A virtual ward information data management method and system

Abstract

This invention discloses a virtual ward information data management method and system, relating to the field of smart healthcare technology. The method includes: collecting ward information data through data acquisition equipment; generating a virtual ward based on patient basic information; encrypting the ward information data using a hybrid encryption algorithm; recording the encrypted ward information data in the virtual ward; uploading a portion of the encrypted key ward information data to an online expert interaction platform to generate expert evaluation information; matching the encrypted complete ward information data and the expert evaluation information to generate a processing signal, and sending the processing signal to the medical staff in the corresponding ward; storing the encrypted ward information data, expert evaluation information, and processing signal data in a database, and constructing a data index for rapid data retrieval. By constructing a virtual ward, combined with encrypted data transmission and an online expert interaction platform, the reliability and security of data information transmission are improved.

Description

Technical Field

[0001] This invention relates to the field of smart healthcare technology, specifically to a method and system for managing virtual ward information data. Background Technology

[0002] Currently, hospital ward information management mainly relies on on-call doctors' rounds and registrations, which is inefficient in obtaining patient health information and ward details. For example, in traditional inpatient blood glucose management, nurses perform 3-4 bedside finger-prick blood tests daily, and on-call doctors adjust medications the following day. This leads to a delay in doctors obtaining information about patients' health status, resulting in information processing lag and affecting the accuracy of doctors' management of patient treatment. Existing medical data management uses local area networks to achieve data management of the medical and nursing systems, but lacks comprehensive data management and protection measures for ward information transmission, making it prone to leakage of patient and medical data. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of the prior art. This invention provides a virtual ward information data management method and system. By constructing a virtual ward and combining encrypted data transmission with an online expert interaction platform, it is possible to promptly assess and treat patient information and improve the reliability and security of data information transmission.

[0004] This invention provides a method for managing virtual ward information data, comprising the following steps:

[0005] S1. Collect basic patient information, medical equipment operation data, and environmental monitoring data in the ward using data acquisition equipment;

[0006] S2. A virtual ward is generated based on the patient's basic information. The ward information data collected in step S1 is encrypted using a hybrid encryption algorithm. The encrypted ward information data is recorded in the virtual ward. The hybrid encryption algorithm combines symmetric and asymmetric encryption and updates the encryption key periodically or through event-triggered updates via a dynamic key update mechanism.

[0007] S3. Upload the encrypted key ward information data from step S2 to the online expert interaction platform. The platform will assign experts in the corresponding fields based on the knowledge graph matching mechanism, and use a multi-expert cross-evaluation and adversarial consensus game strategy to analyze and evaluate the data, generating expert evaluation information including a divergence report.

[0008] S4. Match the complete ward information data encrypted in step S2 with the expert evaluation information generated in step S3 to generate a processing signal, and send the processing signal to the medical staff in the corresponding ward.

[0009] S5. The encrypted ward information data, the expert evaluation information, and the processed signal data are stored in a database. A dual-channel isolated storage architecture is used to separate and store patient identity information from medical data. A data index is constructed using the virtual ward ID as the unique index keyword. The patient identity information is encrypted and stored using a ciphertext structure compatible with homomorphic encryption algorithms.

[0010] Furthermore, step S2 includes:

[0011] A dynamic key factor is generated based on a combination of the patient's unique identification code, current timestamp, and the patient's real-time critical vital signs.

[0012] When a patient's vital signs are detected to change from stable to critical, a forced key update is triggered, and a brand new key pair is immediately generated and distributed, ignoring the preset update cycle.

[0013] When the patient is in a stable state, a lazy update strategy based on a smooth window is adopted to update the key at preset time intervals;

[0014] The updated key is synchronized to the data acquisition device, virtual ward management terminal and expert online interaction platform through an encrypted channel. The old key is destroyed within a preset expiration time after the update is completed.

[0015] Furthermore, the generation of virtual wards based on the patient's basic information includes:

[0016] Extract unique identification data from the patient's basic information, including the patient's medical record number, ID card number, and biometric information;

[0017] A unique identification code is generated by performing a hash operation on the unique identification data, and the unique identification code is used as the virtual ward ID of the virtual ward;

[0018] Establish a mapping relationship between virtual ward IDs and patient basic information and data acquisition device numbers, and construct a single transmission channel for virtual wards and corresponding patient ward information data.

[0019] Furthermore, when a patient's basic information changes, the virtual ward data synchronization mechanism is triggered, automatically extracting the changed information and re-encrypting it, overwriting the original corresponding data in the virtual ward.

[0020] Simultaneously, a data change log is generated, recording the change time, change content, and operator identity information. The change log is stored in conjunction with the virtual ward ID.

[0021] 5. The virtual ward information data management method according to claim 1, wherein the hybrid encryption algorithm in step S2 specifically includes:

[0022] S21. The ward information data collected in step S1 is encrypted using the AES-128 symmetric encryption algorithm to generate ciphertext data.

[0023] S22. Use the RSA-2048 asymmetric encryption algorithm to encrypt the AES-128 symmetric encryption key to generate the key ciphertext;

[0024] S23. Combine the data ciphertext and the key ciphertext to form an encrypted data packet. When recording the encrypted data packet in the virtual ward, store the data ciphertext and the key ciphertext in different data partitions, and the key ciphertext partition can only be accessed by terminals with key management permissions.

[0025] Furthermore, step S3 includes:

[0026] S31. Construct a dynamic knowledge graph engine to associate medical entities such as diseases, symptoms, examinations, and drugs into a semantic network, and build a personal knowledge graph for each expert based on their published literature and historical medical records.

