Medical equipment remote control system based on Internet of Things data

Abstract

The invention discloses a medical equipment remote control system based on Internet of Things data, relates to the technical field of equipment remote control, and realizes efficient safety protection of medical equipment remote control through four links of multi-source data acquisition and behavior baseline construction, real-time anomaly detection, risk grading evaluation and linkage processing, and adaptive safety strategy execution. Firstly, an operation log is collected and cleaned, a multi-dimensional label is added, and a baseline model M is constructed to judge the deviation degree; then, deviation degree calculation is carried out on the instructions in real time, suspicious instructions are screened, and a structured suspicious instruction set is formed; in the risk grading stage, performing multi-factor grading on suspicious instructions and screening medium-high risk objects; and finally, according to the risk vector, calculating the priority of a strategy and matching a predefined interval, implementing security protection through measures such as multi-factor authentication, secondary approval or temporary freezing, and continuously writing back and updating a disposal result to form a closed loop to improve the overall security management and control level.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment remote control, and in particular to a medical equipment remote control system based on Internet of Things data. Background Art

[0002] With the widespread penetration of IoT technology in the medical industry, various types of intelligent medical equipment have gradually realized networking, dataization and remote management, helping hospitals improve resource utilization efficiency and provide patients with more timely and accurate diagnosis and treatment services. In existing applications, key medical equipment such as infusion pumps, monitors, imaging equipment and surgical assistance robots are often connected to hospital information systems through wireless networks or wired private networks, and interconnected with remote terminals or mobile devices to achieve cross-regional operations and data exchange. However, medical scenarios often involve different departments, medical staff at different levels and diversified clinical processes. There are also differences in models and functional complexity between devices. In addition, the dynamic changes in patients' conditions and the uncertainty of the network environment make remote control and real-time monitoring processes face higher security and reliability challenges. Although the remote control architecture based on IoT data can significantly improve the efficiency of diagnosis and treatment, how to ensure the accuracy and security of data transmission and the compliance of operation processes while maintaining high availability of equipment has become an important issue that needs to be solved in the current deepening of medical IoT applications.

[0003] In the Chinese invention patent with the authorization announcement number CN118534832B, a medical equipment remote control method and system are disclosed, which includes the following steps: collecting real-time operating parameters of remote medical equipment, collecting data through sensors, screening key parameters of temperature, power consumption and response time, identifying abnormal data points through preset thresholds, and generating a key operating parameter abnormality list. In the present invention, sensors are used to comprehensively monitor key operating parameters and actively identify abnormalities, timely discover potential problems of medical equipment and allow rapid intervention, prevent small problems from turning into major failures, not only ensure that medical equipment continues to operate in the optimal state, but also through in-depth analysis of equipment adjustment logs, accurate performance evaluation records are made, and operations are further optimized. By implementing data-based decisions, remote operation standards and protocols are updated to ensure the high quality and safety of medical services, especially the response speed and efficiency when dealing with acute diseases and emergency medical situations.

[0004] How to identify and handle possible abnormal operation behaviors in real time and in detail when remotely controlling medical equipment. Specifically, medical equipment often involves multi-dimensional data collection and control instructions, including patient physiological parameters, equipment operating status, clinical use scenarios, etc., and these data are easily misjudged due to delays, poor synchronization, inconsistent data, or lack of context information during network transmission and cross-system docking. Once medical staff or external systems issue abnormal instructions to the equipment (such as sudden large-scale parameter modifications, unauthorized personnel operating equipment during non-working hours, etc.), if they cannot be detected and corresponding measures are taken in time, it may cause serious risks to patient safety or stable operation of the equipment. At present, the general monitoring mechanism mostly stays at the level of simple threshold alarm or event tracking based on fixed rules. It is difficult to achieve flexible real-time identification and dynamic security protection when complex and multi-dimensional changes occur, which brings great hidden dangers to the controllability and reliability of remote control of the medical Internet of Things.

