Intelligent hospital data management method and system based on cloud computing

Abstract

The present invention discloses a smart hospital data management method and system based on cloud computing, including: the present invention proposes a smart hospital data management method and system based on cloud computing, obtains patient appointment registration data, medical process data, medical service data, drug management data and rehabilitation guidance data in the smart hospital to be managed, and pre-processes the acquired data; performs improved encryption and first structured operations on the pre-processed first data, and uploads the first data after the improved encryption and first structured operations to a first cloud terminal; the first cloud terminal performs a first analysis and mining on the first data after the improved encryption and first structured operations according to a built-in first processing logic; combines the second data after the first analysis and mining with the first data for second structured processing; stores the data after the second structured processing to the first cloud terminal, completes the smart hospital data management, and provides strong support for hospital managers.

Description

Technical Field

[0001] The present invention relates to the technical field of cloud computing management of patient-specific data, and in particular to a cloud computing-based intelligent hospital data management method and system. Background Art

[0002] With the rapid development of information technology, smart hospitals, as an important form of modern medical services, have gradually entered people's daily lives. By cleverly using information technology, smart hospitals have not only optimized various aspects such as patient appointment registration, medical treatment process, medical services, drug management, and rehabilitation guidance, but also significantly improved the hospital's service efficiency and service quality.

[0003] In smart hospitals, comprehensive data management becomes the core. Patient information, medical records, drug inventory, equipment usage and other types of data are efficiently integrated and analyzed through advanced algorithms and models to provide doctors with scientific decision-making basis and bring more accurate and personalized medical services to patients. In addition, the effective use of data also promotes the collaboration between various departments within the hospital, breaks the information island, and makes medical resources more reasonably allocated.

[0004] However, with the continuous deepening of smart hospital data management, data privacy and security issues have become increasingly prominent. Once sensitive data such as patients' personal information and health status are leaked or abused, it will cause irreparable losses to patients. Therefore, how to achieve effective management and utilization of smart hospital data under the premise of ensuring data security and privacy has become an urgent problem to be solved. Summary of the invention

[0005] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0006] In view of the above existing problems, the present invention is proposed.

[0007] Therefore, the present invention provides a cloud computing-based intelligent hospital data management method and system, which can solve the problems mentioned in the background technology.

[0008] In order to solve the above technical problems, the present invention provides the following technical solutions, a cloud computing-based intelligent hospital data management method, comprising:

[0009] Obtain patient appointment registration data, medical process data, medical service data, drug management data, and rehabilitation guidance data in the smart hospital to be managed, and pre-process the acquired data;

[0010] Performing improved encryption and a first structuring operation on the preprocessed first data, and uploading the first data after the improved encryption and the first structuring operation to the first cloud terminal;

[0011] The first cloud terminal performs a first analysis and mining on the first data after the improved encryption and the first structuring operation according to the built-in first processing logic;

[0012] Performing a second structural processing on the second data after the first analysis and mining combined with the first data;

[0013] The data after the second structured processing is stored in the first cloud terminal to complete the smart hospital data management.

[0014] As a preferred solution of the cloud computing-based smart hospital data management method of the present invention, the step of obtaining patient appointment registration data, medical process data, medical service data, drug management data and rehabilitation guidance data in the smart hospital to be managed, and preprocessing the obtained data includes:

[0015] The patient appointment registration data includes a unique number identifying the patient who has made an appointment, the appointment date of the corresponding patient, the unique number of the doctor who has seen the patient, the name of the appointment department of the corresponding patient, the specific appointment time of the corresponding patient, and the unique number identifying the patient who has canceled the appointment;

[0016] The medical consultation process data includes the unique identification of the patient, the check-in time of the patient, the medical consultation process sequence of the patient, and the actual time spent on the medical consultation service for the patient;

[0017] The medical service data includes the unique identifier of the medical treatment item of the patient, the doctor's diagnosis result of the patient, the doctor's treatment plan for the patient, and the detailed cost of the medical treatment;

[0018] The drug management data includes the unique number of the drug prescribed by the doctor, the prescription number associated with the drug, the name of the drug, the given dosage of the drug, and the frequency of use of the drug;

[0019] The rehabilitation guidance data includes a unique identifier of the patient's rehabilitation plan, a rehabilitation guidance text for the patient, and the start and end dates of the patient's rehabilitation plan;

[0020] The pre-processing includes data unique indexing processing including the unique identification of the patient, the unique identification of the patient's treatment program, the unique number of the doctor who treated the patient, the unique number of the patient who canceled the appointment, the unique number of the drug prescribed by the doctor, the associated prescription number of the drug, and the unique identification of the patient's rehabilitation plan;

[0021] The missing values ​​of the corresponding patient's appointment date, the corresponding patient's specific appointment time, the patient's check-in time, the actual time spent on the patient's medical service, and the start and end dates of the patient's rehabilitation plan are processed and then formatted in a unified manner;

[0022] The word segmentation and named entity recognition are performed on the name of the appointment department for the corresponding patient, the patient's medical treatment process sequence, the doctor's diagnosis result for the patient, the doctor's treatment plan for the patient, the detailed cost of this diagnosis and treatment, the name of the drug, the given dosage of the drug, the frequency of use of the drug, and the patient's rehabilitation guidance text.

[0023] As a preferred solution of the cloud computing-based intelligent hospital data management method of the present invention, the improved encryption and first structured operation on the pre-processed first data includes:

[0024] The data processed by the unique index of the data is encrypted by an improved first algorithm, wherein the first algorithm includes firstly constructing the combined data and adding salt, which is expressed as follows:

[0025]

[0026] Among them, C initial represents the initial combination value, Pj represents the unique identifier of the patient, Tj represents the unique identifier of the patient's treatment item, Dj represents the unique number of the doctor who treated the patient, Rj represents the unique number of the drug prescribed by the doctor, Ej represents the associated prescription number of the drug, Cj represents the unique identifier of the patient's rehabilitation plan, S g Indicates the global salt value, represents the splicing operation, j represents the jth patient;

[0027] Second, perform multiple iterations of hashing on the combined salted data:

[0028] H1=H(C initial ,S g )

[0029] H2=H(H1,S g ∥0) ...

[0030] H final =H(H iterations-1 ,S g ∥(iterations-1))

[0031] Where H represents the hash function, H final It represents the final hash value obtained after multiple rounds of iterations of the combined salted data. iterations-1It represents the intermediate hash value before the last iteration of the combined salted data, iterations represents the number of hash iterations in the combined salted data, and ∥ represents the string concatenation operation;

[0032] Finally, the context salt is combined to generate the final index hash:

[0033]

[0034] Among them, I represents the final index hash result, S c Represents the client salt value, and P represents the message or data segment to be verified.

[0035] As a preferred solution of the cloud computing-based intelligent hospital data management method of the present invention, the improved encryption and first structuring operation on the pre-processed first data further includes:

[0036] The data after the missing values ​​are processed and then formatted uniformly are encrypted using a second algorithm, wherein the second algorithm includes firstly recording the appointment date of the corresponding patient as Asj, the specific appointment time of the corresponding patient as Ayj, the check-in time of the patient as Csj, the actual time consumed by the patient's medical service as Ssj, the start and end dates of the patient's rehabilitation plan as Rsj and Rej, and converting all time data uniformly into a Unix timestamp format;

[0037] Secondly, differential privacy noise is injected into the time data in the unified timestamp format, as shown below:

[0038]

[0039] Where D[j] represents the appointment date or appointment time or check-in time or the actual time spent on medical services or the start and end dates of the rehabilitation plan of the jth patient after differential privacy processing, S[j] represents the original appointment date or appointment time or check-in time or the actual time spent on medical services or the start and end dates of the rehabilitation plan corresponding to the jth patient, L represents the function sampled from the Laplace distribution, Δf represents the maximum difference between the appointment date or appointment time or check-in time or the actual time spent on medical services or the start and end dates of the rehabilitation plan between two different patients, ∈ represents the differential privacy budget, the smaller ∈ is, the higher the degree of privacy protection is;