[0027] S32. Extract key medical entities from the encrypted data to be evaluated, generate a local knowledge subgraph, and calculate the semantic similarity between the local knowledge subgraph and the expert's personal knowledge graph as the matching degree.

[0028] S33. Select experts with a matching degree greater than or equal to a preset threshold to form a pool of candidate experts. Sort them in descending order of matching degree and allocate data to the expert ranked first. If the expert does not respond within the preset response time, the data will be automatically allocated to the next expert until the data allocation is successful.

[0029] Furthermore, step S3 also includes:

[0030] Key ward information data is simultaneously distributed to multiple candidate experts to obtain guidance from multiple experts.

[0031] Calculate the consensus coefficient of expert opinions. If the consensus coefficient is greater than or equal to the first threshold, then directly integrate them to form expert evaluation information.

[0032] If the consistency coefficient is less than the first threshold, the differences in different opinions will be automatically extracted, quantified as the focus of the dispute, and sent anonymously to all experts in the candidate expert pool for a new round of independent evaluation. The opinion revision trajectory of each expert will be recorded and their opinion migration vector will be calculated.

[0033] By integrating the final consensus reached after multiple rounds of negotiation with diverse problem-solving approaches to the points of contention, expert evaluation information including a disagreement report is generated.

[0034] Furthermore, step S4 includes:

[0035] S41. Detect the complete ward information data after encryption processing in step S2, and obtain the abnormality value A of the ward information data.

[0036] The urgency value B of the expert evaluation information is obtained by analyzing the expert evaluation information.

[0037] The uncertainty of continuous monitoring data is calculated to obtain the data stream information entropy C. The data stream information entropy C is used to quantify the degree to which the real-time fluctuations of multiple vital signs parameters of the patient deviate from the Gaussian distribution of their personal historical baseline.

[0038] S42. Calculate the comprehensive risk score based on the weighted fuzzy logic algorithm:

[0039] ;

[0040] in , , The weighting coefficients are dynamically configurable, and the processing signal level is set according to the comprehensive risk score.

[0041] S43. Perform medical procedures according to the level of the processed signal.

[0042] Furthermore, the dual-channel isolated storage architecture described in step S5 includes:

[0043] Patient identity information is encrypted and stored separately in the first storage area. It is encrypted using a ciphertext structure compatible with the fully homomorphic encryption algorithm. The encryption key is divided into multiple independent fragments and stored in physical encryption devices held by different authorized parties. The key fragments are combined only when the data is decrypted with authorization from multiple parties.

[0044] Medical data, expert evaluation information, and processed signal data are stored in the second storage area. The first storage area and the second storage area are physically isolated using different server clusters.

[0045] When building the data index, only the virtual ward ID is used as the index keyword, and the plaintext fields of patient identity are excluded from the index entries. The retrieval interface only accepts the virtual ward ID as the query condition.

[0046] The present invention also provides a virtual ward information data management system, the management system comprising:

[0047] Data acquisition module: Used to collect basic patient information, medical equipment operation data, and environmental monitoring data in the ward through data acquisition devices;

[0048] Virtual ward construction module: used to generate virtual wards based on patient basic information. It uses a hybrid encryption algorithm to encrypt the ward information data collected by the data acquisition module and records the encrypted ward information data in the virtual ward. The hybrid encryption algorithm combines symmetric encryption and asymmetric encryption and has a built-in dynamic key update mechanism. The dynamic key update mechanism generates dynamic key factors based on the patient's unique identity code, timestamp, and the patient's real-time critical vital signs value. When the patient's state changes from stable to critical, a forced key update is triggered.

[0049] Data evaluation module: used to upload encrypted partial key ward information data to the online expert interaction platform. The online expert interaction platform has a built-in dynamic knowledge graph engine, which performs expert matching by calculating the semantic similarity between the local knowledge subgraph of the data to be evaluated and the individual knowledge graph of the experts, and generates expert evaluation information including a divergence report by adopting a multi-expert cross-evaluation and adversarial consensus game strategy.

[0050] Analysis and processing module: This module is used to input the encrypted complete ward information data and the expert evaluation information output by the data evaluation module into the multidimensional entropy value perception quantitative decision-making model to generate a processing signal. The model comprehensively calculates the abnormality value of the ward information data, the urgency value of the evaluation information, and the entropy of the data flow. It outputs a comprehensive risk score through a weighted fuzzy logic algorithm and sets the processing signal level. The processing signal is then sent to the medical staff in the corresponding ward.

[0051] Data indexing module: used to store the encrypted ward information data, the expert evaluation information and the processed signal data in the database. It adopts a dual-channel isolated storage architecture to physically isolate the patient identity information from the medical data and constructs a data index with the virtual ward ID as the unique index keyword. The patient identity information is encrypted and stored using a ciphertext structure compatible with homomorphic encryption algorithms.

[0052] This invention provides a virtual ward information data management method and system. By constructing a virtual ward and combining encrypted data transmission with an online expert interaction platform, it is possible to promptly assess and treat patient information and improve the reliability and security of data transmission. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1This is a flowchart of the virtual ward information data management method in an embodiment of the present invention;

[0055] Figure 2 This is a flowchart illustrating the data encryption process for ward information data in an embodiment of the present invention;

[0056] Figure 3 This is a data processing flowchart of the expert online interaction platform in this embodiment of the invention;

[0057] Figure 4 This is a flowchart of the evaluation information processing in an embodiment of the present invention;

[0058] Figure 5 This is a schematic diagram of the virtual ward information data management system in an embodiment of the present invention. Detailed Implementation

[0059] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0060] Example 1:

[0061] Figure 1 A flowchart of a virtual ward information data management method according to an embodiment of the present invention is shown. The management method includes the following steps:

[0062] S1. Collect basic patient information, medical equipment operation data, and environmental monitoring data in the ward using data acquisition equipment;

[0063] S2. A virtual ward is generated based on the patient's basic information. The ward information data collected in step S1 is encrypted using a hybrid encryption algorithm. The encrypted ward information data is recorded in the virtual ward. The hybrid encryption algorithm combines symmetric and asymmetric encryption and updates the encryption key periodically or through event-triggered updates via a dynamic key update mechanism.

