Drug resistance prediction method, drug resistance prediction apparatus, electronic device, and storage medium

Abstract

The present disclosure relates to a drug resistance prediction method, a drug resistance prediction apparatus, an electronic device, and a storage medium. The method comprises: acquiring prediction data of a target patient; and on the basis of the prediction data and a pre-constructed prediction model, determining drug resistance data of a bacterium infecting the target patient against a target antibiotic, wherein the prediction model is pre-constructed on the basis of historical data of sample patients, the prediction data comprises a plurality of pieces of indicator data of the target patient as model features of the prediction model, and the drug resistance data of the bacterium infecting the target patient against the target antibiotic is determined on the basis of a drug resistance value corresponding to each piece of indicator data among the plurality of pieces of indicator data and output by the prediction model, and a threshold corresponding to the piece of indicator data. Therefore, drug resistance data of a bacterium infecting a patient against an antibiotic can be quickly and accurately predicted.

Description

Drug resistance prediction method, drug resistance prediction device, electronic equipment and storage medium TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of prediction, in particular to a drug resistance prediction method, a drug resistance prediction device, an electronic equipment and a storage medium. BACKGROUND

[0002] Drug-resistant bacteria infection is a major threat to global public health. Inappropriate use of broad-spectrum antibacterial drugs is one of the causes of the emergence of drug-resistant bacteria. The hospital pneumonia diagnosis and treatment guidelines advocate that after a clear diagnosis, pathogenic examination should be sent as soon as possible, and antibacterial drugs should be used as soon as possible to start empirical anti-infection treatment. Due to the need to consider the risk of the emergence of drug-resistant bacteria, the guidelines propose that subsequent drug use should be adjusted in time according to the patient's treatment response, pathogenic examination results and drug sensitivity test results, and individualized treatment and target treatment should be carried out.

[0003] However, in actual clinical practice, the results of antibacterial drug sensitivity tests can only be obtained after several days of diagnosis, so there is a risk of improper selection of antibacterial drugs during empirical anti-infection treatment. In the related art, research shows that the probability of inappropriate use of antibacterial drugs in empirical anti-infection treatment is about 37.3% to 72.5%.

[0004] Therefore, how to improve the accuracy of drug selection in empirical anti-infection treatment is a problem to be solved. SUMMARY

[0005] Therefore, the present disclosure proposes a drug resistance prediction method, a drug resistance prediction device, an electronic equipment and a storage medium, which can quickly and accurately predict the drug resistance data of the bacteria infected by the patient to the antibiotic, and accurately select the antibacterial drugs to be used by the patient based on the quickly obtained drug resistance data, thereby improving the accuracy of drug selection in empirical anti-infection treatment.

[0006] According to a first aspect of the present disclosure, a drug resistance prediction method is provided for predicting drug resistance data of bacteria infected by a target patient to a target antibiotic, the drug resistance prediction method comprising: an acquisition step of acquiring prediction data of the target patient; a determination step of determining drug resistance data of bacteria infected by the target patient to the target antibiotic according to the prediction data and a prediction model constructed in advance, wherein the prediction model is constructed in advance according to historical data of sample patients, the prediction data comprises a plurality of index data of the target patient as model features of the prediction model, and the drug resistance data of bacteria infected by the target patient to the target antibiotic is determined according to a drug resistance value corresponding to each index data in the plurality of index data output by the prediction model and a threshold value corresponding to the index data.

[0007] In a possible implementation, the prediction model comprises a first prediction model and a second prediction model, the first prediction model is configured to make a prediction about bacteria that the patient is infected with, and the second prediction model is configured to make a prediction about drug resistance of the bacteria that the patient is infected with to an antibiotic, and accordingly, the determining comprises: inputting the prediction data into the first prediction model, the first prediction model outputs a first prediction result, the first prediction result comprises whether the target patient is infected with each of the bacteria in the list of bacteria and / or a probability of the target patient being infected with the bacteria; inputting the prediction data and the first prediction result into the second prediction model, the second prediction model outputs a second prediction result, the second prediction result comprises whether each of the bacteria is resistant to the target antibiotic and / or a probability of each of the bacteria being resistant to the target antibiotic.

[0008] In a possible implementation, the prediction model comprises a third prediction model, the third prediction model is configured to make a prediction about bacteria that the patient is infected with and drug resistance of the bacteria that the patient is infected with to an antibiotic, and accordingly, the determining comprises: inputting the prediction data into the third prediction model, the third prediction model outputs a third prediction result, the third prediction result comprises whether the target patient is infected with each of the bacteria in the list of bacteria and / or a probability of the target patient being infected with the bacteria, and whether each of the bacteria is resistant to the target antibiotic and / or a probability of each of the bacteria being resistant to the target antibiotic.

[0009] In a possible implementation, the list of bacteria comprises a list of bacteria that the target patient is likely to be infected with or a list of bacteria that sample patients are infected with.

[0010] In a possible implementation, whether the target patient is infected with each of the bacteria in the list of bacteria comprises at least one of whether the target patient is infected with bacteria of a corresponding bacterial class, whether the target patient is infected with bacteria of a corresponding bacterial family, whether the target patient is infected with bacteria of a corresponding bacterial genus, and whether the target patient is infected with bacteria of a corresponding bacterial species.

[0011] In a possible implementation, the plurality of index data comprises a use intensity of the target antibiotic within a predetermined time before the onset of the disease of the target patient.

[0012] In a possible implementation, the use intensity of the target antibiotic is calculated according to a number of days in which the target patient uses the target antibiotic within the predetermined time, a dose of a single target antibiotic, a frequency of using the target antibiotic per day, and a total number of days in which the target patient uses antibiotics within the predetermined time.

