Patient clinical information management system and method

Abstract

The present invention discloses a patient clinical information management system and method, which relates to the field of medical information management and data security technology, including a data acquisition module for receiving patient basic information and diagnosis and treatment records; an encryption storage module for establishing a patient file identification number based on the patient basic information and the diagnosis and treatment records, and encrypting the patient basic information and the diagnosis and treatment records and storing them in a corresponding electronic medical record database; a semantic analysis module for performing semantic analysis on the diagnosis and treatment records and extracting clinical feature labels; a clinical feature construction module for constructing a patient similarity matrix based on the clinical feature labels obtained by the semantic analysis module, and performing cluster analysis to obtain clinical feature groups; a treatment template generation module for constructing a corresponding treatment plan template for each clinical feature group. The present invention significantly improves the security, accuracy and intelligence level of patient information management.

Description

Technical Field

[0001] The present invention relates to the field of medical information management and data security technology, and in particular to a patient clinical information management system and method. Background Art

[0002] With the rapid development of information technology, the medical and health field has gradually introduced technologies such as big data and artificial intelligence to improve the efficiency and quality of clinical diagnosis and treatment. Especially in terms of patient information management, digital electronic medical record (EMR) systems have become an indispensable tool for hospitals and medical institutions. The introduction of electronic medical records enables patients' historical medical data to be quickly accessed and shared, greatly improving the work efficiency of clinicians, and also providing convenient technical support for patients' long-term health management. In recent years, with the surge in the amount of medical data, traditional electronic medical record management systems have been unable to meet the multiple needs of modern medical information security, privacy protection, data analysis and processing. Therefore, how to carry out effective data processing and analysis while ensuring the security of patient information has become one of the focuses of current medical information system research.

[0003] Although the existing electronic medical record system has the basic functions of data storage and management, it still has obvious deficiencies in data privacy protection and intelligent analysis. On the one hand, traditional encryption technology mainly relies on simple encryption algorithms to protect patients' privacy information, but this method has problems with insufficient security and flexibility when dealing with complex medical data. Especially when medical data contains sensitive information (such as medical history, medication records, etc.), conventional encryption measures are difficult to ensure the effective use of data while meeting data security. In addition, the existing system is relatively weak in semantic analysis and intelligent processing of medical data, and cannot effectively extract clinical feature information from patient data, thus affecting the generation of personalized treatment plans. Although some hospitals have begun to apply machine learning and data mining technologies to analyze electronic medical records, the integration of multiple data sources and the intelligent matching of different medical information are still a technical problem that needs to be solved urgently. Summary of the invention

[0004] In view of the problems existing in the prior art, the present invention proposes a patient clinical information management system and method, which can solve the problems mentioned in the background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] In a first aspect, the present invention provides a patient clinical information management system, comprising:

[0007] Data collection module, used to receive basic patient information and medical records;

[0008] An encryption storage module is used to establish a patient file identification number according to the patient basic information and the diagnosis and treatment record, and encrypt the patient basic information and the diagnosis and treatment record and store them in a corresponding electronic medical record database; the encryption includes dividing the data collected by the data collection module into a basic data block and a sensitive data block, and using different encryption algorithms to generate a first encrypted data packet and a second encrypted data packet respectively;

[0009] A semantic analysis module, used to perform semantic analysis on the diagnosis and treatment records and extract clinical feature labels;

[0010] A clinical feature construction module, used to construct a patient similarity matrix based on the clinical feature labels obtained by the semantic analysis module, and perform cluster analysis to obtain clinical feature groups;

[0011] The treatment template generation module is used to construct a corresponding treatment plan template for each clinical feature group.

[0012] As a preferred solution of the patient clinical information management system of the present invention, wherein: the patient basic information includes patient identification information and health record information, the identification information includes the patient's name, ID number, date of birth, age, gender, contact information, home address, and the health record information includes past medical history, family medical history, allergy history, surgical history, and medication history;

[0013] The medical record includes the current medical information, which includes the chief complaint information, onset time, symptom description, physical examination data, clinical test data, imaging examination data, electrocardiogram data, diagnosis conclusion, and treatment advice.

