Prediction model and prediction method for infection of patient in perioperative period

Abstract

The invention provides a perioperative patient infection prediction model and prediction method. The prediction model comprises a patient information acquisition module; the first index evaluation sub-model is used for outputting a first index about the infection probability of the patient according to the clinical information of the patient; an operating room information acquisition module; the second index evaluation sub-model is used for outputting a second index of the infection indexes in the operating room based on the environmental parameters in the operating room; an operation information acquisition module; the third index evaluation sub-model is used for outputting a third index of the intraoperative infection probability based on the operation related information; the fourth index evaluation sub-model is used for outputting a fourth index of the infection probability according to the use conditions of the antibiotics before and during the operation; and the prediction module is used for fitting the first index, the second index, the third index and the fourth index, and outputting a patient infection probability score or an infection grade according to an algorithm. According to the invention, the infection rate is reduced, and the medical cost of patients and the resource waste of hospitals are effectively reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of big data analysis, and in particular to a perioperative patient infection prediction model and a prediction method. Background Art

[0002] For surgical patients, strengthening perioperative infection control can prevent patients from being infected with bacteria and viruses during surgery, reduce the risk of postoperative infection, and thus ensure the health and safety of patients. The existing perioperative patient infection prediction is only based on the daily experience of doctors and nurses, and there is no scientific and reasonable evaluation method. Therefore, it will cause the following problems:

[0003] 1. Regarding the surgical environment, the staff only cleans or disinfects based on experience, and there is no way to evaluate the cleaning and disinfection effects in a targeted manner. There may be over-disinfection or over-cleaning, resulting in a waste of medical resources;

[0004] 2. Different patients have different physical constitutions and drug resistance, but doctors often only prescribe drugs based on their own experience and tend to prevent infection to the greatest extent possible, while ignoring individual differences, resulting in heavy drugs for minor illnesses and drug abuse, which is not conducive to the health of patients.

[0005] Therefore, there is an urgent need for a reasonable perioperative patient infection prediction mechanism to provide decision-making support for infection control and improve the safety of the surgical process and the rationality of medication. Summary of the invention

[0006] To solve the problems in the prior art, the present invention provides a perioperative patient infection prediction model, and also provides a prediction method based on the perioperative patient infection prediction model. Through scientific evaluation of patients, medication can be guided according to individual differences, thereby effectively reducing patients' medical costs and hospital resource waste.

[0007] The perioperative patient infection prediction model of the present invention comprises:

[0008] Patient information acquisition module: used to obtain the patient's clinical information;

[0009] The first indicator evaluation sub-model is used to output the first indicator of the patient's infection probability based on the patient's clinical information;

[0010] Operating room information acquisition module: used to obtain environmental parameters in the operating room;

[0011] The second indicator evaluation sub-model is used to output the second indicator of infection index in the operating room based on the environmental parameters in the operating room;

[0012] Surgery information acquisition module: used to obtain surgery-related information, including surgery type, surgery site, surgery duration, intraoperative vital signs, and intraoperative input and output;

[0013] The third indicator evaluation sub-model is used to output the third indicator of the probability of intraoperative infection based on the operation-related information;

[0014] The fourth indicator evaluation sub-model is used to output the fourth indicator of infection probability according to the use of antibiotics before and during surgery;

[0015] Prediction module: used to fit the first indicator, the second indicator, the third indicator and the fourth indicator, and output the patient's infection probability score or infection level according to the algorithm.

[0016] Furthermore, the perioperative patient infection prediction model also includes:

[0017] Postoperative monitoring and evaluation sub-model: used to obtain analysis and evaluation results based on monitored vital signs, wound conditions and / or microbiological data;

[0018] Infection prediction result dynamic adjustment module: used to adjust the patient's infection probability score or infection level based on the analysis and evaluation results and integrated big data.

