Security encryption method and system in medical data collection

Abstract

The present application relates to a kind of security encryption method and system in medical data acquisition, belong to medical data security technical field, solve the problem that medical data acquisition in prior art is difficult to meet the security needs of high real-time, strong privacy under medical scene.Method includes the medical record generated by acquisition medical equipment, and each medical record is executed: based on current medical record, extract semantic information, generate four tuples, the four tuples include four semantic factors;Based on the four tuples, a key derivation method is used to dynamically derive the record-level encryption key of current medical record;Based on record-level encryption key, current medical record is encrypted in real time streaming block, generates encrypted medical data, and the encrypted medical data and corresponding four tuples are transmitted and stored;If medical record needs to be revoked, the logical access of medical record is revoked based on the lightweight revocation mechanism constructed.The efficient encryption transmission in medical data acquisition is realized.

Description

Technical Field

[0001] This invention relates to the field of medical data security technology, and in particular to a secure encryption method and system for medical data acquisition. Background Technology

[0002] Healthcare informatization has widely covered hospital information systems, electronic medical record systems, and inter-institutional data sharing platforms, containing a large amount of highly sensitive patient identity information, physiological parameter data, and medical records. After collection, this data often needs to be transmitted and stored within hospital or inter-institutional networks, and its security and privacy protection have become key constraints on the development of healthcare informatization.

[0003] The existing medical data collection process suffers from several vulnerabilities: First, the collection phase is highly vulnerable, as medical data is often transmitted in plaintext or with simple encryption after collection, making it susceptible to man-in-the-middle attacks and theft. Second, traditional encryption schemes are disconnected from medical operations; key generation does not incorporate medical semantics, resulting in isolated encryption of data for the same patient in different contexts, making it difficult to achieve dynamic access control based on business logic. Third, there is a lack of efficient and compliant data revocation mechanisms. Once a patient exercises their "right to be forgotten," existing systems often require full decryption and reencryption, leading to low processing efficiency and failing to meet legal requirements for timely response. These shortcomings collectively hinder the efficient sharing and utilization of medical data while ensuring security. Summary of the Invention

[0004] Based on the above analysis, the embodiments of the present invention aim to provide a secure encryption method for medical data acquisition, in order to solve the problem that existing medical data acquisition methods cannot meet the security requirements of high real-time performance and strong privacy in medical scenarios.

[0005] On one hand, embodiments of the present invention provide a secure encryption method for medical data acquisition, comprising the following steps:

[0006] Collect medical records generated by medical devices and perform the following actions on each medical record:

[0007] Based on current medical records, semantic information is extracted to generate a quadruple, which includes four semantic factors: identity factor, physiological factor, time factor, and device factor.

[0008] Based on the aforementioned quadruple, a key derivation method is used to dynamically derive the record-level encryption key for the current medical record.

[0009] The current medical record is encrypted in real time using a record-level encryption key to generate encrypted medical data, and the encrypted medical data and the corresponding quadruple are transmitted and stored.

[0010] If a medical record needs to be revoked, the logical access to the medical record will be revoked based on the constructed revocation mechanism.

[0011] Furthermore,

[0012] The identity factor is generated based on the unique identifier of the patient corresponding to the medical record;

[0013] The physiological factors are generated based on the types and values ​​of physiological parameters in medical records;

[0014] The time factor is generated based on the type of medical service and the collection time in the medical records;

[0015] The device factor is generated based on the security level of the medical record acquisition device.

[0016] Furthermore, the identity factor is obtained by hashing the unique identifier of the patient corresponding to the medical record;

[0017] The physiological factors are obtained by jointly encoding the types and values ​​of physiological parameters in medical records; wherein, the values ​​of physiological parameters are mapped to predefined interval codes after segmented discretization.

[0018] The time factor is obtained by rounding down the collection time of the medical record based on the time window determined by the type of medical service in the medical record, and then converting it into an encoding.