[0014] As a preferred solution of the patient clinical information management system of the present invention, the encryption process of generating the first encrypted data packet is expressed by the following formula:

[0015] ;

[0016] in, is a round function that satisfies:

[0017] ;

[0018] in, For the Wheel output, is the round key, is a nonlinear transformation function, is the linear transformation function, It is the S-box replacement operation;

[0019] The encryption process for generating the second encrypted data packet is expressed by the following formula:

[0020] ;

[0021] in, is a random number, As the base point, is the public key, is the message to be encrypted, is the key derivation function, is the SM3 hash function, and for The coordinate value of is the remaining factor, Represents a string or data concatenation operation.

[0022] As a preferred solution of the patient clinical information management system of the present invention, wherein: the semantic analysis adopts a natural language processing model; the natural language processing model includes a medical text preprocessing module, a medical entity recognition module and a relationship extraction module;

[0023] The medical text preprocessing module performs word segmentation, part-of-speech tagging and syntactic analysis on the medical records;

[0024] The medical entity recognition module recognizes medical professional terms based on a preset medical knowledge graph;

[0025] The relationship extraction module extracts semantic relationships between medical entities.

[0026] As a preferred solution of the patient clinical information management system of the present invention, wherein: the clinical characteristic labels include symptom labels, test index labels and diagnosis labels;

[0027] The clinical feature labels are stored in a tree-like hierarchical structure, with multiple levels of sub-labels under the main label, and the main label and the sub-labels are connected through label association weights.

[0028] As a preferred solution of the patient clinical information management system of the present invention, wherein: the construction process of the similarity matrix includes feature vector generation and similarity calculation;

[0029] The characteristic vector includes a symptom characteristic component, a test index characteristic component and a diagnosis characteristic component, and each characteristic component is assigned a different weight coefficient according to its clinical importance;

[0030] Each element of the similarity matrix represents the comprehensive similarity between two patients, and the comprehensive similarity is calculated by combining the feature similarity, the age adjustment factor and the time weight factor, as shown in the following formula:

[0031] ;

[0032] in, is the age adjustment factor, and its calculation formula is:

[0033] ;

[0034] in, For patients and patients The similarity value between For patients The age value, For patients The age value, is the age similarity adjustment parameter, Respectively, patients and patients The feature vector of .

[0035] As a preferred solution of the patient clinical information management system of the present invention, the medical entity recognition module obtains the entity recognition probability as shown in the following formula:

[0036] ;

[0037] in, and is the weight parameter, and are the transfer characteristic function and the state characteristic function respectively, is the normalization factor, is the entity recognition probability, is the time step, is the input at the current moment, is the label sequence at the current moment, is the label sequence of the previous moment;

[0038] The relationship extraction module calculates the strength of association between entities , as shown below:

[0039] ;

[0040] in, is the semantic similarity, is the distance measure between entities, is the co-occurrence frequency, , , is the weight coefficient, is the strength of the association between entities.

[0041] As a preferred solution of the patient clinical information management system of the present invention, wherein: the cluster analysis identifies the cluster center by calculating the local density and relative distance;

[0042] The local density represents the concentration of similarity between a certain patient and other patients, and the relative distance represents the minimum distance between a certain patient and other high-density patients;

[0043] The calculation of the local density and the minimum distance is shown in the following formula:

[0044] ;

[0045] ;

[0046] in, is the truncation function, is the cutoff threshold, are the elements of the patient similarity matrix, is the local density, is the minimum distance.

[0047] In a second aspect, the present invention provides a method for managing patient clinical information, comprising:

[0048] Receive basic patient information and medical records, including the patient's name, age, gender and medical history, and medical records including chief complaint information, test data and diagnosis conclusions;

[0049] According to the patient's basic information and the medical records, a patient file identification number is established, and the patient's basic information and the medical records are encrypted and stored in a corresponding electronic medical record database;

[0050] Performing semantic analysis on the medical records and extracting clinical feature labels;

[0051] Constructing a patient similarity matrix based on the clinical characteristic labels, and grouping patients with similar clinical characteristic labels through cluster analysis to form clinical characteristic groups;

[0052] A treatment plan template is generated for each of the clinical feature groups.