[0040] Again, the time data after differential privacy noise injection is hashed and encrypted;

[0041] Finally, the hashed time data is subjected to improved Paillier homomorphic encryption, and the improved Paillier homomorphic encryption is expressed as:

[0042] Generate the Paillier public key n = pq and g, where p and q represent two prime numbers, n represents the public modulus in the Paillier encryption system, that is, one of the public keys, and the public key is (n, g);

[0043] Pre-compute the 2 w th power of g and store it in the array G w where w represents the window width, which is used to divide the number of bits in the binary representation during fast exponentiation calculation, mod represents the modulo operation, i represents the window bit number, and i = 0, 1, 2..., w - 1;

[0044] Convert the plaintext m to an integer, 0 ≤ m < n, and divide the binary representation of m into windows of width w;

[0045] Initialize g m′ ← 1 mod n 2 , and m' represents the exponential form of the plaintext m obtained after fast exponentiation calculation;

[0046] For each window b w-1 ...b0 of the plaintext m, traverse from the high-order bit to the low-order bit of the window, and for each bit b i in the window, perform the following operations:

[0047] If b i = 0, then g m′ ← (g m′ ) 2 mod n 2

[0048] If b i = 1, then g m′ ← (g m′ ) 2 · g mod n 2

[0049] Randomly select an integer r (0 < r < n) and perform the encryption calculation process of the improved Paillier encryption. The encryption calculation process is expressed as:

[0050] c = (g m′ · r n ) mod n 2

[0051] where c represents the encrypted ciphertext, and g m′ represents a part of the public key after the modulo operation update.

[0052] As a preferred solution of the intelligent hospital data management method based on cloud computing described in the present invention, wherein: the improved encryption and the first structuring operation on the preprocessed first data further include:

[0053] The data after word segmentation and named entity recognition are encrypted by a third algorithm, and the third algorithm encryption includes firstly quantifying the name of the appointment department of the corresponding patient, the patient's medical process sequence, the doctor's diagnosis result of the patient, the doctor's treatment plan of the patient, the cost details of this diagnosis and treatment, the name of the drug, the given dose of the drug, the frequency of use of the drug, and the rehabilitation guidance text of the patient, and setting Dj={Slj, Plj, Clj, Tlj, Flj, Mlj, Doselj, Frequencylj, Rlj} represents the data set of the jth patient, wherein Slj represents the word vector representation of the appointment department name, Plj represents the encoding of the medical process sequence, Clj represents the word vector representation of the diagnosis result, Tlj represents the word vector representation of the treatment plan, Flj represents the numerical list of the medical cost details, Mlj represents the word vector representation of the drug name, Doselj represents the numerical value of the given dose of the drug, Frequencylj represents the numerical value of the frequency of use of the drug, and Rlj represents the word vector representation of the rehabilitation guidance text;

[0054] Secondly, differential privacy noise is injected into the data quantized data set, as shown below:

[0055] Noisy D =Dj+N(0,σ 2 I)

[0056] Among them, Noisy D represents the data set after adding noise, Dj represents the original data set of the jth patient, N(0,σ 2 I) represents the multivariate normal distribution, σ 2 I represents the covariance matrix;

[0057] Finally, the improved Paillier homomorphic encryption algorithm is used for encryption operation, which is expressed as follows:

[0058] Encrypted D =E((n,g),Dj)

[0059] =[E((n,g),Slj),E((n,g),Plj),E((n,g),Clj),...,E((n,g),Rlj)]

[0060] Among them, E represents the improved Paillier homomorphic encryption algorithm, (n,g) represents the public key, and Encrypted D Represents the encrypted overall data set.

[0061] As a preferred solution of the cloud computing-based intelligent hospital data management method of the present invention, wherein: the first cloud terminal performs a first analysis and mining on the first data after the improved encryption and the first structuring operation according to the built-in first processing logic, including: the first processing logic includes using a trained machine learning model for logic processing;

[0062] The trained machine learning models include a hospital service demand prediction model, a drug consumption trend prediction model, a drug abuse prediction model, an unreasonable drug prescribing prediction model, a frequent appointment cancellation prediction model, and a department resource allocation prediction model;

[0063] The input of the hospital service demand prediction model is the encrypted appointment registration data of the previous cycle, and the output is the overall service demand of the hospital in a specified time period in the future;

[0064] The input of the drug consumption trend prediction model is the drug management data after improved encryption and structured processing in the previous period, and the output is the estimated total consumption of drugs in a specified time period in the future;

[0065] The input of the drug abuse prediction model is the encrypted and structured given dose of the drug in the previous cycle, the doctor's diagnosis of the patient, and the doctor's treatment plan for the patient, and the output is a classified prediction of whether the patient has a risk of drug abuse;

[0066] The input of the unreasonable prescription prediction model is the encrypted diagnosis result of the doctor on the patient in the previous cycle, the unique number of the drug prescribed by the doctor for the patient's treatment plan, the associated prescription number of the drug, the name of the drug, the given dosage of the drug, and the frequency of use of the drug. The output is an evaluation of the rationality of the drug prescribed by the doctor.

[0067] The input of the frequent appointment cancellation prediction model is the unique number identifying the patient who made an appointment in the previous period and the unique number identifying the patient who canceled the appointment, and the output is the prediction result of whether the patient will cancel the current appointment;

[0068] The input of the department resource allocation prediction model is the unique number of the patient who made an appointment in the previous period, the unique number of the doctor who saw the patient, the name of the corresponding patient's appointment department, the unique number of the patient who canceled the appointment, the patient's check-in time, the actual time spent on the patient's medical service, the patient's rehabilitation guidance text, and the start and end dates of the patient's rehabilitation plan. The output is the hospital's material resource allocation plan.

[0069] As a preferred solution of the cloud computing-based smart hospital data management method of the present invention, wherein: the first structured processing includes performing a first structured processing on the first data by identity card number, and the second structured processing includes performing a second structured processing on the first data and the second data by identity card number as the first priority and consultation date as the second priority;

[0070] The data after the second structured processing is stored in the first cloud terminal to complete the smart hospital data management. If it is necessary to retrieve patient appointment registration data, medical process data, medical service data, drug management data and rehabilitation guidance data, the data is decrypted and obtained according to the decryption operation corresponding to the encryption operation of different data.

[0071] The intelligent hospital data management system based on cloud computing is characterized by including:

[0072] The data acquisition and processing module is used to acquire the patient appointment registration data, medical process data, medical service data, drug management data and rehabilitation guidance data in the smart hospital to be managed, and pre-process the acquired data;

[0073] A first structuring module, used to perform improved encryption and a first structuring operation on the preprocessed first data, and upload the first data after the improved encryption and the first structuring operation to a first cloud terminal;

[0074] An analysis and mining module, used for the first cloud terminal to perform a first analysis and mining on the first data after the improved encryption and the first structuring operation according to a built-in first processing logic;

[0075] A second structuring module, used to combine the second data after the first analysis and mining with the first data to perform a second structuring process;

[0076] The management module is used to store the data after the second structured processing in the first cloud terminal to complete the smart hospital data management.

[0077] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and wherein the processor implements the steps of the above method when executing the computer program.

[0078] A computer-readable storage medium stores a computer program thereon, wherein the computer program implements the steps of the method described above when executed by a processor.

