Artificial intelligence-based antibacterial drug use purpose judgment method and automatic judgment system

Abstract

According to the artificial intelligence-based antibacterial drug administration purpose judgment method and the automatic judgment system disclosed by the invention, after the acquired medical data of the patient, the clinical application guide and literature of the antibacterial drug and the clinical application guide principle of the antibacterial drug are preprocessed, the historical medical data set of the patient is constructed; a historical medical data set of a patient is randomly divided into a training set, a verification set and a test set according to a set proportion for training, and a trained antibacterial drug use purpose judgment model is converted into an antibacterial drug use purpose judgment rule. After an antibacterial drug order making request is received, an antibacterial drug medication purpose judgment rule is called, a matched antibacterial drug medication purpose is generated according to collected antibacterial drug order making information and current medical data of a corresponding patient, and automatic judgment of the antibacterial drug medication purpose is achieved. And selecting a corresponding antibacterial drug management and control rule and path according to the matched medication purpose.

Description

Technical Field

[0001] The present invention relates to the technical field of medical data processing, and in particular to an artificial intelligence-based method and automatic judgment system for determining the medication purpose of an antibacterial drug. Background Art

[0002] At present, the purpose of clinical use of antibiotics can be divided into surgical preventive medication (including preoperative prevention, intraoperative prevention, postoperative prevention, etc.), non-surgical prevention, therapeutic medication and discharge medication. The rational use of antibiotics is very important for improving efficacy and safety and reducing the risk of drug resistance. In order to curb the development of bacterial resistance, it is necessary to carry out scientific management of the clinical application of antibiotics. The first step of the management process is to accurately determine the purpose of antibiotic use. The judgment result directly determines the subsequent management process.

[0003] However, the current purpose of antibiotic use is determined by the doctor's personal subjective judgment. This judgment process should be based on a comprehensive assessment of the patient's genetic background, underlying diseases, surgical details, immune status, complications, infection status, previous use of antibiotics, pathogen test results, inflammatory indicators, etc. However, due to the doctor's own insufficient ability to diagnose and treat infections and large differences in clinical experience, it is difficult to make a consistent judgment. In addition, in order to evade the supervision and control of the management department, it is also common for some doctors to deliberately take the wrong clinical pathway, such as choosing the purpose of preventive medication for the purpose of therapeutic medication to avoid microbial specimen sampling, changing the purpose of preventive medication to the purpose of therapeutic medication to extend the time of postoperative preventive medication or to use antibiotics that exceed the surgical / non-surgical prevention control. Incorrect use of antibiotics will directly lead to changes in the subsequent antibiotic management pathway, thereby losing reasonable supervision of doctors' antibiotic prescriptions, and then causing a decline in patient medical quality, excessive use of antibiotics, development of bacterial resistance, and waste of medical resources. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention discloses a method for determining the purpose of antibacterial drug use based on artificial intelligence, comprising the following steps: S1, pre-process the acquired patient medical data, clinical application guidelines and literature of antimicrobial drugs, and guiding principles for clinical application of antimicrobial drugs, and construct a patient historical medical data set, wherein the patient medical data includes the patient's basic information, disease diagnosis information, surgery application information, medical records, surgery records, etiology test results and other laboratory test results, historical antimicrobial drug use and treatment / prevention effects; S2, randomly dividing the patient historical medical data set into a training set, a validation set and a test set according to a set ratio, using the training set to train the initial antimicrobial drug use purpose judgment model, and then using the validation set to adjust parameters and optimize, and using the test set to evaluate and screen to obtain the trained antimicrobial drug use purpose judgment model, and convert it into an antimicrobial drug use purpose judgment rule; S3, after receiving the request for antibiotic prescription, calls the antibiotic purpose judgment rule, generates a matching antibiotic purpose based on the collected antibiotic prescription information and the current medical data of the corresponding patient, and selects corresponding control rules and paths based on the matched antibiotic purpose.

[0005] Preferably, step S2 also includes: before using the patient history medical data set to train the initial antimicrobial drug purpose judgment model, preprocessing the collected patient history medical data set, the preprocessing including but not limited to data cleaning, data labeling, missing value filling, numerical standardization and encoding of categorical data.

