mainly used to predict the trend of the pressure error, and adjusts the control quantity in advance to reduce the dynamic error of the system.

[0054] The adaptive adjustment coefficients α and β will be dynamically adjusted according to the body mass index BMI and the respiratory rate RR of the patient. The body mass index reflects the body fatness of the patient, and patients with different body mass indexes may have different pressure and flow requirements for medical equipment. The respiratory rate reflects the respiratory state of the patient, and the change of the respiratory rate will affect the pressure-flow control requirements of the equipment. By dynamically adjusting α and β according to BMI and RR, the pressure-flow control algorithm can better adapt to the physiological characteristics of different patients.

[0055] In practical applications, the adaptive PID control strategy monitors the equipment's operating status and pressure-flow control in real time. When the equipment encounters different patient physiological parameters or extreme operating conditions, it automatically adjusts the parameters of the pressure-flow control algorithm based on performance evaluation reports and real-time feedback information. This enables the equipment to respond quickly and accurately to changes in pressure and flow, ensuring stable and efficient operation under various conditions.

[0056] As an optional embodiment of the present invention, optionally, the expression of the Monte Carlo algorithm in step S101 is: in, Representing a spatial point Radiation dose at the location, This represents the total number of particles in the Monte Carlo simulation. Indicates the first The initial flux of each particle. Representing a spatial point The mass energy absorption coefficient, Indicates the first The length of the particle trajectory Represents the linear attenuation coefficient. Representing a spatial point The quality of the target area Representing a spatial point The density of matter, Represents the Dirac delta function, Indicates the first A spatial point of a particle.

[0057] As an optional embodiment of the present invention, optionally, in step S2, during the equipment operation phase, the device usage data, maintenance records, and consumable replacement information are stored in the consortium blockchain. Real-time data processing is performed through UWB positioning and edge computing, and combined with an LSTM prediction model, equipment failures are given early warning. The smart contract is then triggered to automatically generate maintenance work orders, including: S201. A consortium blockchain is built based on three nodes: hospital, manufacturer and regulatory agency. The PBFT consensus mechanism is used to perform multi-party signature notarization of equipment usage data, maintenance records and consumable replacement information to prevent data from being tampered with and to meet UDI traceability requirements. In step S201, it needs to be explained in detail that the alliance chain is a blockchain jointly managed and maintained by multiple participants. In this embodiment, the alliance chain is constructed based on three-party nodes of hospitals, manufacturers and regulatory agencies. Hospitals, as users of equipment, will generate a large amount of equipment usage data, such as equipment start time, running time, usage frequency, etc. Manufacturers have access to production information, technical parameters and maintenance recommendations of equipment. Regulatory agencies are responsible for supervising and managing the compliance and safety of equipment. Through the construction of the alliance chain by the three-party nodes, data sharing and interconnection can be achieved, and the transparency and credibility of the data can be improved.

[0058] The PBFT (Practical Byzantine Fault Tolerance) consensus mechanism is a high-efficiency Byzantine fault-tolerant consensus algorithm that can achieve fast consensus under the premise of ensuring system security. In this embodiment, the PBFT consensus mechanism is used to multi-signature evidence of equipment usage data, maintenance records and consumable replacement information. When new data is generated, the relevant nodes will verify and sign the data, and only the data jointly signed by the three-party nodes will be recorded on the alliance chain. This can prevent data from being tampered with, ensure the authenticity and integrity of the data, and also meet the UDI (Unique Device Identification) traceability requirements, making it easier to trace and manage the entire life cycle of the equipment.

[0059] During the operation of the alliance chain, different nodes have different access permissions, and only authorized nodes can access and operate related data. For example, hospital nodes can only access and record data related to the use of equipment in their own hospital; manufacturer nodes can view the overall use of equipment and maintenance records to improve technology and upgrade products; regulatory agency nodes can supervise and review all data to ensure that the use and management of equipment comply with relevant regulations and standards.

[0060] In addition, in order to ensure the efficient operation of the alliance chain, the performance of the alliance chain will also be optimized. For example, distributed storage technology is used to store data on multiple nodes, improving data storage capacity and read / write speed; regular data cleaning and compression of the alliance chain reduces data redundancy and improves system efficiency. Through these measures, the alliance chain can be ensured to run stably and reliably.

[0061] S202, deploy UWB positioning base station array and edge computing node based on alliance chain, real-time collect equipment position data and sensor index, based on equipment position data and sensor index, pre-process and extract features through MEC architecture, obtain feature data; In step S202, it needs to be explained in detail that the UWB (Ultra-Wideband) positioning technology is a high-precision wireless positioning technology that determines the position of a device by measuring the time of flight of a signal. In this embodiment, a UWB positioning base station array is deployed based on a consortium chain, and these base stations will be installed at various key locations in the hospital to form a positioning network covering the entire hospital area. Each medical device is equipped with a UWB tag, and when the device moves within the hospital, the UWB positioning base station can collect real-time position data of the device and accurately determine the specific location of the device.