[0019] The device factor is obtained by hashing the identity credentials of the medical record collection device.

[0020] Furthermore, the physiological factors are obtained by jointly encoding the types and values ​​of physiological parameters in medical records, including:

[0021] Extract the type names and values ​​of physiological parameters from medical records;

[0022] The extracted physiological parameter type names are mapped to a standard medical terminology database to generate type codes;

[0023] Based on the extracted physiological parameter type, a preset numerical division interval is determined, and based on the extracted physiological parameter value, the corresponding numerical division interval is determined, and then the corresponding numerical division interval is mapped to a predefined interval code.

[0024] Physiological factors are obtained by combining the type encoding and interval encoding.

[0025] Furthermore, the record-level encryption key for the current medical record is dynamically derived in the following manner:

[0026] The quadruples are combined to generate an enhanced seed;

[0027] A hash-based key derivation function is used, with the system master key as the entropy source, the dynamic salt value as the salt, and the enhanced seed as the information parameter, to derive a record-level encryption key.

[0028] Furthermore, the quadruplets are combined to generate a reinforced seed in the following manner:

[0029] Each semantic factor in the quadruple is encoded using a structured format that includes type, length, value, and checksum, generating a structured encoded segment for each semantic factor.

[0030] The structured encoded segments of each semantic factor are connected in a predetermined order to obtain intermediate data;

[0031] Head and tail identifiers are added before and after the intermediate data to generate an enhanced seed.

[0032] Furthermore, the dynamic salinity is obtained through the following method:

[0033] Obtain the hash value of the latest block on the public blockchain network and generate a local true random number;

[0034] The quadruple, the obtained hash value, and the local true random number are combined according to a predetermined format to generate input data;

[0035] Using the system master key as the key, the input data is hashed to calculate the message authentication code, generating a dynamic salt value.

[0036] Furthermore, the revocation mechanism includes:

[0037] When a medical record needs to be revoked, the revocation message containing the record-level encryption key is encrypted using a post-quantum cryptography algorithm to generate a revocation token. Subsequently, if access to the medical record is required, the revocation token must be obtained to realize the logical revocation of access to the medical record.

[0038] Furthermore, the secure encryption method further includes:

[0039] During the streaming block encryption process, the combined message of the quadruple and the medical record data block count is used to calculate the message authentication code using an audit key that is separate from the system master key, and an audit tag is generated.

[0040] On the other hand, embodiments of the present invention provide a secure encryption system for medical data acquisition, comprising:

[0041] The data acquisition module is used to collect medical records generated by medical devices in real time.

[0042] The semantic parsing module is used to extract semantic information from each medical record and generate a four-tuple, which includes four semantic factors: identity factor, physiological factor, time factor, and device factor.

[0043] The key derivation module is used to dynamically derive the record-level encryption key of the current medical record based on the quadruple and using the key derivation method.

[0044] The real-time encryption and transmission module is used to perform real-time streaming block encryption on the current medical record based on the record-level encryption key, generate encrypted medical data, and transmit and store the encrypted medical data and the corresponding quadruple;

[0045] The revocation management module is used to perform logical access to and revocation of medical records based on the constructed revocation mechanism when medical records need to be revoked.

[0046] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects:

[0047] This invention provides a secure encryption method and system for medical data acquisition. By dynamically generating a unique key for each medical record, record-level encryption is achieved. The encryption key is directly associated with the clinical semantics of the data, allowing security policies to be naturally integrated with medical business logic and supporting context-based intelligent access control. Streaming block encryption is employed, completing encryption simultaneously with data generation, meeting the high real-time and low-latency requirements of medical scenarios, while parallel encryption and transmission improve efficiency. A cryptographic token mechanism enables logical access revocation, eliminating the need for physical deletion or reencryption of massive amounts of data, quickly responding to compliance requirements such as patient privacy withdrawal, and significantly reducing operational costs. Encrypted data is stored in association with a four-tuple, ensuring security while preserving key semantic information, providing a structured foundation for subsequent privacy computation, cross-institutional collaboration, or data analysis. Furthermore, key derivation, real-time encryption, secure transmission, and dynamic revocation are integrated into a closed-loop process, forming a fully controllable security system from data generation to destruction, significantly improving the security, flexibility, and compliance of medical data in acquisition, transmission, storage, and lifecycle management.