[0006] Preferably, the step S1 specifically includes: S11, obtain and record the patient's basic information B, which includes but is not limited to identity information, date of birth, gender, medical record number, admission date and discharge date: ; in, represents the basic information set of the i-th patient, J is the number of basic information attributes, is the jth basic information attribute, is the data source from which information is extracted, From the data source Functions to extract information, represents the set of all basic information extracted for the i-th patient; S12, obtaining the medical record M of the corresponding patient, wherein the medical record includes the patient's clinical symptoms, historical medical conditions, surgical information, diagnosed diseases, treatment process records and complication records: ; in, represents the weighted sum of the medical record information of the i-th patient, K is the total number of medical record entries, is the numerical representation of the kth medical record entry, is the weight of the medical record entry; S13, collecting the corresponding patient's etiology test results and other laboratory test results L, wherein the etiology test results and other laboratory test results include recording the type of pathogen detected, the detection time, and the antibiotic sensitivity level: ; ; ; in, is the comprehensive laboratory test results of the i-th patient, and are the weights of bacterial culture and drug sensitivity testing, It is the comprehensive result of bacterial culture. It is the comprehensive result of drug sensitivity test. and are the relevant pathogens and drug factors, is the time decay factor, It is the detection time; is the sensitivity coefficient of the mth factor for the jth specific test result, is the response value of the nth drug or chemical substance associated with drug sensitivity test k; S14, summarize the historical use of antibiotics and treatment / prevention effects D of the corresponding patient, wherein the historical use of antibiotics and treatment / prevention effects include drug name, dosage, duration of use, start and end dates, and corresponding treatment / prevention effects: ; in, is the weighted sum of historical drug use of the ith patient, is the code of the drug name, It's the dosage. The duration of medication use, , and is the adjustment coefficient, L is the drug list containing all the considered antimicrobial drugs, and l is a drug in the drug list; S15: Determine the integration weight of each patient based on the clinical application guidelines and literature of antimicrobial drugs and the guiding principles for the clinical application of antimicrobial drugs , which reflects the degree of compliance of the patient's historical medical process with the clinical application guidelines and literature of antimicrobial drugs, as well as the guiding principles for the clinical application of antimicrobial drugs; the data of all collected research samples are combined to construct a patient historical medical data set H: ; in is the integrated weight of the ith patient, N is the number of all patients, represents the basic information set of the i-th patient, represents the weighted sum of the medical record information of the i-th patient, is the result of the etiological test and other laboratory tests of the i-th patient, It is the weighted sum of the historical antibiotic use and treatment / prevention effects of the i-th patient.

[0007] Preferably, the antimicrobial drug use purpose judgment model adopts a multi-classification model formed by combining multiple target support vector machines, wherein step S2 specifically includes: S21, receiving a patient's historical medical data set H, setting an input variable set X of a machine learning algorithm for an antibiotic use purpose judgment model, which consists of each patient's basic information B, medical records M, etiology test results and other laboratory test results L, and historical antibiotic use and treatment / prevention effects D, and defining a target variable Y to represent the patient's antibiotic use purpose; S22, using training set data Train each target support vector machine by finding a maximum margin hyperplane , to achieve the goal of minimizing empirical risk and structural risk while maximizing the classification interval, expressed as: ; in is the eigenvector, is the category label, w is the weight vector, is the norm of vector w, b is the bias term, and m is the number of training data points; The constraint ensures that each sample in the training data is outside the margin of its category, and the objective function To maximize the classification interval; introduce the Lagrange multiplier Construct the Lagrangian function and dualize it: ; The dual problem has the form: ;in is the objective function of the dual problem, and are the eigenvectors of the ith and jth data points respectively, and are the category labels of the i-th and j-th data points, respectively. is the vector representation of the Lagrange multiplier, is the Lagrange multiplier associated with the ith data point and satisfies the Carlo-Kuhn-Tucker condition; after training, the optimal hyperplane is determined, which is the preliminary multi-classification model; S23, using the validation set data to adjust and optimize the parameters of the preliminary multi-classification model; using a grid search or random search method to cross-validate within a predetermined range to determine the best parameter combination; S24, using the test set data to evaluate the performance of the optimized multi-classification model, selecting the best model from each optimized model as the trained model, and converting the trained multi-classification model into a rule for judging the purpose of antimicrobial drug use.

[0008] Preferably, the step S22 further comprises: using a set kernel function to map the data to a higher dimensional space by introducing a slack variable and penalty parameters To control the cost of misclassification, the optimization problem becomes: ; After training, the optimal hyperplane is determined as the trained preliminary multi-classification model.

[0009] Preferably, the step S24 specifically includes: S241, using the rule extraction algorithm to parse the trained multi-classification model and formalize each condition and decision output into a rule for judging the purpose of antimicrobial medication; S242, each antimicrobial drug use purpose judgment rule R is formalized into a conditional expression: ; in It is a condition derived from each patient's basic information B, medical records M, etiology test results and other laboratory test results L, and historical antimicrobial drug use and treatment / prevention effects D. The purpose of the use of the antimicrobial drug in this article; S243, using the trained multi-classification model to analyze the real-time acquired new case data, updated clinical application guidelines and literature of antimicrobial drugs to update the rules for judging the purpose of antimicrobial drug use: ; Where R is the rule for judging the purpose of antimicrobial drug use before the update, It is the updated rule for judging the purpose of using antimicrobial drugs. is the learning rate, The observed purpose of medication use, It is the purpose of medication predicted by the current antimicrobial drug purpose judgment rule R; The purpose of antimicrobial drug use is determined by the rule R about the conditions The partial derivative of When the predicted results change The sensitivity of the data is , and n is the number of antimicrobial drug prescription items obtained in real time.

[0010] Preferably, the purpose of medication includes but is not limited to preoperative preventive medication, intraoperative preventive medication, postoperative preventive medication, non-surgical preventive medication, therapeutic medication and discharge medication.