[0062] At the same time, various sensors are installed on the device to monitor the operating status and performance indicators of the device, such as pressure, flow, temperature, humidity, etc. These sensors will collect various indicator data of the device in real time and transmit them to the edge computing node. The edge computing node has certain computing and processing capabilities and can preliminarily process and analyze the collected data locally.

[0063] MEC (Multi-Access Edge Computing) architecture is a computing mode that sinks cloud computing capabilities to the network edge, which can process and analyze data close to the data source, reduce data transmission delay, and improve system response speed. In this embodiment, the collected device position data and sensor indicator data are preprocessed and feature extracted based on the MEC architecture. The preprocessing process includes data cleaning, filtering, normalization, etc. to remove noise and outliers in the data and improve the quality of the data. Feature extraction is to extract representative features from the preprocessed data, such as device operating status, performance trends, etc.

[0064] Through these operations, the final feature data is obtained. For example, based on the device position data and operating status features, the scheduling and allocation of the device can be optimized to improve the efficiency of the device; based on the performance trend features of the device, the possible faults of the device can be predicted in advance, and timely maintenance and maintenance can be performed to reduce the downtime of the device.

[0065] S203, outputting a fault probability value based on the feature data using a pre-trained LSTM prediction model, and generating a warning signal; In step S203, it needs to be explained in detail that LSTM (Long Short-Term Memory) is a special recurrent neural network that can handle and predict long-term dependencies in sequence data. In this embodiment, a pre-trained LSTM prediction model is used to analyze the feature data to output the fault probability value of the device.

[0066] The pre-trained LSTM prediction model is trained by a large amount of historical equipment operation data. These historical data contain various feature information of the equipment in normal operation and when a fault occurs, such as changes in the location of the equipment, fluctuations in sensor indicators, etc. During the training process, the model learns the correlation between these features and equipment failures.

[0067] When the feature data is obtained, it is input into the pre-trained LSTM prediction model. The model will calculate and analyze the input feature data and output a fault probability value. This fault probability value represents the likelihood of the equipment failing in its current state.

[0068] If the fault probability value exceeds the pre-set threshold, the system will generate a warning signal. The warning signal can be communicated in various ways, such as issuing an alarm prompt in the hospital's monitoring system, or notifying the relevant equipment maintenance personnel through SMS, email, etc.

[0069] To ensure the accuracy and reliability of the LSTM prediction model, the model will also be periodically evaluated and updated. As equipment operation data continues to accumulate, new data will be used to retrain the model to adapt to changes in equipment performance and new fault patterns. At the same time, the parameters of the model will be adjusted and optimized to improve the prediction accuracy of the model.

[0070] S204, based on the warning signal and the fault probability value, an intelligent contract is used to automatically generate a maintenance work order, which includes the device ID, the fault level, the maintenance action list, and the responsible team.

[0071] In step S204, it needs to be specified that the smart contract is an automatically executed contract whose terms are stored in the form of code on the blockchain. In this embodiment, when the system generates a warning signal based on the fault probability value output by the LSTM prediction model, the smart contract will automatically generate a maintenance work order based on this information. The device ID is the unique identifier of each medical device, and through the device ID, the specific device that needs to be maintained can be accurately located. The fault level is evaluated based on the fault probability value, as well as the importance of the equipment, the impact of the fault, etc. For example, it can be divided into a first-level serious fault, a second-level moderate fault, and a third-level minor fault, etc. Different fault levels correspond to different processing priorities.

[0072] The maintenance action list is a series of maintenance operation steps formulated for possible failures of the device, which are based on the technical manual of the device, historical maintenance experience and expert knowledge. For example, if the fault level shows that the pressure sensor of the device is abnormal, the maintenance action list may include checking whether the connection line of the sensor is loose, calibrating the parameters of the sensor, replacing the damaged sensor components, etc. The responsible team is the personnel or department responsible for executing the maintenance work order. According to the fault level and the type of the device, the responsible team may be composed of the device maintenance department of the hospital, the after-sales technical support personnel of the manufacturer or the third-party professional maintenance organization.

[0073] After the smart contract generates the maintenance work order, it automatically distributes the work order information to the corresponding responsible team. After receiving the work order, the responsible team can view the detailed use data, maintenance records and other information of the device through the consortium chain, so as to better understand the condition of the device and formulate a more reasonable maintenance plan. At the same time, during the maintenance process, the responsible team needs to feed back the progress and results of the maintenance to the consortium chain, and these information will be recorded to form a complete device maintenance history.