[0048] In this invention, the above-described technical solutions can be combined with each other to achieve more preferred combinations. Other features and advantages of this invention will be set forth in the following description, and some advantages may become apparent from the description or be learned by practicing the invention. The objects and other advantages of this invention can be realized and obtained from what is particularly pointed out in the description and drawings. Attached Figure Description

[0049] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts.

[0050] Figure 1 This is a flowchart illustrating the secure encryption method for medical data acquisition provided in Embodiment 1 of the present invention. Detailed Implementation

[0051] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0052] Example 1

[0053] A specific embodiment of the present invention discloses a secure encryption method for medical data acquisition, such as... Figure 1 As shown, it includes the following steps:

[0054] Collect medical records generated by medical devices and perform the following actions on each medical record:

[0055] Based on current medical records, semantic information is extracted to generate a quadruple, which includes four semantic factors: identity factor, physiological factor, time factor, and device factor.

[0056] Based on the aforementioned quadruple, a key derivation method is used to dynamically derive the record-level encryption key for the current medical record.

[0057] The current medical record is encrypted in real time using a record-level encryption key to generate encrypted medical data, and the encrypted medical data and the corresponding quadruple are transmitted and stored.

[0058] If a medical record needs to be revoked, the logical access to the medical record will be revoked based on the constructed revocation mechanism.

[0059] Specifically, medical equipment includes, but is not limited to, electrocardiogram monitors, blood pressure monitors, thermometers, and ventilators.

[0060] During implementation,

[0061] The identity factor is generated based on the unique identifier of the patient corresponding to the medical record;

[0062] The physiological factors are generated based on the types and values ​​of physiological parameters in medical records;

[0063] The time factor is generated based on the type of medical service and the collection time in the medical records;

[0064] The device factor is generated based on the security level of the medical record acquisition device.

[0065] In practice, the identity factor is obtained by hashing the unique identifier of the patient corresponding to the medical record, and is used to achieve privacy-de-sensitized identity representation.

[0066] Specifically, the unique identifier of the patient corresponding to the medical record is extracted, and the identity factor is obtained by encrypting and hashing with the system master key. This ensures both uniqueness and privacy desensitization, preventing direct spying on the patient's identity. The patient's unique identifier is the hospital number, medical record number, or social security number.

[0067] Preferably, a three-level identifier mapping is used to obtain the identity factor: For the patient's unique identifier, HMAC is performed using the institution's master key to obtain an internal pseudo-identifier; then, HMAC is performed based on the internal pseudo-identifier and the shared consortium key to obtain a cross-institutional shared identifier, i.e., the identity factor. It is understandable that the same patient may generate different identity factors in different medical institutions. Cross-institutional patient association can be achieved through the consortium key, but multi-party authorization is required.

[0068] In specific implementation, the physiological factors are obtained by jointly encoding the types and values ​​of physiological parameters in medical records; wherein, the values ​​of physiological parameters are mapped to predefined interval codes after segmented discretization processing.

[0069] Specifically, the physiological factors are obtained by jointly encoding the types and values ​​of physiological parameters in medical records, including:

[0070] Extract the type names and values ​​of physiological parameters from medical records;

[0071] The extracted physiological parameter type names are mapped to a standard medical terminology database to generate type codes;

[0072] Based on the extracted physiological parameter type, a preset numerical division interval is determined, and based on the extracted physiological parameter value, the corresponding numerical division interval is determined, and then the corresponding numerical division interval is mapped to a predefined interval code.

[0073] Physiological factors are obtained by combining the type encoding and interval encoding.