[0011] The present invention also discloses an automatic judgment system for the purpose of antibiotic use, including: a data construction module, which is used to pre-process the acquired patient medical data, antibiotic clinical application guidelines and literature, and antibiotic clinical application guiding principles, and construct a patient historical medical data set, wherein the patient medical data includes the patient's basic information, disease diagnosis information, surgery application information, medical records, surgery records, etiology test results and other laboratory test results, historical antibiotic use and treatment / prevention effects. A model training module, which is used to randomly divide the patient historical medical data set into a training set, a validation set and a test set according to a set ratio, use the training set to train the initial antibiotic use purpose judgment model, then use the validation set to adjust parameters and optimize, use the test set to evaluate and screen, obtain the trained antibiotic use purpose judgment model, and convert it into an antibiotic use purpose judgment rule. The rule application module is used to call the antibiotic drug purpose judgment rule after receiving the antibiotic drug prescription request, generate a matching antibiotic drug purpose based on the collected antibiotic drug prescription information and the corresponding patient's current medical data, and select corresponding control rules and paths based on the matched antibiotic drug purpose.

[0012] The present invention also discloses a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any of the above methods when executing the computer program.

[0013] The present invention also discloses a computer program product, comprising a computer program / instruction, which implements the steps of any of the above methods when executed by a processor.

[0014] The invention discloses an artificial intelligence-based method and automatic judgment system for judging the purpose of antimicrobial drug use, which automatically judges the purpose of antimicrobial drug use through data processing and machine learning technology, overcomes the drawback of relying on the subjective judgment of doctors in traditional medical processes, and realizes intelligent and refined management of the use of antimicrobial drugs. By collecting the patient's basic information, disease diagnosis information, surgical application information, medical records, surgical records, etiology test results and other laboratory test results, historical antimicrobial drug use and treatment / prevention effects, as well as clinical application guidelines and literature for antimicrobial drugs, and clinical application guidelines for antimicrobial drugs, a detailed patient history medical data set is formed, and the initial machine learning model is trained to construct an accurate antimicrobial drug use purpose judgment model and convert it into an antimicrobial drug use purpose judgment rule. It provides the premise and basis for realizing the standardized control of surgical antimicrobial drug preventive use and the sampling control of microbial specimens before antimicrobial drug treatment. Thereby, after receiving the request for the prescription of antimicrobial drugs, the purpose of antimicrobial drug use can be automatically judged according to the collected antimicrobial drug prescription information and the current patient's medical data.

[0015] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings described herein are used to provide further understanding of the present invention and constitute a part of this application. The illustrative embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0017] Figure 1 This is a schematic diagram of the steps of a method for determining the purpose of using an antibacterial drug based on artificial intelligence according to an embodiment of the present invention.

[0018] Figure 2 This is a structural block diagram of an automatic determination system for the purpose of use of antibacterial drugs disclosed in another embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings of the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the described embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] Unless otherwise defined, the technical or scientific terms used herein shall have the common meanings understood by persons with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the patent application specification and claims of the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, words such as "one" or "a" do not indicate a quantity limitation, but indicate the existence of at least one.

[0021] The purposes of antimicrobial drugs in clinical use are diverse, and can be divided into surgical preventive drugs, non-surgical preventive drugs, therapeutic drugs, and discharge drugs. Surgical preventive drugs include preoperative preventive drugs, intraoperative preventive drugs, intraoperative additional drugs, postoperative preventive drugs, and postoperative extended preventive drugs. The rational use of antimicrobial drugs is very important for improving efficacy and safety and reducing the risk of drug resistance. Therefore, it is necessary to carry out scientific management of the clinical application of antimicrobial drugs, and the basis of the management process is to accurately judge the purpose of antimicrobial drug use. The judgment result directly determines the subsequent management process. However, the current determination of the purpose of antimicrobial drug use is a personal subjective judgment of the doctor. This judgment process should be based on a comprehensive evaluation of the patient's genetic background, underlying diseases, surgical details, immune status, complications, infection status, previous use of antimicrobial drugs, pathogen test results, inflammatory indicators, etc., but due to the doctor's own insufficient infection diagnosis and treatment capabilities and large differences in clinical experience, it is difficult to make a consistent judgment. In addition, it is common for some doctors to deliberately follow the wrong clinical pathways in order to evade supervision and control by administrative departments. For example, they choose preventive medication instead of therapeutic medication to avoid sampling of microbial specimens, change preventive medication to therapeutic medication to extend the postoperative prevention time, or use antimicrobial drugs that are beyond the control of surgical / non-surgical prevention.

[0022] Therefore, in order to overcome the above problems, this embodiment discloses a method for determining the purpose of antibacterial drug use based on artificial intelligence, such as Figure 1 As shown, the following steps are included.

[0023] Step S1, pre-processing the acquired patient medical data, clinical application guidelines and literature of antibiotics, and guiding principles for clinical application of antibiotics to construct a patient historical medical data set, wherein the patient medical data includes the patient's basic information, disease diagnosis information, surgery application information, medical records, surgery records, etiology test results and other laboratory test results, historical antibiotic use and treatment / prevention effects.