[0074] In addition, the smart contract can also set some conditions and rules, such as stipulating the response time and completion time of the maintenance work order. If the responsible team fails to respond or complete the maintenance task within the specified time, the smart contract can automatically trigger the corresponding punishment mechanism, such as deducting certain credit points or economic penalties, to ensure that the maintenance work can be completed in a timely and efficient manner. In this way, the use of smart contract to automatically generate maintenance work orders realizes the automation and intelligentization of medical device maintenance management, improves the efficiency and quality of device maintenance, and ensures the stable operation of medical devices.

[0075] As an optional embodiment of the present application, the step S204 of automatically generating a maintenance work order by a smart contract based on the early warning signal and the fault probability value comprises: S2041, triggering the execution of the smart contract by the early warning signal, determining the fault level based on the preset three-level fault threshold, and associating the device UDI code with the historical maintenance data; In step S2041, it needs to be specified that the preset three-level fault threshold is set according to the performance characteristics of the device, historical fault data and the importance of the device in the medical process, etc. This three-level fault threshold divides the fault into different severity levels, such as minor fault, moderate fault and serious fault. When the early warning signal triggers the execution of the smart contract, the smart contract compares the fault probability value with the three-level fault threshold, so as to accurately determine the fault level.

[0076] A device's UDI code is a unique identifier for medical devices, containing key information such as the device's model, batch number, and manufacturer. Linking the device's UDI code with historical maintenance data allows maintenance personnel to quickly understand the device's past maintenance history, including types of malfunctions that occurred, maintenance measures taken, and maintenance dates. This association provides maintenance personnel with a more comprehensive understanding of the device's condition.

[0077] Meanwhile, linking equipment UDI codes with historical maintenance data facilitates traceability and management throughout the equipment's entire lifecycle. When equipment malfunctions, regulatory agencies and manufacturers can quickly access the equipment's usage history and maintenance records using the UDI code, allowing them to assess the equipment's quality and reliability. Furthermore, for hospitals, this also helps optimize equipment procurement and management decisions; for example, for equipment models that frequently malfunction, changing suppliers or adjusting procurement plans can be considered.

[0078] Furthermore, during the process of associating device UDI codes with historical maintenance data, smart contracts ensure data security and privacy. Only authorized personnel can access and manipulate this data, preventing data leakage and misuse. Moreover, the storage and transmission of data on the consortium blockchain employs encryption technology, further guaranteeing data integrity and confidentiality.

[0079] S2042. Based on the fault level, a maintenance action list is generated using a dynamic template engine. The maintenance action list includes equipment-specific maintenance steps and the corresponding responsible team. In step S2042, it should be explained in detail that the dynamic template engine is a tool that can dynamically generate text content based on different input parameters. In this embodiment, based on the fault level, the dynamic template engine will refer to various information such as the device's technical manual, historical maintenance experience, and expert knowledge to generate a targeted maintenance action list.

[0080] The maintenance action list will vary significantly depending on the severity of the fault. For example, when a fault is determined to be minor, the maintenance actions may be relatively simple. The equipment-specific maintenance steps may only involve some routine checks and adjustments, such as cleaning the equipment surface, checking the connection wiring, and calibrating some basic parameters. The corresponding responsible team may be junior equipment maintenance personnel within the hospital who have received some training and are able to handle such relatively simple faults.

[0081] If the fault level is moderate, the device-specific maintenance steps in the maintenance action list will be more complex and in-depth. Detailed inspection, testing, and adjustment of some critical components of the device may be required, such as replacing some wear-and-tear parts, upgrading or repairing the software system of the device, etc. At this time, the responsible team may need to be composed of both professional technicians from the hospital equipment maintenance department and after-sales technical support personnel from the manufacturer to ensure accurate diagnosis and problem solving.

[0082] When the fault level is determined to be a serious fault, the maintenance action list involves a comprehensive overhaul of the device or even replacement of core components. Device-specific maintenance steps may include disassembly of the device, performance testing and evaluation of each component, replacement of damaged core components, reassembly and debugging of the device, etc. In this case, the responsible team usually needs to be composed of senior technical experts from the manufacturer and professionals from third-party professional maintenance institutions, who have rich experience and professional skills to deal with complex fault conditions.

[0083] By generating the maintenance action list according to the fault level through the dynamic template engine, the pertinence and effectiveness of maintenance work can be ensured, and the quality and efficiency of equipment maintenance can be improved. At the same time, clearly defining the corresponding responsible team also helps to implement responsibilities and coordinate work, ensuring that medical equipment can resume normal operation as soon as possible.

[0084] S2043, generate a standardized maintenance work order through the fault probability value and the UDI code, and realize three-party node signature evidence through the PBFT consensus mechanism of the alliance chain.

[0085] The detailed description in step S2043 is that the standardized maintenance work order is an important document for standardizing and uniformly managing maintenance work, which contains device ID, fault level, maintenance action list, and responsible team, etc. The standardized maintenance work order is generated through the fault probability value and the UDI code, which can ensure the accuracy and completeness of the work order information. The fault probability value provides a quantitative indicator for the possibility of device failure for maintenance personnel, which helps them more accurately judge the fault condition of the device; and the UDI code can uniquely identify the device, which facilitates maintenance personnel to quickly locate the specific device and obtain detailed technical information and historical maintenance records of the device.