[0064] S3. Upload the encrypted key ward information data from step S2 to the online expert interaction platform. The platform will assign experts in the corresponding fields based on the knowledge graph matching mechanism, and use a multi-expert cross-evaluation and adversarial consensus game strategy to analyze and evaluate the data, generating expert evaluation information including a divergence report.

[0065] S4. Match the complete ward information data encrypted in step S2 with the expert evaluation information generated in step S3 to generate a processing signal, and send the processing signal to the medical staff in the corresponding ward.

[0066] S5. The encrypted ward information data, the expert evaluation information, and the processed signal data are stored in a database. A dual-channel isolated storage architecture is used to separate and store patient identity information from medical data. A data index is constructed using the virtual ward ID as the unique index keyword. The patient identity information is encrypted and stored using a ciphertext structure compatible with homomorphic encryption algorithms.

[0067] The intensive care unit is equipped with various data acquisition devices, including medical equipment such as vital sign monitors, infusion pumps, and ventilators, as well as environmental monitoring equipment such as temperature and humidity sensors and air quality detectors. These devices use Internet of Things (IoT) technology to collect real-time data on patients' vital signs, medication administration, respiratory parameters, and the ward environment.

[0068] The system automatically generates virtual wards based on patient admission information and assigns a unique identifier. A hybrid encryption algorithm, combining AES-128 and RSA-2048 asymmetric encryption, is used to encrypt the collected data. This hybrid encryption scheme incorporates a dynamic key update mechanism, binding the key's lifecycle to the patient's treatment status and data operation behavior, thus constructing an adaptive security model. The encrypted data is updated in real-time to the virtual ward, forming a digital file for the patient. This virtual ward is a digital file generated by combining the patient's identity information with the ward information they are admitted to, ensuring a one-to-one correspondence between the virtual ward and the patient and preventing errors in patient information.

[0069] When an abnormality in a patient's vital signs is detected, the system automatically uploads encrypted key data to an online expert interaction platform. Based on data characteristics, such as abnormal heart rate and decreased blood oxygen saturation, the platform matches the patient with experts in relevant fields such as cardiology and respiratory medicine. The platform's built-in dynamic knowledge graph engine associates multi-dimensional medical entities such as "disease-symptom-examination-medication" into a semantic network. By calculating the semantic similarity between the local knowledge subgraph of the data to be evaluated and the expert's personal knowledge graph, it accurately matches relevant experts. Experts then view the decrypted data through the remote consultation system and provide treatment suggestions.

[0070] Specifically, generating a virtual ward based on the patient's basic information includes:

[0071] Extract unique identification data from the patient's basic information, including the patient's medical record number, ID card number, and biometric information;

[0072] The unique identification data is verified by extracting at least one of the following: medical record number, ID card number, and biometric information. The hash operation uses the SHA-256 algorithm to concatenate multiple sets of identification data to generate a fixed-length hash value as a unique identity code. For example, the medical record number and ID card number are concatenated into a string and then hashed. The association mapping relationship is achieved by establishing a mapping table between the unique identifier of the virtual ward and the patient's basic information fields. The mapping table also records a list of authorized data collection device numbers, which are uniformly assigned by the hospital and pre-set in the device firmware.

[0073] A unique identification code is generated by hashing the unique identification data. This unique identification code is used as the virtual ward ID, and a one-to-one mapping is established between the virtual ward ID and the patient's basic information such as name, age, and gender. Simultaneously, the virtual ward ID is associated with the unique serial numbers of medical equipment such as vital sign monitors and infusion pumps within the ward.

[0074] This association mapping ensures that the virtual ward only loads the ward information data of the corresponding patient. Furthermore, the system sets up access control policies so that only the associated data acquisition devices can transmit data to the virtual ward.

[0075] Establish a mapping relationship between virtual ward IDs and patient basic information and data acquisition device numbers, and construct a single transmission channel for virtual wards and corresponding patient ward information data.

[0076] Specifically, when a patient is admitted, the system automatically extracts their medical record number, ID number, and biometric information. These three elements are concatenated and input into a hash function to generate a 64-bit hexadecimal string as a unique identification code. This code serves as the unique identifier for the virtual ward, ensuring complete isolation between virtual wards for different patients. Simultaneously, the system creates a mapping relationship between the unique virtual ward identifier and the patient's basic information in the database, such as associating the code with fields like patient name, age, and department. Before transmitting data, data acquisition devices must verify that their device number exists in the authorized device list for that virtual ward; if not authorized, transmission is refused. For example, a blood glucose monitor in a ward might have the device number GLU-2024-001. Blood glucose data can only be uploaded to the corresponding virtual ward if this device number is bound to the unique virtual ward identifier in the mapping table. By avoiding the direct storage of plaintext identification information through hash operations and restricting data sources through a device number whitelist mechanism, the system provides double protection for the accurate matching of patient data with virtual wards and the security of data transmission.

[0077] Specifically, when a patient's basic information changes, the virtual ward data synchronization mechanism is triggered, automatically extracting the changed information and re-encrypting it, overwriting the original corresponding data in the virtual ward. Changes in the patient's basic information may include updates to diagnosis results or adjustments to medication. The triggering mechanism monitors the modification status of the patient's electronic medical record in real time through a preset medical information system interface, and activates the synchronization process when a change in the value of the diagnosis result field or medication record field is detected.