[0024] Specifically, in the hospital working environment, doctors face a large number of patients who need to use antibiotics every day. In the process of handling cases, doctors need to minimize the irrational use of antibiotics while ensuring the efficacy to avoid the development of drug resistance. In addition, it is necessary to consider the specific conditions of the patient, such as surgical information, infection type, pathogen test results and drug sensitivity test results, details of antibiotic use, clinical treatment response, adverse reactions, etc., to make an individualized judgment on the purpose of medication. The method for judging the purpose of antibiotic use disclosed in this embodiment can automatically collect detailed historical medical data of patients. After data cleaning and screening based on the clinical application guidelines and literature of antibiotics, as well as the guiding principles for the clinical application of antibiotics, these data are used to train machine learning models and generate accurate rules for judging the purpose of medication, which provides powerful assistance for doctors when prescribing antibiotics, not only reducing the probability of unreasonable decisions by doctors, ensuring the personalization and optimization of treatment, but also promoting the rational allocation of medical resources.

[0025] In this embodiment, step S1 may specifically include the following contents.

[0026] Step S11, obtain and record the patient's basic information B, which includes but is not limited to identity information, date of birth, gender, medical record number, admission date and discharge date: ; in, represents the basic information set of the i-th patient, J is the number of basic information attributes, is the jth basic information attribute, is the data source from which information is extracted, From the data source Functions to extract information, Represents the set of all basic information extracted for the i-th patient.

[0027] Step S12, obtaining the medical record M of the corresponding patient, wherein the medical record includes the patient's clinical symptoms, historical medical conditions, surgical information, diagnosed diseases, treatment process records and complication records: ; in, represents the weighted sum of the medical record information of the i-th patient, is the numerical representation of the kth medical record entry, is the weight of the medical record entry.

[0028] Step S13, collecting the corresponding patient's etiology test results and other laboratory test results L, wherein the etiology test results and other laboratory test results include recording the type of pathogen detected, the detection time and the antibiotic sensitivity level: ; ; ; in, is the comprehensive laboratory test results of the i-th patient, and are the weights of bacterial culture and drug sensitivity testing, It is the comprehensive result of bacterial culture. It is the comprehensive result of drug sensitivity test. and are the relevant pathogens and drug factors, is the time decay factor, It is the detection time; is the sensitivity coefficient of the mth factor for the jth specific test result, is the response value of the nth drug or chemical substance associated with drug sensitivity test k.

[0029] Step S14, summarize the historical use of antibiotics and treatment / prevention effects D of the corresponding patient, wherein the historical use of antibiotics and treatment / prevention effects include drug name, dosage, duration of use, start and end dates, and corresponding treatment / prevention effects: ; in, is the weighted sum of historical drug use of the ith patient, is the code of the drug name, It's the dosage. The duration of medication use, , and is the adjustment coefficient, L is the drug list containing all the considered antimicrobial drugs, and l is a drug in the drug list.

[0030] Step S15, determine the integration weight of each patient according to the clinical application guidelines and literature of antibiotics, and the guiding principles for the clinical application of antibiotics. This coefficient reflects the degree of compliance of the patient's historical medical process with the clinical application guidelines and literature of antibiotics, and the guiding principles for the clinical application of antibiotics; merge the data of all collected research samples to construct a patient historical medical data set H: ; in is the integrated weight of the ith patient, N is the number of all patients, represents the basic information set of the i-th patient, represents the weighted sum of the medical record information of the i-th patient, is the result of the etiological test and other laboratory tests of the i-th patient, It is the weighted sum of the historical antibiotic use and treatment / prevention effects of the i-th patient.

[0031] Step S2, randomly divide the patient's historical medical data set into a training set, a validation set and a test set according to a set ratio, use the training set to train the initial antibiotic medication purpose judgment model, then use the validation set to adjust parameters and optimize, use the test set for evaluation and screening to obtain the trained antibiotic medication purpose judgment model, and convert it into an antibiotic medication purpose judgment rule.

[0032] The purpose of medication includes but is not limited to preoperative preventive medication, intraoperative preventive medication, postoperative preventive medication, non-surgical preventive medication, therapeutic medication and discharge medication. Surgical preventive medication can include preoperative preventive medication, intraoperative preventive medication, additional intraoperative medication, postoperative preventive medication, and extended postoperative preventive medication.

[0033] Furthermore, to ensure the quality and effectiveness of the training data, the collected patient history medical data set is preprocessed before using the patient history medical data set to train the initial antimicrobial drug use purpose judgment model. The preprocessing includes but is not limited to data cleaning, data labeling, missing value filling, numerical standardization, and encoding of categorical data. The data preprocessing process can significantly improve the quality and availability of the patient history medical data set, and provide strong support for the subsequent training and selection of the antimicrobial drug use purpose judgment model. In the specific implementation process, this embodiment uses medical data of more than 50,000 patients. The data is preprocessed by using advanced data preprocessing technology and machine learning algorithms, which further improves the accuracy of the constructed antimicrobial drug use purpose judgment model while ensuring improved data processing efficiency.

[0034] Specifically, the patient history medical data set can be randomly divided into a training set, a validation set, and a test set in a ratio of 6:2:2. After the initial antimicrobial drug use purpose judgment model is trained using the training set, the validation set is used for parameter adjustment and optimization, and finally the optimized model is evaluated using the test set, and the model with the best comprehensive performance is selected and converted into a drug use purpose judgment rule.