[0079] Beneficial effects of the present invention: The present invention proposes a smart hospital data management method and system based on cloud computing, obtains patient appointment registration data, medical process data, medical service data, drug management data and rehabilitation guidance data in the smart hospital to be managed, and pre-processes the acquired data; performs improved encryption and first structured operation on the first data after pre-processing, and uploads the first data after improved encryption and first structured operation to the first cloud terminal; the first cloud terminal performs a first analysis and mining on the first data after improved encryption and first structured operation according to the built-in first processing logic; combines the second data after the first analysis and mining with the first data for second structured processing; stores the data after the second structured processing in the first cloud terminal to complete the smart hospital data management. In summary, the smart hospital data management method and system based on cloud computing proposed by the present invention ensure the security and availability of data through improved encryption and structured processing; realizes the analysis and mining of hospital data through machine learning models and deep learning models; and improves the security and visualization of data through blockchain technology and data visualization tools. These advantages enable the present invention to better meet the needs of smart hospital data management and provide strong support for hospital managers. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0081] Figure 1 A method flow chart of a cloud computing-based intelligent hospital data management method and system provided in accordance with an embodiment of the present invention;

[0082] Figure 2 An internal structural diagram of a computer device of a cloud computing-based intelligent hospital data management method and system provided in accordance with an embodiment of the present invention. DETAILED DESCRIPTION

[0083] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0084] Example 1

[0085] Reference Figure 1-2 , which is the first embodiment of the present invention, and provides a cloud computing-based intelligent hospital data management method and system, including a cloud computing-based intelligent hospital data management method and a cloud computing-based intelligent hospital data management system, wherein the cloud computing-based intelligent hospital data management method includes:

[0086] S101, obtaining patient appointment registration data, medical process data, medical service data, drug management data, and rehabilitation guidance data in the smart hospital to be managed, and pre-processing the obtained data;

[0087] Among them, the patient appointment registration data, medical process data, medical service data, drug management data and rehabilitation guidance data in the smart hospital to be managed are obtained, and the obtained data are pre-processed, including: the patient appointment registration data includes a unique number that identifies the patient who has made an appointment, the appointment date of the corresponding patient, the unique number of the doctor who sees the patient, the name of the appointment department of the corresponding patient, the specific appointment time of the corresponding patient, and the unique number that identifies the patient who cancels the appointment;

[0088] Specifically, the unique number of the patient making the appointment and the unique number of the doctor who sees the patient can be data of similar numbers such as a unique identifier, ID card, medical insurance account number, etc. generated by the smart hospital;

[0089] Furthermore, the medical process data includes the unique identification of the patient, the check-in time of the patient, the medical process sequence of the patient, and the actual time spent on the medical service for the patient;

[0090] It should be noted that the patient's medical treatment process sequence can be a serial number of the day generated according to the date, such as the ****th number on the **th month of the **th day of the **th year of 2***, or it can be a patient serial number generated in sequence since the smart hospital was put into use, for example, 1, 2, ..., *******, ..., ******* represents the number of digits, that is, the *******th patient since the smart hospital was put into use, or it can be a fixed starting identification code + a patient serial number generated in sequence, for example, 123456+*******, or patients with the same ID card are always given only one serial number, and other generating methods.

[0091] Furthermore, the diagnosis and treatment service data includes the unique identifier of the diagnosis and treatment item of the patient, the doctor's diagnosis result of the patient, the doctor's treatment plan for the patient, and the detailed cost of the diagnosis and treatment;

[0092] Furthermore, the drug management data includes a unique number of a drug prescribed by a doctor, a prescription number associated with the drug, the name of the drug, a given dosage of the drug, and a frequency of use of the drug;

[0093] Furthermore, the rehabilitation guidance data includes a unique identifier of the patient's rehabilitation plan, a text of the patient's rehabilitation guidance, and the start and end dates of the patient's rehabilitation plan;

[0094] It should be noted that the above data are important information in the operation of smart hospitals, which are of great significance to improving the management efficiency and service quality of hospitals. By collecting, sorting and analyzing these data, hospitals can more accurately understand the needs and conditions of patients, optimize the allocation of medical resources, and improve the pertinence and effectiveness of medical services.

[0095] First, patient appointment registration data can help hospitals understand patients' medical habits and patterns, optimize doctors' shifts and admission plans, and avoid patients' dissatisfaction due to long waiting times. At the same time, for patients who cancel their appointments, hospitals can analyze the reasons, improve the appointment process and service quality, and increase patients' appointment success rate and medical satisfaction.

[0096] Secondly, the medical process data can help hospitals understand the patient's medical experience and process optimization. By analyzing data such as the patient's check-in time, medical process sequence, and actual medical service time, the hospital can find bottlenecks and problems in the process, make timely improvements and optimizations, and improve patient medical efficiency and satisfaction.

[0097] Thirdly, diagnosis and treatment service data is the core information for hospitals to provide medical services. By mining and analyzing these data, hospitals can understand the types of diseases, treatment methods, treatment effects, etc. of patients, providing important basis for hospitals to provide clinical decision support, improve treatment plans and improve the quality of medical services.

[0098] In addition, drug management data also plays an important role in the hospital's drug procurement, inventory management and drug safety. By monitoring and analyzing data such as the drug's unique number, associated prescription number, drug name, given dosage and medication frequency, the hospital can ensure the rational use and safe management of drugs and avoid problems such as drug abuse and misuse.

[0099] Finally, rehabilitation guidance data is an important basis for hospitals to provide rehabilitation services. By tracking and analyzing data such as patients' rehabilitation plans, rehabilitation guidance texts, and the start and end dates of rehabilitation plans, hospitals can provide patients with more personalized and scientific rehabilitation services, promote patients' rehabilitation process, and improve rehabilitation effects.

[0100] To sum up, by collecting, organizing and analyzing patient appointment registration data, medical process data, medical service data, drug management data and rehabilitation guidance data, hospitals can comprehensively improve management efficiency and service quality, and provide patients with better-quality, more efficient and safer medical services.

[0101] Furthermore, the pre-processing includes unique identification of the patient, unique identification of the patient's treatment program, unique number of the doctor who treated the patient, unique number of the patient who canceled the appointment, unique number of the drug prescribed by the doctor, the associated prescription number of the drug, and unique identification of the patient's rehabilitation plan for data unique index processing;

[0102] It should be noted that the standard for unique data index processing is the fixed standard of unique identification of patients visiting the smart hospital, unique identification of patients' medical treatment items, unique number of the doctor who sees the patient, unique number of the patient who cancels the appointment, unique number of the medicine prescribed by the doctor, the associated prescription number of the medicine and unique identification of the patient's rehabilitation plan. For example, if the format of the unique identification of the patients' medical treatment items is ****-****-****, and each * takes a value of 1-5, then the unique identification of the patients' medical treatment items in the data unique index processing is not ****-****-****, or the * has a value other than 1-5, which are all incorrect identifications and need to be filled with missing values.

[0103] In the embodiment of the present application, the data unique index processing includes the first step of generating a unique identifier, which specifically includes:

[0104] Patients: Each patient is assigned a globally unique patient ID, identifying them with the same ID no matter how many appointments, medical records or rehabilitation programs they have.

[0105] Treatment items: Each treatment item will also have its own unique item ID to ensure that there is no confusion within the system.

[0106] Doctor: The doctor's unique number can be the work number assigned within the system or an encrypted identifier of the personal ID number.

[0107] Cancel appointment registration: The record of canceled appointment should be associated with the original appointment registration record, marked with the same patient ID and appointment serial number, and an additional cancellation mark.

[0108] Drugs: Each drug and each prescription it is written for has a unique drug ID and prescription number.

[0109] Rehabilitation plan: Each rehabilitation plan is assigned a unique rehabilitation plan ID, which is associated with the patient ID.

[0110] The second step is index creation:

[0111] At the database level, create a unique index for each unique field to ensure that no duplicate records are generated when data is inserted or updated.

[0112] For cases of compound uniqueness, for example, a combination such as "(patient ID, appointment date, department)" may also require the creation of a compound unique index to ensure that the same patient will not have duplicate appointments in the same department on the same date.

[0113] The third step is data cleaning and normalization:

[0114] During the preprocessing phase, the data is checked for duplicate or inconsistent unique identifiers and are removed or corrected.

[0115] Missing unique identifiers may need to be filled in by looking back at historical records or interfacing with other systems.

[0116] The fourth step is to maintain reference integrity:

[0117] When designing a database, foreign key constraints are used to ensure the integrity of data references, such as ensuring that patient ID, doctor ID, drug ID, etc. must be valid when referencing the primary table in the foreign table.

