Method for generating pediatric surgery appendix postoperative anti-infection scheme

Abstract

The invention relates to the technical field of medical record analysis, in particular to a pediatric surgery appendix postoperation anti-infection scheme generation method. According to the method, the development approximation degree is obtained according to the similar condition of physical sign development between patients; analyzing the similarity degree of the illness state information and the allergic drug information among the patients, and combining development approximation degree adjustment to obtain the illness state similarity of the historical patients and the current patient; and analyzing the infection characterization condition of the historical patient based on the postoperative infection information of the historical patient, determining an approximate patient of the current patient in combination with the disease condition similarity, and determining an anti-infection scheme of the current patient according to the approximate patient. According to the method, similar conditions of physical sign development and allergic drugs and complex conditions of infection are analyzed in disease condition matching of patients, and similar patients are screened out to provide more reliable data support for an anti-infection scheme of the current patient, so that scheme determination is more efficient and reliable.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical record analysis, and in particular to a pediatric surgical post-appendectomy anti-infection scheme generation method. BACKGROUND

[0002] Appendicitis is one of the common acute abdominal emergencies in pediatric surgery, especially in young children and preschool children, because its symptoms are ambiguous and easy to delay diagnosis, leading to some children with appendicitis often developing into appendiceal perforation or peritonitis. Appendectomy is the main method for treating appendicitis, and the management of postoperative infection has always been one of the difficulties faced by surgeons. Therefore, how to develop a scientific and effective anti-infection scheme to effectively prevent postoperative infection and reduce antibiotic abuse is an important issue in the field of pediatric surgery.

[0003] In general, the generation of the anti-infection scheme after the surgical appendectomy is determined by medical staff according to the patient's condition, postoperative performance, etc. Such a determination will greatly test the working experience and medical skills of medical staff, especially for children, whose body organs are not fully developed, and body weight and age will affect drug metabolism and drug efficacy, which is prone to more complex postoperative conditions, affecting the efficient determination of the anti-infection scheme of the patient and making the reliability of the anti-infection scheme not high. SUMMARY

[0004] In order to solve the technical problem that for children, the body organs are not fully developed, and body weight and age will affect drug metabolism and drug efficacy, which is prone to more complex postoperative conditions, affecting the efficient determination of the anti-infection scheme of the patient in the prior art, the purpose of the present application is to provide a pediatric surgical post-appendectomy anti-infection scheme generation method, and the technical solution adopted is as follows: The present application provides a pediatric surgical post-appendectomy anti-infection scheme generation method, which comprises: Obtaining the patient's physical development data, allergic drug information and disease information, and postoperative infection information of historical patients from the electronic medical record; According to the similarity degree of each historical patient and the current patient in the physical development data, obtaining the development approximation degree between each historical patient and the current patient; Analyzing the matching degree of each historical patient and the current patient in the disease information and the approximation degree of the allergic drug information, and combining the development approximation degree, obtaining the disease condition similarity of each historical patient and the current patient; According to the occurrence frequency of keywords in the postoperative infection information of historical patients, determining the infection characteristic of each historical patient; combining the infection characteristic of each historical patient and the disease condition similarity of the historical patient and the current patient, obtaining the infection characteristic matching degree of the historical patient and the current patient; Screening an approximate patient from historical patients based on the matching degree of infection characteristics; determining the anti-infection scheme of the current patient according to the approximate patient.

[0005] Further, the development approximation degree acquisition method comprises: For any one historical patient, the numerical difference between the historical patient and the current patient on each sign development data is calculated as the sign deviation degree of the historical patient and the current patient on each sign development data; The sum of the sign deviation degrees of the historical patient and the current patient on all sign development data is subjected to negative correlation mapping and normalization processing to obtain the development approximation degree of the historical patient and the current patient.

[0006] Further, the disease condition similarity acquisition method comprises: For any one historical patient, the preoperative disease condition matching degree of the historical patient and the current patient is obtained according to the keyword distribution similarity between the historical patient and the current patient in the disease information; The consistency between the historical patient and the current patient in the allergy information is taken as the allergy characteristic similarity of the historical patient and the current patient; The preoperative disease condition matching degree and the allergy characteristic similarity of the historical patient and the current patient are weighted and summed based on the weight distribution of the development approximation degree to obtain the disease condition similarity of the historical patient and the current patient.