[0074] It should be noted that the division range of physiological parameter values ​​is preset according to the physiological parameter type, and the interval code of the physiological parameter value division range is also preset, which can be directly called when using it.

[0075] More specifically, the standard medical terminology database adopts the IEEE 11073 terminology database or LOINC code, and the type codes for each physiological parameter type are preset; for example, the standard medical term for "systolic blood pressure" is IEEE: MDC_PRESS_BLD_SYS, and the corresponding type code is 0x0101.

[0076] More specifically, physiological parameters are categorized into vital signs, laboratory test results, and continuous waveforms. For example, vital signs include blood pressure, heart rate, and blood oxygen; laboratory test results include blood glucose and white blood cell count; and continuous waveforms include electrocardiogram (ECG) waveforms, electroencephalogram (EEG) waveforms, and respiratory waveforms.

[0077] Furthermore, for vital signs and other physiological parameters, numerical values ​​are divided using standard clinical medical intervals.

[0078] For example, the standard range for systolic blood pressure is 90~140 mmHg, with a range width of 5 mmHg, and the range can be divided into 90~94, 95~99, 100~104, ..., 135~139; the standard range for heart rate is 60~100 bpm, with a range width of 5 bpm, and the range can be divided into 60~64, 65~69, ..., 95~99.

[0079] Furthermore, for physiological parameters tested in laboratories, numerical values ​​are divided using medical reference ranges.

[0080] For example, the normal range of blood glucose medical reference values ​​is 3.9~6.1 mmol / L, and the medical reference value range includes: <3.0 (hypoglycemia), 3.0~3.8 (low), 3.9~6.1 (normal), 6.2~7.0 (prediabetes), 7.1~11.0 (diabetes), and >11.0 (hyperglycemia).

[0081] Furthermore, for continuous waveform physiological parameters, numerical division is performed using eigenvalue intervals.

[0082] For example, the heart rate characteristics of the electrocardiogram waveform are divided into intervals of <40 (severe bradycardia), 40~59 (bradycardia), 60~100 (normal heart rate), 101~120 (tachycardia), and >120 (severe tachycardia).

[0083] Understandably, segmenting and discretizing physiological parameter values ​​and mapping specific values ​​to predefined interval codes can keep physiological factors generated by the same patient stable and avoid drastic changes in keys due to minor physiological fluctuations.

[0084] In practice, the time factor is obtained by rounding down the collection time of the medical record based on the time window determined by the type of medical business in the medical record for time alignment, and then encoding and converting the aligned time.

[0085] Specifically, the encoding conversion can use Unix timestamp encoding.

[0086] Specifically, a corresponding time window is preset according to the type of medical service, and is set according to specific needs; for example, in this embodiment, if the type of medical service is continuous monitoring, the time window is 1 second; if the type of medical service is real-time vital signs, the time window is 10 seconds; if the type of medical service is nursing records, the time window is 1 minute; if the type of medical service is examination and testing, the time window is 1 hour; if the type of medical service is medical records, the time window is 1 day.

[0087] For example, if the collection time is 2024-05-20 10:30:25.123 and the time window is 10 seconds, then rounding down is: floor(second-level timestamp / time window) × time window = floor(25.123 / 10) × 10 = 2 × 10 = 20 seconds, resulting in the time-aligned time 2024-05-20 10:30:20; where floor represents rounding down.

[0088] Preferably, for multiple medical records in the same window, an incremental sequence number or random number is introduced as a supplementary factor to ensure that the record-level key of each record is unique even within the same time window, thus avoiding conflicts.

[0089] In practice, the device factor is obtained by hashing the identity credentials of the medical record collection device.

[0090] Specifically, based on the type of medical record collection equipment, it is divided into high-trust equipment, medium-trust equipment, and basic-trust equipment. Among them, high-trust equipment is intelligent medical equipment with a complete X.509 digital certificate, medium-trust equipment is conventional equipment with a unique hardware identifier (MAC address, IMEI), and basic-trust equipment is low-cost sensor with only a software identifier.

