Intelligent medical equipment management system and method

Abstract

The invention discloses an intelligent medical equipment management system and method, and the method comprises the steps: achieving the decentralized credible storage of an equipment use record through a block chain technology, and guaranteeing the safety and traceability of data through an SHA-256 salted Hash and Merkle tree structure; an intelligent contract is integrated to realize automatic execution of an inventory strategy, and an intelligent scheduling system is constructed in combination with an ARIMA-LSTM hybrid model and a linear programming algorithm. The system covers core modules of data acquisition, block chain storage, smart contract, prediction optimization and the like, solves the problems of data potential safety hazards, low inventory efficiency and scheduling blindness in traditional management, is suitable for intelligent management of the medical equipment in the whole life cycle, and remarkably improves the medical resource utilization efficiency and management compliance.

Description

Technical Field

[0001] The present invention belongs to the technical field of medical device management, and particularly relates to an intelligent medical device management system and method. Background Art

[0002] In the medical industry, the management of the usage records of medical devices, inventory scheduling, and resource optimization are key links to ensure medical safety and efficiency. The traditional management methods have the following defects: Insufficient data security and traceability: Relying on a centralized database to store device usage records is vulnerable to risks of tampering and leakage, and it is difficult to share across institutions, unable to meet the high security requirements of medical data.

[0003] Inefficient inventory management: Inventory management dominated by manual experience leads to overstocking or shortages, lacking dynamic adjustment driven by real-time data, and low utilization rate of devices.

[0004] Lack of intelligent scheduling strategies: Existing systems rely on manual intervention and cannot perform intelligent scheduling based on historical data and real-time requirements, resulting in unreasonable resource allocation.

[0005] Existing technologies such as electronic health record (EHR) systems, medical asset management systems, and enterprise resource planning (ERP) systems, although they achieve basic data management, have significant deficiencies in data security, intelligent analysis, and automated processes. Therefore, there is an urgent need for a medical device management system that integrates blockchain technology and intelligent algorithms to solve the above problems. Summary of the Invention

[0006] The purpose of the present invention is to provide an intelligent medical device management system and method to solve the problems of significant deficiencies in data security, intelligent analysis, and automated processes in the existing technologies as mentioned in the background art.

[0007] To solve the above technical problems, the technical solution adopted by the present invention is: An intelligent medical device management system, comprising: A data acquisition module, configured to collect the usage data of medical devices in real time through Internet of Things devices, where the usage data includes device ID, timestamp, operating status, usage duration, and fault records; and generate the basic information of medical devices through a device management module, where the basic information includes device ID, type, and usage status; A blockchain storage module, configured to encrypt and store the usage data and basic information in a blockchain. The encryption process includes appending a randomly generated salt value to each record and then performing hash encryption using the SHA-256 algorithm, and encapsulating the encrypted data into blocks, and managing the data on the blockchain using a Merkle tree structure; The smart contract module is used to define the rules for medical device usage records and inventory management, and to automate the creation and verification of device usage records and the automated execution of inventory scheduling strategies. The rules include the format definition of device usage records, the setting of minimum inventory levels, replenishment strategies, and scheduling strategies; The inventory management module is used to manage the inventory information of medical devices based on the rules set by the smart contract module, including device type, quantity, and storage location. It optimizes the inventory structure through the first-in, first-out algorithm and the minimum inventory level algorithm, and triggers a replenishment operation when the inventory is below the minimum inventory level; The scheduling optimization module is used to optimize the device scheduling strategy through the linear programming algorithm according to the real-time data obtained by the data acquisition module and the prediction results of the prediction analysis module, so as to achieve the optimal allocation of device resources; The prediction analysis module is used to predict the change of device demand by using historical device usage data through time series analysis algorithms and machine learning algorithms, and to assist in inventory management and scheduling decisions; The data query and traceability module is used to allow authorized users to query the usage records and inventory scheduling records of medical devices, and to support traceability according to conditions such as time, device type, and device ID; The user management module is used to implement user permission management to ensure that only authorized users can access and operate the system, and combines an authentication mechanism to ensure data access security; Furthermore, the intelligent medical device management system also includes a consensus mechanism module; the consensus mechanism module is used to ensure the real-time update and consistency of data in the blockchain network by adopting an improved proof-of-authority (PoA) mechanism. It randomly selects a primary verification node to verify and broadcast transactions, so as to achieve balanced participation and data synchronization of each node.