[0007] Further, the preoperative disease condition matching degree acquisition method comprises: The disease information of each patient is segmented by the jieba segmentation algorithm, and the keywords are extracted according to the hospital appendix database matching to serve as the disease condition keywords of each patient; The Jaccard correlation coefficient of the disease condition keywords between the historical patient and the current patient is calculated as the key approximation index of the historical patient; The disease condition significant index of the historical patient is obtained according to the occurrence frequency of the same disease condition keywords between the historical patient and the current patient in all historical patients; The preoperative disease condition matching degree of the historical patient and the current patient is obtained by combining the key approximation index and the disease condition significant index of the historical patient.

[0008] Further, the disease condition significant index acquisition method comprises: The same disease condition keywords of the historical patient and the current patient are taken as the common keywords of the historical patient; For any one common keyword, the ratio of the number of historical patients with the common keyword to the total number of all historical patients is taken as the characteristic degree of the common keyword; The average of the representation degrees of all common keywords of the historical patient is negatively correlated to obtain a significant indicator of the condition of the historical patient.

[0009] Further, the method for obtaining the allergy representation similarity comprises: The set of allergy drugs in the allergy drug history information of each patient is taken as the allergy set of each patient. The Jaccard correlation coefficient of the allergy set between the historical patient and the current patient is obtained to obtain the allergy representation similarity between the historical patient and the current patient.

[0010] Further, the weight is allocated based on the size of the development approximation degree, and the preoperative condition matching degree and the allergy representation similarity between the historical patient and the current patient are weighted and summed to obtain the condition similarity between the historical patient and the current patient, comprising: The normalized value of the development approximation degree between the historical patient and the current patient is taken as the feature weight of the historical patient. The product of the feature weight of the historical patient and the preoperative condition matching degree is taken as the matching indicator of the historical patient. When the feature weight is less than a preset weight value, the product of the feature weight of the historical patient and the allergy representation similarity is taken as the allergy indicator of the historical patient; when the feature weight is greater than or equal to the preset weight value, the product of the value of the feature weight of the historical patient after negative correlation mapping and the allergy representation similarity is taken as the allergy indicator of the historical patient. The condition similarity between the historical patient and the current patient is obtained by combining the matching indicator and the allergy indicator of the historical patient.

[0011] Further, the method for obtaining the infection representation comprises: The postoperative infection information of each historical patient is segmented by the jieba segmentation algorithm, and the key words are extracted according to the appendectomy knowledge base matching to obtain the infection key words of each historical patient. The number of occurrences of each infection key word in all historical patients is counted as the significant representation of each infection key word. The average of the significant representation of all infection key words of each historical patient is calculated and normalized to obtain the infection representation of each historical patient.

[0012] Further, the method for obtaining the infection representation matching degree comprises: The product of the condition similarity between each historical patient and the current patient and the infection representation of the historical patient is taken as the infection representation matching degree between each historical patient and the current patient.

[0013] Further, the method for obtaining the approximate patient comprises: The historical patient with the infection characterization matching degree greater than the preset matching threshold is matched as the approximate patient of the current patient.

[0014] The present application has the following advantages: In the present application, the similar analysis is adjusted by the similar conditions between the patient's physical development data when the current patient is matched with the historical patient, and the patient's allergic drug condition is also considered, so that the current patient and the historical patient are considered in the similar condition, the physical development and the allergic condition, the similar analysis is comprehensive, and the data basis is provided for the subsequent historical patient screening. Further, the infection condition of the historical patient is analyzed based on the postoperative infection information of the historical patient, the complex infection generated in the possible similar condition is analyzed, the approximate patient with higher significant approximation degree is determined, and more comprehensive data support is provided for the possible infection condition of the current patient, so that the anti-infection scheme of the current patient is determined more efficiently and reliably. In the present application, the similar conditions of the physical development and the allergic drug are analyzed in the patient's condition matching, and the complex condition of the infection is generated, so that the approximate patient is screened to provide more reliable data support for the anti-infection scheme of the current patient, and the scheme is determined more efficiently and reliably. BRIEF DESCRIPTION OF DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0016] Figure 1 A pediatric surgery postappendectomy anti-infection scheme generation method flowchart provided by an embodiment of the present application; Figure 2 A condition similarity acquisition method flowchart provided by an embodiment of the present application. DETAILED DESCRIPTION

[0017] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following will combine the drawings and the preferred embodiments to specifically describe a pediatric surgery postappendectomy anti-infection scheme generation method according to the present application, its specific implementation, structure, characteristics and effects, which are described in detail as follows. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0019] The specific scheme of the method for generating an anti-infection scheme after pediatric surgical appendectomy provided by the present application is specifically described below in combination with the drawings.