[0005] To this end, the present invention provides a medical equipment remote control system based on Internet of Things data. Summary of the invention

[0006] 1. Technical issues to be resolved In view of the shortcomings of the prior art, the present invention provides a remote control system for medical equipment based on Internet of Things data, which realizes efficient and safe protection of remote control of medical equipment through four links: multi-source data collection and behavior baseline construction, real-time anomaly detection, risk classification assessment and linkage processing, and adaptive security policy execution. First, collect and clean the operation logs and add multi-dimensional labels to build a baseline model M to determine the deviation; then, calculate the deviation of the instructions in real time, screen suspicious instructions and form a structured suspicious instruction set; in the risk classification stage, perform multi-factor scoring on suspicious instructions and screen medium and high risk objects; finally, calculate the strategy priority based on the risk vector and match the predefined interval, implement security protection through multi-factor authentication, secondary approval or temporary freezing, and continuously write back and update the disposal results to form a closed loop to improve the overall security management level, thereby solving the technical consultant recorded in the background technology.

[0007] (II) Technical solution To achieve the above objectives, the present invention is implemented through the following technical solutions: a medical equipment remote control system based on IoT data, collects medical equipment operation logs, removes duplicates and verifies fields to form a cleaned log data set, adds labels and normalizes to generate a log preprocessing data set, counts the distribution of each dimension and uses Rényi divergence to determine the degree of deviation, and writes the results into a baseline model M; Receive the original data stream of real-time instructions and align them to generate real-time instruction data, calculate the deviation from the baseline model M, determine suspicious or normal instructions and output them to the label instruction result set, filter suspicious instructions and encapsulate them into a structured suspicious instruction set, and realize multi-dimensional and highly sensitive anomaly recognition; Read the structured suspicious instruction set and supplement the context to synthesize a temporary data set, perform multi-factor risk scoring on each suspicious instruction, generate risk scoring output results, screen medium and high risk instructions to obtain risk linkage processing output objects, and notify the administrator or start the linkage process based on the scoring results; Obtain medium and high risk instructions from the risk linkage processing output object, form a temporary risk instruction data set, calculate the policy priority based on the risk vector and match the predefined interval, output it to the policy mapping output object, and execute the corresponding security measures based on the policy mapping output object.

[0008] Further, each log record in the cleaned log data set is read, and an operation role label, an instruction type label, and a time segment label are added to it; Perform normalization mapping on logs containing numeric fields and store new values ​​in additional fields; add hierarchical labels to each record, output the final log preprocessing dataset, and create an empty baseline model M based on established medical processes and historical experience, which contains a multi-dimensional distribution structure.

[0009] Furthermore, the actual frequency distribution of each key dimension is counted from the log preprocessing data set, denoted as , using Renyi divergence to compare with the reference distribution and The difference If the dimension If the value exceeds the sensitivity threshold, Write the baseline model M as the new reference distribution, otherwise keep the original reference distribution unchanged, where ; Where: For the An operation mode under a dimension The actual frequency value of is the corresponding frequency of the reference distribution in the same dimension, is the sensitivity control parameter, For the The set of all possible values ​​for a dimension.

[0010] Furthermore, each field in the raw data stream of the real-time instruction is preliminarily checked for consistency and aligned with the dimensions required by the baseline model M. If a missing or abnormal field is encountered, the record is marked with a field abnormality but its core information is still retained; Take each instruction from the aligned real-time instruction data , locate the role, time period, instruction type, the corresponding reference center vector m and sensitivity factor in M parameter.

[0011] Furthermore, define collaborative anomaly formula measurement instructions Deviation from the reference center vector m , as follows: ; in, For the current instruction Numerical features on key dimensions, is the reference center value vector in the baseline model M, For the The sensitivity factor of the dimension, and is the power coefficient, : Synergistic anomaly amplification factor; If the deviation , mark the instruction as suspicious; otherwise, it is considered normal and the judgment result is stored in the label instruction result set A1. Values ​​and suspicious or normal labels; instructions marked as normal are only written into the normal operation log for future reference, and no longer trigger the subsequent high-intensity risk assessment process.

[0012] Further, all instruction sets determined to be suspicious are read from the structured suspicious instruction set, the target fields are parsed out, and the sets are summarized into a temporary data set; if any suspicious instruction in the structured suspicious instruction set is missing a required field, a field missing mark is made for it in the temporary data set; Based on the information such as the device type, the operator's department, the current diagnosis and treatment link, etc. of each suspicious instruction in the temporary data set, the internal clinical context database is queried to supplement the environmental factors.