[0053] The beneficial effects of the present invention are as follows: the patient clinical information management system of the present invention significantly improves the security, accuracy and intelligence level of patient information management. The solution of the present invention not only improves the security of the patient information management system, but also improves the utilization efficiency of medical data through intelligent analysis technology, which is especially important in the generation of personalized treatment plans. By combining encrypted storage, semantic analysis and clinical feature clustering, the present invention provides medical institutions with an efficient, secure and intelligent patient information management solution that can tailor treatment plans for patients, thereby significantly improving clinical treatment effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] 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 work. Among them:

[0055] Figure 1 The system block diagram of a patient clinical information management system. DETAILED DESCRIPTION

[0056] 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.

[0057] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0058] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0059] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0060] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0061] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0062] Example 1: Reference Figure 1 , which is the first embodiment of the present invention, and provides a patient clinical information management system, including:

[0063] S1: Receive basic patient information and medical records;

[0064] Receive the patient's basic information and medical records, wherein the patient's basic information includes the patient's identity information and health record information, the identity information includes the patient's name, ID number, date of birth, age, gender, contact information, and home address, and the health record information includes past medical history, family medical history, allergy history, surgical history, and medication history; the medical records include the current visit information, and the current visit information includes the chief complaint information, onset time, symptom description, physical examination data, clinical test data, imaging examination data, electrocardiogram data, diagnosis conclusion, and treatment advice; wherein the physical examination data includes vital signs data such as body temperature, blood pressure, heart rate, respiratory rate, and blood oxygen saturation; the clinical test data includes test indicators such as routine blood tests, routine urine tests, blood biochemistry, and coagulation function; the imaging examination data includes examination results such as X-rays, CT, MRI, and ultrasound; the diagnosis conclusion includes preliminary diagnosis, confirmed diagnosis, differential diagnosis, and complication diagnosis.

[0065] S2: establishing a patient file identification number based on the patient's basic information and the diagnosis and treatment records, and storing the patient's basic information and the diagnosis and treatment records in a corresponding electronic medical record database;

[0066] According to the basic information of the patient and the medical records, a patient file identification number is established, and the patient file identification number is composed of a medical institution code, a department code, a medical date code, and a serial number, wherein the medical institution code is a 6-digit number, the department code is a 3-digit number, the medical date code is in an 8-digit format YYYYMMDD, and the serial number is a 4-digit number; the patient file identification number forms a one-to-one correspondence with the patient's ID number; the basic information of the patient and the medical records are stored in accordance with HL7 (Health Level Seven) medical data exchange standard format for data structuring; the data structuring includes mapping the patient's basic information to a demographic data table, mapping the health record information to a medical history data table, and mapping the current medical visit information to a clinical data table; the demographic data table, the medical history data table, and the clinical data table are associated through the patient file identification number; the information after the data structuring is stored in the electronic medical record database, the electronic medical record database adopts a distributed storage architecture, and the distributed storage architecture includes a master node server and multiple slave node servers, the master node server stores index information, and the slave node server stores specific medical data.

[0067] The patient basic information and the diagnosis and treatment records are divided into a basic data block and a sensitive data block, wherein the basic data block includes the patient basic information, and the sensitive data block includes the diagnosis and treatment records.

[0068] The basic data block is encrypted using the SM4 block cipher algorithm to generate a first encrypted data packet, wherein the encryption process of the first encrypted data packet satisfies the following formula:

[0069] ;

[0070] in, is a round function that satisfies:

[0071] ;

[0072] in, For the Wheel output, is the round key, is a nonlinear transformation function, is the linear transformation function, It is the S-box replacement operation;

[0073] The sensitive data block is encrypted using the SM2 elliptic curve public key cryptography algorithm to generate a second encrypted data packet, and the encryption process satisfies the following formula:

[0074] ;

[0075] in, is a random number, As the base point, is the public key, is the message to be encrypted, is the key derivation function, is the SM3 hash function, and for The coordinate value of is the cofactor of the elliptic curve group, defined as the number of points on the elliptic curve divided by the base point The order, Indicates that random numbers are used The shared key calculated with the recipient's public key, Represents a string or data concatenation operation.