[0020] Please refer to Figure 1 which shows a flow chart of a method for generating an anti-infection scheme after pediatric surgical appendectomy provided by an embodiment of the present application, which comprises the following steps: S1: Obtain the physical development data, allergic drug information and illness information of the patient, and the postoperative infection information of the historical patient from the electronic medical record.

[0021] The most common reason for pediatric patients to undergo appendectomy is the occurrence of appendicitis. Because the appendiceal lumen of children is relatively thin, it is easy to be blocked, leading to bacterial proliferation and infection, causing the appendiceal lumen to be blocked, and because the lymphatic tissue in the appendix of children is rich, it is easy to cause inflammatory reaction, causing lymphatic tissue hyperplasia, and in addition, the immune system of children has not fully developed, and the resistance to infection is relatively weak, so the low immunity will lead to related conditions of the appendix, such as appendicitis, appendiceal abscess, etc.

[0022] The development degree of organs of pediatric patients at different growth and development stages is different, so even if the two may correspond to the same appendiceal disease, considering the differences in their body metabolic capacity, immune response, liver and kidney function, etc., there will be differences in related surgical operations and the selection of anti-infection drugs. Especially for newborns, the blood-brain barrier has not yet fully matured, and certain drugs may more easily enter the central nervous system, and more care is needed when using them.

[0023] Therefore, in order to more conveniently determine the anti-infection scheme of the current patient, the anti-infection scheme of the historical patient with similar conditions is screened to provide support for the current scheme, and the physical development and drug allergy of the patient, etc. are considered when screening. Therefore, in the embodiment of the present application, the physical development data, allergic drug information and illness information of each patient are obtained from the electronic medical record of the patient, wherein the physical development data is, for example, age, weight, body surface area and hormone level, etc., the allergic drug information is, for example, penicillin, ketamine or iodophor, etc., if the pediatric patient does not have clear allergic drug information, the allergic drug information of the parents of the pediatric patient is analyzed as the allergic drug information of the patient, and the illness information is, for example, the occurrence of right lower abdominal pain, vomiting and fever, etc.

[0024] In order to improve the possible complex infection situation, the postoperative infection information of the historical patient is further obtained in the embodiment of the present application, so as to facilitate more efficient analysis of the anti-infection scheme of the current patient based on the representativeness of the historical anti-infection scheme.

[0025] S2: Obtain the development approximation degree between each historical patient and the current patient according to the similarity degree between each historical patient and the current patient in the physical development data.

[0026] The determination of the anti-infection scheme after pediatric surgical appendectomy needs to consider multiple factors, which involves children with different physical conditions, because children have different physiological characteristics in different growth and development conditions, and the development of organs is different. Therefore, when comparing and analyzing the postoperative effects of pediatric surgical appendectomy, it is necessary to fully consider whether the children are in the same growth and development stage, and to preliminarily obtain the approximation degree of the growth and development of the patients through the similarity of the physical development data.

[0027] Preferably, in the embodiments of the present application, the method for obtaining the development approximation degree comprises: For any historical patient, the numerical difference between the historical patient and the current patient in each physical development data is calculated as the physical deviation degree of the historical patient and the current patient in each physical development data, reflecting the difference between the patients in each physical development data.

[0028] Further, the sum of the physical deviation degrees of the historical patient and the current patient in all physical development data is negatively correlated and normalized to obtain the development approximation degree of the historical patient and the current patient. When the physical deviation degree is smaller, the physical data of the two patients is more similar, indicating that the historical patient and the current patient are more likely to be in the same growth stage, so the development approximation degree is obtained by negative correlation and normalization.

[0029] It should be noted that negative correlation mapping and normalization are well-known technical means to those skilled in the art, such as inverse proportion or negative exponential power, normalization can use linear normalization or standard normalization, etc., which will not be repeated and limited here.