[0086] The PBFT (Practical Byzantine Fault Tolerance) consensus mechanism of the consortium chain is an efficient distributed consensus algorithm that can achieve fast consensus under the premise of ensuring system security and reliability. In this embodiment, the PBFT consensus mechanism is used to realize the signature evidence of three nodes, which are the hospital node, the manufacturer node and the regulatory agency node. After the intelligent contract generates a standardized maintenance work order, it sends the work order information to the three nodes. The three nodes will review and confirm the work order information, and then sign the work order through their respective private keys. The signed work order information is stored on the consortium chain, forming an unalterable evidence record.

[0087] The hospital node signature represents the recognition of the equipment failure situation and maintenance needs in the work order, and also means that the hospital will cooperate with the responsibility team to carry out equipment maintenance work. The manufacturer node signature indicates its confirmation of the maintenance action list and the arrangement of the responsibility team in the work order, reflecting the manufacturer's technical support and responsibility in the equipment maintenance process. The regulatory agency node signature means supervision and recognition of the entire maintenance work order generation process, ensuring that the maintenance work complies with relevant regulations and standards.

[0088] The PBFT consensus mechanism realizes the signature evidence of the three nodes, which not only guarantees the authenticity and effectiveness of the maintenance work order, but also improves the transparency and traceability of the maintenance work. In the maintenance process, if any disputes or problems arise, they can be queried and verified through the evidence record on the consortium chain. At the same time, this signature evidence method also helps to promote cooperation and trust among the three nodes.

[0089] As an optional embodiment of the present application, the expression of the feature data obtained in step S203 is: , ; ; wherein, represents the feature data, represents the average position coordinate of the equipment, represents the position change rate, represents the arithmetic mean value of the temperature sensor within the sampling window, represents the temperature standard deviation, represents the temperature change rate slope, represents the vibration peak value, represents the electrical parameter standard deviation, represents the temperature-vibration correlation coefficient, represents the temperature-humidity joint entropy, represents the covariance of temperature and vibration sampling values, represents the temperature, a vibration sample value, a standard deviation of a vibration sample value, a humidity, a joint distribution probability of temperature and humidity.

[0090] As an optional embodiment of the present application, optionally, in step S3, in the trace and security phase, the integrated UDI and blockchain build a full-process trace system, and based on the full-process trace system, the intelligent audit module automatically screens abnormal use patterns to trigger security review and associate patient-device-diagnosis data, including: S301, based on the three nodes of hospital, manufacturer and regulatory agency, a UDI-blockchain full-process trace system is built, and based on the above-mentioned UDI-blockchain full-process trace system, a multi-party signature storage of device full life cycle data is carried out through PBFT consensus mechanism, the device full life cycle data includes design parameters, production batch number, use record, maintenance log and consumable replacement information, and the UDI is integrated as the unique identifier of the device; In step S301, it needs to be specified that the UDI-blockchain full-process trace system is built to realize the transparent, traceable and secure management of the medical device full life cycle data. The three nodes of hospital, manufacturer and regulatory agency play important roles in this system. The hospital node is responsible for uploading the use record of the device, including the use department, use time, use patient information and the like of the device, which can reflect the running of the device in the actual medical scene. The manufacturer node provides the design parameters and production batch number and the like of the device, which are the basic attributes of the device and are crucial for understanding the performance and quality of the device. The regulatory agency node supervises and manages the data of the whole system to ensure the authenticity and compliance of the data.

[0091] The introduction of PBFT consensus mechanism enables the three nodes to reach efficient consensus in the signature storage process of device full life cycle data. When new data is generated, such as maintenance log and consumable replacement information of the device, the data will be sent to the three nodes for audit. The three nodes will verify the data according to the preset rules, and only when the three nodes all sign to confirm the data, the data will be stored on the alliance chain. This multi-party signature storage method ensures the data unalterable and traceable.

[0092] The integration of UDI as a unique device identifier further enhances the accuracy and reliability of the system. The UDI code contains key information about the device, and by associating UDI with device lifecycle data, it can quickly and accurately locate detailed information about each device. Whether it is in the design stage, production stage, use stage or maintenance stage of the device, it can be queried and traced through the UDI code. This is of great significance to the quality control, troubleshooting and recall management of medical devices. For example, when a batch of devices is found to have quality problems, the UDI code can quickly locate the hospitals and patients using the batch of devices, and take appropriate measures in a timely manner to ensure patient safety. At the same time, for regulatory agencies, it is also more convenient to supervise medical devices, ensuring the standardization and orderly development of the medical device market.