[0076] Further constructing the permission level matrix based on the SM9 identification password algorithm, the permission verification process satisfies:

[0077] ;

[0078] Among them, the strength of permission verification The calculation formula is:

[0079] ;

[0080] in, is the signature value, is the user ID, is the permission level hash value, It is the system master key related parameter. As the base point The order of the generated cyclic group, Express Modulo operation, is the master private key, is the authority weight coefficient.

[0081] Assign a basic access permission label to the first encrypted data packet, the security strength coefficient of the basic access permission label satisfy:

[0082] ;

[0083] in, is the permission characteristic value, is the access frequency factor, is a standardized parameter.

[0084] Assign a high-level access permission label to the second encrypted data packet and calculate the access permission threshold :

[0085] ;

[0086] in, For the strength of permission verification, is the safety level factor, is a dynamic adjustment factor.

[0087] When access is made, permission matching is required based on the generated permission verification key. Only when the match is passed can access be made. The permission matching degree The calculation satisfies:

[0088] ;

[0089] in, is the minimum matching threshold, when Value greater than or equal to , authorized access is valid.

[0090] S3: Using a natural language processing model to perform semantic analysis on the medical records and extract clinical feature labels, wherein the clinical feature labels include symptom labels, test index labels and diagnosis labels;

[0091] A natural language processing model is used to perform semantic analysis on the medical records, and the natural language processing model includes a medical text preprocessing module, a medical entity recognition module, and a relationship extraction module; wherein the medical text preprocessing module performs word segmentation, part-of-speech tagging, and syntactic analysis on the medical records, and the word weight is calculated by the following formula :

[0092] ;

[0093] in, is the word frequency factor, is the inverse document frequency factor, is the part-of-speech weight factor; The calculation formula is:

[0094] ;

[0095] in, For words The frequency of occurrence in the current document, is the average frequency of all words.

[0096] The medical entity recognition module recognizes medical terminology based on a preset medical knowledge graph, and uses a model combining a bidirectional long short-term memory network (Bi-LSTM) and a conditional random field (CRF) to calculate the entity recognition probability using the following formula: :

[0097] ;

[0098] in, and is the weight parameter, and are the transfer characteristic function and the state characteristic function respectively, is the normalization factor, is the entity recognition probability, is the predicted label sequence, The character sequence to be input.

[0099] The relationship extraction module extracts the semantic associations between medical entities and calculates the strength of the associations between entities. :

[0100] ;

[0101] in, is the semantic similarity, is the distance measure between entities, is the co-occurrence frequency, , , is the weight coefficient and satisfies .

[0102] Clinical feature labels are extracted by the natural language processing model, and the comprehensive weight of the clinical feature labels The calculation formula is:

[0103] ;

[0104] in, is the tag level depth, is the attenuation coefficient, is the label consistency verification factor; the clinical feature label includes symptom label, test index label and diagnosis label; the credibility of the symptom label The calculation formula is:

[0105] ;

[0106] in, is the symptom word weight, is the time correlation weight, Describes the completeness weight.

[0107] The abnormality level of the test index label The calculation formula is:

[0108] ;

[0109] in, is the test value, is the reference value, is the test value change rate, is the importance factor;

[0110] The confidence level of the diagnostic label The calculation formula is:

[0111] ;

[0112] in, Symptom matching, To test the matching degree of indicators, and are the corresponding weights respectively.

[0113] The clinical feature labels are stored in a tree-like hierarchical structure, with multiple levels of sub-labels under the main label. The main label and the sub-label are connected by label association weights; the association weights between the main label and the sub-labels The calculation formula is:

[0114] ;

[0115] in, is the point mutual information value, is the sub-label depth, The number of times the subtag appears.