[0013] Furthermore, the baseline model M or the configuration center reads the risk-related weight set , magnification factor and several nonlinear parameters (such as After numerically expressing the feature vector z of each suspicious instruction in the temporary data set, the following risk score is defined: : ; Where: z is a multidimensional feature vector, For the The numerical expression of the risk factor, is the weight factor, is the nonlinear amplification coefficient of a single factor, is the aggregation coefficient, is a nonlinear parameter, Used to control the impact of synergistic product terms on overall risk; If , recorded as low risk; if , recorded as medium risk; if , recorded as high risk, and the risk score and risk level of each suspicious instruction are packaged into the new object risk score output result, and the necessary context traceback information is also saved.

[0014] Furthermore, instructions with medium or high risk levels are screened out from the risk scoring output results, and it is further determined whether it is necessary to immediately notify the medical supervisor or security administrator. For instructions marked as low risk, they are only retained for audit and do not enter the mandatory security policy process. For suspicious instructions with medium or high risks, their core information is encapsulated into the risk linkage processing output object, and the current server timestamp is attached to trigger notification or security administrator intervention in a timely manner.

[0015] Further, the data is merged to form a temporary risk instruction data set, and the record contains at least the following fields: risk score S, risk level and emergency mark U; divide the records in the temporary risk instruction data set T4 into medium risk group and high risk group, and retain the value of U. If the record is an emergency scene, mark it with emergency priority mark; Combine the key elements of each record into a vector , define the following priorities , comprehensively evaluate the final strategy strength: ; Where: for of -norm, often taken to represent the Euclidean length, Used to gradually increase the risk from 0 to The process of amplification, Then measure The synergistic lifting effect of each component, All of them are control coefficients greater than 0. is the identity matrix of the corresponding dimension.

[0016] Further, after calculating the priority Then, compare the result with several predefined intervals. , follow a lower-intensity protection process, additional identity verification or warning reminders; if , follow the medium-intensity protection process, and the multi-factor authentication and approval process is high-intensity; If the priority High-intensity protection procedures are adopted, such as temporarily freezing operations, forcing secondary approval, or allowing operations in emergency scenarios but recording and auditing the entire process; the final priority of each record is Writes the policy-map output object with the selected policy-map results.

[0017] (III) Beneficial effects The present invention provides a medical equipment remote control system based on Internet of Things data, which has the following beneficial effects: 1. Through multi-source data collection and baseline model construction, a reference standard for subsequent anomaly detection was successfully established, providing a quantifiable benchmark for real-time monitoring and accurate identification; unqualified data was screened out in advance and legal data was labeled in multiple dimensions, laying a solid foundation for subsequent real-time or batch detection and analysis, and improving the system's adaptability to complex scenarios.

[0018] 2. The use of nonlinear deviation formula can provide more sensitive detection capabilities when there are multi-dimensional anomalies in the operation characteristics; small but continuous deviations can also be discovered in time. After determining whether the instruction is suspicious or normal, the suspicious instruction will be further packaged immediately, shortening the transmission cycle of suspicious information from detection to subsequent response, and achieving more timely risk isolation.

[0019] 4. Output the test results in a structured manner, retaining key fields such as operator, timestamp, and deviation, which is convenient for quick reading by the risk assessment module in the third step and for subsequent analysis. Only logs are retained for normal instructions to improve overall processing efficiency, and the main calculations and resources are concentrated on higher-risk suspicious instructions to improve resource utilization.

[0020] 5. Through the combined scoring of multi-dimensional factors (role, abnormal type, clinical scenario, etc.), the operational risk can be quantified more precisely, no longer limited to a single indicator. The scoring results are mapped to low, medium, and high levels, and combined with the threshold strategy pre-set by third parties or in-hospital experts, suspicious instructions can be quickly classified into different risk levels, facilitating subsequent differentiated processing.

[0021] 6. Through detailed risk assessment and graded disposal of suspicious instructions, a one-size-fits-all approach is avoided, and precise and hierarchical security management is achieved. In addition to improving the accuracy of identifying abnormal risks, it also provides complete context and graded information for subsequent security policy execution, ensuring that security measures are more targeted and feasible.