[0093] In step S302, an intelligent audit module is deployed based on the UDI-blockchain full-process traceability system, and an abnormal use mode is screened through a pre-set three-level audit rule engine. The rule engine includes threshold rules, dynamic rules and compliance rules, and outputs structured alert information. In step S302, the intelligent audit module relies on the pre-set three-level audit rule engine to efficiently and accurately screen the abnormal use mode of the medical device. Among them, the threshold rule is based on the setting of a clear threshold range for each parameter of the device. For example, for the key parameters such as the operating temperature, pressure and voltage of the device, the upper and lower limits are set according to the technical specifications and safety standards of the device. When the actual operating parameters of the device exceed this threshold range, an alert is triggered. This rule is simple and direct, and can quickly identify obvious abnormal conditions.

[0094] Dynamic rules are more flexible and intelligent. They consider various factors such as the use environment, use time and use frequency of the device, and dynamically adjust based on historical and real-time data. For example, some devices may have different normal operating modes at different times, and dynamic rules will automatically adjust the judgment criteria according to these changes. If the use frequency of the device suddenly increases significantly within a certain time period, or the performance indicators of the device show abnormal fluctuations in a certain environment, dynamic rules can capture these changes in a timely manner and determine whether they belong to an abnormal use mode.

[0095] Compliance rules mainly focus on whether the use of the device complies with relevant regulations, standards and internal regulations of the hospital. For example, whether the operation process of the medical device is standardized, whether the maintenance records are complete, and whether the use of the device is in compliance, etc. Compliance rules will conduct a comprehensive review of the use of the device to ensure that the use of the device meets various requirements.

[0096] When the intelligent audit module detects abnormal usage patterns through the rule engine, it outputs structured alert information. These alert information details the type of abnormality, the time of occurrence, the devices involved, and other key information. Structured alert information allows hospital managers and technical personnel to quickly understand the severity and impact of abnormal conditions, so that appropriate measures can be taken quickly. For example, for minor abnormalities, technical personnel can be arranged for simple inspection and debugging; for serious abnormalities, the use of the device may need to be stopped immediately, and comprehensive maintenance and troubleshooting may be required. In this way, the intelligent audit module can effectively improve the safety and reliability of medical devices, reducing the risks caused by abnormal use of devices.

[0097] S303, based on the alarm information, trigger the safety review process, through the patient-device-diagnosis data association module stored in the blockchain, hash the association of device UDI, patient diagnosis record and device parameters, generate a safety review report containing abnormal mode details, associated data path and review action list; In step S303, after receiving the structured alert information output by the intelligent audit module, the safety review process is triggered immediately. This process uses the patient-device-diagnosis data association module stored in the blockchain to achieve the hash association of device UDI, patient diagnosis record and device parameters. This association method is efficient and secure, ensuring accurate data correspondence and difficult tampering. Through the association, the patient's diagnosis and the specific parameters of the device during the abnormal use of the device can be understood. For example, when the device diagnosis is abnormal, the patient's diagnosis record and the device parameters during that period can be quickly found.

[0098] After completing the data association, the system generates a detailed safety review report containing abnormal mode details, associated data paths and review action lists. The review action list is formulated according to the severity and type of abnormality. Minor abnormalities may only require simple inspection and calibration of the device, while serious abnormalities may require comprehensive testing, repair or even device shutdown to ensure patient safety.

[0099] After the report is generated, it is sent to hospital device management personnel, technical experts and regulatory agency staff, who can analyze and evaluate the abnormal situation in depth based on the report and develop reasonable solutions. At the same time, the report is stored in the blockchain as a historical record. Through this safety review process, the safe use of medical devices can be ensured, and medical risks can be reduced.

[0100] S304, based on the safety review report, execute closed-loop feedback, and optimize the audit rule threshold through machine learning.

[0101] In step S304, it needs to be explained in detail that when the safety review report is generated and sent to the relevant personnel, the relevant personnel will take action according to the review action list in the report. For slight abnormalities, technicians will check and calibrate the equipment according to the list requirements; for serious abnormalities, the equipment will be comprehensively detected and repaired, or even suspended.

[0102] In the process of implementing these measures, relevant data of the equipment will be collected, such as performance parameters before and after maintenance, whether the abnormal situation occurs again, etc. These data will be fed back to the system and integrated with the previous device life cycle data for analysis. Through the analysis of these feedback data, the effectiveness and accuracy of the current audit rules can be evaluated.

[0103] Machine learning will learn and mine a large amount of device data and feedback information to find out potential rules and patterns. According to these rules and patterns, machine learning algorithms (such as decision trees, neural networks) will optimize the audit rule thresholds. For example, if it is found that the normal operating parameter range of some equipment under certain conditions deviates from the current threshold, machine learning will adjust the threshold to make it more consistent with the actual situation.