[0078] Simultaneously, a data change log is generated, recording the change time, change content, and operator identity information. This change log is stored in conjunction with the virtual ward ID. Specifically, when a patient's medication regimen is adjusted, after the nurse's workstation updates the electronic medical record system, the system automatically identifies the modification markers in the medication record fields, triggering the virtual ward data synchronization mechanism. The changed medication information is encrypted using a hybrid AES-128 and RSA-2048 encoding to generate a new data packet, replacing the corresponding historical encrypted data in the virtual ward. During the synchronization process, a change log is generated containing the adjustment time, new medication dosage, operating nurse's ID, and digital signature. This log, after generating a unique identifier through hash calculation, is written to the distributed database along with the virtual ward ID.

[0079] Specifically, this invention constructs a zero-trust dynamic key co-existence model, and the dynamic key update mechanism in step S2 includes:

[0080] A dynamic key factor is generated based on a combination of the patient's unique identification code, current timestamp, and the patient's real-time critical vital signs. The dynamic key factor is generated by concatenating these three elements, ensuring the uniqueness of each generated key factor.

[0081] When a patient's vital signs are detected to transition from stable to critical, a forced key update is triggered, immediately generating and distributing a new key pair regardless of the preset update cycle. For example, when the vital signs monitor detects that the patient's heart rate variability or the rate of sudden drop in blood oxygen exceeds a preset threshold, the system automatically triggers a forced key update to address the potential exposure risks associated with high-frequency data exchange in critical situations.

[0082] When the patient is in a stable state, a lazy update strategy based on a smooth window is adopted to update the key at preset time intervals (such as 15 minutes) to reduce the computing power overhead of encryption and decryption. This "static-dynamic" elastic key update mechanism allows the key's lifecycle to resonate completely with the urgency of the medical event, achieving a dynamic optimal balance between security strength and system efficiency.

[0083] The updated key is synchronized to the data acquisition device, virtual ward management terminal, and online expert interaction platform via an encrypted channel. The old key is destroyed within a preset expiration time (e.g., 10 minutes) after the update. This mechanism ensures that even if the key is cracked during a certain period, the impact is limited to the data within that period, and the old key cannot be used to decrypt subsequent data, significantly improving data security.

[0084] Specifically, Figure 2 The flowchart of the data encryption process for ward information data in an embodiment of the present invention is shown. Step S2 further includes:

[0085] S21. The ward information data collected in step S1 is encrypted using the AES-128 symmetric encryption algorithm to generate ciphertext data.

[0086] S22. Use the RSA-2048 asymmetric encryption algorithm to encrypt the AES-128 symmetric encryption key to generate the key ciphertext;

[0087] S23. Combine the data ciphertext and the key ciphertext to form an encrypted data packet. When recording the encrypted data packet in the virtual ward, store the data ciphertext and the key ciphertext in different data partitions, and the key ciphertext partition can only be accessed by terminals with key management permissions.

[0088] The AES-128 symmetric encryption algorithm is used for rapid encryption of ward information data, while the RSA-2048 asymmetric encryption algorithm is used to protect the secure transmission of the symmetric encryption key. After the data ciphertext and key ciphertext are combined into an encrypted data packet, physical isolation between the key and data is achieved through partitioned storage. Access to the key ciphertext partition is restricted to the key management terminal, further reducing the risk of key leakage. In the data encryption process, the ward information data is first encrypted using the AES-128 algorithm to generate data ciphertext. This algorithm is suitable for real-time processing of large amounts of medical data, and its encryption efficiency meets the high-frequency acquisition requirements of ward information. Subsequently, the AES key is encrypted using the RSA-2048 algorithm. The public-key mechanism of asymmetric encryption is used to achieve secure key transmission, preventing the key from being intercepted and cracked during transmission. After the encrypted data packet is formed, the data ciphertext and key ciphertext are stored separately in different data partitions of the virtual ward. The key ciphertext partition has independent access control, allowing access only to the key management terminal after authentication, preventing unauthorized key acquisition operations. This hybrid encryption method ensures both the efficiency of medical data encryption processing and enhances the overall security of the encryption system through a hierarchical key management mechanism.

[0089] Specifically, the encrypted ward information data also includes a dynamic key update step during transmission:

[0090] A dynamic key factor is generated based on the combination of the patient's unique identification code and the current timestamp, and the AES-128 symmetric encryption key is updated according to the dynamic key factor at preset time intervals.

[0091] The updated key is synchronized to the data acquisition device, virtual ward management terminal, and expert online interaction platform via an encrypted channel. When recording encrypted data packets in the virtual ward, the encrypted data and key are stored separately in different data partitions. The key partition is accessible only to terminals with key management permissions. For example, the encrypted data is stored in the "Patient Data" partition of the virtual ward, while the key is stored in the "System Management" partition. Furthermore, an access control list can be set up to stipulate that only authorized system administrator accounts can access the key in the "System Management" partition.

[0092] Therefore, even if an attacker obtains the encrypted data, they cannot recover the original data without the corresponding decryption key, thus ensuring the security of ward information data. Furthermore, storing the encrypted key separately and restricting access further enhances the system's security.

[0093] Specifically, this invention also proposes a dynamic key update step: a dynamic key factor is generated based on the patient's unique identification code and the current timestamp. Every 15 minutes, the AES-128 symmetric encryption key is updated according to the dynamic key factor. The updated key is synchronized to the data acquisition device, the virtual ward management terminal, and the online expert interaction platform via an encrypted channel. The old key becomes invalid and is destroyed within 10 minutes of the update. This mechanism ensures that even if the key is cracked during a certain period, the impact is limited to data within 15 minutes, and the old key cannot be used to decrypt subsequent data, significantly improving data security.