[0118] Furthermore, the missing values ​​of the corresponding patient's appointment date, the corresponding patient's specific appointment time, the patient's check-in time, the actual time spent on the patient's medical service, and the start and end dates of the patient's rehabilitation plan are processed and then formatted in a unified manner;

[0119] Furthermore, word segmentation and named entity recognition are performed on the name of the appointment department for the corresponding patient, the patient's medical treatment process sequence, the doctor's diagnosis result for the patient, the doctor's treatment plan for the patient, the detailed cost of this diagnosis and treatment, the name of the drug, the given dosage of the drug, the frequency of use of the drug, and the patient's rehabilitation guidance text.

[0120] It should be noted that word segmentation includes: word segmentation is the process of dividing continuous natural language text into word units with independent meanings. For example, sentences such as "internal medicine outpatient appointment", "cardiac ultrasound examination", and "acute pneumonia" are divided into individual words.

[0121] Named entity recognition includes: NER identifies entities with specific meanings in the text and classifies them into categories such as names of people, places, names of organizations, or names of diseases, department names, drug names, dosage units, etc. in the medical field.

[0122] In an optional embodiment, in a medical scenario, this step may identify:

[0123] Department name: such as "Internal Medicine", "Surgery", "Radiology", etc.

[0124] Diagnostic results: nouns of diseases or symptoms such as "hypertension", "diabetes", "fracture", etc.

[0125] Treatment plan: Identify specific treatment methods, surgical names, drug names, etc.

[0126] Drug information: including drug name, dosage (such as "5 mg", "twice a day") and frequency of use (such as "bid" for twice a day and "qd" for once a day).

[0127] Key actions, precautions, recovery goals, etc. in the rehabilitation guidance text.

[0128] It should be noted that through data unique index processing, the medical system can ensure that each patient, each diagnosis and treatment, each drug and each rehabilitation plan has a unique identifier, thereby avoiding data confusion and duplication. This processing method not only improves the quality of data, but also provides a solid foundation for subsequent data analysis and decision-making.

[0129] Data cleaning and normalization steps ensure data integrity and accuracy. By removing duplicate data, correcting errors, and filling missing values, the system is able to provide more accurate and consistent information, helping doctors make more informed decisions.

[0130] Referential integrity maintenance ensures the association and consistency between data through foreign key constraints. This helps prevent errors and misunderstandings caused by inconsistent data and improves the reliability and safety of medical services.

[0131] In addition, word segmentation and named entity recognition technologies further improve the readability and analyzability of medical text data. By applying these technologies to medical text data, the system can automatically identify and classify key information, such as department names, diagnosis results, treatment plans, and drug information. This helps doctors quickly and accurately obtain key information and improve work efficiency, while also helping hospitals to manage the quality of medical services more scientifically and meticulously.

[0132] In summary, through the application of unique index processing of medical data, data cleaning and normalization, reference integrity maintenance, word segmentation processing and named entity recognition, the medical system can provide more accurate, consistent and reliable data support, and provide strong guarantees for improving the quality of medical services and ensuring patient safety.

[0133] S102, performing improved encryption and a first structuring operation on the preprocessed first data, and uploading the first data after the improved encryption and the first structuring operation to a first cloud terminal;

[0134] The improved encryption and the first structural operation on the preprocessed first data include: performing an improved first algorithm encryption on the data after the data unique index processing, and the first algorithm includes firstly constructing the combined data and adding salt, which is expressed as follows:

[0135]

[0136] Among them, C initial represents the initial combination value, Pj represents the unique identifier of the patient, Tj represents the unique identifier of the patient's treatment item, Dj represents the unique number of the doctor who treated the patient, Rj represents the unique number of the drug prescribed by the doctor, Ej represents the associated prescription number of the drug, Cj represents the unique identifier of the patient's rehabilitation plan, S g Indicates the global salt value, represents the splicing operation, j represents the jth patient;

[0137] Second, perform multiple iterations of hashing on the combined salted data:

[0138] H1=H(C initial ,S g )

[0139] H2=H(H1,S g ∥0) ...

[0140] H final =H(H iterations-1 ,S g ∥(iterations-1))

[0141] Where H represents the hash function, H final It represents the final hash value obtained after multiple rounds of iterations of the combined salted data. iterations-1 It represents the intermediate hash value before the last iteration of the combined salted data, iterations represents the number of hash iterations in the combined salted data, and ∥ represents the string concatenation operation;

[0142] Finally, the context salt is combined to generate the final index hash:

[0143]

[0144] Among them, I represents the final index hash result, S c Represents the client salt value, and P represents the message or data segment to be verified.

[0145] Furthermore, the improved encryption and the first structuring operation on the preprocessed first data further include: performing second algorithm encryption on the data after missing value processing and format unification. The second algorithm includes: First, record the appointment date of the corresponding patient as Asj, the specific appointment time of the corresponding patient as Ayj, the check-in time of the patient as Csj, the actual time consumed for the patient's medical service as Ssj, and the start and end dates of the patient's rehabilitation plan as Rsj and Rej, and uniformly convert all time data into the Unix timestamp format;

[0146] Second, perform differential privacy noise injection on the time data after the unified timestamp format, which is expressed as follows:

[0147]

[0148] where D[j] represents the appointment date or specific appointment time or check-in time or actual time consumed for medical service or start and end dates of the rehabilitation plan of the j-th patient after differential privacy processing, S[j] represents the original appointment date or specific appointment time or check-in time or actual time consumed for medical service or start and end dates of the rehabilitation plan of the corresponding j-th patient, L represents a function sampled from the Laplace distribution, Δf represents the maximum difference in the appointment date or specific appointment time or check-in time or actual time consumed for medical service or start and end dates of the rehabilitation plan between two different patients, and ∈ represents the differential privacy budget. The smaller ∈ is, the higher the degree of privacy protection;

[0149] Third, perform hash encryption on the time data after differential privacy noise injection;

[0150] Finally, perform improved Paillier homomorphic encryption on the hash-encrypted time data. The improved Paillier homomorphic encryption is expressed as:

[0151] Generate the Paillier public key n = pq and g, where p and q represent two prime numbers, n represents the public modulus in the Paillier encryption system, that is, one of the public keys, and the public key is (n, g);

[0152] Furthermore, pre-compute the 2 w -th power of g and store it in the array G w where w represents the window width, which is used to divide the number of bits in the binary representation in fast exponentiation calculation, , mod represents the modulo operation, i represents the window bits, and i = 0, 1, 2..., w - 1;

[0153] Furthermore, convert the plaintext m into an integer, 0 ≤ m < n, and divide the binary representation of m into windows with a width of w bits;

[0154] Furthermore, initialize gm′ ←1 mod n 2 , where m' represents the exponential form of the plaintext m obtained after fast exponentiation calculation;

[0155] Furthermore, for each window b of the plaintext m w-1 ...b0, traverse from the high-order bit to the low-order bit of the window, and for each bit b within the window i , perform the following operations:

[0156] If b i = 0, then g m′ ←(g m′ ) 2 mod n 2

[0157] If b i = 1, then g m′ ←(g m′ ) 2 ·g mod n 2

[0158] Furthermore, randomly select an integer r (0 < r < n), and perform the encryption calculation process of the improved Paillier encryption. The encryption calculation process is expressed as:

[0159] c = (g m′ ·r n ) mod n 2

[0160] where c represents the encrypted ciphertext, and g m′ represents a part of the public key updated by the modulo operation.