[0019] Furthermore, the clinical information includes the patient's previous surgical history, chronic disease history, allergy history, the patient's physical signs and various relevant preoperative examination indicators. The first indicator evaluation submodel obtains the influencing factors related to the infection based on the previous surgical history, chronic disease history, allergy history, the patient's physical signs and various relevant preoperative examination indicators, and scores each influencing factor to obtain the first indicator including the total score of all influencing factors.

[0020] Furthermore, it also includes a correction module: based on the data big model and the prediction results, the weights of various influencing factors are set and updated iteratively to obtain the weighted total score of the first indicator.

[0021] Furthermore, the environmental parameters are obtained through sensors and monitoring equipment, including air quality, temperature and humidity, pressure difference and / or microbial content.

[0022] The present invention also provides a perioperative patient infection prediction method, which is implemented based on the perioperative patient infection prediction model and includes the following steps:

[0023] S1: Obtain the first indicator of the patient's infection probability based on the patient's clinical information;

[0024] S2: Based on the environmental parameters in the operating room, a second indicator of infection index in the operating room is obtained;

[0025] S3: Based on the surgery-related information, output a third indicator of the probability of intraoperative infection;

[0026] S4: Output the fourth indicator of infection probability based on the use of antibiotics before and during surgery;

[0027] S5: Fit the first indicator, the second indicator, the third indicator and the fourth indicator, and output the patient's infection probability score or infection level according to the algorithm.

[0028] Furthermore, the method further comprises step S6: providing guidance on medication to the patient according to the infection probability score or infection level.

[0029] Furthermore, the method further comprises step S7: obtaining the recovery status of the patient after the operation and evaluating the effects of different drugs on reducing infection;

[0030] And according to the patient's recovery situation, the perioperative patient infection prediction model is optimized.

[0031] Furthermore, the disinfection method evaluation step is also included:

[0032] The disinfection method used for infection control is used as a variable to input into the perioperative patient infection prediction model;

[0033] Evaluate the difference in effect with previous disinfection methods, and determine the influencing factors of the model based on the difference in effect;

[0034] Based on the influencing factors, determining a disinfection method evaluation sub-model;

[0035] Based on the disinfection method evaluation sub-model, obtaining a fourth indicator affecting infection prediction;

[0036] The fourth indicator is added to the perioperative patient infection prediction model to obtain an infection prediction result.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] 1. Relying on big data analysis, a patient infection prediction model is established to provide decision-making support for infection control;

[0039] 2. Ensure sterility requirements by monitoring environmental parameters in the operating room, improve surgical safety and hospital operating efficiency;

[0040] 3. Ability to monitor the patient's condition throughout the entire process according to the model, promptly detect and treat signs of infection, and help medical staff develop personalized surgery and care plans based on the patient's condition;

[0041] 4. The perioperative patient infection prediction model can optimize the workflow of medical staff, including disinfection and aseptic operation procedures, effectively prevent patients from being infected with bacteria and viruses during surgery, and reduce the risk of postoperative infection;

[0042] 5. Guiding patients on medication and reducing infection rates can effectively reduce drug abuse, reduce patients' medical expenses and waste of hospital resources; through model prediction results, strengthen communication and learning among doctors, improve doctors' work ability, and enhance the overall strength of the hospital. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0044] Figure 1 This is a structural diagram of the prediction model of the present invention;

[0045] Figure 2 This is a flow chart of the prediction method in the present invention. DETAILED DESCRIPTION

[0046] Unless otherwise defined, all technical and scientific terms used in the present invention have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs; the terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention; the terms "including" and "having" and any variations thereof in the specification and claims of the present invention and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present invention or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0047] Reference to "embodiments" in the present invention means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it mutually exclusive, independent, or alternative to other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described in the present invention may be combined with other embodiments.

[0048] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings.