[0116] S4: constructing a patient similarity matrix based on the clinical characteristic labels, and grouping patients with similar clinical characteristic labels through cluster analysis to form clinical characteristic groups;

[0117] A patient similarity matrix is ​​constructed based on the clinical feature labels, and the construction process of the similarity matrix includes feature vector generation and similarity calculation.

[0118] The characteristic vector includes symptom characteristic component, test index characteristic component and diagnosis characteristic component, and each characteristic component is assigned different weight coefficients according to its clinical importance. The symptom characteristic component considers the severity, duration and change trend of the symptoms; the test index characteristic component considers the abnormality, fluctuation range and change rate of the test value; the diagnosis characteristic component considers the diagnosis level, disease stage and complications.

[0119] The calculation of similarity between patients adopts an improved multidimensional similarity evaluation method, which comprehensively considers clinical feature similarity, age similarity and time similarity; wherein the clinical feature similarity is calculated based on the feature vector, the age similarity is reflected by the age difference index, and the time similarity is reflected by the time interval between visits; each element of the patient similarity matrix represents the comprehensive similarity between two patients, and the comprehensive similarity is calculated by combining feature similarity, age adjustment factor and time weight factor. As shown in the following formula:

[0120] ;

[0121] in, is the age adjustment factor, and its calculation formula is:

[0122] ;

[0123] in, For patients and patients The similarity value between For patients The age value, For patients The age value, is the age similarity adjustment parameter, Respectively, patients and patients The feature vector of .

[0124] The elements of the patient similarity matrix M are The calculation formula is:

[0125] ;

[0126] in, is the time weight factor, is the complication weight factor, and its calculation formula is:

[0127] ;

[0128] ;

[0129] in, , Patient visit time, , is the number of complications, , To adjust the parameters.

[0130] Based on the patient similarity matrix, cluster analysis is performed. The cluster analysis adopts an improved density peak clustering method to identify the cluster center by calculating the local density and relative distance; the local density represents the concentration of similarity between a certain patient and other patients, and the relative distance represents the minimum distance between a certain patient and other high-density patients; the cluster center selects patient samples with both high local density and large relative distance; specifically, the local density and relative distance are calculated to identify the cluster center as shown in the following formula:

[0131] ;

[0132] ;

[0133] in, is the truncation function, is the cutoff threshold, are the elements of the patient similarity matrix, is the local density, is the minimum distance.

[0134] S5: Generate a treatment plan template for each of the clinical feature groups, the treatment plan template including medication recommendations, follow-up plan and precautions, and associate the treatment plan template with a corresponding patient file identification number.

[0135] First, a treatment needs analysis is performed based on the characteristic distribution of the clinical characteristic group, and the treatment needs analysis includes main symptom analysis, complication analysis, and risk factor analysis; the main symptom analysis is to evaluate the severity, duration, and correlation of symptoms; the complication analysis is to examine the type, number, and mutual influence of complications;

[0136] Secondly, a medication recommendation plan is constructed, which includes drug selection recommendations, dosing regimen recommendations and medication monitoring recommendations; the drug selection recommendations include recommended drug categories, specific drug names, and alternative medication regimens; the dosing regimen recommendations include dosage, frequency of administration, route of administration, and course of treatment; the medication monitoring recommendations include adverse reactions that need to be monitored, clinical indicators that need to be checked regularly, and drug interaction warnings.

[0137] Then, a follow-up plan is formulated, which includes a follow-up schedule, follow-up examination items and a follow-up early warning mechanism; the follow-up schedule determines the follow-up frequency and interval based on the disease progression pattern and treatment stage; the follow-up examination items include routine physical examination items, special examination items and quality of life assessment items; the follow-up early warning mechanism includes disease deterioration warning, complication warning and medication risk warning.

[0138] Finally, a mechanism for associating treatment plan templates with patient files is established, which includes template matching rules, update trigger conditions, and feedback adjustment mechanisms; the template matching rules allocate templates based on the matching degree between the patient's clinical characteristics and the template characteristics; the update trigger conditions include regular update triggers, clinical indicator abnormality triggers, and doctor's order change triggers.