[0094] The dynamic key factor is generated by concatenating the patient's unique identification code with a timestamp accurate to the second, ensuring that each generated key factor is unique; the key update cycle is set to 15 minutes, which avoids the system burden caused by frequent updates and effectively shortens the validity period of the key; the key synchronization adopts encrypted channel transmission to prevent the key from being intercepted during transmission; the old key is destroyed immediately after it expires, eliminating the risk of historical keys being reused.

[0095] Specifically, Figure 3 The following is a flowchart of the data processing of the online expert interaction platform in an embodiment of the present invention. Step S3, which involves assigning corresponding domain experts based on a knowledge graph matching mechanism, includes:

[0096] S31. Construct a dynamic knowledge graph engine to associate medical entities such as diseases, symptoms, examinations, and drugs into a semantic network, and build a personal knowledge graph for each expert based on their published literature and historical medical records.

[0097] S32. Extract key medical entities from the encrypted partial key ward information data, generate a local knowledge subgraph, and calculate the semantic similarity between the local knowledge subgraph and the expert's personal knowledge graph as the matching degree.

[0098] The matching degree calculation formula is as follows:

[0099]

[0100] Where K is the matching degree, This is a local knowledge subgraph of the data to be evaluated. For individual expert knowledge graphs, The subgraph semantic similarity is calculated based on the graph neural network, and μ is the expert domain weight. The weight value μ can be set according to the expert's professional title and years of clinical experience, and the value of μ ranges from 0.6 to 1.0.

[0101] S33. Select experts with a matching degree ≥ 0.8 to form a candidate expert pool, sort them in descending order of matching degree, and prioritize the allocation of data to the expert ranked first. If the expert does not respond within the preset response time (e.g., 30 minutes), the data will be automatically allocated to the next expert until the data allocation is successful.

[0102] By accurately matching and efficiently allocating expert resources, the relevance and timeliness of expert guidance were improved, thereby enhancing the quality and efficiency of virtual ward information data management. Simultaneously, by setting matching thresholds and implementing automatic allocation mechanisms, it was ensured that critical ward information data received timely professional evaluation and guidance, effectively reducing medical risks.

[0103] Specifically, step S3 further includes:

[0104] Based on the encrypted partial key ward information data, a multi-expert cross-evaluation and adversarial consensus game strategy is adopted. The key ward information data is simultaneously distributed to multiple candidate experts to obtain more than one expert guidance opinion. The consistency coefficient of expert opinions is calculated. If the consistency coefficient is ≥85%, it is directly integrated to form expert evaluation information. If the consistency coefficient is <85%, the differences in opinions are fed back to each expert, and experts are organized to hold online consultations until a consensus is reached, and then the expert evaluation information is output.

[0105] A multi-expert cross-evaluation mechanism is adopted. When key ward information data involves difficult and complex diseases, it is simultaneously assigned to 2-3 experts in the same field to collect guidance from multiple experts. The consistency coefficient of expert opinions is calculated as the number of overlapping clauses in expert opinions divided by the total number of clauses in expert opinions multiplied by 100%. If the consistency coefficient is ≥85%, it is directly integrated to form the final expert guidance. If the consistency coefficient is <85%, the differences in opinions are fed back to each expert, and online expert consultation is organized until a consensus is reached. The differences in different opinions are automatically extracted, quantified as the focus of controversy, and sent anonymously to all experts in the candidate expert pool for a new round of independent evaluation. The opinion revision trajectory of each expert is recorded and their viewpoint migration vector is calculated. The final consensus formed after multiple rounds of game theory and the diversified problem-solving approaches for the focus of controversy are integrated to generate expert evaluation information including a divergence report.

[0106] Specifically, when the system identifies key ward information data containing characteristics of complex cases, it automatically triggers a multi-expert allocation mechanism. For example, for a complex case of acute myocardial infarction complicated by respiratory failure, the system simultaneously sends the data to cardiology expert A, respiratory expert B, and critical care medicine expert C. The three experts submit guidance opinions containing 8, 7, and 9 treatment suggestions, respectively. The system extracts and compares the clauses of each opinion, finding that 6 suggestions are completely overlapping, and calculates the consistency coefficient as (6 / 8) × 100% = 75%. Instead of simply organizing a consultation, the system automatically extracts the discrepancies, quantifies them as pending issues, and anonymously sends them to all experts in the candidate expert pool as "points of contention" for a new round of independent evaluation. The system records the opinion revision trajectory of each expert and calculates their "viewpoint migration vector." After discussion, the experts unanimously add two common suggestion clauses, ultimately forming guidance opinions containing 8 consistent opinions. At the same time, the system generates a divergence report, recording the diversified problem-solving approaches for each point of contention. The system generates a confidence score of 98 based on an expert matching degree of 0.92, a consistency coefficient of 100%, and a past adoption rate of 95%, marking it as a high-confidence opinion and triggering a first-level medical and nursing response priority.

[0107] Specifically, Figure 4 The flowchart of the evaluation information processing in an embodiment of the present invention is shown. The multidimensional entropy value perception quantification decision model in step S4 includes:

[0108] S41. Detect the complete ward information data after encryption processing in step S2, and obtain the abnormality value A of the ward information data.

[0109] The urgency value B of the expert evaluation information is obtained by analyzing the expert evaluation information.