[0008] According to the above technical solution, the encrypted device usage records and inventory change records are used as leaf nodes, and the root hash value is generated by layer-by-layer hash merging. By verifying the root hash value and the hash values of sibling nodes on the path from the leaf node to the root node, the fast verification of data integrity is realized.

[0009] According to the above technical solution, the device usage record format defined by the smart contract module includes device ID, operator ID, usage time, and operation type. The operation type includes start, stop, maintenance, scheduling, and replenishment; the smart contract module automatically receives a record request when the device operation is completed, uses a mapping data structure to map the device ID to the corresponding operation record list, encrypts the record by hashing, inserts it into the blockchain through a Merkle tree, and triggers a synchronous update to push the data to all nodes.

[0010] According to the above technical solution, the Internet of Things device of the data acquisition module includes sensors for real-time collection of the power-on and power-off times, operating status, operating duration, and fault records of the device; the data acquisition module cleans and preprocesses the original data, including removing duplicate data, filling in missing values, detecting and removing outliers, and data standardization.

[0011] According to the above technical solution, the time series analysis algorithm adopted by the prediction and analysis module includes the ARIMA model, and the machine learning algorithm includes the LSTM network. The model is trained with historical device usage data to predict future device requirements.

[0012] According to the above technical solution, the linear programming algorithm of the scheduling optimization module takes minimizing the device scheduling cost as the objective function, and the device availability and demand as the constraints, and solves to obtain the optimal scheduling plan.

[0013] According to the above technical solution, the authentication mechanism of the user management module includes biometric recognition, password verification, or digital certificate verification to ensure that only authorized personnel can access device data and perform inventory scheduling operations.

[0014] A medical device management method includes the following steps: Data acquisition step: Real-time collect the usage data of medical devices through Internet of Things devices, and generate the basic information of medical devices through the device management module; Data processing step: Clean and preprocess the collected usage data, and identify device usage patterns through clustering algorithms; Data storage step: After appending a salt value to the processed usage data and basic information, perform SHA-256 hash encryption, encapsulate it into a block, and store it in the blockchain through a Merkle tree structure; Smart contract execution step: Define device usage record rules and inventory management rules through smart contracts, automatically create and verify device usage records, and trigger replenishment operations according to inventory levels; Prediction and analysis step: Use historical data to predict device requirements through time series analysis and machine learning algorithms; Scheduling optimization step: Generate an optimal scheduling plan through a linear programming algorithm based on real-time data and prediction results; Query and trace step: Authorized users can query device usage records and inventory scheduling records through the data query and trace module.

[0015] Furthermore, the medical device management method further includes a consensus verification step: The consensus verification step ensures the real-time update and consistency of data in the blockchain network through an improved PoA mechanism. The main verification node broadcasts to other nodes after verifying the transaction to complete synchronization.

[0016] According to the above technical solution, the clustering algorithm adopts the K-means algorithm, and clusters according to features such as the usage duration, frequency, and number of failures of the devices, to identify frequently used devices, idle devices, and moderately used devices.

[0017] According to the above technical solution, in the data storage step, the root node of the Merkle tree contains the hash values of all device records. By hashing the records into leaf nodes and gradually merging them upward to generate the root hash value, data integrity verification is achieved by verifying the root hash value and the verification path.

[0018] Compared with the prior art, the present invention has the following beneficial effects: In the present invention, through the decentralized storage of the blockchain and the salted hash algorithm, the anti-tampering and anti-repudiation of device records are realized, meeting the medical data security compliance requirements such as HIPAA; the Merkle tree structure reduces the data verification complexity to O(logN), improving the integrity verification efficiency in large-scale data scenarios.

[0019] Based on the minimum inventory dynamic monitoring of real-time data and the intelligent replenishment of the EOQ model, the inventory turnover rate is increased by 40%-60%, effectively reducing the inventory holding cost; the combination of the FIFO algorithm and the expiration date management controls the equipment expiration loss rate below 1%. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic diagram of the system structure of the present invention; Figure 2 It is a flowchart of the medical device management method of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0022] Embodiment 1 As Figure 1 shown, an intelligent medical device management system includes: A data acquisition module, configured to collect the usage data of medical devices in real time through Internet of Things devices, and generate the basic information of medical devices; A blockchain storage module, configured to encrypt and store the usage data and basic information in the blockchain, and manage the data on the blockchain using the Merkle tree structure; An intelligent contract module, which is used to define the rules for medical device usage records and inventory management, and implement the automated creation and verification of device usage records as well as the automated execution of inventory scheduling strategies; An inventory management module, which is used to manage the inventory information of medical devices based on the rules set by the intelligent contract module; A scheduling optimization module, which is used to according to the real-time data obtained by the data acquisition module and the prediction results of the prediction analysis module; A prediction analysis module, which is used to utilize historical device usage data, and predict the change of device demand through time series analysis algorithms and machine learning algorithms to assist inventory management and scheduling decisions; A data query and traceability module, which is used to allow authorized users to query the usage records and inventory scheduling records of medical devices; A user management module, which is used to implement user permission management, ensure that only authorized users can access and operate the system, and combine an authentication mechanism to ensure data access security.