[0104] The optimized audit rule thresholds will be applied to the intelligent audit module again, so that the intelligent audit module can more accurately screen abnormal use patterns. In this way, the entire system forms a closed-loop feedback mechanism, continuously learning from the actual use and maintenance of the equipment, optimizing its own audit rules and management strategies.

[0105] With the passage of time, through the continuous implementation of closed-loop feedback and optimization of audit rule thresholds, the medical equipment life cycle intelligent management system will be able to better adapt to various complex medical environments and equipment operation conditions, improve the safety and reliability of medical equipment, further reduce medical risks, and provide better and safer medical services for patients. At the same time, this continuous optimization mechanism also helps to improve the cooperation efficiency and management level among hospitals, manufacturers and regulatory agencies, and promotes the continuous development and progress of the medical equipment management industry.

[0106] As an optional embodiment of the present application, optionally, the method further comprises, in the device scrapping and recycling stage, collecting the state data of the scrapped device by using the Internet of Things sensing layer and associating the UDI, storing the scrapping whole process data through the alliance chain, evaluating the recycling value combined with the machine learning model, triggering the smart contract to allocate the recycling responsibility and generate the recycling work order, and realizing the closed-loop management of the scrapping process.

[0107] It should be noted that in the device retirement recycling phase, the Internet of Things perception layer collects the state data of the retired device comprehensively and accurately through various sensors such as temperature sensors, pressure sensors, vibration sensors, etc. These data cover the physical state, performance parameters, operating conditions and other aspects of the device, which can truly reflect the actual situation of the device at the time of retirement.

[0108] The collected state data is associated with the UDI, which serves as the unique identifier of the device, like a "ID card" for the device. Through this association, it can ensure that the data of each retired device can be accurately matched, achieving precise management and traceability of data. Whether it is the design parameters, production information of the device, or maintenance records, fault conditions during use, etc. can be combined with the UDI and the retired state data to form a complete device lifecycle data chain.

[0109] The alliance chain stores the whole process data of the retired device, and the alliance chain has the characteristics of decentralization, non-tamperability, traceability, etc. The whole process data of the retired device is stored on the alliance chain, which can prevent data from being tampered with or lost. From the moment the device decides to retire, to the entire process of recycling the device, the data of each link, including the recycling time, location, recycling method, processing results, etc. of the device, are recorded and stored on the alliance chain. In this way, relevant departments and personnel can query and verify these data at any time, ensuring the transparency and standardization of the retirement process.

[0110] Combined with the machine learning model to evaluate the recycling value, it is a scientific and efficient method. Machine learning models can learn and analyze a large amount of device retirement data, considering factors such as device brand, model, service life, damage degree, market demand, etc. to accurately evaluate the recycling value of the retired device. For example, for some high-end medical devices, if some of their parts still have high value, the machine learning model can identify and evaluate their potential value in the secondary market or remanufacturing field.

[0111] Triggering the smart contract to allocate recycling responsibilities and generate recycling work orders, realizing the automation and intelligent management of the retirement process. Smart contract is an automatic contract based on blockchain technology, which can automatically execute according to pre-set rules and conditions. When the machine learning model evaluates the recycling value, the smart contract will automatically allocate the recycling responsibility to the corresponding subject, such as the hospital, manufacturer or professional recycling company, according to the ownership and responsibility division of the device. At the same time, the smart contract will generate detailed recycling work orders, including specific requirements, time nodes, recycling methods and other information of the recycling task. The recycling work order will be sent to the relevant recycling subject in time to ensure that the recycling work can be carried out smoothly according to the predetermined process and standard.

[0112] Through such a closed-loop management of the scrapping process, the efficiency and quality of medical device scrapping and recycling can be effectively improved. On the one hand, it can ensure that the scrapped devices are properly handled, reducing pollution to the environment and waste of resources; on the other hand, it can tap the potential value of the scrapped devices, realizing resource recycling and the development of circular economy. At the same time, such a closed-loop management mode is also helpful to strengthen the cooperation and communication among hospitals, manufacturers and regulatory agencies, and to improve the operation efficiency and management level of the entire medical device management system.

[0113] Embodiment 2 An Internet of Things-based medical device full-life-cycle intelligent management system, comprising a processor and a memory for storing processor-executable instructions; wherein the processor is configured to implement an Internet of Things-based medical device full-life-cycle intelligent management method when executing the executable instructions.

[0114] It should be noted that the computer device includes a processor, a memory, and can further include one or more of a multimedia component, an input / output (I / O) interface, and a communication component.

[0115] The processor is configured to control the overall operation of the computer device to complete all or part of the steps of the above-mentioned Internet of Things-based medical device full-life-cycle intelligent management method.

[0116] The memory is used to store various types of data to support the operation of the computer device, which can include, for example, instructions for operating any application or method on the computer device, and application-related data; the memory can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0117] The multimedia component can include a screen, which can be a touch screen for example, and an audio component for outputting and / or inputting audio signals; for example, the audio component can include a microphone for receiving external audio signals, which can be further stored in the memory or transmitted through the communication component; the audio component also includes at least one speaker for outputting audio signals.