[0091] Specifically, for highly trusted devices, their certificate digest is extracted as identity credentials; for medium trusted devices, their unique hardware identifier is used as identity credentials; and for basic trusted devices, their software identifier and a pre-set device group key are used as identity credentials.

[0092] It should be noted that existing technologies, such as natural language processing, rule engines, and knowledge graphs, can be used to extract semantic information from medical records, which will not be elaborated here.

[0093] Understandably, fine-grained, semantically aware data security protection is achieved through record-level keys dynamically derived from the quadruple of identity, physiology, time, and device.

[0094] During implementation, the record-level encryption key for the current medical record is dynamically derived using the following methods:

[0095] A1. Combine the quadruples to generate an enhanced seed;

[0096] In practice, the enhanced seed is generated by combining the quadruplets in the following way:

[0097] Each semantic factor in the quadruple is encoded using a structured format that includes type, length, value, and checksum, generating a structured encoded segment for each semantic factor.

[0098] The structured encoded segments of each semantic factor are connected in a predetermined order to obtain intermediate data;

[0099] Head and tail identifiers are added before and after the intermediate data to generate an enhanced seed.

[0100] Understandably, generating enhanced seeds using the above method can effectively prevent seed collisions caused by overlapping field values.

[0101] A2. Using a hash-based key derivation function, the system master key is used as the entropy source, the dynamic salt value is used as the salt, and the enhanced seed is used as the information parameter to derive a record-level encryption key.

[0102] In practice, the dynamic salt value is obtained through the following methods:

[0103] Obtain the hash value of the latest block on the public blockchain network and generate a local true random number;

[0104] The quadruple, the obtained hash value, and the local true random number are combined according to a predetermined format to generate input data;

[0105] Using the system master key as the key, the input data is hashed to calculate the message authentication code, generating a dynamic salt value.

[0106] It should be noted that blockchain hashes provide a global randomness reference, and in scenarios with extremely high real-time requirements, a high-precision clock entropy source can be used as an alternative.

[0107] Specifically, the system master key is generated and stored by the Hardware Security Module (HSM) or the Trusted Execution Environment (TEE). The master key follows a strict rotation strategy (such as quarterly or annually). During rotation, the old master key is sealed for decrypting historical data, and the new master key is used to derive record-level encryption keys for new data, thus achieving forward security.

[0108] In practice, the revocation mechanism includes:

[0109] When a medical record needs to be revoked, the revocation message containing the record-level encryption key is encrypted using a post-quantum cryptography algorithm to generate a revocation token. Subsequently, if access to the medical record is required, the revocation token must be obtained to realize the logical revocation of access to the medical record.

[0110] Specifically, the revocation message also includes a predefined revocation flag, such as 0x01; the revocation message is represented as: revocation message = revocation flag || quadruple hash || corresponding record-level key; || indicates concatenation.

[0111] Specifically, the revocation token is a structured message encrypted with the RLWE public key.

[0112] Understandably, when any application (including the server) attempts to access revoked data, it must first obtain the revocation token. Since decrypting the token requires the corresponding RLWE private key (securely held by the data controller), neither the cloud nor any unauthorized party can decrypt the original record-level encryption key. In this embodiment, when a patient exercises their right to revoke consent according to regulations, there is no need to re-encrypt or delete the massive cloud data; instead, logical destruction is achieved through a cryptographic revocation command.

[0113] Preferably, this embodiment also includes lifecycle management of revocation tokens. The post-quantum cryptographic key pair used to encrypt the revocation token is updated periodically. When the key pair is updated, the revocation token previously generated based on the old public key automatically becomes invalid. For data that is still within the revocation period, the revocation token is regenerated using the new public key, which limits the validity period of the token and reduces the long-term risk of key exposure.