[0049] like Figure 1As shown, the perioperative patient infection prediction model of the present invention includes:

[0050] Patient information acquisition module: used to obtain the patient's clinical information;

[0051] The first indicator evaluation sub-model is used to output the first indicator of the patient's infection probability based on the patient's clinical information;

[0052] Operating room information acquisition module: used to obtain environmental parameters in the operating room;

[0053] The second indicator evaluation sub-model is used to output the second indicator of infection index in the operating room based on the environmental parameters in the operating room;

[0054] Surgery information acquisition module: used to obtain surgery-related information, including surgery type, surgery site, surgery duration, intraoperative vital signs, and intraoperative input and output;

[0055] The third indicator evaluation sub-model is used to output the third indicator of the probability of intraoperative infection based on the operation-related information;

[0056] The fourth indicator evaluation sub-model is used to output the fourth indicator of infection probability according to the use of antibiotics before and during surgery;

[0057] Prediction module: used to fit the first indicator, the second indicator, the third indicator and the fourth indicator, and output the patient's infection probability score or infection level according to the algorithm.

[0058] Doctors are guided in medication based on the patient's confirmed infection probability score or infection level.

[0059] Preferably, the perioperative patient infection prediction model further includes:

[0060] Postoperative monitoring and evaluation sub-model: used to obtain analysis and evaluation results based on monitored vital signs, wound conditions and / or microbiological data;

[0061] Infection prediction result dynamic adjustment module: used to adjust the patient's infection probability score or infection level based on the analysis and evaluation results and integrated big data.

[0062] Preferably, this example is also provided with a closed-loop control optimization program, and the prediction model of this example also includes a correction module: based on the data big model and the prediction results, the weights of the influencing factors in each indicator evaluation sub-model are set and iteratively updated to obtain the optimized weighted total score, so as to make the results of the prediction model more accurate.

[0063] The model of the present invention is described in detail below.

[0064] This case relies on the perioperative information management system. When the patient visits the hospital, detailed information of the patient is collected and uploaded to the perioperative information management system. In the above information, the influencing factors related to the risk of infection are evaluated and then used as the input of the first indicator evaluation sub-model. The influencing factors in this case include previous surgical history, chronic disease history, allergy history, patient's physical signs and various related preoperative examination indicators, such as:

[0065] Age: The patient's age is an important factor in predicting the risk of infection. The risk of infection may vary among patients of different age groups;

[0066] Gender: Gender may also affect infection risk, with certain infections being more common in certain genders;

[0067] Nutritional status: The patient's nutritional status has an important impact on their ability to resist infection, and malnutrition may increase the risk of infection;

[0068] Immune function: The patient's immune function status, such as whether there is immunodeficiency or immunosuppression, has a direct impact on the risk of infection.

[0069] In addition, the present invention also considers the patient's behavior and living habits, such as smoking history, drinking history, medical history, etc., which may also increase the risk of infection. In this case, the medical history is considered, that is, the patient's past medical history, because chronic diseases such as diabetes and cardiovascular disease may affect their recovery after surgery and the risk of infection.

[0070] In addition, the patient will be examined before surgery, and various laboratory examination indicators are also part of the patient's clinical information, mainly including:

[0071] Routine blood test: including white blood cell count, neutrophil ratio and other indicators, which can reflect the patient's inflammatory state and immune function;

[0072] Urinalysis: Urinalysis can reflect the patient's urinary system condition and help find potential sources of infection;

[0073] Biochemical indicators: Biochemical indicators such as blood sugar and albumin can reflect the nutritional status and metabolic status of patients, and have certain reference value for assessing the risk of infection.

[0074] The hospital's laboratory test results can be obtained directly by scanning the QR code on the test result sheet, or they can be obtained directly through the interface provided by the perioperative information management system after the test results are uploaded to the applicant's perioperative information management system platform.

[0075] Of course, the impact factors can be added or deleted as needed. After collecting all relevant information in this example, the first indicator evaluation sub-model assigns values ​​according to the values ​​of each impact factor. In this example, the values ​​of each impact factor are set to 1-5, and then they are scored within the range of 1-5 according to different information. Finally, all the numerical scores are calculated, and then averaged according to the amount of data of the impact factor to obtain the first indicator.