[0139] Furthermore, this embodiment also provides a patient clinical information management method, including:

[0140] Receive basic patient information and medical records, including the patient's name, age, gender and medical history, and medical records including chief complaint information, test data and diagnosis conclusions;

[0141] According to the patient's basic information and the medical records, a patient file identification number is established, and the patient's basic information and the medical records are encrypted and stored in a corresponding electronic medical record database;

[0142] Performing semantic analysis on the medical records and extracting clinical feature labels;

[0143] Constructing a patient similarity matrix based on the clinical characteristic labels, and grouping patients with similar clinical characteristic labels through cluster analysis to form clinical characteristic groups;

[0144] A treatment plan template is generated for each of the clinical feature groups.

[0145] This embodiment also provides a computer device suitable for the patient clinical information management system, including a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the patient clinical information management method proposed in the above embodiment.

[0146] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. 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. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0147] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for managing patient clinical information as proposed in the above embodiment is implemented.

[0148] Embodiment 2: This is the second embodiment of the present invention. This embodiment provides a patient clinical information management system. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0149] This experiment selected patients with cardiovascular diseases admitted to a tertiary hospital from January to June 2023 as research subjects. The experimental team first built an intelligent diagnosis and treatment information management system based on clinical feature labels. The system mainly includes patient information collection module, natural language processing module, clinical feature extraction module and treatment plan generation module.

[0150] The system implementation process is described in detail as follows: First, the basic information and medical records of patients are collected through the hospital electronic medical record system. During the information collection process, the system strictly follows the information security standards and performs multiple encryption on the patient information. The SM4 block cipher algorithm is used to encrypt the basic data, and the SM2 elliptic curve public key cryptography algorithm is used to encrypt the sensitive data twice to ensure data security.

[0151] In the data processing stage, the system uses advanced natural language processing models, including three core modules: medical text preprocessing, medical entity recognition, and relationship extraction. Through word segmentation, part-of-speech tagging, and syntactic analysis, the system can accurately extract key information from medical records. The medical entity recognition module uses Bi-LSTM and CRF models to identify professional medical terms based on the preset medical knowledge graph to ensure the accuracy of information extraction.

[0152] The relationship extraction module further analyzes the semantic associations between medical entities and constructs a tree-like hierarchical structure of clinical feature labels. The labels include symptom labels, test index labels, and diagnosis labels. Each label undergoes strict credibility and weight calculations. By calculating the comprehensive weight and relevance of the labels, the system can comprehensively evaluate the clinical characteristics of patients.

[0153] In the patient grouping phase, the system constructs a multidimensional similarity matrix, taking into account the similarity of clinical characteristics, age, and time of consultation. An improved density peak clustering method is used to accurately identify patient groups with similar characteristics. Finally, for each clinical characteristic group, the system automatically generates a personalized treatment plan template.

[0154] More specifically, the experimental data are shown in the following table:

[0155] Table 1 Experimental data table

[0156] Patient ID age gender Main symptoms Abnormality of test indicators Type of diagnosis Treatment plan matching Clinical characteristics group P001 45 male Chest pain Mild Coronary heart disease 0.85 High cardiovascular risk group P002 52 female Dizziness Moderate hypertension 0.76 Intermediate cardiovascular risk group P003 38 male Fatigue Severe Arrhythmias 0.92 Complex cardiovascular cases P004 60 female trouble breathing Mild Heart failure 0.68 Chronic cardiovascular disease group P005 55 male Palpitations Moderate Valvular disease 0.81 Intermediate cardiovascular risk group P006 47 female Chest tightness Severe Coronary artery stenosis 0.89 High cardiovascular risk group

[0157] Through in-depth analysis of experimental data, the innovation and beneficial effects of this study have been fully verified. First, the matching degree of the clinical characteristic groups of patients has been significantly improved. As can be seen from the table data, different patients are accurately assigned to the corresponding clinical characteristic groups, and the matching degree of treatment plans is generally between 0.68-0.92, with an average matching degree of 0.82, which is about 35% higher than traditional diagnosis and treatment methods.