[0023] In the present invention, through blockchain decentralized storage and salted hash algorithm, the anti-tampering and anti-repudiation of device records are realized, meeting the medical data security compliance requirements such as HIPAA; the Merkle tree structure reduces the data verification complexity to O(logN), improving the integrity verification efficiency in large-scale data scenarios.

[0024] The intelligent replenishment based on the minimum inventory dynamic monitoring of real-time data and the EOQ model increases the inventory turnover rate by 40%-60%, effectively reducing the inventory holding cost; the combination of the FIFO algorithm and expiration date management controls the device expiration loss rate below 1%.

[0025] Embodiment 2 This embodiment is a further refinement of Embodiment 1.

[0026] System architecture overview: The intelligent medical device management system constructed in this embodiment integrates the functions of multiple modules such as data acquisition, storage, analysis, scheduling, and user management, aiming to improve the efficiency, security, and intelligence level of medical device management. The system mainly includes a data acquisition module, a blockchain storage module, an intelligent contract module, an inventory management module, a scheduling optimization module, a prediction analysis module, a data query and traceability module, and a user management module. Each module works together to achieve the intelligent management of the entire life cycle of medical devices.

[0027] In a specific implementation manner, the data acquisition module is specifically: It is implemented in a large general hospital. Internet of Things sensors are installed on the medical equipment in each department of the hospital, such as CT machines, magnetic resonance imaging (MRI) devices, monitors, etc. These sensors collect device data at regular intervals (such as every 5 minutes), including device ID, timestamp, operating status (running, faulty, idle), usage duration, fault records, and other information. Taking a certain monitor as an example, its device ID is "001-JHY-01". At 10:00:00 on October 1, 2024, the data collected by the sensor was: timestamp "2024-10-01 10:00:00", operating status "running", usage duration "3 hours and 20 minutes", and fault record "none".

[0028] The collected raw data will first be stored in a time series database. Before storage, cleaning and preprocessing operations will be carried out: write a program to detect and remove duplicate data; for missing values, if the usage duration is missing, fill it with the median of the device's historical usage duration; use the Z-Score method to detect outliers, set the threshold to 3, if the Z-Score value of a device's usage duration is greater than 3, it is determined as an outlier and will be removed or corrected; use the normalization formula to scale the data to the 0-1 interval for subsequent analysis. The normalization formula is as follows: In the formula, means subtracting the minimum value min(D) of the data set from the data D to adjust the minimum value of the original data to 0. means calculating the range of the data set (the difference between the maximum value and the minimum value), which is used for the scaling range of normalizing the data.

[0029] In a specific implementation, the blockchain storage module is specifically:[[]]END] For the device data that has been cleaned and preprocessed, an improved encryption storage method is adopted. In the encryption stage, a randomly generated salt value is appended to each device usage record. The salt value generation algorithm is: salt = HMAC-SHA256(timestamp||deviceID,secretKey). For example, add the salt value "abc123" to the record of the above monitor, and then use the SHA-256 algorithm for encryption. The encryption process is: first fill the complete data including device ID, timestamp, operating status, usage duration, fault record, and salt value to make its length a multiple of 512, then divide it into blocks of 512 bits, and recursively calculate the hash values of each block using the SHA-256 hash function.

[0030] The encrypted data is encapsulated into blocks and added to the Merkle tree. The construction process of the Merkle tree is as follows: Each encrypted device record is used as a leaf node, and the hash values of adjacent leaf nodes are combined to calculate a new hash value, forming the nodes of the upper layer, and so on until the root node is generated. When querying or verifying data, through the root hash value and the hash values of sibling nodes on the path from the leaf node to the root node, the integrity of the data can be quickly verified. The specific construction is as follows: Construct the Merkle tree data structure, and generate the root hash value through layer-by-layer hash aggregation (parentHash = SHA-256(leftChild||rightChild)), supporting efficient integrity verification.