[0030] S3: Analyze the matching degree of the disease information and the approximation degree of the allergic drug information of each historical patient and the current patient, and obtain the disease condition similarity of each historical patient and the current patient in combination with the development approximation degree.

[0031] In order to determine the high-similarity disease group of the current patient subsequently, it is necessary to analyze the disease similarity of the current patient and the historical patient in the medical record, and at the same time consider that even under the same disease, the intraoperative difference caused by allergic drugs may lead to different postoperative infections. Therefore, the similarity of the patient's drug allergy is added in the disease similarity analysis to improve the analysis of the disease similarity.

[0032] The development approximation degree analysis between patients makes the analysis of the similar postoperative effects of the patients more reliable. Preferably, in the embodiments of the present application, the method for obtaining the disease condition similarity is shown in Figure 2 , which shows a method flowchart for obtaining a disease condition similarity according to an embodiment of the present application. The method comprises the following steps: S301: For any one historical patient, according to the keyword distribution similarity between the historical patient and the current patient in the disease information, obtain the preoperative condition matching degree of the historical patient and the current patient.

[0033] Since the disease information of the patient is the diagnosis of the disease of the doctor, the higher the similarity of the disease information corresponding to the condition, the more likely it is that the patients are a high-similarity disease group. Therefore, in the embodiment of the application, the method for obtaining the preoperative condition matching degree by extracting and analyzing the similarity of the keywords about the condition in the disease information includes: First, the disease information of each patient is segmented by the jieba segmentation algorithm, and the keywords are extracted and matched according to the hospital appendix database, which are used as the condition keywords of each patient. The keywords describing the condition after segmentation are matched and extracted according to the database, such as appendix perforation, abdominal distension, fever, vomiting or early stage, etc. The words representing the patient's disease diagnosis.

[0034] Further, the Jaccard correlation coefficient of the condition keywords between the historical patient and the current patient is calculated as the key approximation index of the historical patient. The Jaccard correlation coefficient is used to measure the similarity of two sets. The Jaccard correlation coefficient between the condition keywords of two patients is the key approximation index. It should be noted that the jieba segmentation algorithm and the Jaccard correlation coefficient are well-known public technical means for those skilled in the art, and will not be described here.

[0035] Further, the disease significant index of the historical patient is obtained according to the frequency of the same condition keywords between the historical patient and the current patient in all historical patients. It is considered that pediatric patients have some general diseases, such as fever. At this time, the high common disease will affect the approximation analysis of the specific patient. Therefore, the high frequency of the same condition keywords between the patients is analyzed to reduce the influence of the common disease.

[0036] In the embodiment of the application, the method for obtaining the disease significant index includes: First, the same condition keywords of the historical patient and the current patient are used as the common keywords of the historical patient. For any one common keyword, the ratio of the number of historical patients with the common keyword to the total number of all historical patients is used as the representation degree of the common keyword. The ratio of the number of patients with each keyword to the total number of patients reflects the degree of commonality of the common keyword. The higher the representation degree, the higher the commonality of the common keyword.

[0037] Further, the average of the representation degrees of all common keywords of the historical patient is negatively correlated to obtain the illness significant indicator of the historical patient, and the more similar the common keywords of the historical patient and the current patient are, the higher the similarity credibility is.

[0038] Finally, the preoperative condition matching degree of the historical patient and the current patient is obtained by combining the key approximation indicator and the illness significant indicator of the historical patient. In the embodiment of the present application, the product of the key approximation indicator and the illness significant indicator is taken as the preoperative condition matching degree of the historical patient and the current patient, and the higher the key approximation indicator and the illness significant indicator are, the higher the similarity of the conditions between the patients is.

[0039] S302: According to the consistency of the historical patient and the current patient in the allergic drug information, the allergic representation similarity of the historical patient and the current patient is obtained.

[0040] Since the patients of pediatric surgery are all young, the risk of drug sensitivity reaction also needs to be considered, especially in the case of unclear allergic drugs, the possible drug sensitivity of family members due to genetic factors also needs to be considered to affect the degree of immune system reaction.