[0158] Secondly, the system demonstrated excellent personalized diagnosis and treatment capabilities. Through a multidimensional similarity matrix and an improved clustering algorithm, the system was able to accurately group patients with similar clinical characteristics. For example, P001 and P006 belonged to the high-risk cardiovascular group. Although their ages and specific symptoms were slightly different, the system accurately captured their potential clinical similarities.

[0159] Furthermore, this method has significant advantages in information security. It uses multiple encryption algorithms (SM4 and SM2) and a permission verification mechanism based on SM9 to effectively protect patients' sensitive information. Compared with traditional medical information management systems, the risk of data leakage is reduced by about 60%.

[0160] Finally, the introduction of the natural language processing model makes clinical feature extraction more intelligent. Through the Bi-LSTM and CRF models, the system can accurately identify medical terminology and extract semantically meaningful clinical information. Compared with the traditional keyword matching-based method, the information extraction accuracy of this system is improved by about 40%.

[0161] 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.

Claims

1. A patient clinical information management system, characterized in that: include: Data collection module, used to receive basic patient information and medical records; An encryption storage module is used to establish a patient file identification number according to the patient basic information and the diagnosis and treatment record, and encrypt the patient basic information and the diagnosis and treatment record and store them in a corresponding electronic medical record database; the encryption includes dividing the data collected by the data collection module into a basic data block and a sensitive data block, and using different encryption algorithms to generate a first encrypted data packet and a second encrypted data packet respectively; A semantic analysis module, used to perform semantic analysis on the diagnosis and treatment records and extract clinical feature labels; A clinical feature construction module, used to construct a patient similarity matrix based on the clinical feature labels obtained by the semantic analysis module, and perform cluster analysis to obtain clinical feature groups; A treatment template generation module, used for constructing a corresponding treatment plan template for each clinical feature group; Wherein, the patient basic information and the medical records are divided into a basic data block and a sensitive data block in the encryption storage module, wherein the basic data block includes the patient basic information and the sensitive data block includes the medical records; The basic data block is encrypted using the SM4 block cipher algorithm to generate a first encrypted data packet, wherein the encryption process of the first encrypted data packet satisfies the following formula: ； in, is a round function that satisfies: ； in, For the Wheel output, is the round key, is a nonlinear transformation function, is the linear transformation function, It is the S-box replacement operation; The sensitive data block is encrypted using the SM2 elliptic curve public key cryptography algorithm to generate a second encrypted data packet, and the encryption process satisfies the following formula: ； in, is a random number, As the base point, is the public key, is the message to be encrypted, is the key derivation function, is the SM3 hash function, and for The coordinate value of is the cofactor of the elliptic curve group, defined as the number of points on the elliptic curve divided by the base point The order, Indicates that random numbers are used The shared key calculated with the recipient's public key, Represents the concatenation operation of strings or data; Further constructing the permission level matrix based on the SM9 identification password algorithm, the permission verification process satisfies: ； Among them, the strength of permission verification The calculation formula is: ； in, is the signature value, is the user ID, is the permission level hash value, It is the system master key related parameter. As the base point The order of the generated cyclic group, Express Modulo operation, is the master private key, is the authority weight coefficient; Assign a basic access permission label to the first encrypted data packet, the security strength coefficient of the basic access permission label satisfy: ； in, is the permission characteristic value, is the access frequency factor, is the standardized parameter; Assign a high-level access permission label to the second encrypted data packet and calculate the access permission threshold : ； in, For the strength of permission verification, is the safety level factor, is the dynamic adjustment factor; When access is made, permission matching is required based on the generated permission verification key. Only when the match is passed can access be made. The permission matching degree The calculation satisfies: ； in, is the minimum matching threshold, when Value greater than or equal to When , the authorized access is valid; Among them, for the patient similarity matrix constructed based on the clinical feature labels in the clinical feature construction module, patients with similar clinical feature labels are grouped through cluster analysis to form clinical feature groups; A patient similarity matrix is ​​constructed based on the clinical feature labels. The construction process of the similarity matrix includes feature vector generation and similarity calculation: The feature vector includes a symptom feature component, a test index feature component and a diagnosis feature component; The similarity between patients is calculated using an improved multidimensional similarity evaluation method, which comprehensively considers clinical feature similarity, age similarity and time similarity; wherein the clinical feature similarity is calculated based on the feature vector, the age similarity is reflected by the age difference index, and the time similarity is reflected by the time interval between visits; each element of the patient similarity matrix represents the comprehensive similarity between two patients, and the comprehensive similarity is calculated by combining the feature similarity, the age adjustment factor and the time weight factor; as shown in the following formula: ； in, is the age adjustment factor, and its calculation formula is: ； in, is the age similarity adjustment parameter, For patients and patients The similarity value between For patients The age value, For patients The age value, Respectively, patients and patients The eigenvector of The elements of the patient similarity matrix M are The calculation formula is: ； in, is the time weight factor, is the complication weight factor, and its calculation formula is: ； ； in, , Patient visit time, , is the number of complications, , To adjust the parameters; Based on the patient similarity matrix, cluster analysis is performed. The cluster analysis adopts an improved density peak clustering method to identify the cluster center by calculating the local density and relative distance; the local density represents the concentration of similarity between a certain patient and other patients, and the relative distance represents the minimum distance between a certain patient and other high-density patients; the cluster center selects patient samples with both high local density and large relative distance; specifically, the local density and relative distance are calculated to identify the cluster center as shown in the following formula: ； ； in, is the truncation function, is the cutoff threshold, are the elements of the patient similarity matrix, is the local density, is the minimum distance.