[0114] Specifically, post-quantum cryptography key pairs are updated periodically (e.g., monthly). When a key pair is updated, the revocation token generated based on the old public key automatically becomes invalid. For data still within the revocation period, a new revocation token is generated using the new public key. The data controller encrypts a structured revocation message using the current public key, generates a revocation token, and broadcasts it. Before attempting decryption in the cloud, the token must first be decrypted using the current private key to obtain the record-level key. Understandably, by managing the lifecycle of the revocation token, its validity period is limited, reducing the long-term risk of key exposure and allowing the system to flexibly manage data state.

[0115] Understandably, in this embodiment, the complex data reconstruction problem is transformed into a simple key revocation problem, and a post-quantum cryptography algorithm (RLWE) is used to protect the revocation token. This is intended to address the security threats brought about by future quantum computing, which is an important enhancement to the existing revocation mechanism. At the same time, the entire process does not require the transmission or processing of the ciphertext data itself, making it extremely efficient and meeting the timeliness requirements for responding to the right to be forgotten in GDPR, personal information protection laws, and other laws.

[0116] Specifically, the streaming block encryption process is as follows: the data stream collected from the device is piped through the operating system kernel layer or driver layer, the data blocks are encrypted in real time using the record-level encryption key, and the ciphertext is written to the output stream.

[0117] Preferably, the secure encryption method further includes:

[0118] During the streaming block encryption process, the combined message of the quadruple and the medical record data block count is used to calculate the message authentication code using an audit key that is separate from the system master key, and an audit tag is generated.

[0119] Understandably, the audit tags generated by the above process cannot be reversed to derive the original four-tuple semantic information. However, authorized system personnel can use the same audit key to verify the logs, achieving verifiable but untraceable security auditing, confirming which encryption operation a certain audit record corresponds to, and greatly reducing the risk of privacy leakage while meeting audit requirements.

[0120] Preferably, the data transmission in this embodiment adopts a two-level security protection system. At the transmission layer, a secure channel is established through two-way certificate authentication. The client certificate of the acquisition device includes certificate extension information generated based on its device factors. At the encryption layer, a record-level encryption key dynamically generated by a quadruple is used to encrypt the data, so that the device identity authenticated at the transmission layer is used as one of the generation factors of the encryption layer key, thus forming a closed-loop self-verification security system.

[0121] Specifically, at the transport layer, TLS 1.3 and the Chinese national cryptographic standard SM2 are used. When establishing a secure channel, the client certificate in the two-way certificate authentication includes certificate extension information generated based on its device factors; that is, not only must the device prove its identity, but its identity information is also directly used as one of the factors for upper-layer encryption. At the encryption layer, AES-256-GCM and record-level encryption keys are used.

[0122] Understandably, the record-level encryption key used by the encryption layer is dynamically generated entirely from the device factors and other business semantics certified by the transport layer. The two security layers are no longer isolated, but constitute a closed-loop, self-verifying security system. Any security violation at any layer (such as device certificate expiration or transmission hijacking) will directly result in the upper layer data being unable to be decrypted, thus achieving dynamic and in-depth defense.

[0123] It should be noted that when an authorized doctor needs to view patient data, the system first verifies the doctor's identity and permissions. Then, based on the patient information, data type, time window, and device identifier, it reconstructs a four-tuple semantic factor that is exactly the same as the one used when the data was encrypted. Next, it requests the hardware security module to recalculate a unique record-level encryption key for that record using the corresponding master key version, salt value, and the same key derivation algorithm. Before decryption, it checks the revocation token to ensure that the data has not been withdrawn by the patient. Finally, it uses this key to decrypt the ciphertext, verifies its integrity, and returns the original medical data to the doctor. Throughout the entire process, the record-level encryption key material never leaves the secure hardware, and all operations are recorded by an irreversible audit tag.