[0161] Furthermore, the improved encryption and the first structural operation of the pre-processed first data also include: performing a third algorithm encryption on the data after word segmentation and named entity recognition, and the third algorithm encryption includes firstly quantifying the name of the appointment department of the corresponding patient, the patient's medical process sequence, the doctor's diagnosis result of the patient, the doctor's treatment plan for the patient, the cost details of this diagnosis and treatment, the name of the drug, the given dosage of the drug, the frequency of use of the drug, and the patient's rehabilitation guidance text, assuming Dj = {Slj, Plj, Clj, Tlj, Flj, Mlj, Doselj, Frequencylj, Rlj} represents the data set of the jth patient, where Slj represents the word vector representation of the appointment department name, Plj represents the encoding of the medical process sequence, Clj represents the word vector representation of the diagnosis result, Tlj represents the word vector representation of the treatment plan, Flj represents the numerical list of medical expenses, Mlj represents the word vector representation of the drug name, Doselj represents the numerical value of the given dose of the drug, Frequencylj represents the numerical value of the frequency of drug use, and Rlj represents the word vector representation of the rehabilitation guidance text;

[0162] Secondly, differential privacy noise is injected into the data quantized data set, as shown below:

[0163] Noisy D =Dj+N(0,σ 2 I)

[0164] Among them, Noisy D represents the data set after adding noise, Dj represents the original data set of the jth patient, N(0,σ 2 I) represents the multivariate normal distribution, σ 2 I represents the covariance matrix;

[0165] Finally, the improved Paillier homomorphic encryption algorithm is used for encryption operation, which is expressed as follows:

[0166] Encrypted D =E((n,g),Dj)

[0167] =[E((n,g),Slj),E((n,g),Plj),E((n,g),Clj),...,E((n,g),Rlj)]

[0168] Among them, E represents the improved Paillier homomorphic encryption algorithm, (n,g) represents the public key, and Encrypted D Represents the encrypted overall data set.

[0169] It should be noted that by preprocessing the original data, including missing value processing, format unification, differential privacy noise injection and other steps, patient privacy and data security can be protected, while improving the availability and accuracy of the data; by hashing the data and improving Paillier homomorphic encryption, the security and privacy protection of the data can be further enhanced, while ensuring the integrity and verifiability of the data; by performing word segmentation, named entity recognition and data quantification on the data, the structured degree of the data can be improved, which facilitates subsequent data analysis and mining; by encrypting the data with a third algorithm, the security and privacy protection of the data can be further enhanced, while ensuring the confidentiality and integrity of the data. In summary, the above steps can effectively protect patient privacy and data security, improve the availability and accuracy of the data, and provide strong support for subsequent data analysis and mining.

[0170] In addition, the above steps also have good flexibility and scalability. For example, for different data types and scenarios, different preprocessing methods, encryption algorithms and parameter settings can be selected as needed to meet different privacy protection and data security requirements. At the same time, the above steps can also be combined with other data processing and analysis methods to form a more complete and efficient data processing and analysis system, providing more reliable and secure support for medical informatization and intelligence.

[0171] In the embodiment of the present application, a corresponding decryption table is generated and placed in a local server while encrypting. The hash functions in the present application are all reversible hash functions. The differential privacy noise injection in the present application can be set according to the parameters of the differential privacy algorithm to provide a certain confidence interval or estimation range for reverse estimation of the original data.

[0172] In the embodiment of the present application, a copy of the unprocessed data needs to be saved in the local server for subsequent retrieval of the corresponding original data according to the decryption table.

[0173] S103, the first cloud terminal performs a first analysis and mining on the first data after the improved encryption and the first structuring operation according to the built-in first processing logic;

[0174] The first cloud terminal performs a first analysis and mining on the first data after the improved encryption and the first structuring operation according to the built-in first processing logic, including: the first processing logic includes performing logic processing using a trained machine learning model;

[0175] Specifically, the trained machine learning models include hospital service demand prediction model, drug consumption trend prediction model, drug abuse prediction model, unreasonable drug prescription prediction model, frequent appointment cancellation prediction model, and department resource allocation prediction model;

[0176] Furthermore, the input of the hospital service demand prediction model is the encrypted appointment registration data of the previous cycle, and the output is the overall service demand of the hospital in a specified time period in the future;

[0177] Furthermore, the input of the drug consumption trend prediction model is the drug management data that has been improved, encrypted and structured in the previous period, and the output is the estimated total consumption of drugs in a specified time period in the future;

[0178] Furthermore, the input of the drug abuse prediction model is the encrypted and structured given dosage of the drug in the previous cycle, the doctor's diagnosis of the patient, and the doctor's treatment plan for the patient, and the output is a classified prediction of whether the patient has a risk of drug abuse;

[0179] Furthermore, the input of the unreasonable prescription prediction model is the encrypted diagnosis result of the doctor on the patient in the previous cycle, the unique number of the drug prescribed by the doctor for the patient's treatment plan, the associated prescription number of the drug, the name of the drug, the given dosage of the drug, and the frequency of use of the drug, and the output is an evaluation of the rationality of the drug prescribed by the doctor;

[0180] Furthermore, the input of the frequent cancellation prediction model is the unique number identifying the patient who made an appointment in the previous period and the unique number identifying the patient who canceled the appointment, and the output is the prediction result of whether the patient will cancel the current appointment;

[0181] Furthermore, the input of the department resource allocation prediction model is the unique number of the patient who made an appointment in the previous period, the unique number of the doctor who saw the patient, the name of the corresponding patient's appointment department, the unique number that identifies the patient who canceled the appointment, the patient's check-in time, the actual time spent on the patient's medical service, the patient's rehabilitation guidance text, and the start and end dates of the patient's rehabilitation plan. The output is the hospital's material resource allocation plan.

[0182] In the embodiments of the present application, the prediction model is established using the following steps, which are illustrated below using a hospital service demand prediction model as an example:

[0183] Step 1: Data collection and preprocessing:

[0184] Obtain the encrypted appointment registration data of the previous period, including but not limited to the number of patients, department distribution, appointment time and other information.

[0185] Obtain plaintext data through decryption (under the premise of legality and security), and perform necessary preprocessing, such as missing value filling, outlier processing, data standardization, time series adjustment, etc.

[0186] Step 2: Feature Engineering:

[0187] Combined with time series analysis, useful features can be extracted, such as the number of registrations per day, week, and month, peak periods, trough periods, holiday effects, etc.

[0188] Perform security analysis on encrypted data and use the encrypted data to extract statistical features, such as calculating the distribution and correlation of the encrypted data.

[0189] Step 3: Model Design:

[0190] Select model architecture: Since this is a time series prediction problem, you can consider using models such as ARIMA, LSTM, GRU, Prophet, or ensemble learning methods such as random forest and gradient boosting machine for time series prediction.

[0191] Encryption-friendly design: Since the data is encrypted, it is necessary to design a model that can run on encrypted data or predict features derived from encrypted data. Currently, homomorphic encryption technology supports some calculations on encrypted data, but not all models can be directly applied to encrypted data. Model design and optimization need to be carried out based on the characteristics of homomorphic encryption.

[0192] Step 4: Model training and validation:

[0193] Divide the model into training and testing sets, make sure the model performs well on the training set, and evaluate the performance on the validation set.

[0194] Securely train models on encrypted data, such as training model parameters in a homomorphic encryption environment or training based on encrypted features.

[0195] Step 5: Forecast of future service demand:

[0196] Use the trained model to predict the demand for hospital services in a specified time period in the future (such as the next week or month).

[0197] If the model can make predictions directly on encrypted data, then predictions can be made directly on the encrypted future appointment registration data; if not, then the data needs to be decrypted first and then predicted, while ensuring that the decryption process complies with privacy protection regulations.

[0198] Step 6: Decrypt and interpret the prediction results:

[0199] The encrypted results predicted by the model are legally and securely decrypted to obtain the plaintext prediction results.

[0200] Based on the decrypted forecast results, analyze the overall trend of future hospital service demand and the specific needs of each department.

[0201] It should be noted that by improving the encryption and structured operation of the first data and using machine learning models for analysis and mining, the prediction and evaluation functions of multiple hospital operations management can be realized, thereby helping hospitals to better allocate resources, improve service quality, and reduce operating costs. Specifically, these prediction models can bring the following beneficial effects:

[0202] By predicting the overall demand for hospital services within a specified time period in the future, the hospital can plan human resources, material preparation, etc. in advance to meet the needs of patients and avoid insufficient services or waste of resources.

[0203] Predicting the total expected consumption of drugs in the future can help hospitals optimize drug inventory management, reduce drug backlogs and expiration losses, and ensure the stability of drug supply.