[0022] 7. Differentiated security measures are adopted for different risk levels and degrees of urgency. On the one hand, interference with the normal operation of the hospital is reduced. On the other hand, extreme or malicious operation incidents are strictly prevented. The disposal information is traceable and auditable throughout the process, and the actual disposal results are applied to the subsequent upgrades and iterations of the system itself, so as to continuously improve the overall security management level. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a schematic diagram of the process of the medical equipment remote control system based on Internet of Things data of the present invention. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] See also Figure 1 The present invention provides a medical equipment remote control system based on Internet of Things data, including: Step 1: Collect medical equipment operation logs, remove duplicates and verify fields to form a cleaned log data set, add labels and normalize to generate a log preprocessing data set, count the distribution of each dimension and use Rényi divergence to determine the degree of deviation, and write the results into the baseline model M; The step 1 includes the following contents: Step 101: Obtain various operation logs through the log interface, remove duplicate records and verify field integrity. The output result is the cleaned log data set D1. If serious format errors or missing key fields are found, the corresponding records will be marked as abnormal. Read each log record in the cleaned log data set D1, add operation role labels (such as doctor, nurse), instruction type labels (such as equipment adjustment, access query) and time period labels (such as working hours, night time) to it, and generate log preprocessing data set D2. For logs containing numerical fields (such as network delay and operation frequency), normalization mapping is performed in the log preprocessing data set D2. To reduce the impact of extreme values ​​on subsequent analysis, logarithmic mapping or interval mapping is used to store new values ​​in additional fields. Add department or section-level labels to each record to distinguish the operational differences of different medical functional groups in the future. After the processing is completed, the final log preprocessing data set D2 is output to provide dimensional cleaning and annotation data for the subsequent baseline model construction; When in use, by performing deduplication and field verification on the original log data, duplicate records can be effectively eliminated and obvious format errors can be captured, improving the accuracy and completeness of subsequent analysis data, and abnormal or missing information can be separately marked and retained to ensure that subsequent links can take protective measures or supplementary corrections based on the mark, thereby reducing missed detections or false detections caused by source data problems. Adding multiple tags (role, instruction type, time period, etc.) to each log record can more accurately characterize operational behavior characteristics in subsequent dimensional analysis to avoid important information being omitted. Using a normalization algorithm to map numeric fields can alleviate the interference of extreme values ​​on the overall statistical model, making the subsequent baseline model more stable in a variety of clinical and operational environments.

[0026] Step 102: Based on the established medical process and historical experience, create an empty baseline model M, which contains a multi-dimensional distribution structure (role-time period-instruction type-department); The actual frequency distribution of each key dimension (such as role, time period, instruction type, and department) is counted from the log preprocessing data set D2, which is recorded as In order to improve the sensitivity of identifying distribution differences, the following Rényi divergence is used to compare the reference distribution (derived from historical statistics) and The differences: ; Where: For the An operation mode under a dimension The actual frequency value of is the corresponding frequency of the reference distribution in the same dimension, is a sensitivity control parameter, with a value greater than 1. For the The set of all possible values ​​of a dimension. The larger the divergence, the more obvious the deviation between the current distribution and the historical reference baseline. A certain dimension If the value exceeds the sensitivity threshold, Write the new reference distribution into the baseline model M, otherwise keep the original reference distribution unchanged. In this way, the baseline can be updated in time when the medical work mode changes, ensuring that the judgment of abnormal operations is more timely. Finally, the updated baseline model M is obtained, and the distribution calculation results and related thresholds are archived, so that the next step of real-time anomaly detection can determine the deviation of new operation instructions; When in use, by calculating the statistical distribution of different dimensions and judging the degree of deviation based on advanced metrics (such as Rényi divergence), it can flexibly adapt to changes in operating modes in medical scenarios and improve sensitivity to abnormal trends. The generated behavioral baseline model records reference distribution information in multiple dimensions. Once subsequent operations deviate significantly, it can quickly determine whether there are potential risks or abnormalities, greatly reducing the burden of manual inspections.

[0027] When in use, combine the contents of steps 101 and 102: through multi-source data collection and baseline model construction, successfully establish a reference standard for subsequent anomaly detection, provide a quantifiable benchmark for real-time monitoring and accurate identification; screen out unqualified data in advance and perform multi-dimensional annotation on legal data, lay a solid foundation for subsequent real-time or batch detection and analysis, and enhance the system's adaptability to complex scenarios.