[0110] The uncertainty of continuously monitored data is calculated to obtain the data stream information entropy C. The data stream information entropy C is used to quantify the degree to which the real-time fluctuations of multiple vital signs parameters of a patient deviate from the Gaussian distribution of their personal historical baseline. For example, when the real-time fluctuations of multiple vital signs parameters of a patient deviate from the Gaussian distribution of their personal historical baseline, even if the indicators do not reach the absolute danger threshold, the surge in information entropy C can still capture the early decompensation trend.

[0111] S42. Calculate the comprehensive risk score based on the weighted fuzzy logic algorithm:

[0112] ;

[0113] in , , The weighting coefficients are dynamically configurable, and the processing signal level is set according to the comprehensive risk score.

[0114] S43. Perform medical procedures according to the level of the processed signal.

[0115] The signal processing levels are divided into Level 1, Level 2, and Level 3 signals, corresponding to critical vital signs with emergency resuscitation recommendations, abnormal vital signs with routine adjustment recommendations, and no abnormal data with routine nursing recommendations, respectively. The delivery method is configured according to the level: Level 1 signals use a combination of terminal pop-ups, voice prompts, and SMS notifications; Level 2 signals use both terminal pop-ups and SMS notifications; and Level 3 signals only trigger terminal pop-ups. A virtual ward ID link is embedded in the processed signal, allowing medical staff to directly access the complete encrypted data of the corresponding virtual ward by clicking the link.

[0116] By introducing data flow information entropy as a third-dimensional adjustment factor, this scheme breaks through the limitations of traditional single-threshold judgment. It can generate early warning-level processing signals in advance before the patient's vital signs show an early decompensation trend but have not yet reached a critical value, thus significantly advancing the window for medical intervention.

[0117] Specifically, when the system detects a patient's heart rate exceeding 120 beats per minute and the expert recommendation includes "initiate resuscitation procedures," a Level 1 signal is automatically generated. At this time, a warning pop-up appears on the medical staff's terminal, the device issues a voice alarm, and an SMS message containing a virtual ward ID link is sent to the on-duty doctor's mobile phone. Medical staff can click the link to immediately view encrypted, stored real-time ECG monitoring data and details of the expert recommendation. If the patient's blood oxygen saturation is below 90% but there is no emergency resuscitation recommendation, a Level 2 signal is generated, with the terminal pop-up and SMS notification sent in parallel to avoid excessive interference with medical staff's work in non-emergency situations. For routine nursing recommendations without abnormal data, only a pop-up notification is sent, reducing unnecessary communication channel usage. Through this tiered sending mechanism, high-priority signals are ensured to reach multiple channels, while low-priority signals only retain necessary prompts, achieving a balance between information transmission efficiency and resource consumption.

[0118] When the system detects that the patient's electrocardiogram shows a heart rate of 135 beats / min and an oxygen saturation of 88%, the abnormal data extraction module automatically identifies this abnormal vital sign data and retrieves the emergency treatment recommendation marked "immediately initiate oxygen therapy intervention" from the expert guidance. Simultaneously, the data stream information entropy C module detects a significant increase in the deviation of multiple vital sign parameters from their historical baseline, resulting in a surge in information entropy values ​​and a further increase in the comprehensive risk score. The processing signal generation module classifies this type of data as a Level 1 signal according to preset rules, triggering a Level 3 alarm mechanism: A red warning pop-up window appears on the medical staff's terminal device, playing a "Level 1 Alert" voice prompt, while an encrypted SMS containing the virtual ward number is sent to the responsible physician's mobile terminal; the pop-up window interface includes a virtual ward access portal, which, when clicked, redirects to the encrypted data interface to view the patient's real-time ventilator parameters and the full text of the expert recommendations.

[0119] Through the above technical solution, this application achieves intelligent hierarchical and precise delivery of medical processing signals, effectively solving the problem of delayed emergency response in traditional ward information management. By quantifying and matching the degree of abnormal vital signs and data information entropy, a multi-level alarm triggering mechanism is established to ensure that critical situations receive priority response from medical resources. The embedded encrypted virtual ward link enables medical staff to quickly obtain complete clinical data, avoiding the operational delays required by traditional systems that require multiple interface switching to query information, and significantly improving the timeliness of emergency patient rescue.

[0120] Specifically, step S5 includes:

[0121] A dual-channel isolated storage architecture is adopted, with patient identity information stored separately in encrypted form using a ciphertext structure compatible with fully homomorphic encryption algorithms. Sensitive identity data such as patient ID numbers and biometric information are stored in the first storage area, encrypted using the national cryptographic algorithm SM4. The encryption key is divided into multiple independent fragments, stored separately on physical encryption devices held by different authorized parties, such as the head of the hospital's information department, the system security administrator, and the data center manager. The key fragments are only recombined during data decryption with authorization from all parties. The first storage area and the second storage area, which stores medical data, expert evaluation information, and processed signal data, are physically isolated using different server clusters and access is only permitted through a dedicated line within the hospital's intranet.

[0122] When constructing the data index, only the virtual ward ID is used as the index keyword. Index entries strictly exclude plaintext fields such as patient names and ID numbers, and do not contain any sensitive patient information to prevent the index data from leaking patient identities. When performing data retrieval operations, the retrieval interface only accepts the virtual ward ID as the query condition and returns the associated encrypted data storage path after access permission verification.

[0123] The system's storage architecture features a forward-looking "homomorphic encryption-ready" characteristic. The encrypted structure of patient identity information is designed to be compatible with fully homomorphic encryption algorithms. This means that when the big data analytics engine needs to perform de-identified trend analysis on massive amounts of virtual ward data, statistical calculations can be completed directly on the encrypted domain without decryption or exposing any original patient information or medical data. This architecture provides a solid data security foundation for the training of future medical AI models, ensuring data sovereignty and privacy while unlocking the potential value of medical big data.