[0035] The evaluation criteria include but are not limited to accuracy, precision, recall, and harmonic mean F value. and the area under the ROC curve AUC. The machine learning algorithms used for the antimicrobial drug use purpose judgment model include but are not limited to decision tree algorithm, logistic regression algorithm, target support vector machine algorithm, naive Bayes algorithm, K-nearest neighbor algorithm, neural network algorithm, Bayesian network algorithm, linear discriminant analysis algorithm, maximum entropy model, random forest algorithm, adaptive boosting algorithm, gradient boosting decision tree algorithm GBDT, extreme gradient boosting algorithm XGBoost, etc.

[0036] In this embodiment, the antimicrobial drug purpose judgment model adopts a multi-classification model formed by a combination of multiple target support vector machines. Among them, the input layer of the multi-classification model is composed of multiple target support vector machine models SVM, which are the same as the number of medication purpose categories to be identified, and each target support vector machine model is used to identify a corresponding category of medication purpose. The output layer of the multi-classification model receives the predicted probability of the corresponding medication purpose category output by each target support vector machine model, and selects the medication purpose category with the highest predicted probability as the final output classification result. The training of the multi-classification model can include the following contents.

[0037] Step S21, receiving the patient's historical medical data set H, setting the input variable set X of the machine learning algorithm of the antibiotic medication purpose judgment model, which consists of each patient's basic information B, medical records M, etiology test results and other laboratory test results L, and historical antibiotic use and treatment / prevention effects D, and defining the target variable Y to represent the patient's antibiotic medication purpose.

[0038] Step S22, using training set data Train each target support vector machine model; by finding a maximum margin hyperplane , to achieve the goal of minimizing empirical risk and structural risk while maximizing the classification interval, expressed as: ; in is the eigenvector, is the category label, w is the weight vector, is the norm of vector w, b is the bias term, and m is the number of training data points; The constraint ensures that each sample in the training data is outside the margin of its category, and the objective function To maximize the classification interval; introduce the Lagrange multiplier Construct the Lagrangian function and dualize it: ; The dual problem has the form: ;in is the objective function of the dual problem, and are the eigenvectors of the ith and jth data points respectively, and are the category labels of the i-th and j-th data points, respectively. is the vector representation of the Lagrange multiplier, is the Lagrange multiplier associated with the ith data point and satisfies the Carlo-Kuhn-Tucker condition, and >0; after training, the optimal hyperplane is determined, which is the preliminary multi-classification model.

[0039] In this embodiment, step S22 may also include: using a set kernel function to map the data to a higher dimensional space, by introducing a slack variable and penalty parameters To control the cost of misclassification, the optimization problem becomes: ; After training, the optimal hyperplane is determined as the trained preliminary multi-classification model.

[0040] Specifically, in other embodiments, the kernel functions used may include but are not limited to linear kernel LinearKernel, polynomial kernel Polynomial Kernel, radial basis kernel function Radial Basis Function, exponential kernel Exponential Kernel, Laplacian Kernel, ANOVA kernel, quadratic rational kernel RationalQuadratic Kernel, Multiquadric Kernel, inverse Multiquadric Kernel, etc.

[0041] Step S23, use the validation set data to adjust and optimize the preliminary multi-classification model, and use a grid search or random search method to cross-validate within a predetermined range to determine the best parameter combination.

[0042] Specifically, the adjusted parameters include but are not limited to the penalty coefficient C of the objective function, the selection of the kernel function kernel, the highest power degree of the polynomial kernel function, the coefficient gamma of the kernel function, the independent term coef0 in the kernel function, whether to use the probability estimate probability, whether to perform heuristic shrinking, the accuracy of the support vector machine model termination standard, the memory required for training, the weights of each sample category, the maximum number of iterations, the classification decision, the seed of the pseudo-random number generator when rearranging data for probability estimation, etc.

[0043] Step S24, using the test set data to evaluate the performance of the optimized multi-classification model, selecting the best model from each optimized model as the trained model, and converting the trained multi-classification model into a rule for judging the purpose of antimicrobial drug use.

[0044] Specifically, the evaluation methods used include but are not limited to performance indicators such as accuracy, precision, recall, harmonic mean F value, and area under the ROC curve. The model is fine-tuned and the above steps are repeated until satisfactory performance is achieved.

[0045] In this embodiment, step S24 specifically includes the following contents.

[0046] Step S241, using a rule extraction algorithm to parse the trained multi-classification model, and formalizing each condition and decision output into a rule for judging the purpose of antimicrobial drug use.

[0047] Step S242, formalize each antimicrobial drug use purpose judgment rule R into a conditional expression: ;in It is a condition derived from each patient's basic information B, medical records M, etiology test results and other laboratory test results L, and historical antimicrobial drug use and treatment / prevention effects D. This is the purpose of using this antibacterial drug.