[0118] The I / O interface provides an interface between the processor and other interface modules, which can be a keyboard, a mouse, a button, etc.; these buttons can be virtual buttons or physical buttons.

[0119] The communication component is used for wired or wireless communication between the computer device and other devices; wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, 4G or 5G, or a combination of one or more of them, so the corresponding communication component can include a Wi-Fi module, a Bluetooth module, an NFC module, and a mobile communication module.

[0120] As a preferred scheme of the present embodiment, the computer device can be implemented by one or more Application Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), controller, microcontroller, microprocessor or other electronic elements, for executing the above-mentioned Internet of Things-based medical device whole life cycle intelligent management method.

[0121] Although the embodiments of the present application have been shown and described, those skilled in the art can understand that various changes, modifications, replacements and variations can be made to the embodiments without departing from the principles and spirit of the present application, and the scope of the present application is defined by the claims and their equivalents.

Claims

1. A method for intelligent management of the entire lifecycle of medical devices based on the Internet of Things, characterized in that, The method includes: S1. During the equipment design and production phase, a three-dimensional model of the equipment is constructed using a digital twin system. CAE simulation data and clinical scenario simulation are integrated into the three-dimensional model of the equipment, and process parameters are optimized simultaneously using assembly line sensor data. S2. During the equipment operation phase, the consortium blockchain is used to store equipment usage data, maintenance records and consumable replacement information. Real-time data processing is performed through UWB positioning and edge computing. Combined with the LSTM prediction model, equipment failures are given early warning and smart contracts are triggered to automatically generate maintenance work orders. S3. In the traceability and security phase, UDI and blockchain are integrated to build a full-process traceability system. Based on the full-process traceability system, the intelligent audit module automatically screens abnormal usage patterns, triggers security reviews, and associates patient-device-treatment data.

2. The method for intelligent management of the entire lifecycle of medical devices based on the Internet of Things as described in claim 1, characterized in that, In step S1, a three-dimensional model of the equipment is constructed using a digital twin system during the equipment design and manufacturing phase. CAE simulation data and clinical scenario simulations are integrated into this three-dimensional model, and assembly line sensor data is used to simultaneously optimize process parameters, including: S101. Construct a three-dimensional digital twin model of the equipment based on the equipment design parameters, integrate multi-physics simulation data into the three-dimensional digital twin model of the equipment, and simulate the radiation dose distribution under different scanning parameters using the Monte Carlo algorithm to verify whether the equipment design meets safety standards. S102. Embed a clinical scenario simulation module into the three-dimensional digital twin model of the device, simulate the performance of the device in actual use based on the patient physiological parameter database, and optimize the pressure-flow control algorithm according to the performance to adapt to extreme working conditions. S103. Deploy an industrial IoT sensor array to collect assembly line data in real time and build a closed-loop optimization system; S104. When the sensor of the closed-loop optimization system detects an assembly error ≥ n or an abnormal temperature rise, the process parameter adjustment algorithm in the closed-loop optimization system is triggered.

3. The method for intelligent management of the entire lifecycle of medical devices based on the Internet of Things as described in claim 2, characterized in that, In step S102, a clinical scenario simulation module is embedded into the three-dimensional digital twin model of the device. Based on a database of patient physiological parameters, the performance of the device in actual use is simulated. The pressure-flow control algorithm is optimized based on the performance to adapt to extreme operating conditions, including: S1021. Collect multi-dimensional clinical characteristic parameters to construct a patient physiological parameter database; S1022. Based on the patient physiological parameter database, a clinical scenario simulation module is embedded in the three-dimensional digital twin model of the device to simulate the operation of the device under different physiological parameters and output the pressure-flow control parameter baseline. S1023. Based on the pressure-flow control parameter baseline, evaluate the performance of the equipment under extreme operating conditions through Monte Carlo simulation and dynamic response analysis, and obtain a performance evaluation report. S1024. Based on the performance evaluation report, the pressure-flow control algorithm is dynamically optimized using an adaptive PID control strategy.

4. The method for intelligent management of the entire lifecycle of medical devices based on the Internet of Things as described in claim 2, characterized in that, The expression for the Monte Carlo algorithm in step S101 is: in, Representing a spatial point Radiation dose at the location, This represents the total number of particles in the Monte Carlo simulation. Indicates the first The initial flux of each particle. Representing a spatial point The mass energy absorption coefficient, Indicates the first The length of the particle trajectory Represents the linear attenuation coefficient. Representing a spatial point The quality of the target area Representing a spatial point The density of matter, Represents the Dirac delta function, Indicates the first A spatial point of a particle.