[0076] Of course, in this example, if a certain influencing factor accounts for a relatively large proportion, such as advanced age, chronic diseases, etc., this example can also set a corresponding weight for each influencing factor according to the proportion of influence, and then use the weighted average method to calculate the final first indicator value. The first indicator in this example can also be obtained by using the standardization method, TOPSIS method, etc., which will not be repeated here.

[0077] The environmental sanitation and air cleanliness of the operating room are crucial to preventing surgical site infection. Therefore, in the operating room information acquisition submodule of this example, sensors and monitoring equipment are set up in this example. The output ends of the sensors and monitoring equipment are connected to the input interface of the perioperative management system. The prediction model of this example can directly obtain data from the perioperative management system, which is more convenient. The environmental parameters of this example include air quality, temperature and humidity, pressure difference, microbial content and other indicators to ensure that the surgical environment meets the sterile requirements and the disinfection process meets the specified requirements.

[0078] In addition, this example also sets up a high-efficiency filtering system and air purification equipment to filter and purify the air in the operating room to remove dust, bacteria, viruses and other harmful substances in the air. The present invention scores the environmental quality through environmental parameters to form a second indicator.

[0079] The surgical information obtained in this case includes surgical instrument disinfection information, medical staff operation standard information and surgical information. In this case, sensors and cameras are used to monitor the disinfection process of surgical instruments and the degree of medical staff operation standardization in real time, and AI recognition is used to score the various operations of medical staff, and finally obtain the scores of each indicator. The surgical instrument disinfection information in this case includes indicators such as disinfection time, temperature, and humidity; the medical staff operation standard information is the degree of medical staff's operation standardization during the operation, which is also an important factor affecting the risk of infection.

[0080] Of course, the AI ​​recognition model in this example can be built using the YOLO model algorithm, or other machine vision algorithms such as neural network algorithms. The AI ​​recognition model can be trained through a large number of actual standard operation pictures and operation videos. During the actual operation process, the camera and sensor obtain image and video information, and then compare it with the standard operation to obtain the vector value, and then the operation score is obtained based on the vector value.

[0081] The surgical information for this case includes:

[0082] Type of surgery: Different types of surgery carry different risks of infection. For example, complex surgeries or those involving multiple organs may be more likely to result in infection.

[0083] Duration of surgery: The longer the surgery, the longer the patient is exposed to potential sources of infection, increasing the risk of infection

[0084] Surgical site: The surgical site is also an important factor affecting the risk of infection. Some parts (such as the intestines, urinary tract, etc.) may be more susceptible to infection due to their special physiological environment;

[0085] Surgical vital signs: hypothermia during surgery can also increase the chance of postoperative infection;

[0086] Intraoperative blood loss: Large amounts of intraoperative bleeding may lead to a decrease in the patient's immunity, thereby increasing the risk of infection.

[0087] Among them, since the disinfection of surgical instruments is generally completed under standard procedures, the impact on infection is relatively small, and the most important thing is the duration of the operation, that is, the wound exposure time, which has the greatest impact on infection control. In this example, a certain weight is given to each piece of information, and then each piece of information is weighted to obtain the third indicator of the probability of intraoperative infection.

[0088] Preferably, the present invention further comprises a fourth indicator evaluation sub-model: used for outputting a fourth indicator of infection probability according to the use of antibiotics before and during surgery.