[0124] This invention provides a method for managing information data in a virtual ward. By constructing a virtual ward and combining asymmetric data encryption transmission and an online expert interaction platform, a zero-trust dynamic key co-existence encryption mechanism, a knowledge graph-guided adversarial consensus game evaluation mechanism, and a multi-dimensional entropy value perception quantitative decision-making model, it is possible to evaluate and treat patient information in a timely manner and improve the reliability and security of data information transmission.

[0125] Example 2:

[0126] Figure 5 A schematic diagram of a virtual ward information data management system according to an embodiment of the present invention is shown. The management system includes:

[0127] Data acquisition module 10: Used to collect basic patient information, medical equipment operation data and environmental monitoring data in the ward through data acquisition equipment;

[0128] Virtual ward construction module 20: used to generate virtual wards based on patient basic information, and to encrypt the ward information data collected by data acquisition module 10 using a hybrid encryption algorithm (asymmetric encryption algorithm). The encrypted ward information data is recorded in the virtual ward. The hybrid encryption algorithm combines symmetric and asymmetric encryption and has a built-in dynamic key update mechanism. The dynamic key update mechanism generates dynamic key factors based on the patient's unique identity code, timestamp, and the patient's real-time critical vital signs value. When the patient's state changes from stable to critical, a forced key update is triggered.

[0129] Data evaluation module 30: used to upload encrypted partial key ward information data to the online expert interaction platform, where the platform assigns corresponding domain experts. The online expert interaction platform has a built-in dynamic knowledge graph engine, which performs expert matching by calculating the semantic similarity between the local knowledge subgraph of the data to be evaluated and the individual knowledge graph of the experts, and generates expert evaluation information, including expert evaluation information with a divergence report, by adopting a multi-expert cross-evaluation and adversarial consensus game strategy.

[0130] Analysis and processing module 40: This module is used to match the encrypted complete ward information data and the expert evaluation information output by data evaluation module 30 with the input of a multidimensional entropy value perception quantification decision model to generate a processing signal. The model comprehensively calculates the abnormality value of the ward information data, the urgency value of the evaluation information, and the entropy of the data flow. It outputs a comprehensive risk score through a weighted fuzzy logic algorithm and sets the processing signal level. The processing signal is then sent to the medical staff in the corresponding ward.

[0131] Data indexing module 50: used to store the encrypted ward information data, the expert evaluation information and the processed signal data in the database, and to use a dual-channel isolated storage architecture to physically isolate the patient identity information from the medical data, and to build a data index using the virtual ward ID as the unique index keyword for fast data retrieval; the patient identity information is encrypted and stored using a ciphertext structure compatible with homomorphic encryption algorithms.

[0132] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, which may include: read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, etc.

[0133] Furthermore, the embodiments of the present invention have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for managing virtual ward information data, characterized in that, The management method includes the following steps: S1. Collect basic patient information, medical equipment operation data, and environmental monitoring data in the ward using data acquisition equipment; S2. A virtual ward is generated based on the patient's basic information. The ward information data collected in step S1 is encrypted using a hybrid encryption algorithm. The encrypted ward information data is recorded in the virtual ward. The hybrid encryption algorithm combines symmetric and asymmetric encryption and updates the encryption key periodically or through event-triggered updates via a dynamic key update mechanism. S3. Upload the encrypted key ward information data from step S2 to the online expert interaction platform. The platform will assign experts in the corresponding fields based on the knowledge graph matching mechanism, and use a multi-expert cross-evaluation and adversarial consensus game strategy to analyze and evaluate the data, generating expert evaluation information including a divergence report. S4. Match the complete ward information data encrypted in step S2 with the expert evaluation information generated in step S3 to generate a processing signal, and send the processing signal to the medical staff in the corresponding ward. S5. The encrypted ward information data, the expert evaluation information, and the processed signal data are stored in a database. A dual-channel isolated storage architecture is used to separate and store patient identity information from medical data. A data index is constructed using the virtual ward ID as the unique index keyword. The patient identity information is encrypted and stored using a ciphertext structure compatible with homomorphic encryption algorithms.

2. The virtual ward information data management method according to claim 1, characterized in that, Step S2 includes: A dynamic key factor is generated based on a combination of the patient's unique identification code, current timestamp, and the patient's real-time critical vital signs. When a patient's vital signs are detected to change from stable to critical, a forced key update is triggered, and a brand new key pair is immediately generated and distributed, ignoring the preset update cycle. When the patient is in a stable state, a lazy update strategy based on a smooth window is adopted to update the key at preset time intervals; The updated key is synchronized to the data acquisition device, virtual ward management terminal and expert online interaction platform through an encrypted channel. The old key is destroyed within a preset expiration time after the update is completed.

3. The virtual ward information data management method according to claim 1, characterized in that, The process of generating a virtual ward based on the patient's basic information includes: Extract unique identification data from the patient's basic information, including the patient's medical record number, ID card number, and biometric information; A unique identification code is generated by performing a hash operation on the unique identification data, and the unique identification code is used as the virtual ward ID of the virtual ward; Establish a mapping relationship between virtual ward IDs and patient basic information and data acquisition device numbers, and construct a single transmission channel for virtual wards and corresponding patient ward information data.

4. The virtual ward information data management method according to claim 3, characterized in that, When a patient’s basic information changes, the virtual ward data synchronization mechanism is triggered, automatically extracting the changed information and re-encrypting it, overwriting the original corresponding data in the virtual ward. Simultaneously, a data change log is generated, recording the change time, change content, and operator identity information. The change log is stored in conjunction with the virtual ward ID.