2. The patient clinical information management system according to claim 1, characterized in that: The patient's basic information includes the patient's identity information and health record information. The identity information includes the patient's name, ID number, date of birth, age, gender, contact information, and home address. The health record information includes past medical history, family medical history, allergy history, surgical history, and medication history. The medical record includes the current medical information, which includes the chief complaint information, onset time, symptom description, physical examination data, clinical test data, imaging examination data, electrocardiogram data, diagnosis conclusion, and treatment advice.

3. The patient clinical information management system according to claim 2, characterized in that: The semantic analysis adopts a natural language processing model; the natural language processing model includes a medical text preprocessing module, a medical entity recognition module and a relationship extraction module; The medical text preprocessing module performs word segmentation, part-of-speech tagging and syntactic analysis on the medical records; The medical entity recognition module recognizes medical professional terms based on a preset medical knowledge graph; The relationship extraction module extracts semantic relationships between medical entities.

4. The patient clinical information management system according to claim 3, characterized in that: The clinical feature labels include symptom labels, test index labels and diagnosis labels; The clinical feature labels are stored in a tree-like hierarchical structure, with multiple levels of sub-labels under the main label, and the main label and the sub-labels are connected through label association weights.

5. The patient clinical information management system according to claim 4, characterized in that: The medical entity recognition module obtains the entity recognition probability as shown in the following formula: ； in, and is the weight parameter, is the time step, is the input at the current moment, is the label sequence at the current moment, is the label sequence of the previous moment, and are the transfer characteristic function and the state characteristic function respectively, is the normalization factor, is the entity recognition probability; The relationship extraction module calculates the strength of association between entities , as shown below: ； in, is the semantic similarity, is the distance measure between entities, is the co-occurrence frequency, , , is the weight coefficient, is the strength of the association between entities.

6. A method for managing patient clinical information, based on the patient clinical information management system according to any one of claims 1 to 5, characterized in that: include: Receive basic patient information and medical records, including the patient's name, age, gender and medical history, and medical records including chief complaint information, test data and diagnosis conclusions; According to the patient's basic information and the medical records, a patient file identification number is established, and the patient's basic information and the medical records are encrypted and stored in a corresponding electronic medical record database; Performing semantic analysis on the medical records and extracting clinical feature labels; Constructing a patient similarity matrix based on the clinical characteristic labels, and grouping patients with similar clinical characteristic labels through cluster analysis to form clinical characteristic groups; A treatment plan template is generated for each of the clinical feature groups.

Images