[0089] The use of preoperative antibiotics is mainly based on the purpose of preventing surgical site infection and needs to strictly follow medical standards. The following is a detailed summary of the use of preoperative antibiotics:

[0090] 1. Indications for the use of antibiotics

[0091] 1. High-risk surgeries such as those expected to take a long time, involve a large surgical scope, or involve important organs or foreign body implantation;

[0092] 2. The patient is older, has low immunity, or has high-risk factors for infection;

[0093] 3. Specific surgical sites (such as colon, rectum, etc.) are susceptible to infection;

[0094] 2. Choice of Antibiotics

[0095] 1. Antibiotics that are sensitive to Gram-positive bacteria are usually selected, such as penicillins or first- and second-generation cephalosporins;

[0096] 2. The antimicrobial spectrum of the drug should cover the types of bacteria that may contaminate the surgical site;

[0097] 3. Consider the patient's history of drug allergies and avoid using allergy medications.

[0098] 3. Usage time

[0099] 1. Preoperative use time: Antibiotics should be used 0.5~2 hours before surgery (or 30 minutes to 1 hour before surgery) to ensure that the blood drug concentration is in the rising stage and has not reached the peak value during surgery. There is also a saying that preoperative antibiotics can be used about half an hour before surgery, but usually not more than two days before surgery;

[0100] 2. Intraoperative additional administration: If the operation time exceeds 3 hours, the half-life of the drug exceeds 2 times, or the amount of blood loss in adults exceeds 1500 ml, an additional antibiotic should be considered during the operation;

[0101] 3. Postoperative use: The duration of postoperative antibiotic use is usually no more than 24 hours. For clean-contaminated and contaminated surgeries, the duration of preventive medication is also 24 hours, which can be extended to 48 hours if necessary. However, excessively extending the duration of medication will not further improve the preventive effect and may increase the chance of infection with drug-resistant bacteria.

[0102] IV. Dosage Method

[0103] 1. Antibiotics should be dissolved in normal saline and injected into the body through blood vessels.

[0104] 2. The dosage and frequency of administration should be determined based on the drug instructions and factors such as the patient's weight and age.

[0105] In summary, the use of preoperative antibiotics involves multiple indicators, including indications, selection, use time, administration method and precautions. Medical personnel should follow the requirements in a timely manner during the operation to ensure the safety of the surgical patients and the success of the operation. The above indicators are uploaded to the model of the present invention as parameter information of the fourth indicator, and then the information of each parameter is comprehensively evaluated to obtain the fourth indicator value.

[0106] The present invention fits the first indicator, the second indicator, the third indicator and the fourth indicator by a weighted average method, and outputs a patient infection probability score or an infection level according to an algorithm.

[0107] Through the above process, it can be seen that the data required for the perioperative patient infection prediction model and prediction method are multifaceted, including basic patient information, surgery-related information, hospital and environmental factors, laboratory test results, microbiological data and other relevant data. These data together constitute the basis of the prediction model and provide scientific decision-making support for doctors.

[0108] Preferably, the postoperative monitoring care and the patient's living environment in this example often play a decisive role in the patient's infection control. Therefore, this example uses professional medical equipment to monitor the patient's body temperature, heart rate, respiration, blood pressure and other vital signs in real time, and regularly checks the redness, swelling, pain, exudation and other conditions of the wound. By setting up a postoperative monitoring and evaluation sub-model, the monitored vital signs and wound conditions are analyzed and evaluated in detail, and then the infection prediction result dynamic adjustment module is used to timely adjust the patient's infection probability score or infection level. This enables patients and doctors to more accurately judge the patient's condition and treat the symptoms. If an abnormality is found, appropriate nursing measures can be taken in a timely manner, such as drug treatment, adjustment of treatment plans, etc.

[0109] In addition, the microbiological data detected by the postoperative monitoring and evaluation sub-model in this case include:

[0110] Bacterial culture: If there is an infection, bacterial culture can be used to determine whether the patient has a bacterial infection and the type of bacteria that infects the patient.

[0111] Drug sensitivity test results: Drug sensitivity test results can guide doctors to choose appropriate antibiotics for anti-infection treatment.