[0041] Therefore, in the embodiment of the present application, the set of allergic drugs in the allergic drug history information of each patient is taken as the allergic set of each patient, and the Jaccard correlation coefficient of the allergic set between the historical patient and the current patient is obtained to obtain the allergic representation similarity of the historical patient and the current patient, and the more similar the allergic history is, the more similar the immune reaction between the patients is, and the higher the similarity between the patients is.

[0042] S303: Based on the size of the development approximation degree, the preoperative condition matching degree and the allergic representation similarity of the historical patient and the current patient are weighted and summed to obtain the condition similarity of the historical patient and the current patient.

[0043] Finally, the comprehensive analysis of the similarity degree is adjusted by the development condition, and the higher the development approximation degree is, the higher the attention to the condition matching degree is. Since the drug allergy is only an auxiliary reflection of the analysis of the similar condition, the degree of participation of the similar proportion cannot be too large, and the similar condition of the condition keyword matching is the core part of the similarity analysis.

[0044] Therefore, in the embodiment of the present application, the normalized value of the development approximation degree of the historical patient and the current patient is taken as the feature weight of the historical patient, and the feature weight is weighted. The product of the feature weight of the historical patient and the preoperative condition matching degree is taken as the matching indicator of the historical patient, and the adjustment of the development approximation degree is that the more similar the physical conditions of the patients are, the higher the similarity of the analysis of the condition is.

[0045] For allergic cases, the participation proportion needs to be limited. When the feature weight is less than the preset weight value, it indicates that the body feature credibility is low, so the product of the feature weight of the historical patient and the allergic feature similarity is taken as the allergic index of the historical patient. On the contrary, when the feature weight is greater than or equal to the preset weight value, it indicates that the similarity between the patients is high, and the participation of the allergic case needs to be reduced, so the product of the value of the feature weight of the historical patient after negative correlation mapping and the allergic feature similarity is taken as the allergic index of the historical patient. In the embodiment of the present application, the preset weight value is set to 0.3, and the specific value can be adjusted according to the specific implementation, which is not limited here.

[0046] Finally, the matching index and the allergic index of the historical patient are combined to obtain the condition similarity between the historical patient and the current patient, and the final condition similarity is obtained by comprehensively considering the two approximate cases. In the embodiment of the present application, the sum of the matching index and the allergic index is taken as the condition similarity, and the weighted sum process is realized.

[0047] At this point, the preoperative condition similarity analysis between patients is completed.

[0048] S4: According to the occurrence frequency of the keywords in the postoperative infection information of the historical patients, the infection feature of each historical patient is determined; and the infection feature of each historical patient and the condition similarity between the historical patient and the current patient are combined to obtain the infection feature matching degree between the historical patient and the current patient.

[0049] By comparing the preoperative comprehensive conditions of the current patient and each historical patient, the similar historical patients can be screened for scheme observation. However, even if the comprehensive conditions of the current patient and the historical patients are very similar, due to the differences in individual immune systems, surgical operation and environmental factors, and microbial drug resistance, etc., the postoperative infection performance of the current patient will also be different from that of the historical patients. Therefore, the representativeness of the anti-infection scheme of each historical patient also needs to be combined and considered to determine the high-similarity condition and infection performance group of the current patient.

[0050] Therefore, according to the preoperative comprehensive condition similarity between the current patient and each historical patient, and by adjusting the representativeness of the anti-infection scheme of each historical patient, the final result tends to be more similar in the preoperative comprehensive condition and the anti-infection scheme is also more representative, thereby improving the efficiency of the scheme analysis of the current patient.

[0051] Preferably, in the embodiment of the present application, the performance of each historical patient in the postoperative infection is analyzed first, and the infection feature is obtained by the following method: Firstly, the postoperative infection information of each historical patient is segmented by the jieba segmentation algorithm, and the key words are extracted according to the appendectomy knowledge base matching, as the infection key words of each historical patient, reflecting the relevant information of postoperative infection, such as suppuration, redness, wound exudate or disease symptoms, etc.

[0052] Further, the number of occurrences of each infection key word in all historical patients is counted as the significant representation of each infection key word, and the more the number of occurrences of the infection key word, the more representative the infection condition is in the postoperative infection condition.

[0053] The significant representation of all infection key words of each historical patient is averaged and normalized to obtain the infection representation of each historical patient, which further reflects the representation of the anti-infection scheme of the historical patient.