[0204] By monitoring and analyzing patients' drug use and predicting whether they are at risk of drug abuse, doctors can adjust treatment plans in a timely manner and reduce the health risks caused by drug abuse.

[0205] Assessing the rationality of drugs prescribed by doctors will help standardize doctors' prescribing behavior, reduce irrational drug use, and improve medical quality and patient satisfaction.

[0206] Predicting whether a patient will cancel their current appointment helps hospitals manage appointments in advance, avoiding waste of medical resources and inconvenience to patients.

[0207] By comprehensively analyzing patients' appointment, check-in, medical treatment and other information, we can predict the hospital's material resource allocation plan, help hospitals optimize department layout and resource allocation, and improve operational efficiency and patient experience.

[0208] In addition, these prediction models are all processed based on encrypted data, ensuring patient privacy and data security. At the same time, by using technologies such as homomorphic encryption, the model can make predictions directly on encrypted data without decryption, further enhancing data security. In short, these prediction models provide hospitals with a scientific and efficient management tool that helps improve the management level and operational efficiency of hospitals.

[0209] S104, performing a second structural processing on the second data after the first analysis and mining combined with the first data;

[0210] It should be noted that the first structured processing includes performing a first structured processing on the first data by using the ID card number, and the second structured processing includes performing a second structured processing on the first data and the second data by using the ID card number as the first priority and the date of consultation as the second priority;

[0211] Furthermore, the data after the second structured processing is stored in the first cloud terminal to complete the smart hospital data management. If it is necessary to retrieve patient appointment registration data, medical process data, medical service data, drug management data and rehabilitation guidance data, the decryption operation corresponding to the encryption operation of different data is used to decrypt and retrieve the original data in the local server.

[0212] It should be noted that the specific implementation process of the first structured processing may include parsing and classifying the first data using the ID number so as to convert it into a structured format to facilitate subsequent data processing and analysis. Through the key information of the ID number, the basic information of the patient can be accurately identified and classified, providing strong data support for hospital management.

[0213] The second structured processing is more complex and sophisticated. In addition to using the ID number as the first priority for structured processing, the date of consultation is also introduced as the second priority. This means that when processing data, the system will first classify and organize the data according to the ID number, and then further subdivide the data according to the date of consultation under the same ID number. This processing method can more accurately reflect the patient's treatment situation and process in the hospital, providing more comprehensive and accurate data support for the hospital.

[0214] Furthermore, storing the data after the second structured processing in the first cloud terminal not only completes the management of the smart hospital data, but also greatly improves the security and reliability of the data. The application of cloud storage technology allows data to be backed up and restored in the cloud, effectively preventing the risk of data loss and damage. At the same time, cloud storage also provides powerful data sharing and collaboration functions, allowing various departments in the hospital to share and use data more conveniently.

[0215] When it is necessary to retrieve patient appointment registration data, medical process data, medical service data, drug management data, and rehabilitation guidance data, the system will perform corresponding decryption operations based on the encryption operations of different data. This process ensures the confidentiality and security of the data and prevents the risk of data leakage and abuse. The decrypted data can be directly retrieved from the local server to ensure the integrity and accuracy of the data.

[0216] In general, this intelligent hospital data management method has brought many beneficial effects. First, it has greatly improved the management efficiency and accuracy of hospital data, allowing hospitals to obtain and use data more conveniently. Secondly, the application of cloud storage technology has enhanced the security and reliability of data and effectively protected the privacy and rights of patients. Finally, this management method has also promoted data sharing and collaboration among various departments of the hospital, providing strong support for the overall operation and decision-making of the hospital.

[0217] S105, storing the second structured data in the first cloud terminal to complete the smart hospital data management.

[0218] In an optional embodiment, a data monitoring mechanism may also be established to monitor and analyze smart hospital data in real time to ensure data quality and accuracy.

[0219] The data monitoring mechanism is a key link in smart hospital data management. It can monitor and analyze data in real time to ensure the quality and accuracy of data. Specifically, the data monitoring mechanism can include the following aspects:

[0220] First, by monitoring the flow and changes of data in real time, the data monitoring mechanism can promptly detect data anomalies and errors. For example, if a sudden surge in appointment registration data is found within a certain period of time, this may mean that there is a loophole in the system or it has been maliciously attacked. At this time, the data monitoring mechanism can immediately issue an alarm to remind the administrator to deal with it in time.

[0221] Secondly, the data monitoring mechanism can also verify and proofread the accuracy of the data. For example, by comparing data from different sources, it can check whether there are any contradictions or inconsistencies between the data. If problems are found in the data, the data monitoring mechanism can automatically correct it or prompt the administrator to manually correct it to ensure the accuracy of the data.

[0222] Finally, the data monitoring mechanism can also perform trend analysis and prediction on the data. By analyzing and mining historical data, the patterns and trends of the data can be discovered, thereby predicting future data changes. This prediction can provide strong support for the hospital's decision-making and planning, helping the hospital better cope with future challenges and changes.

[0223] In short, establishing a data monitoring mechanism is a necessary part of smart hospital data management. It can ensure the quality and accuracy of data and improve the management level and operational efficiency of the hospital. At the same time, the data monitoring mechanism can also provide strong support for the hospital's decision-making and planning, and promote the hospital to achieve more scientific, efficient and sustainable development.

[0224] In an optional embodiment, under the data monitoring mechanism, a data quality assessment system is established to ensure the integrity and reliability of the data.

[0225] The data quality assessment system is an important part of the data monitoring mechanism. It can conduct a comprehensive and objective assessment of the quality of hospital data to ensure the integrity and reliability of the data. Specifically, the data quality assessment system can include the following aspects:

[0226] First, the data quality assessment system needs to clearly define the assessment standards and requirements for data quality. These standards and requirements should be based on the hospital's business needs and actual conditions, including but not limited to data accuracy, completeness, consistency, availability and security. At the same time, corresponding assessment indicators and assessment methods need to be formulated to quantitatively evaluate and analyze hospital data.

[0227] Secondly, the data quality assessment system needs to establish a data quality monitoring mechanism. Through regular or real-time inspection and analysis of hospital data, problems and anomalies in the data can be found and promptly processed and repaired. At the same time, a data quality reporting system needs to be established to regularly report the assessment results and improvement measures to the hospital management and relevant departments, so as to timely adjust and optimize the data management process.

[0228] Finally, the data quality assessment system needs to be continuously improved and optimized. As hospital business continues to develop and change, the data quality assessment system also needs to be continuously updated and adjusted to adapt to new business needs and data characteristics. At the same time, it is also necessary to strengthen data quality education and training, improve hospital staff's awareness and attention to data quality, and jointly promote the improvement of hospital data quality.

[0229] In short, establishing a data quality assessment system is a key link in smart hospital data management. It can comprehensively and objectively evaluate the quality of hospital data and ensure the integrity and reliability of data. At the same time, the data quality assessment system can also promote the continuous improvement and optimization of hospital data management, provide hospitals with better quality and more efficient data services, and promote the digital transformation and high-quality development of hospitals.

[0230] In summary, the present invention proposes a smart hospital data management method based on cloud computing, obtains patient appointment registration data, medical process data, medical service data, drug management data and rehabilitation guidance data in the smart hospital to be managed, and pre-processes the acquired data; performs improved encryption and first structured operation on the pre-processed first data, and uploads the first data after improved encryption and first structured operation to the first cloud terminal; the first cloud terminal performs a first analysis and mining on the first data after improved encryption and first structured operation according to the built-in first processing logic; the second data after the first analysis and mining is combined with the first data for second structured processing; the data after the second structured processing is stored in the first cloud terminal to complete the smart hospital data management. In summary, the smart hospital data management method and system based on cloud computing proposed by the present invention ensure the security and availability of data through improved encryption and structured processing; realizes the analysis and mining of hospital data through machine learning models and deep learning models; and improves the security and visualization of data through blockchain technology and data visualization tools. These advantages enable the present invention to better meet the needs of smart hospital data management and provide strong support for hospital managers.