[0124] Compared with existing technologies, this embodiment provides a secure encryption method for medical data acquisition. By dynamically generating a unique key for each medical record, record-level encryption is achieved. The encryption key is directly associated with the clinical semantics of the data, allowing security policies to be naturally integrated with medical business logic and supporting context-based intelligent access control. Streaming block encryption is employed, completing encryption simultaneously with data generation, meeting the high real-time and low-latency requirements of medical scenarios, while parallel encryption and transmission improve efficiency. Logical access revocation is achieved through a cryptographic token mechanism, eliminating the need for physical deletion or reencryption of massive amounts of data, quickly responding to compliance requirements such as patient privacy withdrawal, and significantly reducing operational costs. Encrypted data is stored in association with a four-tuple, ensuring security while preserving key semantic information, providing a structured foundation for subsequent privacy computation, cross-institutional collaboration, or data analysis. Furthermore, key derivation, real-time encryption, secure transmission, and dynamic revocation are integrated into a closed-loop process, forming a fully controllable security system from data generation to destruction, significantly improving the security, flexibility, and compliance of medical data in acquisition, transmission, storage, and lifecycle management.

[0125] Example 2

[0126] A specific embodiment of the present invention discloses a secure encryption system for medical data acquisition, comprising:

[0127] The data acquisition module is used to collect medical records generated by medical devices in real time.

[0128] The semantic parsing module is used to extract semantic information from each medical record and generate a four-tuple, which includes four semantic factors: identity factor, physiological factor, time factor, and device factor.

[0129] The key derivation module is used to dynamically derive the record-level encryption key of the current medical record based on the quadruple and using the key derivation method.

[0130] The real-time encryption and transmission module is used to perform real-time streaming block encryption on the current medical record based on the record-level encryption key, generate encrypted medical data, and transmit and store the encrypted medical data and the corresponding quadruple;

[0131] The revocation management module is used to perform logical access to and revocation of medical records based on the constructed revocation mechanism when medical records need to be revoked.

[0132] The specific implementation process of this invention can be found in the above method embodiments, and will not be repeated here.

[0133] Since this embodiment is based on the same principle as the above-described method embodiments, this system also has the corresponding technical effects of the above-described method embodiments.

[0134] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware, and the program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0135] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A secure encryption method for medical data acquisition, characterized in that, Includes the following steps: Collect medical records generated by medical devices and perform the following actions on each medical record: Based on current medical records, semantic information is extracted to generate a quadruple, which includes four semantic factors: identity factor, physiological factor, time factor, and device factor. The physiological factor is obtained by jointly encoding the type and value of physiological parameters in the medical record. The physiological parameter values ​​are segmented and discretized, then mapped to a predefined interval code. This process includes: extracting the physiological parameter type name and value from the medical record; mapping the extracted physiological parameter type name to a standard medical terminology database to generate a type code; determining preset intervals for each value based on the extracted physiological parameter type; determining the corresponding value interval based on the extracted physiological parameter value; and mapping the corresponding value interval to a predefined interval code. Based on the type code and interval code, the physiological factor is obtained by combining them. The physiological parameter types are categorized into vital signs, laboratory tests, and continuous waveforms. The time factor is obtained by rounding down the collection time of the medical record based on a time window determined by the medical service type in the medical record, followed by encoding conversion. Based on the aforementioned quadruple, a key derivation method is used to dynamically derive the record-level encryption key for the current medical record; including: Combining the quadruples to generate a reinforced seed includes: encoding each semantic factor in the quadruples using a structured format of type, length, value, and check, generating a structured encoding segment for each semantic factor; connecting the structured encoding segments of each semantic factor in a set order to obtain intermediate data; and adding header and tail identifiers before and after the intermediate data to generate a reinforced seed. A hash-based key derivation function is used, with the system master key as the entropy source, the dynamic salt value as the salt, and the enhanced seed as the information parameter, to derive a record-level encryption key. The dynamic salt value is obtained as follows: the hash value of the latest block in the public blockchain network is obtained, and a local true random number is generated; the quadruple, the obtained hash value, and the local true random number are combined according to a predetermined format to generate input data; using the system master key as the key, a hash message authentication code is calculated on the input data to generate the dynamic salt value. Real-time streaming block encryption of the current medical record is performed based on the record-level encryption key to generate encrypted medical data, and the encrypted medical data and the corresponding quadruple are transmitted and stored; during the streaming block encryption process, the combined message of the quadruple and the medical record data block count is calculated using an audit key separate from the system master key to generate an audit tag. If a medical record needs to be revoked, the logical access to the medical record will be revoked based on the constructed revocation mechanism.