[0028] Step 2: Receive the original data stream of real-time instructions and align it to generate real-time instruction data, calculate the deviation from the baseline model M, determine suspicious or normal instructions and output them to the label instruction result set, filter suspicious instructions and encapsulate them into a structured suspicious instruction set, and realize multi-dimensional and highly sensitive anomaly recognition; The step 2 includes the following contents: Step 201: Receive a real-time instruction stream from an external operation source and aggregate it into an original data stream R0 of real-time instructions, where each record contains (timestamp, operator ID, instruction type, terminal information, etc.); perform a preliminary consistency check on each field in the original data stream R0 of real-time instructions, and align it with the dimensions required by the baseline model M (role, time period, department, etc.), generate aligned real-time instruction data R1, and if there is a missing or abnormal field, mark the record with a field abnormality but still retain its core information; When in use, the newly arrived operation instruction data is quickly aligned with the fields required by the baseline model M to reduce recognition delays or errors caused by inconsistent formats, ensure that the data can be directly used for deviation calculations, and make special marks on missing or abnormal field records to ensure that such incomplete data can be flexibly handled in subsequent detection and risk assessment, rather than simply discarded or ignored.

[0029] Step 202: Take each instruction from the aligned real-time instruction data R1 , locate the reference center vector m and sensitivity factor corresponding to its role, time period, and instruction type in M And other parameters, Define collaborative exception formula measurement instructions Deviation from the reference center vector m , which not only accumulates the deviation of a single dimension, but also introduces the detection of multi-dimensional synergy effects through product terms, as follows: ; in, For the current instruction Numerical features on key dimensions, is the reference center value vector in the baseline model M, For the The sensitivity factor of the dimension is used to adjust the contribution of the dimension to the overall deviation. and is a power coefficient, and its value is greater than 1 to increase the penalty for outliers and multi-dimensional anomalies. : The collaborative anomaly amplification coefficient is greater than 1. When multiple dimensions deviate at the same time, this product term will be significantly amplified. , improve the sensitivity to complex abnormalities, It is used to constrain possible extreme large values ​​and ensure the numerical stability of the deviation results. The deviation threshold from the baseline model M Make comparisons; If the deviation , marking the instruction as suspicious; otherwise, it is considered normal. At the same time, the baseline model M can be allowed to set multiple thresholds (such as ) to facilitate more refined risk grading in the future; store the judgment results in the label instruction result set A1, and attach values ​​and suspicious or normal labels; When used, the nonlinear deviation formula can provide more sensitive detection capabilities when there are multi-dimensional anomalies in the operating characteristics; small but continuous deviations can also be discovered in time. After determining whether the instruction is suspicious or normal, the suspicious instruction will be further packaged immediately, shortening the transmission cycle of suspicious information from detection to subsequent response, and achieving more timely risk isolation.

[0030] Step 203: Filter all suspicious instructions from the label instruction result set A1, together with their , role, time period and other key fields are encapsulated into the structured suspicious instruction set A2. For instructions marked as normal, they are only written into the ordinary operation log for future reference and no longer trigger the subsequent high-intensity risk assessment process. The structured suspicious instruction set A2 (including suspicious instructions and related context) will be sent to the third step of risk classification assessment and linkage processing for the risk assessment module to comprehensively determine the risk level and execute the corresponding security strategy; When in use, the detection results are output in a structured manner, retaining key fields such as the operator, timestamp, and deviation. This is convenient for quick reading by the risk assessment module in the third step and for subsequent analysis. Only logs are retained for normal instructions to improve overall processing efficiency, concentrate main computing and resources on higher-risk suspicious instructions, and improve resource utilization.

[0031] When in use, combined with the contents of steps 201 to 203: real-time and efficient identification of potential suspicious behaviors, not only can risk warnings be issued at the stage of operation, but also lay an information foundation for subsequent in-depth risk assessments. Through the multi-dimensional deviation calculation method, the adaptability to complex abnormal scenarios is greatly improved, reducing the reliance on single thresholds or simple rules.

[0032] Step 3: Read the structured suspicious instruction set and supplement the context to synthesize a temporary data set, perform a multi-factor risk score on each suspicious instruction, generate a risk score output result, screen medium and high risk instructions to obtain a risk linkage processing output object, and notify the administrator or start the linkage process based on the score result; The step three includes the following contents: Step 301: read all suspicious instruction sets from the structured suspicious instruction set A2, parse out the target fields (operator ID, role category, instruction type, deviation value, etc.), and summarize them into a temporary data set C3; if any suspicious instruction in the structured suspicious instruction set A2 lacks a required field (such as role information, time period label), mark it as missing field in the temporary data set C3 for reference in subsequent decision-making; Based on the equipment type, operator department, current diagnosis and treatment link and other information of each suspicious instruction in the temporary data set C3, the internal clinical context database is queried to supplement environmental factors (such as the patient's condition level, whether the equipment is in a critical use stage, etc.). This process packs the context elements of each suspicious instruction into the extended field of C3, so that each instruction has a multi-dimensional feature description (role, equipment criticality, clinical importance, etc.); if the necessary information cannot be obtained during the synthesis process, the null value is retained in C3 and an abnormal mark is attached; When in use, while reading suspicious instructions, supplementing clinical or business context (such as equipment type, department information, and patient condition) can make subsequent risk scoring more in line with medical reality, and explicitly mark instructions that lack key fields, providing a higher tolerance rate for subsequent grading and linkage, and reducing misjudgments or missed judgments.