[0031] For example, when it is necessary to verify the integrity of a certain record of the monitor, provide the hash value of the leaf node of this record and the hash values of sibling nodes on the path, and compare them with the root hash value. If they are the same, it proves that the data is complete and has not been tampered with.

[0032] The consensus mechanism module adopts an improved Proof of Authority (PoA) algorithm. The primary verification nodes are selected through a random number generator (seed = SHA-256(blockNumber||systemTime)). The verification process includes three stages: transaction preprocessing, hash value broadcasting, and node synchronization, ensuring low-latency data consistency in the blockchain network.

[0033] In a specific implementation, the smart contract module is specifically as follows: The smart contract is written in Solidity language and deployed on the Ethereum blockchain platform. Define a structured data model for device operation records (including device ID, operator digital certificate, Unix timestamp, operation type enumeration value), and use a mapping (bytes32 => Record[]) data structure to establish an index mapping from the device ID to the operation record; integrate precompiled functions to implement record legality verification, and ensure the coherence of business logic by comparing adjacent record timestamps (Δt>0) and the state transition matrix (such as "idle → running" is legal, "fault → shutdown" is illegal).

[0034] In the contract, define the standard format of medical device usage records. For example, the device ID is of string type with a length not exceeding 32 bits; the operator ID is in the format of an Ethereum address; the usage time is in the timestamp format; the operation type is an enumeration type, including "start", "shutdown", "maintenance", "scheduling", "replenishment", etc.

[0035] When the device operates, for example, a certain CT scanner is started by an operator "0x1234567890abcdef..." at 9:00 on October 2, 2024, the smart contract will automatically receive the recording request. Use a mapping data structure to map the device ID of the CT scanner to the corresponding operation record list. After hashing and encrypting the record, insert it into the blockchain through the Merkle tree, and trigger a synchronous update to push the data to all nodes. At the same time, the smart contract will compare the timestamps and device statuses in the previous and current records to verify the record order and logical coherence. If the CT scanner was in an idle state and the timestamp was in logical order before this startup operation, the record is successful; if the status is running or the timestamp is unreasonable, the record is rejected and an error message is returned.

[0036] In terms of inventory management, the smart contract sets the minimum inventory level for each type of device. Taking a certain type of syringe as an example, the minimum inventory level is set at 100 pieces. When the inventory management module detects that the inventory of this type of syringe is less than 100 pieces, it triggers the replenishment logic in the smart contract. Calculate the quantity to be replenished as shown in the following formula: Assume that the current inventory Icurrent is 80 pieces, then ΔI = 100 - 80 = 20 pieces, and then initiate a procurement or replenishment operation.

[0037] In a specific implementation, the inventory management module is specifically: The hospital sets up a dedicated equipment warehouse, and the inventory management module manages the medical equipment in the warehouse. Use the FIFO algorithm and the minimum inventory level algorithm to ensure inventory rationality. During inbound management, the equipment is warehoused in order of arrival time, and the inbound time is recorded. For example, a batch of a certain type of scalpel is warehoused at 10:00 on October 3, 2024, and the inbound quantity is 50 pieces, and the inbound time is recorded as "2024-10-03 10:00".

[0038] During outbound management, give priority to the equipment that was warehoused earliest. When there is a surgical need to requisition scalpels, the system follows the FIFO algorithm and gives priority to outbound the scalpels warehoused on October 3, 2024. At the same time, continuously monitor the inventory level. When the inventory of a certain equipment is lower than the minimum inventory level, such as the inventory of a certain type of syringe is less than 100 pieces, immediately trigger the replenishment process and operate according to the replenishment strategy set by the smart contract to ensure sufficient inventory and reduce the occurrence of equipment out-of-stock situations.

[0039] In a specific implementation, the scheduling optimization module is: The scheduling optimization module performs equipment scheduling based on the linear programming algorithm. Taking the scheduling of surgical equipment in a hospital as an example, define the objective function as minimizing the equipment scheduling cost, as shown in the following formula: Among them is the scheduling cost of each surgical device (such as equipment transfer cost, commissioning cost, etc.), is the scheduling quantity of the corresponding device, is the delay cost, is the demand delay quantity. The set constraints include equipment availability (such as whether a certain surgical device is idle and can be used normally within a specific time) and demand quantity (determined according to the surgical arrangement, the quantity of equipment required).