[0231] In a preferred embodiment, the cloud computing-based intelligent hospital data management system includes:

[0232] The data acquisition and processing module is used to acquire the patient appointment registration data, medical process data, medical service data, drug management data and rehabilitation guidance data in the smart hospital to be managed, and pre-process the acquired data;

[0233] A first structuring module, used to perform improved encryption and a first structuring operation on the preprocessed first data, and upload the first data after the improved encryption and the first structuring operation to a first cloud terminal;

[0234] An analysis and mining module, used for the first cloud terminal to perform a first analysis and mining on the first data after the improved encryption and the first structuring operation according to a built-in first processing logic;

[0235] A second structuring module, used to combine the second data after the first analysis and mining with the first data to perform a second structuring process;

[0236] The management module is used to store the data after the second structured processing in the first cloud terminal to complete the smart hospital data management.

[0237] The above-mentioned unit modules may be embedded in or independent of a processor in a computer device in the form of hardware, or may be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to the above-mentioned modules.

[0238] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 2 As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a smart hospital data management method based on cloud computing is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a button, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0239] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0240] Obtain patient appointment registration data, medical process data, medical service data, drug management data, and rehabilitation guidance data in the smart hospital to be managed, and pre-process the acquired data;

[0241] Performing improved encryption and a first structuring operation on the preprocessed first data, and uploading the first data after the improved encryption and the first structuring operation to the first cloud terminal;

[0242] The first cloud terminal performs a first analysis and mining on the first data after the improved encryption and the first structuring operation according to the built-in first processing logic;

[0243] Combining the second data after the first analysis and mining with the first data to perform a second structural processing;

[0244] The data after the second structured processing is stored in the first cloud terminal to complete the smart hospital data management.

[0245] Example 2

[0246] Reference Figure 1-2 , which is an embodiment of the present invention, provides an intelligent hospital data management method and system based on cloud computing. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through comparative experiments.

[0247] Experiment 1: Data Security Verification

[0248] In order to verify the advantages of the present invention in terms of data security, a comparative experiment was designed. A hospital that has implemented traditional data management methods was selected as the control group, and a hospital that uses the method described in the present invention for data management was selected as the experimental group. During the experiment, data leakage incidents in the two groups of hospitals were recorded and analyzed.

[0249] The experimental results show that the control group hospital had multiple data leakage incidents during the experiment, including leakage of patient personal information and medical records. However, the experimental group hospital did not have any data leakage incidents during the experiment. This result shows that the method of the present invention can effectively improve the security of hospital data and reduce the risk of data leakage.

[0250] Experiment 2: Data Quality Assessment

[0251] In order to verify the effect of the present invention on data quality assessment, the data quality of the experimental group hospitals was evaluated. A variety of evaluation indicators were used, including data integrity, accuracy, consistency and timeliness. By comparing the data quality with that of the control group hospitals, it was found that the data quality of the experimental group hospitals was significantly higher.

[0252] The experimental results show that the data integrity of the experimental group hospitals reached 98%, while that of the control group hospitals was only 85%; the data accuracy of the experimental group hospitals reached 95%, while that of the control group hospitals was 80%. This result shows that the method of the present invention can effectively improve the quality of hospital data and provide more accurate and complete data support for hospitals.

[0253] Experiment 3: Data Management and Analysis Efficiency

[0254] In order to verify the advantages of the present invention in data management and analysis efficiency, the data processing time of the experimental group hospitals and the control group hospitals was compared. The same task volume was selected and the time required for the two groups of hospitals to complete the task was recorded.

[0255] The experimental results show that the experimental group hospital completed the data processing task in a significantly shorter time than the control group hospital. Specifically, the experimental group hospital completed 1,000 data processing tasks in 30 minutes, while the control group hospital took 45 minutes. This result shows that the method of the present invention can effectively improve the efficiency of hospital data management and analysis, saving a lot of time costs for the hospital.

[0256] In summary, through comparative experiments, the advantages of the present invention in data security, data quality assessment, and data management and analysis efficiency have been verified. These advantages enable the present invention to better meet the needs of smart hospital data management and provide hospitals with safer and more efficient data support. At the same time, the present invention also helps to promote the digital transformation and high-quality development of hospitals and contribute to the progress of the medical industry.