[0112] like Figure 2 As shown, the present invention also provides a method for predicting infection in perioperative patients, which is implemented based on the perioperative patient infection prediction model. The prediction method of the present invention comprises the following steps:

[0113] S1: Obtain the first indicator of the patient's infection probability based on the patient's clinical information;

[0114] S2: Based on the environmental parameters in the operating room, a second indicator of infection index in the operating room is obtained;

[0115] S3: Based on the surgery-related information, output a third indicator of the probability of intraoperative infection;

[0116] S4: Output the fourth indicator of infection probability based on the use of antibiotics before and during surgery;

[0117] S5: Fit the first indicator, the second indicator, the third indicator and the fourth indicator, and output the patient's infection probability score or infection level according to the algorithm.

[0118] Preferably, the method further comprises step S6: providing guidance on medication to the patient according to the infection probability score or infection level.

[0119] Step S7: Obtain the patient's recovery status after surgery and evaluate the effects of different drugs on reducing infection;

[0120] and / or,

[0121] And according to the patient's recovery situation, the perioperative patient infection prediction model is optimized.

[0122] In this case, by evaluating the performance of each drug and its effect on infection prevention, we can implement gradient medication for patients in a more targeted manner, making the treatment of medical staff more scientific and reasonable.

[0123] Preferably, this example can not only guide patients to take medicine, but also optimize the workflow of medical staff, including disinfection and aseptic operation procedures, through the perioperative patient infection prediction method, to effectively prevent patients from being infected with bacteria and viruses during surgery and reduce the risk of postoperative infection. This example takes the optimization of the disinfection process as an example for explanation.

[0124] This example first evaluates the disinfection method. The disinfection method evaluation steps are:

[0125] The disinfection method used for infection control is used as a variable to input into the perioperative patient infection prediction model;

[0126] Evaluate the difference in effect with previous disinfection methods, and determine the influencing factors of the model based on the difference in effect;

[0127] Based on the influencing factors, determining a disinfection method evaluation sub-model;

[0128] Based on the disinfection method evaluation sub-model, obtaining a fourth indicator affecting infection prediction;

[0129] The fourth indicator is added to the perioperative patient infection prediction model to obtain an infection prediction result.

[0130] The infection prediction results obtained can effectively compare the differences in the effects of disinfection methods, thereby further optimizing the disinfection process. For example, the variables selected in this example are disinfection time, disinfection method, temperature, etc., and then a comprehensive evaluation is conducted based on the environmental parameters collected by sensors and detection devices. By optimizing the disinfection time, disinfection interval, and selection of disinfection methods, the optimal optimization process can be determined to improve the hospital's work efficiency and economic benefits.

[0131] It can be seen from the above description that the present invention establishes a perioperative patient infection prediction model, and performs infection prediction based on the perioperative patient infection prediction model, so as to effectively control the postoperative infection of the patient. The present invention has the following innovative features:

[0132] 1. Relying on big data analysis, a patient infection prediction model is established to provide decision-making support for infection control;

[0133] 2. Ensure sterility requirements by monitoring environmental parameters in the operating room, improve surgical safety and hospital operating efficiency;

[0134] 3. Ability to monitor the patient's condition throughout the entire process according to the model, promptly detect and treat signs of infection, and help medical staff develop personalized surgery and care plans based on the patient's condition;

[0135] 4. The perioperative patient infection prediction model can optimize the workflow of medical staff, including disinfection and aseptic operation procedures, effectively prevent patients from being infected with bacteria and viruses during surgery, and reduce the risk of postoperative infection;

[0136] 5. Guiding patients on medication and reducing infection rates can effectively reduce drug abuse, reduce patients' medical expenses and waste of hospital resources; through model prediction results, strengthen communication and learning among doctors, improve doctors' work ability, and enhance the overall strength of the hospital.

[0137] The specific implementation modes described above are preferred implementation modes of the present invention, and are not intended to limit the specific implementation scope of the present invention. The scope of the present invention includes but is not limited to the specific implementation modes, and all equivalent changes made according to the present invention are within the protection scope of the present invention.