[0033] Step 302: The baseline model M or the configuration center reads the risk-related weight set , magnification factor and several nonlinear parameters (such as ,It should be noted that these parameters are pre-defined by security administrators or ,clinical experts based on different business scenarios and equipment types, ,which aim to reflect differentiated sensitivity in multi-dimensional ,abnormal scenarios; In order to characterize the compound risk effect between multiple factors, after numerically expressing the feature vector z of each suspicious instruction in the temporary data set C3, the following risk score is defined: : ; Where: z is the multidimensional feature vector extracted from C3 (such as operator role sensitivity, equipment criticality, deviation, clinical importance, etc.), For the A numerical expression of a risk factor (e.g., a quantitative value of the sensitivity of a role), is the weight factor, which measures the The importance of each risk factor relative to the whole is the nonlinear amplification coefficient of a single factor, and its value is greater than 1. The aggregation coefficient is the overall power amplification of the weighted sum of all factors. The value is greater than 1, which can improve the overall risk recognition of multiple factors that are simultaneously high. is a nonlinear parameter that processes each factor to a power in the product term, and its value is greater than 1. It is used to control the impact of the synergistic product term on the overall risk. The value is greater than 1. The larger the value, the more attention is paid to the joint increase of multi-dimensional risk factors. Calculating risk scores After that, with multiple configured scoring thresholds (such as , , ) comparison, like , recorded as low risk; if , recorded as medium risk; if , recorded as high risk.

[0034] The risk score and risk level of each suspicious instruction are packaged into the new object risk score output result R2. The risk score output result R2 also saves the necessary context traceback information (such as operator ID, device ID, and timestamp) to ensure the feasibility of subsequent auditing and tracking. Through the combined scoring of multi-dimensional factors (role, abnormal type, clinical scenario, etc.), and the addition of customizable amplification coefficients and synergy effect parameters, the operational risk can be quantified more finely, no longer limited to a single indicator. The scoring results are mapped to low, medium, and high levels, combined with the threshold strategy pre-set by third parties or in-hospital experts, suspicious instructions can be quickly classified into different risk levels, facilitating subsequent differentiated processing.

[0035] Step 303: Filter out instructions with medium or high risk levels in the risk score output result R2, and further determine whether it is necessary to immediately notify the medical director or security administrator. For instructions marked as low risk, only audit retention is performed and they do not enter the mandatory security policy process; For medium or high-risk suspicious instructions, encapsulate their core information (such as instruction identification, risk score, risk level, context data) into the risk linkage processing output object R3 and attach the current server timestamp; For medium- and high-risk instructions, notifications can be triggered in a timely manner or security administrators can intervene, and the instruction context can be integrated and output to the next step to reduce information loss during the transmission process. Only audit records are made for low-risk instructions, allowing resources to focus on handling higher-risk incidents and improve overall efficiency.

[0036] When used, combined with the contents of steps 301 to 303: by conducting detailed risk assessment and hierarchical disposal of suspicious instructions, a one-size-fits-all approach is avoided, and precise and hierarchical security management is achieved. In addition to improving the accuracy of identifying abnormal risks, it also provides complete context and hierarchical information for subsequent security policy execution, ensuring that security measures are more targeted and feasible.

[0037] Step 4: Obtain medium and high risk instructions from the risk linkage processing output object, form a temporary risk instruction data set, calculate the policy priority based on the risk vector and match the predefined interval, output it to the policy mapping output object, and execute the corresponding security measures according to the policy mapping output object; The step 4 includes the following contents: Step 401: read all suspicious instructions that are judged to be medium or high risk from the risk linkage processing output object R3, and merge them to form a temporary risk instruction data set T4. The record contains at least the following fields: risk score S, risk level and emergency mark U; if some records in the risk linkage processing output object R3 are missing necessary fields, data anomaly marks are made for them in the temporary risk instruction data set T4, and protective strategies are adopted in subsequent execution links (such as defaulting to a higher level of approval or freezing the process); Divide the records in the temporary risk instruction data set T4 into medium risk groups ) and high risk group , and retain the value of U. If the record is an emergency scene ( ), it will be marked with an emergency priority mark, which may trigger a special processing path in the subsequent strategy stage; distinguish emergency scenarios in the preliminary screening, prepare for possible emergency priority channels or stricter control in subsequent strategy decisions, and record potential data anomalies at this time to prevent unexpected conflicts or omissions in the strategy execution stage.