2. The secure encryption method for medical data acquisition according to claim 1, characterized in that, The identity factor is generated based on the unique identifier of the patient corresponding to the medical record; The physiological factors are generated based on the types and values ​​of physiological parameters in medical records; The time factor is generated based on the type of medical service and the collection time in the medical records; The device factor is generated based on the security level of the medical record acquisition device.

3. The secure encryption method for medical data acquisition according to claim 2, characterized in that, The identity factor is obtained by hashing the unique identifier of the patient corresponding to the medical record; The device factor is obtained by hashing the identity credentials of the medical record collection device.

4. The secure encryption method for medical data acquisition according to claim 1, characterized in that, The revocation mechanism includes: When a medical record needs to be revoked, the revocation message containing the record-level encryption key is encrypted using a post-quantum cryptography algorithm to generate a revocation token. Subsequently, if access to the medical record is required, the revocation token must be obtained to realize the logical revocation of access to the medical record.

5. A secure encryption system for medical data acquisition, characterized in that, include: The data acquisition module is used to collect medical records generated by medical devices in real time. The semantic parsing module is used to extract semantic information from each medical record and generate a four-tuple, which includes four semantic factors: identity factor, physiological factor, time factor, and device factor. The physiological factor is obtained by jointly encoding the type and value of physiological parameters in the medical record. The values ​​of the physiological parameters are segmented and discretized and mapped to predefined interval codes. This includes: extracting the type name and value of physiological parameters from the medical record; mapping the extracted physiological parameter type name to a standard medical terminology database to generate a type code; determining preset intervals for each value based on the extracted physiological parameter type; determining the corresponding value interval based on the extracted physiological parameter value; and mapping the corresponding value interval to a predefined interval code. Based on the type code and interval code, the physiological factor is obtained by combining them. The physiological parameter types are categorized into vital signs, laboratory tests, and continuous waveforms. The time factor is obtained by rounding down the acquisition time of the medical record based on the time window determined by the medical service type in the medical record, and then encoding and converting the result. A key derivation module is used to dynamically derive the record-level encryption key of the current medical record based on the quadruple and using a key derivation method; including: Combining the quadruples to generate a reinforced seed includes: encoding each semantic factor in the quadruples using a structured format of type, length, value, and check, generating a structured encoding segment for each semantic factor; connecting the structured encoding segments of each semantic factor in a set order to obtain intermediate data; and adding header and tail identifiers before and after the intermediate data to generate a reinforced seed. A hash-based key derivation function is used, with the system master key as the entropy source, the dynamic salt value as the salt, and the enhanced seed as the information parameter, to derive a record-level encryption key. The dynamic salt value is obtained as follows: the hash value of the latest block in the public blockchain network is obtained, and a local true random number is generated; the quadruple, the obtained hash value, and the local true random number are combined according to a predetermined format to generate input data; using the system master key as the key, a hash message authentication code is calculated on the input data to generate the dynamic salt value. The real-time encryption and transmission module is used to perform real-time streaming block encryption on the current medical record based on the record-level encryption key, generate encrypted medical data, and transmit and store the encrypted medical data and the corresponding quadruple; during the streaming block encryption process, the combined message of the quadruple and the medical record data block count is used to calculate the message authentication code using an audit key that is separate from the system master key, and generate an audit tag; The revocation management module is used to perform logical access to and revocation of medical records based on the constructed revocation mechanism when medical records need to be revoked.

Images