[0048] Step S243, using the trained multi-classification model to analyze the new case data acquired in real time, the updated clinical application guidelines and literature of antimicrobial drugs to update the antimicrobial drug use purpose judgment rules: ; Where R is the rule for judging the purpose of antimicrobial drug use before the update, It is the updated rule for judging the purpose of using antimicrobial drugs. is the learning rate, The observed purpose of medication use, It is the purpose of medication predicted by the current antimicrobial drug purpose judgment rule R; The purpose of antimicrobial drug use is determined by the rule R about the conditions The partial derivative of When the predicted results change The sensitivity of the data is , and n is the number of antimicrobial drug prescription items obtained in real time.

[0049] Specifically, by introducing machine learning algorithms for dynamic adjustment, the adaptability and individualization of the rules can be further improved. By building a rule engine and integrating the rule engine into the hospital information system, the real-time updated rules for judging the purpose of antibiotic use can automatically judge the purpose of antibiotic use in the electronic medical record system and enter the corresponding management path. The electronic system can automatically judge the purpose of the antibiotic use based on the patient's current health status and historical medical data, and activate the corresponding management path for targeted management.

[0050] Step S3, after receiving the request for antibiotic prescription, call the antibiotic purpose judgment rule, generate a matching antibiotic purpose based on the collected antibiotic prescription information and the current medical data of the corresponding patient, and select corresponding control rules and paths based on the matched antibiotic purpose.

[0051] In a specific embodiment, a hospital information system that is deployed with the above-mentioned artificial intelligence-based method for determining the purpose of antibiotic use can automatically collect detailed historical medical data of patients, including basic information, disease diagnosis, surgical application, medical records, surgical records, laboratory test results such as bacterial culture and drug sensitivity test results, and historical antibiotic use. After undergoing a preprocessing process, these data can be used to train an antibiotic use purpose judgment model to generate accurate antibiotic use purpose judgment rules. When a doctor needs to prescribe antibiotics, the system relies on an integrated rule engine to quickly determine the purpose of drug use and automatically match applicable management rules and processes, effectively eliminating the interference of subjective judgment and simplifying the process of clinicians prescribing antibiotics.

[0052] In the first six months of the system implementation, medication data before and after implementation were collected and analyzed. The results showed that before the system was deployed, only 33.45% of the hospital's doctors collected pathogen specimens before antimicrobial treatment; after the system was implemented, the proportion jumped to 81.81%. In the first year of the system implementation, the number of patients who were treated with antimicrobial drugs without infection indications in the hospital decreased by 82.34%. At the same time, the number of patients who were treated with antimicrobial drugs for surgery or non-surgical prevention without prevention indications also decreased by 79.98%. Due to the significant increase in the rate of etiological inspection, there has been an increase in clinical cases of precise antimicrobial treatment for specific pathogens, and the unnecessary use of high-level, broad-spectrum antimicrobial drugs has been reduced. In addition, surgical and non-surgical preventive medications have become more standardized, and the hospital has also observed a reduction in the average length of stay of patients, which has directly promoted a more efficient allocation of medical resources and a reduction in overall medical costs.

[0053] The method for judging the purpose of antimicrobial drug use based on artificial intelligence disclosed in the above embodiment automatically judges the purpose of antimicrobial drug use through data processing and machine learning technology, overcomes the drawbacks of relying on the subjective judgment of doctors in traditional medical processes, and realizes intelligent and refined management of the use of antimicrobial drugs. By collecting the patient's basic information, disease diagnosis information, surgery application information, medical records, surgery records, etiology test results and other laboratory test results, historical antimicrobial drug use and treatment / prevention effects, as well as antimicrobial drug clinical application guidelines and literature, and antimicrobial drug clinical application guidelines, a detailed patient historical medical data set is formed, the initial machine learning model is trained, and an accurate antimicrobial drug use purpose judgment model is constructed and converted into antimicrobial drug use purpose judgment rules. Provide the premise and basis for realizing the standardized control of surgical antimicrobial drug preventive use and the sampling control of microbial specimens before antimicrobial drug treatment. Thereby, after receiving the request for the prescription of antimicrobial drugs, the purpose of antimicrobial drug use can be automatically judged based on the collected antimicrobial drug prescription information and the current patient's medical data.

[0054] In another embodiment, an automatic determination system for the purpose of using an antibacterial drug is also disclosed. Figure 2 As shown, it includes a data construction module 1, a model training module 2 and a rule application module 3. The data construction module 1. The data construction module 1 is used to pre-process the acquired patient medical data, clinical application guidelines and literature of antibiotics, and clinical application guidelines of antibiotics to construct a patient historical medical data set, wherein the patient medical data includes the patient's basic information, disease diagnosis information, surgery application information, medical records, surgery records, etiology test results and other laboratory test results, historical antibiotic use and treatment / prevention effects. The model training module 2 is used to randomly divide the patient's historical medical data set into a training set, a validation set and a test set according to a set ratio, use the training set to train the initial antibiotic use purpose judgment model, then use the validation set to adjust parameters and optimize, use the test set to evaluate and screen, obtain the trained antibiotic use purpose judgment model, and convert it into an antibiotic use purpose judgment rule. The rule application module 3, after receiving the request for the prescription of antibiotics, calls the antibiotic purpose judgment rule, generates a matching antibiotic purpose based on the collected antibiotic prescription information and the current medical data of the corresponding patient, and selects corresponding control rules and paths based on the matching antibiotic purpose.