Claims

1. A perioperative patient infection prediction model, characterized in that: include: Patient information acquisition module: used to obtain the patient's clinical information; The first indicator evaluation sub-model is used to output the first indicator of the patient's infection probability based on the patient's clinical information; Operating room information acquisition module: used to obtain environmental parameters in the operating room; The second indicator evaluation sub-model is used to output the second indicator of infection index in the operating room based on the environmental parameters in the operating room; Surgery information acquisition module: used to obtain surgery-related information, including surgery type, surgery site, surgery duration, intraoperative vital signs, and intraoperative input and output; The third indicator evaluation sub-model is used to output the third indicator of the probability of intraoperative infection based on the operation-related information; The fourth indicator evaluation sub-model is used to output the fourth indicator of infection probability according to the use of antibiotics before and during surgery; Prediction module: used to fit the first indicator, the second indicator, the third indicator and the fourth indicator, and output the patient's infection probability score or infection level according to the algorithm.

2. The perioperative patient infection prediction model according to claim 1, characterized in that: The perioperative patient infection prediction model also includes: Postoperative monitoring and evaluation sub-model: used to obtain analysis and evaluation results based on monitored vital signs, wound conditions and / or microbiological data; Infection prediction result dynamic adjustment module: used to adjust the patient's infection probability score or infection level based on the analysis and evaluation results and integrated big data.

3. The perioperative patient infection prediction model according to claim 1 or 2, characterized in that: The clinical information includes the patient's previous surgical history, chronic disease history, allergy history, the patient's physical signs and various relevant preoperative examination indicators. The first indicator evaluation submodel obtains infection-related influencing factors based on the previous surgical history, chronic disease history, allergy history, the patient's physical signs and various relevant preoperative examination indicators, and scores each influencing factor to obtain a first indicator including the total score of all influencing factors.

4. The perioperative patient infection prediction model according to claim 3, characterized in that: It also includes a correction module: based on the data big model and prediction results, the weights of various influencing factors are set and updated iteratively to obtain the weighted total score of the first indicator.

5. The perioperative patient infection prediction model according to claim 1 or 2, characterized in that: The environmental parameters are obtained through sensors and monitoring equipment, including air quality, temperature and humidity, pressure difference and / or microbial content.

6. A method for predicting infection in perioperative patients, based on the perioperative patient infection prediction model according to any one of claims 1 to 5, characterized in that: The perioperative patient infection prediction method comprises the following steps: S1: Obtain the first indicator of the patient's infection probability based on the patient's clinical information; S2: Based on the environmental parameters in the operating room, a second indicator of infection index in the operating room is obtained; S3: Based on the surgery-related information, output a third indicator of the probability of intraoperative infection; S4: Output the fourth indicator of infection probability based on the use of antibiotics before and during surgery; S5: Fit the first indicator, the second indicator, the third indicator and the fourth indicator, and output the patient's infection probability score or infection level according to the algorithm.

7. The method for predicting perioperative patient infection according to claim 6, characterized in that: The method further includes step S6: providing guidance on medication to the patient according to the infection probability score or infection level.

8. The method for predicting perioperative patient infection according to claim 7, characterized in that: The step S7 is also included: obtaining the patient's recovery status after surgery and evaluating the effects of different drugs on reducing infection; and / or, The perioperative patient infection prediction model is optimized according to the patient's recovery status.

9. The method for predicting perioperative patient infection according to claim 6, characterized in that: Also included are steps for evaluating disinfection methods: The disinfection method used for infection control is used as a variable to input into the perioperative patient infection prediction model; Evaluate the difference in effect with previous disinfection methods, and determine the influencing factors of the model based on the difference in effect; Based on the influencing factors, determining a disinfection method evaluation sub-model; Based on the disinfection method evaluation sub-model, obtaining a fourth indicator affecting infection prediction; The fourth indicator is added to the perioperative patient infection prediction model to obtain an infection prediction result.

Images