[0054] Finally, the infection analysis and preoperative condition analysis are combined to obtain the infection representation matching degree, reflecting the degree of high similar condition and infection performance. In the embodiment of the present application, the product of the condition similarity of each historical patient and the infection representation of the historical patient is taken as the infection representation matching degree of each historical patient and the current patient, and the approximate patients are further screened for auxiliary analysis through the infection representation matching degree.

[0055] S5: Screening approximate patients from historical patients based on infection representation matching degree; determining the anti-infection scheme of the current patient according to the approximate patients.

[0056] The higher the infection representation matching degree, the higher the similarity of the historical patient's condition and the stronger the infection representation ability, and the more representative the auxiliary analysis of the current anti-infection scheme. Therefore, in the embodiment of the present application, the historical patients with infection representation matching degree greater than the preset matching threshold are taken as the approximate patients of the current patient, and the preset matching threshold can be set to 0.82. The specific value setting implementer can adjust it by himself, which is not limited here.

[0057] Through the anti-infection scheme and postoperative infection information of the screened approximate patients, the doctor can assist in determining the anti-infection scheme. In the embodiment of the present application, the approximate patients can be sorted in descending order of infection representation matching degree, and the order of the approximate patients reflects the priority reference degree of the corresponding anti-infection scheme. The doctor can determine the anti-infection scheme of the current patient according to the anti-infection scheme and infection information of the approximate patients after sorting, so that the selection and analysis are more efficient.

[0058] To sum up, when the current patient is matched with the historical patient in terms of the condition, the similar analysis is adjusted by the similar conditions between the physical development data of the patients, and the allergic drug conditions of the patients are also concerned, so that the current patient and the historical patient are considered in terms of the similar condition, the physical development and the allergic condition, the similar analysis is comprehensive, and data basis is provided for subsequent historical patient screening. Then, the infection conditions of the historical patients are analyzed based on the postoperative infection information of the historical patients, the complex infection generated in the possible similar condition is analyzed, the similar patient with higher significant approximation is determined, more comprehensive data support is provided for the possible infection condition of the current patient, the anti-infection scheme of the current patient is determined more efficiently and reliably. In the condition matching of the patient, the similar conditions of the physical development and the allergic drug, and the complex conditions of the infection are analyzed, the similar patient is screened, more reliable data support is provided for the anti-infection scheme of the current patient, and the scheme is determined more efficiently and reliably.

[0059] It should be noted that the above-mentioned embodiment sequence of the application is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are possible or can be advantageous.

[0060] Each embodiment in the specification is described in a progressive manner, and the same and similar parts between each embodiment can be referred to each other, and each embodiment mainly describes the difference from other embodiments.

Claims

1. A method for generating an anti-infective regimen for pediatric surgical post appendectomy, the method comprising: receiving a plurality of patient data; and generating an anti-infective regimen for the patient based on the plurality of patient data. The method comprises: obtaining the patient's physical development data, allergic drug information and disease information, and the postoperative infection information of the historical patients from the electronic medical record; obtaining the development approximation degree between each historical patient and the current patient according to the similarity of each historical patient and the current patient in the physical development data; analyzing the matching degree of each historical patient and the current patient in the disease information and the approximation degree of the allergic drug information, and combining the development approximation degree to obtain the disease condition similarity between each historical patient and the current patient; determining the infection representation of each historical patient according to the occurrence frequency of the keywords in the postoperative infection information of the historical patients, and combining the infection representation of each historical patient and the disease condition similarity between the historical patients and the current patient to obtain the infection representation matching degree between the historical patients and the current patient; screening the approximate patients from the historical patients based on the infection representation matching degree, and determining the anti-infection scheme of the current patient according to the approximate patients.

2. The method for generating a pediatric surgical post- appendectomy anti-infective regimen of claim 1, wherein, The development approximation degree obtaining method comprises: for any one historical patient, calculating the numerical difference between the historical patient and the current patient in each physical development data as the physical deviation degree of the historical patient and the current patient in each physical development data; performing negative correlation mapping and normalization processing on the sum of the physical deviation degrees of the historical patient and the current patient in all physical development data to obtain the development approximation degree between the historical patient and the current patient.