[0055] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Similar parts between the various embodiments can be referred to each other. For the antibacterial drug use purpose judgment system based on artificial intelligence disclosed in the embodiment, since it corresponds to the antibacterial drug use purpose judgment method based on artificial intelligence disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the aforementioned method part description.

[0056] In other embodiments, a computer device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method for determining the purpose of antimicrobial drug use based on artificial intelligence as described in the above embodiments when executing the computer program. The server may include, but is not limited to, a processor and a memory. It will be understood by those skilled in the art that the schematic diagram is merely an example of a server and does not constitute a limitation on the server, and may include more or fewer components than shown in the figure, or a combination of certain components, or different components.

[0057] If the artificial intelligence-based antibacterial drug use purpose judgment system is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned embodiments of the antibacterial drug use purpose judgment method based on artificial intelligence can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device that can carry the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory, random access memory, electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0058] Finally, 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 aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

[0059] In conclusion, the above is only a preferred embodiment of the present invention, and all equivalent changes and modifications made according to the scope of the patent application of the present invention should fall within the scope of the patent of the present invention.

Claims

1. A method for determining the purpose of antibacterial drug use based on artificial intelligence, characterized in that: The steps include: S1, pre-process the acquired patient medical data, clinical application guidelines and literature of antimicrobial drugs, and guiding principles for clinical application of antimicrobial drugs, and construct a patient historical medical data set, wherein the patient medical data includes the patient's basic information, disease diagnosis information, surgery application information, medical records, surgery records, etiology test results and other laboratory test results, historical antimicrobial drug use and treatment / prevention effects; S2, randomly dividing the patient historical medical data set into a training set, a validation set and a test set according to a set ratio, using the training set to train the initial antimicrobial drug use purpose judgment model, and then using the validation set to adjust parameters and optimize, and using the test set to evaluate and screen to obtain the trained antimicrobial drug use purpose judgment model, and convert it into an antimicrobial drug use purpose judgment rule; S3, after receiving the request for antibiotic prescription, call the antibiotic purpose judgment rule, generate a matching antibiotic purpose based on the collected antibiotic prescription information and the current medical data of the corresponding patient, and select corresponding control rules and paths based on the matched antibiotic purpose.

2. The method for determining the purpose of antibacterial drug use based on artificial intelligence according to claim 1, characterized in that: The step S2 further comprises: Before using the patient history medical data set to train the initial antimicrobial drug use purpose judgment model, the collected patient history medical data set is preprocessed, and the preprocessing includes but is not limited to data cleaning, data labeling, missing value filling, numerical standardization and encoding of categorical data.

3. The method for determining the purpose of antibacterial drug use based on artificial intelligence according to claim 2, characterized in that: The step S1 specifically includes: S11, obtain and record the patient's basic information B, which includes but is not limited to identity information, date of birth, gender, medical record number, admission date and discharge date: ； in, represents the basic information set of the i-th patient, J is the number of basic information attributes, is the jth basic information attribute, is the data source from which information is extracted, From the data source Functions to extract information, represents the set of all basic information extracted for the i-th patient; S12, obtaining the medical record M of the corresponding patient, wherein the medical record includes the patient's clinical symptoms, historical medical conditions, surgical information, diagnosed diseases, treatment process records and complication records: ； in, represents the weighted sum of the medical record information of the i-th patient, K is the total number of medical record entries, is the numerical representation of the kth medical record entry, is the weight of the medical record entry; S13, collecting the corresponding patient's etiology test results and other laboratory test results L, wherein the etiology test results and other laboratory test results include recording the type of pathogen detected, the detection time, and the antibiotic sensitivity level: ； ； ； in, is the comprehensive laboratory test results of the i-th patient, and are the weights of bacterial culture and drug sensitivity testing, It is the comprehensive result of bacterial culture. It is the comprehensive result of drug sensitivity test. and are the relevant pathogens and drug factors, is the time decay factor, It is the detection time; is the sensitivity coefficient of the mth factor for the jth specific test result, is the response value of the nth drug or chemical substance associated with drug sensitivity test k; S14, summarize the historical use of antibiotics and treatment / prevention effects D of the corresponding patient, wherein the historical use of antibiotics and treatment / prevention effects include drug name, dosage, duration of use, start and end dates, and corresponding treatment / prevention effects: ； in, is the weighted sum of historical drug use of the ith patient, is the code of the drug name, It's the dosage. The duration of medication use, , and is the adjustment coefficient, L is the drug list containing all the considered antimicrobial drugs, and l is a drug in the drug list; S15: Determine the integration weight of each patient based on the clinical application guidelines and literature of antimicrobial drugs and the guiding principles for the clinical application of antimicrobial drugs , which reflects the degree of compliance of the patient's historical medical process with the clinical application guidelines and literature of antimicrobial drugs, as well as the guiding principles for the clinical application of antimicrobial drugs; the data of all collected research samples are combined to construct a patient historical medical data set H: ； in is the integrated weight of the ith patient, N is the number of all patients, represents the basic information set of the i-th patient, represents the weighted sum of the medical record information of the i-th patient, is the result of the etiological test and other laboratory tests of the i-th patient, It is the weighted sum of the historical antibiotic use and treatment / prevention effects of the i-th patient.