[0257] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

[0258] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0259] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0260] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0261] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0262] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0263] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A cloud computing-based intelligent hospital data management method, characterized in that: include: Obtain patient appointment registration data, medical process data, medical service data, drug management data, and rehabilitation guidance data in the smart hospital to be managed, and pre-process the acquired data; Performing improved encryption and a first structuring operation on the preprocessed first data, and uploading the first data after the improved encryption and the first structuring operation to the first cloud terminal; The first cloud terminal performs a first analysis and mining on the first data after the improved encryption and the first structuring operation according to the built-in first processing logic; Performing a second structural processing on the second data after the first analysis and mining combined with the first data; The data after the second structured processing is stored in the first cloud terminal to complete the smart hospital data management; The obtaining of patient appointment registration data, medical process data, medical service data, drug management data, and rehabilitation guidance data in the smart hospital to be managed, and preprocessing of the obtained data includes: The patient appointment registration data includes a unique number identifying the patient who has made an appointment, the appointment date of the corresponding patient, the unique number of the doctor who has seen the patient, the name of the appointment department of the corresponding patient, the specific appointment time of the corresponding patient, and the unique number identifying the patient who has canceled the appointment; The medical consultation process data includes the unique identification of the patient, the check-in time of the patient, the medical consultation process sequence of the patient, and the actual time spent on the medical consultation service for the patient; The medical service data includes the unique identifier of the medical treatment item of the patient, the doctor's diagnosis result of the patient, the doctor's treatment plan for the patient, and the detailed cost of the medical treatment; The drug management data includes the unique number of the drug prescribed by the doctor, the prescription number associated with the drug, the name of the drug, the given dosage of the drug, and the frequency of use of the drug; The rehabilitation guidance data includes a unique identifier of the patient's rehabilitation plan, a rehabilitation guidance text for the patient, and the start and end dates of the patient's rehabilitation plan; The pre-processing includes data unique indexing processing including the unique identification of the patient, the unique identification of the patient's treatment program, the unique number of the doctor who treated the patient, the unique number of the patient who canceled the appointment, the unique number of the drug prescribed by the doctor, the associated prescription number of the drug, and the unique identification of the patient's rehabilitation plan; The missing values ​​of the corresponding patient's appointment date, the corresponding patient's specific appointment time, the patient's check-in time, the actual time spent on the patient's medical service, and the start and end dates of the patient's rehabilitation plan are processed and then formatted in a unified manner; Perform word segmentation and named entity recognition on the name of the appointment department for the corresponding patient, the patient's consultation process sequence, the doctor's diagnosis of the patient, the doctor's treatment plan for the patient, the detailed cost of this diagnosis and treatment, the name of the drug, the given dosage of the drug, the frequency of use of the drug, and the patient's rehabilitation guidance text; The performing improved encryption and first structural operation on the preprocessed first data includes: The data processed by the unique index of the data is encrypted by an improved first algorithm, wherein the first algorithm includes firstly constructing the combined data and adding salt, which is expressed as follows: Among them, C initial represents the initial combination value, Pj represents the unique identifier of the patient, Tj represents the unique identifier of the patient's treatment item, Dj represents the unique number of the doctor who treated the patient, Rj represents the unique number of the drug prescribed by the doctor, Ej represents the associated prescription number of the drug, Cj represents the unique identifier of the patient's rehabilitation plan, S g Indicates the global salt value, represents the splicing operation, j represents the jth patient; Secondly, perform multiple iterative hashing on the data after combined salting: H1=H(C initial ,S g ) H2=H(H1,S g ||0) ... H final =H(H iterations-1 ,S g ||(iterations-1)) Where H represents the hash function, H final It represents the final hash value obtained after multiple rounds of iterations of the combined salted data. iterations-1 It represents the intermediate hash value before the last iteration of the combined salted data, iterations represents the number of hash iterations in the combined salted data, and || represents the string concatenation operation; Finally, generate the final index hash in combination with the context salt value: Among them, I represents the final index hash result, S c represents the client salt value, and P represents the message or data segment to be verified; The improvement encryption and the first structuring operation on the preprocessed first data further include: Perform second algorithm encryption on the data with missing values processed and then format unified. The second algorithm includes: First, record the appointment date of the corresponding patient as Asj, the specific appointment time of the corresponding patient as Ayj, the check-in time of this patient as Csj, the actual time consumed for the medical service of this patient as Ssj, and the start and end dates of the rehabilitation plan of this patient as Rsj, Rej, and uniformly convert all time data into Unix timestamp format; Secondly, perform differential privacy noise injection on the time data after unified timestamp format, which is expressed as follows: Where D[j] represents the appointment date or specific appointment time or check-in time or actual time consumed for the medical service or start and end dates of the rehabilitation plan of the j-th patient after differential privacy processing, S[j] represents the original appointment date or specific appointment time or check-in time or actual time consumed for the medical service or start and end dates of the rehabilitation plan of the corresponding j-th patient, L represents a function sampled from the Laplace distribution, Δf represents the maximum difference in the appointment date or specific appointment time or check-in time or actual time consumed for the medical service or start and end dates of the rehabilitation plan between two different patients, and ∈ represents the differential privacy budget. The smaller ∈ is, the higher the degree of privacy protection; Thirdly, perform hash encryption on the time data after differential privacy noise injection; Finally, perform improved Paillier homomorphic encryption on the time data after hash encryption. The improved Paillier homomorphic encryption is expressed as: Generate the Paillier public key n = Pq and g, where p and q represent two prime numbers, n represents the public modulus in the Paillier encryption system, that is, one of the public keys, and the public key is (n, g); Precompute g2 w The powers are stored in array G. w Where w is the window width, which is used to divide the number of bits in the binary representation in the fast exponential calculation. mod represents the modulo operation, i represents the number of window bits, i=0,1,2...,w-1; Convert the plaintext m into an integer, 0 ≤ m < n, and divide the binary representation of m into windows with a width of w bits; Initialize g m′ ←1 mod n 2 , m′ represents the exponential form of the plaintext m obtained after fast exponential calculation; For each window b of plaintext m w-1 ...b0, traverse the window from high to low, for each bit b in the window i , do the following: If b i = 0, then g m′ ←(g m′ ) 2 mod n 2 If b i =1, then g m′ ←(g m′ ) 2 ·g mod n 2 Randomly select an integer r (0 < r < n) and perform the encryption calculation process of the improved Paillier encryption. The encryption calculation process is expressed as: c=(g m′ ·r n )mod n 2 Among them, c represents the encrypted ciphertext, g m′ Represents a portion of the public key updated after the modulo operation; The improvement encryption and the first structuring operation on the preprocessed first data further include: The data after word segmentation and named entity recognition are encrypted by a third algorithm, and the third algorithm encryption includes firstly quantifying the name of the appointment department of the corresponding patient, the patient's medical process sequence, the doctor's diagnosis result of the patient, the doctor's treatment plan for the patient, the cost details of this diagnosis and treatment, the name of the drug, the given dose of the drug, the frequency of use of the drug, and the rehabilitation guidance text of the patient, and setting Dj={Slj, Plj, Clj, Tlj, Flj, Mlj, Doselj, Frequencylj, Rlj} represents the data set of the jth patient, wherein Slj represents the word vector representation of the appointment department name, Plj represents the encoding of the medical process sequence, Clj represents the word vector representation of the diagnosis result, Tlj represents the word vector representation of the treatment plan, Flj represents the numerical list of the medical cost details, Mlj represents the word vector representation of the drug name, Doselj represents the numerical value of the given dose of the drug, Frequencylj represents the numerical value of the frequency of use of the drug, and Rlj represents the word vector representation of the rehabilitation guidance text; Secondly, differential privacy noise is injected into the data quantized data set, as shown below: Noisy D =Dj+N(0,σ 2 I) Among them, Noisy D represents the data set after adding noise, Dj represents the original data set of the jth patient, N(0,σ 2 I) represents the multivariate normal distribution, σ 2 I represents the covariance matrix; Finally, the improved Paillier homomorphic encryption algorithm is used for encryption operation, which is expressed as follows: Encrypted D =E((n,g),Dj) =[E((n,g),Slj),E((n,g),Plj),E((n,g),Clj),...,E((n,g),Rlj)] Among them, E represents the improved Paillier homomorphic encryption algorithm, (n, g) represents the public key, and Encrypted D Represents the encrypted overall data set; The first cloud terminal performs a first analysis and mining on the first data after the improved encryption and the first structuring operation according to the built-in first processing logic, including: the first processing logic includes performing logic processing using a trained machine learning model; The trained machine learning models include a hospital service demand prediction model, a drug consumption trend prediction model, a drug abuse prediction model, an unreasonable drug prescribing prediction model, a frequent appointment cancellation prediction model, and a department resource allocation prediction model; The input of the hospital service demand prediction model is the encrypted appointment registration data of the previous cycle, and the output is the overall service demand of the hospital in a specified time period in the future; The input of the drug consumption trend prediction model is the drug management data after improved encryption and structured processing in the previous period, and the output is the estimated total consumption of drugs in a specified time period in the future; The input of the drug abuse prediction model is the encrypted and structured given dose of the drug in the previous cycle, the doctor's diagnosis of the patient, and the doctor's treatment plan for the patient, and the output is a classified prediction of whether the patient has a risk of drug abuse; The input of the unreasonable prescription prediction model is the encrypted diagnosis result of the doctor on the patient in the previous cycle, the unique number of the drug prescribed by the doctor for the patient's treatment plan, the associated prescription number of the drug, the name of the drug, the given dosage of the drug, and the frequency of use of the drug. The output is an evaluation of the rationality of the drug prescribed by the doctor. The input of the frequent appointment cancellation prediction model is the unique number identifying the patient who made an appointment in the previous period and the unique number identifying the patient who canceled the appointment, and the output is the prediction result of whether the patient will cancel the current appointment; The input of the department resource allocation prediction model is the unique number of the patient who made an appointment in the previous period, the unique number of the doctor who saw the patient, the name of the corresponding patient's appointment department, the unique number of the patient who canceled the appointment, the patient's check-in time, the actual time spent on the patient's medical service, the patient's rehabilitation guidance text, and the start and end dates of the patient's rehabilitation plan. The output is the hospital's material resource allocation plan.

2. The cloud computing-based smart hospital data management method according to claim 1, characterized in that: Also includes: The first structured processing includes performing a first structured processing on the first data by using the ID card number, and the second structured processing includes performing a second structured processing on the first data and the second data by using the ID card number as the first priority and the date of consultation as the second priority; The data after the second structured processing is stored in the first cloud terminal to complete the smart hospital data management. If it is necessary to retrieve patient appointment registration data, medical process data, medical service data, drug management data and rehabilitation guidance data, the data is decrypted and obtained according to the decryption operation corresponding to the encryption operation of different data.

3. A system applied to the cloud computing-based intelligent hospital data management method according to claim 1, characterized in that: include: The data acquisition and processing module is used to acquire the patient appointment registration data, medical process data, medical service data, drug management data and rehabilitation guidance data in the smart hospital to be managed, and pre-process the acquired data; A first structuring module, used to perform improved encryption and a first structuring operation on the preprocessed first data, and upload the first data after the improved encryption and the first structuring operation to a first cloud terminal; An analysis and mining module, used for the first cloud terminal to perform a first analysis and mining on the first data after the improved encryption and the first structuring operation according to a built-in first processing logic; A second structuring module, used to combine the second data after the first analysis and mining with the first data to perform a second structuring process; The management module is used to store the data after the second structured processing in the first cloud terminal to complete the smart hospital data management.

4. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 2 are implemented.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 2 are implemented.

Images