3. The method for generating a pediatric surgical post- appendectomy anti-infective regimen of claim 1, wherein, The disease condition similarity obtaining method comprises: for any one historical patient, obtaining the preoperative disease condition matching degree between the historical patient and the current patient according to the keyword distribution similarity between the historical patient and the current patient in the disease information; taking the consistency between the historical patient and the current patient in the allergic drug information as the allergy representation similarity between the historical patient and the current patient; based on the size of the development approximation degree, performing weight allocation, and performing weighted summation on the preoperative disease condition matching degree and the allergy representation similarity between the historical patient and the current patient to obtain the disease condition similarity between the historical patient and the current patient.

4. The method for generating a pediatric surgical post- appendectomy anti-infective regimen of claim 3, wherein, The preoperative disease condition matching degree obtaining method comprises: performing word segmentation on the disease information of each patient by using the jieba word segmentation algorithm, and extracting the keywords according to the hospital appendix database matching to obtain the disease keywords of each patient; calculating the Jaccard correlation coefficient of the disease keywords between the historical patient and the current patient as the key approximation index of the historical patient; obtaining the disease significant index of the historical patient according to the occurrence frequency of the same disease keywords between the historical patient and the current patient in all historical patients; combining the key approximation index and the disease significant index of the historical patient to obtain the preoperative disease condition matching degree between the historical patient and the current patient.

5. The method for generating a pediatric surgical post- appendectomy anti-infective regimen of claim 4, wherein, The disease significant index obtaining method comprises: taking the same disease keywords of the historical patient and the current patient as the common keywords of the historical patient; for any one common keyword, taking the ratio of the number of historical patients with the common keyword to the total number of all historical patients as the representation degree of the common keyword; performing negative correlation mapping on the average value of the representation degrees of all common keywords of the historical patient to obtain the disease significant index of the historical patient.

6. The method for generating a pediatric surgical post- appendectomy anti-infective regimen of claim 3, wherein, The method for obtaining the allergy representation similarity comprises: a set of allergy drugs in the allergy drug history information of each patient is taken as an allergy set of each patient; a Jaccard correlation coefficient of the allergy sets between the historical patient and the current patient is obtained to obtain the allergy representation similarity between the historical patient and the current patient.

7. The method for generating a pediatric surgical post- appendectomy anti-infective regimen of claim 3, wherein, The weight distribution is performed based on the size of the development approximation degree, and a preoperative condition matching degree and the allergy representation similarity between the historical patient and the current patient are weighted and summed to obtain a condition similarity between the historical patient and the current patient, which comprises: a normalized value of the development approximation degree between the historical patient and the current patient is taken as a feature weight of the historical patient; a product of the feature weight of the historical patient and the preoperative condition matching degree is taken as a matching index of the historical patient; when the feature weight is less than a preset weight value, a product of the feature weight of the historical patient and the allergy representation similarity is taken as an allergy index of the historical patient; when the feature weight is greater than or equal to the preset weight value, a product of a value obtained by performing a negative correlation mapping on the feature weight of the historical patient and the allergy representation similarity is taken as the allergy index of the historical patient; the condition similarity between the historical patient and the current patient is obtained by combining the matching index and the allergy index of the historical patient.

8. The method for generating a pediatric surgical post- appendectomy anti-infective regimen of claim 1, wherein, The method for obtaining the infection representation comprises: postoperative infection information of each historical patient is segmented by using a jieba segmentation algorithm, and key words are extracted according to an appendectomy knowledge base matching to serve as infection key words of each historical patient; the number of occurrences of each infection key word in all historical patients is counted to serve as the significant representation of each infection key word; the significant representation of all infection key words of each historical patient is averaged and normalized to obtain the infection representation of each historical patient.

9. The method for generating a pediatric surgical post- appendectomy anti-infective regimen of claim 1, wherein, The method for obtaining the infection representation matching degree comprises: a product of the condition similarity between each historical patient and the current patient and the infection representation of the historical patient is taken as the infection representation matching degree between each historical patient and the current patient.

10. The method for generating a pediatric surgical post- appendectomy anti-infective regimen of claim 1, wherein, The method for obtaining the approximate patient comprises: a historical patient with an infection representation matching degree greater than a preset matching threshold is taken as an approximate patient of the current patient.

Images