5. The virtual ward information data management method according to claim 1, characterized in that, The hybrid encryption algorithm mentioned in step S2 specifically includes: S21. The ward information data collected in step S1 is encrypted using the AES-128 symmetric encryption algorithm to generate encrypted data. S22. Use the RSA-2048 asymmetric encryption algorithm to encrypt the AES-128 symmetric encryption key to generate the key ciphertext; S23. Combine the data ciphertext and the key ciphertext to form an encrypted data packet. When recording the encrypted data packet in the virtual ward, store the data ciphertext and the key ciphertext in different data partitions, and the key ciphertext partition can only be accessed by terminals with key management permissions.

6. The virtual ward information data management method according to claim 1, characterized in that, Step S3 includes: S31. Construct a dynamic knowledge graph engine to associate medical entities such as diseases, symptoms, examinations, and drugs into a semantic network, and build a personal knowledge graph for each expert based on their published literature and historical medical records. S32. Extract key medical entities from the encrypted data to be evaluated, generate a local knowledge subgraph, and calculate the semantic similarity between the local knowledge subgraph and the expert's personal knowledge graph as the matching degree. S33. Select experts with a matching degree greater than or equal to a preset threshold to form a pool of candidate experts. Sort them in descending order of matching degree and allocate data to the expert ranked first. If the expert does not respond within the preset response time, the data will be automatically allocated to the next expert until the data allocation is successful.

7. The virtual ward information data management method according to claim 6, characterized in that, Step S3 also includes: Key ward information data is simultaneously distributed to multiple candidate experts to obtain guidance from multiple experts; Calculate the consensus coefficient of expert opinions. If the consensus coefficient is greater than or equal to the first threshold, then directly integrate the opinions to form expert evaluation information. If the consistency coefficient is less than the first threshold, the differences in different opinions will be automatically extracted, quantified as the focus of the dispute, and sent anonymously to all experts in the candidate expert pool for a new round of independent evaluation. The opinion revision trajectory of each expert will be recorded and their opinion migration vector will be calculated. By integrating the final consensus reached after multiple rounds of negotiation with diverse problem-solving approaches to the points of contention, expert evaluation information including a disagreement report is generated.

8. The virtual ward information data management method according to claim 1, characterized in that, Step S4 includes: S41. Detect the complete ward information data after encryption processing in step S2, and obtain the abnormality value A of the ward information data. The urgency value B of the expert evaluation information is obtained by analyzing the expert evaluation information. The uncertainty of continuous monitoring data is calculated to obtain the data stream information entropy C. The data stream information entropy C is used to quantify the degree to which the real-time fluctuations of multiple vital signs parameters of the patient deviate from the Gaussian distribution of their personal historical baseline. S42. Calculate the comprehensive risk score based on the weighted fuzzy logic algorithm: ； in , , The weighting coefficients are dynamically configurable, and the processing signal level is set according to the comprehensive risk score. S43. Perform medical procedures according to the level of the processed signal.

9. The virtual ward information data management method according to claim 1, characterized in that, The dual-channel isolated storage architecture described in step S5 includes: Patient identity information is encrypted and stored separately in the first storage area. It is encrypted using a ciphertext structure compatible with the fully homomorphic encryption algorithm. The encryption key is divided into multiple independent fragments and stored in physical encryption devices held by different authorized parties. The key fragments are combined only when the data is decrypted with authorization from multiple parties. Medical data, expert evaluation information, and processed signal data are stored in the second storage area. The first storage area and the second storage area are physically isolated using different server clusters. When building the data index, only the virtual ward ID is used as the index keyword, and the plaintext fields of patient identity are excluded from the index entries. The retrieval interface only accepts the virtual ward ID as the query condition.

10. A virtual ward information data management system, characterized in that, The management system includes: Data acquisition module: Used to collect basic patient information, medical equipment operation data, and environmental monitoring data in the ward through data acquisition devices; Virtual ward construction module: used to generate virtual wards based on patient basic information. It uses a hybrid encryption algorithm to encrypt the ward information data collected by the data acquisition module and records the encrypted ward information data in the virtual ward. The hybrid encryption algorithm combines symmetric encryption and asymmetric encryption and has a built-in dynamic key update mechanism. The dynamic key update mechanism generates dynamic key factors based on the patient's unique identity code, timestamp, and the patient's real-time critical vital signs value. When the patient's state changes from stable to critical, a forced key update is triggered. Data evaluation module: used to upload encrypted partial key ward information data to the online expert interaction platform. The online expert interaction platform has a built-in dynamic knowledge graph engine, which performs expert matching by calculating the semantic similarity between the local knowledge subgraph of the data to be evaluated and the individual knowledge graph of the experts, and generates expert evaluation information including a divergence report by adopting a multi-expert cross-evaluation and adversarial consensus game strategy. Analysis and processing module: This module is used to input the encrypted complete ward information data and the expert evaluation information output by the data evaluation module into the multidimensional entropy value perception quantitative decision-making model to generate a processing signal. The model comprehensively calculates the abnormality value of the ward information data, the urgency value of the evaluation information, and the entropy of the data flow. It outputs a comprehensive risk score through a weighted fuzzy logic algorithm and sets the processing signal level. The processing signal is then sent to the medical staff in the corresponding ward. Data indexing module: used to store the encrypted ward information data, the expert evaluation information and the processed signal data in the database. It adopts a dual-channel isolated storage architecture to physically isolate the patient identity information from the medical data and constructs a data index with the virtual ward ID as the unique index keyword. The patient identity information is encrypted and stored using a ciphertext structure compatible with homomorphic encryption algorithms.

Images