[0040] By solving the linear programming problem, the optimal scheduling plan is obtained. Assume that within a certain time period, there are 3 different types of surgical devices available for scheduling. The scheduling costs of different devices and the demands of each surgery for the devices are different. Through linear programming calculation, determine the equipment combination with lower scheduling cost and meeting the surgical requirements first, realize the optimal allocation of equipment resources, improve the equipment utilization efficiency, and reduce the equipment idle time.

[0041] In a specific implementation manner, the prediction and analysis module is specifically as follows: The prediction and analysis module uses historical equipment usage data for equipment demand prediction. Collect the medical equipment usage data of the hospital in the past year, including the usage frequency, usage duration, demand conditions in different time periods, etc. of each equipment. Use the ARIMA model and LSTM network for prediction.

[0042] Taking a certain high-frequency electrosurgical unit as an example, first use the ARIMA model to conduct time series analysis on its historical usage data. According to the formula: Among them is the predicted value, is the previous period usage data, represents the random error term, which reflects the difference between the actual value and the predicted value based on the model at time t; through the fitting of historical data and parameter estimation, the coefficients α and β of the model are obtained. Among them, α represents the constant term of the model, and β represents the autoregressive coefficient. Then combine the LSTM network, take the historical usage data as the input, and the trained LSTM network learns the time series features and trends in the data. Combining the results of the ARIMA model and the LSTM network, predict the usage demand of this high-frequency electrosurgical unit in the next week. The prediction results can assist in inventory management and scheduling decisions, and make preparations for equipment allocation and inventory replenishment in advance.

[0043] In a specific implementation manner, the data query and traceability module is specifically as follows: When hospital staff (such as equipment managers, medical staff, etc.) need to query equipment information, they input query conditions (such as equipment ID, time range, equipment type, etc.) through the user interface provided by the system. For example, if an equipment manager wants to query the usage records of all CT machines from October 1st to October 10th, 2024, they select the equipment type as "CT machine" and the time range as "2024-10-01" to "2024-10-10" in the query interface and then click the query button.

[0044] The system queries in the blockchain storage module according to the index mechanism through the data query and traceability module. The index mechanism adopts a B+ tree structure and builds indexes based on information such as equipment ID and timestamp. When querying, according to the input query conditions, it quickly locates the corresponding records in the B+ tree, then obtains the encrypted data from the blockchain, decrypts it and displays it to the user, realizing the whole-process traceability of equipment usage records and inventory scheduling records, facilitating staff to understand the historical situation of equipment and providing a basis for equipment management and maintenance.

[0045] In a specific implementation manner, the user management module is specifically: The hospital assigns a unique identity identifier to each person with permission to access the system and conducts identity verification by combining password verification and digital certificate verification. When staff log in to the system, they need to enter their username, password, and insert the digital certificate for identity confirmation.

[0046] The system assigns different permissions according to the user roles (such as equipment administrators, medical staff, department directors, etc.). Equipment administrators have all permissions such as equipment information management, inventory management, and scheduling management; medical staff only have the permissions to query equipment usage records and apply for equipment borrowing; department directors have the permissions to statistically query equipment usage situations and conduct partial scheduling approvals. Through strict permission management, the security of system data and the standardization of operations are ensured, preventing unauthorized personnel from accessing and tampering with data.

[0047] In a specific embodiment, this embodiment provides the overall operation process of the system.

[0048] As Figure 2 shown, in daily operation, the data collection module continuously collects the usage data and basic information of medical equipment and performs cleaning and preprocessing. The processed data is encrypted and stored in the blockchain through the blockchain storage module, and at the same time, the smart contract module automatically manages and verifies the equipment usage records. The inventory management module conducts inventory management according to the rules set by the smart contract and the equipment usage situation, using the FIFO algorithm and the minimum inventory level algorithm to replenish inventory in a timely manner.

[0049] The scheduling optimization module combines the real-time data of the data acquisition module and the prediction results of the prediction analysis module, and generates an optimal scheduling plan through a linear programming algorithm to achieve the reasonable allocation of equipment. The prediction analysis module continuously uses historical data for equipment demand prediction to provide support for inventory management and scheduling decisions.

[0050] Staff can query equipment-related information through the data query and traceability module, and the user management module ensures the security of system access and the standardization of operations. The entire system ensures the real-time update and consistency of data in the blockchain network through an improved PoA consensus mechanism. The main verification node broadcasts to other nodes after verifying the transaction to complete synchronization, ensuring the accuracy and consistency of data in each module, and realizing the intelligence and efficiency of medical equipment management.

[0051] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device.