[0038] Step 402: Combine the key elements of each record into a vector , respectively: : Risk score, the higher the value, the higher the potential threat; : Risk level, 1 is medium risk, 2 is high risk, : Emergency mark, 1 indicates an emergency scenario, 0 indicates a normal scenario; define the following priorities , comprehensively evaluate the final strategy strength: ; Where: for of -norm, often taken to represent the Euclidean length, Used to gradually increase the risk from 0 to The process of amplification, Then measure The synergistic lifting effect of each component, All of them are control coefficients greater than 0 and can be adjusted in the system configuration file. is the identity matrix of the corresponding dimension.

[0039] Calculate the priority Then, compare the result with several predefined intervals. If the priority , follow a lower-intensity protection process, additional identity verification or warning reminders; like , follow the medium-intensity protection process, and the multi-factor authentication and approval process is high-intensity; If the priority High-intensity protection procedures are adopted, such as temporarily freezing operations, forcing secondary approval, or allowing operations in emergency scenarios but recording and auditing the entire process; The final priority of each record Write the selected policy mapping result into the policy mapping output object S4, retaining key fields (such as the selected policy name, policy strength level, etc.); When in use, a more flexible assessment of risk vectors can be performed, which can fully reflect the collaborative amplification or layered attenuation characteristics of multi-dimensional information, map the calculation results to the predefined policy interval, and quickly retrieve the corresponding security protection plan, such as multi-factor authentication, secondary approval, temporary freezing or emergency release, etc., to avoid inconsistent policies caused by fuzzy retrieval. Differentiated security measures are taken for different risk levels and urgency levels. On the one hand, it reduces interference with the normal operation of the hospital, and on the other hand, it strictly prevents extreme or malicious operation events, realizes the traceability and auditability of disposal information throughout the process, and applies the actual disposal results to the subsequent upgrades and iterations of the system itself, and continuously improves the overall security management and control level.

[0040] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0041] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0042] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only some logical function divisions. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0043] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0044] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A remote control system for medical equipment based on IoT data, characterized by: include, Collect medical equipment operation logs, remove duplicates and verify fields to form a cleaned log data set, add labels and normalize to generate a log preprocessing data set, count the distribution of each dimension and use Rényi divergence to determine the degree of deviation, and write the results into the baseline model M; Receive the original data stream of real-time instructions and align them to generate real-time instruction data, calculate the deviation from the baseline model M, determine suspicious or normal instructions and output them to the label instruction result set, filter suspicious instructions and encapsulate them into a structured suspicious instruction set, and realize multi-dimensional and highly sensitive anomaly recognition; Read the structured suspicious instruction set and supplement the context to synthesize a temporary data set, perform multi-factor risk scoring on each suspicious instruction, generate risk scoring output results, screen medium and high risk instructions to obtain risk linkage processing output objects, and notify the administrator or start the linkage process based on the scoring results; Obtain medium and high risk instructions from the risk linkage processing output object, form a temporary risk instruction data set, calculate the policy priority based on the risk vector and match the predefined interval, output it to the policy mapping output object, and execute the corresponding security measures based on the policy mapping output object.

2. The medical equipment remote control system based on Internet of Things data according to claim 1, characterized in that: Read each log record in the cleaned log data set and add operation role labels, instruction type labels, and time segment labels to it; Perform normalization mapping on logs containing numeric fields and store new values ​​in additional fields; add hierarchical labels to each record, output the final log preprocessing dataset, and create an empty baseline model M based on established medical processes and historical experience, which contains a multi-dimensional distribution structure.