4. The method for determining the purpose of antibacterial drug use based on artificial intelligence according to claim 3, characterized in that: The antimicrobial drug use purpose judgment model adopts a multi-classification model formed by combining multiple target support vector machines, wherein step S2 specifically includes: S21, receiving a patient's historical medical data set H, setting an input variable set X of a machine learning algorithm for an antibiotic use purpose judgment model, which consists of each patient's basic information B, medical records M, etiology test results and other laboratory test results L, and historical antibiotic use and treatment / prevention effects D, and defining a target variable Y to represent the patient's antibiotic use purpose; S22, using training set data Train each target support vector machine by finding a maximum margin hyperplane , to achieve the goal of minimizing empirical risk and structural risk while maximizing the classification interval, expressed as: ； in is the eigenvector, is the category label, w is the weight vector, is the norm of vector w, b is the bias term, and m is the number of training data points; The constraint ensures that each sample in the training data is outside the margin of its category, and the objective function To maximize the classification interval; introduce the Lagrange multiplier Construct the Lagrangian function and dualize it: ； The dual problem has the form: ;in is the objective function of the dual problem, and are the eigenvectors of the ith and jth data points respectively, and are the category labels of the i-th and j-th data points, respectively. is the vector representation of the Lagrange multiplier, is the Lagrange multiplier associated with the ith data point and satisfies the Carlo-Kuhn-Tucker condition; after training, the optimal hyperplane is determined, which is the preliminary multi-classification model; S23, using the validation set data to adjust and optimize the parameters of the preliminary multi-classification model; using a grid search or random search method to cross-validate within a predetermined range to determine the best parameter combination; S24, using the test set data to evaluate the performance of the optimized multi-classification model, selecting the best model from each optimized model as the trained model, and converting the trained multi-classification model into a rule for judging the purpose of antimicrobial drug use.

5. The method for determining the purpose of antibacterial drug use based on artificial intelligence according to claim 4, characterized in that: The step S22 further includes: Use the set kernel function to map the data to a higher dimensional space by introducing slack variables and penalty parameters To control the cost of misclassification, the optimization problem becomes: ； After training, the optimal hyperplane is determined as the trained preliminary multi-classification model.

6. The method for determining the purpose of antibacterial drug use based on artificial intelligence according to claim 5, characterized in that: The step S24 specifically includes: S241, using the rule extraction algorithm to parse the trained multi-classification model and formalize each condition and decision output into a rule for judging the purpose of antimicrobial medication; S242, each antimicrobial drug use purpose judgment rule R is formalized into a conditional expression: ； in It is a condition derived from each patient's basic information B, medical records M, etiology test results and other laboratory test results L, and historical antimicrobial drug use and treatment / prevention effects D. The purpose of the use of the antimicrobial drug in this article; S243, using the trained multi-classification model to analyze the real-time acquired new case data, updated clinical application guidelines and literature of antimicrobial drugs to update the rules for judging the purpose of antimicrobial drug use: ; Where R is the rule for judging the purpose of antimicrobial drug use before the update, It is the updated rule for judging the purpose of using antimicrobial drugs. is the learning rate, The observed purpose of medication use, It is the purpose of medication predicted by the current antimicrobial drug purpose judgment rule R; The purpose of antimicrobial drug use is determined by the rule R about the conditions The partial derivative of When the predicted results change The sensitivity of the data is , and n is the number of antimicrobial drug prescription items obtained in real time.

7. The method for determining the purpose of use of antibacterial drugs based on artificial intelligence according to any one of claims 1 to 6, characterized in that: The purposes of medication include but are not limited to preoperative preventive medication, intraoperative preventive medication, postoperative preventive medication, non-surgical preventive medication, therapeutic medication and discharge medication.

8. An automatic determination system for the purpose of antibacterial drug use, characterized in that: include: A data construction module is used to pre-process the acquired patient medical data, clinical application guidelines and literature of antimicrobial drugs, and guiding principles for clinical application of antimicrobial drugs, and construct a patient historical medical data set, wherein the patient medical data includes the patient's basic information, disease diagnosis information, surgery application information, medical records, surgery records, etiology test results and other laboratory test results, historical antimicrobial drug use and treatment / prevention effects; A model training module is used to randomly divide the patient's historical medical data set into a training set, a validation set and a test set according to a set ratio, use the training set to train the initial antimicrobial drug purpose judgment model, then use the validation set to adjust parameters and optimize, use the test set to evaluate and screen, obtain the trained antimicrobial drug purpose judgment model, and convert it into an antimicrobial drug purpose judgment rule; The rule application module, after receiving the request for antibiotic prescription, calls the antibiotic purpose judgment rule, generates a matching antibiotic purpose based on the collected antibiotic prescription information and the current medical data of the corresponding patient, and selects corresponding control rules and paths based on the matched antibiotic purpose.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images