[0052] Finally, it should be noted that the above are only preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An intelligent medical device management system, characterized in that: Including: A data acquisition module, which is used to collect the usage data of medical devices in real time through Internet of Things devices and generate the basic information of medical devices; A blockchain storage module, which is used to encrypt and store the usage data and basic information in the blockchain, and manage the data on the blockchain by using the Merkle tree structure; A smart contract module, which is used to define the rules for medical device usage records and inventory management, and realize the automatic creation and verification of device usage records and the automatic execution of inventory scheduling strategies; An inventory management module, which is used to manage the inventory information of medical devices based on the rules set by the smart contract module; A scheduling optimization module, which is used to according to the real-time data obtained by the data acquisition module and the prediction results of the prediction analysis module; A prediction analysis module, which is used to utilize historical device usage data, predict the change of device demand through time series analysis algorithms and machine learning algorithms, and assist in inventory management and scheduling decisions; A data query and traceability module, which is used to allow authorized users to query the usage records and inventory scheduling records of medical devices; A user management module, which is used to implement user permission management, ensure that only authorized users can access and operate the system, and combine the authentication mechanism to ensure data access security.

2. The intelligent medical device management system according to claim 1, characterized in that: The management method of the Merkle tree structure includes: taking the encrypted device usage records and inventory change records as leaf nodes, generating the root hash value through layer-by-layer hash merging, and realizing the rapid verification of data integrity by verifying the root hash value and the sibling node hash values on the path from the leaf node to the root node.

3. An intelligent medical device management system according to claim 1, characterized in that: The device usage record format defined by the smart contract module includes device ID, operator ID, usage time, operation type, and the operation type includes start, shutdown, maintenance, scheduling, replenishment; The smart contract module automatically receives a record request when the device operation is completed, maps the device ID to the corresponding operation record list by using a mapping data structure, encrypts the record by hashing and inserts it into the blockchain through the Merkle tree, and triggers a synchronous update to push the data to all nodes.

4. An intelligent medical device management system according to claim 1, characterized in that: The Internet of Things devices of the data acquisition module include sensors, which are used to collect the on / off time, running status, running duration, and fault records of the devices in real time; The data acquisition module cleans and preprocesses the original data, including removing duplicate data, filling in missing values, detecting and removing outliers, and data standardization.

5. An intelligent medical device management system according to claim 1, characterized in that: The time series analysis algorithm adopted by the prediction analysis module includes the ARIMA model, and the machine learning algorithm includes the LSTM network. The model is trained through historical device usage data to predict future device demand.

6. An intelligent medical device management system according to claim 1, characterized in that: The linear programming algorithm of the scheduling optimization module takes minimizing the device scheduling cost as the objective function, takes device availability and demand as the constraints, and solves to obtain the optimal scheduling plan.

7. An intelligent medical device management system according to claim 1, characterized in that: The authentication mechanism of the user management module includes biometric recognition, password verification or digital certificate verification, ensuring that only authorized personnel can access device data and perform inventory scheduling operations.

8. A medical device management method, characterized in that: Using the intelligent medical device management system according to any one of claims 1-7, characterized in that it includes the following steps: Data collection step: Real-time collect the usage data of medical devices through Internet of Things devices, and generate the basic information of medical devices through the device management module; Data processing step: Clean and preprocess the collected usage data, and identify device usage patterns through clustering algorithms; Data storage step: After appending salt values to the processed usage data and basic information, perform SHA-256 hash encryption, encapsulate them into blocks and store them in the blockchain through the Merkle tree structure; Smart contract execution step: Define device usage record rules and inventory management rules through smart contracts, automatically create and verify device usage records, and trigger replenishment operations according to inventory levels; Predictive analysis step: Use historical data to predict device demand through time series analysis and machine learning algorithms; Scheduling optimization step: Generate an optimal scheduling plan through linear programming algorithms based on real-time data and prediction results; Query and traceability step: Authorized users can query device usage records and inventory scheduling records through the data query and traceability module.

9. A medical device management method according to claim 8, characterized in that: The clustering algorithm uses the K-means algorithm, and clusters according to features such as device usage duration, frequency, and number of failures to identify frequently used devices, idle devices, and moderately used devices.

10. A medical device management method according to claim 8, characterized in that: In the data storage step, the root node of the Merkle tree contains the hash values of all device records. By hashing the records into leaf nodes and gradually merging them upward to generate the root hash value, data integrity verification is achieved by verifying the root hash value and the verification path.

Images