3. The medical equipment remote control system based on Internet of Things data according to claim 2 is characterized in that: The actual frequency distribution of each key dimension is counted from the log preprocessing data set, denoted as , using the Rényi divergence to compare with the reference distribution and The difference If the dimension If the value exceeds the sensitivity threshold, Write the baseline model M as the new reference distribution, otherwise keep the original reference distribution unchanged, where ； Where: For the An operation mode under a dimension The actual frequency value of is the corresponding frequency of the reference distribution in the same dimension, is the sensitivity control parameter, For the The set of all possible values ​​for a dimension.

4. The medical equipment remote control system based on Internet of Things data according to claim 3 is characterized in that: Perform a preliminary consistency check on each field in the raw data stream of the real-time instruction and align it with the dimensions required by the baseline model M. If a missing or abnormal field is found, the record will be marked with a field abnormality but its core information will still be retained; Take each instruction from the aligned real-time instruction data , locate the role, time period, instruction type, and the corresponding reference center vector in M And the sensitivity factor parameter.

5. The medical equipment remote control system based on Internet of Things data according to claim 4 is characterized in that: Define collaborative exception formula measurement instructions With reference center vector The deviation , as follows: ； in, For the current instruction Numerical features on key dimensions, is the reference center value vector in the baseline model M, For the The sensitivity factor of the dimension, and is the power coefficient, : Synergistic anomaly amplification factor; If the deviation , mark the instruction as suspicious; otherwise, it is considered normal and the judgment result is stored in the label instruction result set A1. Values ​​and suspicious or normal labels; instructions marked as normal are only written into the normal operation log for future reference, and no longer trigger the subsequent high-intensity risk assessment process.

6. The medical equipment remote control system based on Internet of Things data according to claim 5, characterized in that: Read all instruction sets that are determined to be suspicious from the structured suspicious instruction set, parse out the target fields, and summarize them into a temporary data set; if any suspicious instruction in the structured suspicious instruction set is missing a required field, mark it as missing in the temporary data set; Based on the device type, operator department, and current diagnosis and treatment information of each suspicious instruction in the temporary data set, the internal clinical context database is queried to supplement environmental factors.

7. The medical equipment remote control system based on Internet of Things data according to claim 6, characterized in that: The baseline model M or the configuration center reads the risk-related weight set , magnification factor As well as several nonlinear parameters, after numerically expressing the feature vector z of each suspicious instruction in the temporary data set, the following risk score is defined: : ； Where: z is a multidimensional feature vector, For the The numerical expression of the risk factor, is the weight factor, is the nonlinear amplification coefficient of a single factor, is the aggregation coefficient, is a nonlinear parameter, Used to control the impact of synergistic product terms on overall risk; like , recorded as low risk; if , recorded as medium risk; if , recorded as high risk, and the risk score and risk level of each suspicious instruction are packaged into the new object risk score output result, and the necessary context traceback information is also saved.

8. The medical equipment remote control system based on Internet of Things data according to claim 7, characterized in that: In the risk scoring output results, instructions with medium or high risk levels are screened out, and further judgment is made as to whether it is necessary to notify the medical supervisor or security administrator immediately. For instructions marked as low risk, they are only retained for audit and do not enter the mandatory security policy process. For suspicious instructions with medium or high risk, their core information is encapsulated into the risk linkage processing output object, and the current server timestamp is attached to trigger notification or security administrator intervention in a timely manner.

9. The medical equipment remote control system based on Internet of Things data according to claim 8, characterized in that: Merge to form a temporary risk instruction data set, the record contains at least the following fields: risk score S, risk level and emergency mark U; divide the records in the temporary risk instruction data set T4 into medium risk group and high risk group, and retain the value of U. If the record is an emergency scene, mark it with emergency priority mark; Combine the key elements of each record into a vector , define the following priorities , comprehensively evaluate the final strategy strength: ； Where: for of -norm, often taken to represent the Euclidean length, Used to gradually increase the risk from 0 to The process of amplification, Then measure The synergistic lifting effect of each component, All of them are control coefficients greater than 0. is the identity matrix of the corresponding dimension.

10. The medical equipment remote control system based on Internet of Things data according to claim 9, characterized in that: Calculate the priority Then, compare the result with several predefined intervals. , follow a lower-intensity protection process, additional identity verification or warning reminders; if , follow the medium-intensity protection process, and the multi-factor authentication and approval process is high-intensity; If the priority High-intensity protection procedures are adopted, such as temporarily freezing operations, forcing secondary approval, or allowing operations in emergency scenarios but recording and auditing the entire process; the final priority of each record is Writes the policy-map output object with the selected policy-map results.

Images