5. The method for intelligent management of the entire lifecycle of medical devices based on the Internet of Things as described in claim 1, characterized in that, In step S2, during the equipment operation phase, the consortium blockchain is used to store equipment usage data, maintenance records, and consumable replacement information. Real-time data processing is performed using UWB positioning and edge computing, and an LSTM prediction model is used to provide early warnings of equipment failures. This triggers a smart contract to automatically generate maintenance work orders, including: S201. A consortium blockchain is built based on three nodes: hospital, manufacturer and regulatory agency. The PBFT consensus mechanism is used to perform multi-party signature notarization of equipment usage data, maintenance records and consumable replacement information to prevent data from being tampered with and to meet UDI traceability requirements. S202. Deploy UWB positioning base station array and edge computing nodes based on the consortium blockchain, collect device location data and sensor indicators in real time, and perform data preprocessing and feature extraction based on the device location data and sensor indicators through MEC architecture to obtain feature data. S203. Based on the feature data, the pre-trained LSTM prediction model is used to output the fault probability value and generate an early warning signal; S204. Based on the warning signal and the fault probability value, a maintenance work order is automatically generated using a smart contract. The maintenance work order includes the device ID, fault level, maintenance action list, and responsible team.

6. The method for intelligent management of the entire lifecycle of medical devices based on the Internet of Things as described in claim 5, characterized in that, Step S204, which involves automatically generating a maintenance work order using a smart contract based on the warning signal and the fault probability value, includes: S2041. Trigger the execution of the smart contract through the warning signal, determine the fault level based on the preset three-level fault threshold, and associate the device UDI code with historical maintenance data; S2042. Based on the fault level, a maintenance action list is generated using a dynamic template engine. The maintenance action list includes specific maintenance steps for the equipment and the corresponding responsible team. S2043. A standardized maintenance work order is generated using the fault probability value and the UDI code, and the three-party nodes sign and store the evidence through the PBFT consensus mechanism of the consortium blockchain.

7. The method for intelligent management of the entire lifecycle of medical devices based on the Internet of Things as described in claim 5, characterized in that, The expression for obtaining the feature data in step S203 is: in, Representing feature data, Indicates the average position coordinates of the equipment. Indicates the rate of change of position. This represents the arithmetic mean of the temperature sensor readings within the sampling window. Indicates the standard deviation of temperature. The slope represents the rate of change of temperature. Indicates the peak value of vibration. Indicates the standard deviation of electrical parameters. This represents the temperature-vibration correlation coefficient. Represents the joint entropy of temperature and humidity. This is expressed as the covariance of temperature and vibration sample values. Indicates temperature. Indicates vibration sampling value, The standard deviation of the vibration sample values ​​is represented. Indicates humidity. This represents the probability of the joint distribution of temperature and humidity.

8. The method for intelligent management of the entire lifecycle of medical devices based on the Internet of Things as described in claim 1, characterized in that, In step S3, during the traceability and security phase, a full-process traceability system is constructed by integrating UDI and blockchain. Based on this system, an intelligent audit module automatically screens for abnormal usage patterns, triggers security reviews, and associates patient-device-treatment data, including: S301. A UDI-blockchain full-process traceability system is constructed based on three-party nodes: hospital, manufacturer, and regulatory agency. Based on the above UDI-blockchain full-process traceability system, the device's full life cycle data is stored through multi-party signature using the PBFT consensus mechanism. The device's full life cycle data includes design parameters, production batch number, usage records, maintenance logs, and consumable replacement information, and integrates UDI as the device's unique identifier. S302. Deploy an intelligent audit module based on the UDI-blockchain full-process traceability system, and screen abnormal usage patterns through a preset three-level audit rule engine. The rule engine includes threshold rules, dynamic rules and compliance rules, and outputs structured alarm information. S303. Based on the alarm information, a security review process is triggered. Through the patient-device-treatment data association module stored on the blockchain, the device UDI is hashed and associated with the patient's treatment records and device parameters to generate a security review report containing details of abnormal patterns, associated data paths, and a list of review actions. S304. Perform closed-loop feedback based on the security review report and optimize the audit rule thresholds through machine learning.

9. The method for intelligent management of the entire lifecycle of medical devices based on the Internet of Things as described in claim 1, characterized in that, The method also includes, during the equipment scrapping and recycling stage, using the Internet of Things sensing layer to collect status data of scrapped equipment and associate it with UDI, storing the entire scrapping process data through the consortium blockchain, combining it with a machine learning model to evaluate the recycling value, triggering a smart contract to allocate recycling responsibilities and generate a recycling work order.

10. An intelligent management system for the entire lifecycle of medical devices based on the Internet of Things, characterized in that, The system includes: processor; Memory used to store processor-executable instructions; The processor is configured to implement the IoT-based intelligent management method for the entire lifecycle of medical devices as described in any one of claims 1 to 9 when executing the executable instructions.

Images