[0013] In a possible implementation, the target antibiotic includes a carbapenem antibiotic, a quinolone antibiotic and a third-generation cephalosporin antibiotic, and the determining step includes: inputting the prediction data into the prediction model, determining the drug resistance data of the bacteria currently infected by the target patient to the carbapenem antibiotic according to the drug resistance value corresponding to each index data in the first plurality of index data corresponding to the carbapenem antibiotic output by the prediction model and the threshold value corresponding to the index data, determining the drug resistance data of the bacteria currently infected by the target patient to the quinolone antibiotic according to the drug resistance value corresponding to each index data in the second plurality of index data corresponding to the quinolone antibiotic output by the prediction model and the threshold value corresponding to the index data, and determining the drug resistance data of the bacteria currently infected by the target patient to the third-generation cephalosporin antibiotic according to the drug resistance value corresponding to each index data in the third plurality of index data corresponding to the third-generation cephalosporin antibiotic output by the prediction model and the threshold value corresponding to the index data.

[0014] In a possible implementation, the target antibiotic is a carbapenem antibiotic, and the first plurality of index data includes the minimum value of C-reactive protein (CRP) of the target patient before the onset of the disease.

[0015] In a possible implementation, the first plurality of index data further includes the length of time of the target patient in an intensive care unit (ICU) before the onset of the disease.

[0016] In a possible implementation, the first plurality of index data further includes the length of hospitalization of the target patient before the onset of the disease and the length of time of the target patient using mechanical ventilation before the onset of the disease.

[0017] In a possible implementation, the first plurality of index data further includes the number of days of the carbapenem antibiotic used by the target patient before the onset of the disease.

[0018] In a possible implementation, the target antibiotic is a quinolone antibiotic, and the second plurality of index data includes the length of time of the target patient in an intensive care unit (ICU) before the onset of the disease and the length of hospitalization of the target patient before the onset of the disease.

[0019] In a possible implementation, the second plurality of index data further includes the incidence of drug-resistant bacteria to the quinolone antibiotic in the hospital where the target patient is treated within the predetermined time.

[0020] In a possible implementation, the second plurality of index data further includes the minimum value of serum albumin of the target patient before the onset of the disease.

[0021] In a possible implementation, the second plurality of index data further comprises a length of time that the target patient used mechanical ventilation before the onset of the disease.

[0022] In a possible implementation, the target antibiotic is a third-generation cephalosporin antibiotic, and the third plurality of index data comprises a length of time that the target patient stayed in an intensive care unit (ICU) before the onset of the disease and a length of hospital stay of the target patient before the onset of the disease.

[0023] In a possible implementation, the third plurality of index data further comprises a number of days that the target patient used antibacterial drugs before the onset of the disease.

[0024] In a possible implementation, the third plurality of index data further comprises an incidence of drug-resistant bacteria that produces the carbapenem antibiotic in a hospital where the target patient is treated within the predetermined time.

[0025] In a possible implementation, the third plurality of index data further comprises a maximum value of white blood cell count of the target patient before the onset of the disease.

[0026] In a possible implementation, the incidence of drug-resistant bacteria is calculated according to a number of drug-resistant bacteria and a number of non-drug-resistant bacteria, wherein the number of drug-resistant bacteria represents a total number of drug-resistant bacteria monitored by drug sensitivity tests of a plurality of patients in a hospital where the target patient is treated within the predetermined time, and the number of non-drug-resistant bacteria represents a total number of non-drug-resistant bacteria monitored by drug sensitivity tests of the plurality of patients in the hospital where the target patient is treated within the predetermined time.

[0027] In a possible implementation, the target antibiotic further comprises at least one of a first-generation cephalosporin antibiotic, a second-generation cephalosporin antibiotic, a fourth-generation cephalosporin antibiotic, a penicillin antibiotic, a penam antibiotic, a monobactam antibiotic, or a β-lactam-β-lactamase inhibitor compound preparation.

[0028] In a possible implementation, the model comprises a machine learning model.

[0029] In a possible implementation, the machine learning model comprises a neural network model, a logistic regression model, a decision tree model, or a naive Bayes model.

[0030] In a possible implementation, the drug resistance prediction method is used to predict drug resistance data of bacteria currently infected by the target patient to the target antibiotic.

[0031] In a possible implementation, the multiple index data are index data meeting a first condition selected from the prediction data, and the first condition comprises that the importance of the multiple index data output by the prediction model is higher than the importance of other index data of the prediction data.

[0032] In a possible implementation, the target antibiotic is a carbapenem antibiotic, and the first multiple index data comprise at least five of the following index data: the number of days of use of the carbapenem antibiotic by the target patient before the onset of the disease, the length of time of use of mechanical ventilation by the target patient before the onset of the disease, the minimum value of C-reactive protein CRP of the target patient before the onset of the disease, the value of the last procalcitonin PCT of the target patient before the onset of the disease, the length of hospitalization of the target patient before the onset of the disease, the length of time of the target patient in an intensive care unit ICU before the onset of the disease, the number of days of use of antibacterial drugs by the target patient before the onset of the disease, the incidence of drug-resistant bacteria of the carbapenem antibiotic in the hospital where the target patient is treated within the predetermined time, the maximum value of procalcitonin PCT of the target patient before the onset of the disease, the age of the target patient, the last serum albumin number of the target patient before the onset of the disease, the minimum value of serum albumin number of the target patient before the onset of the disease, the incidence of drug-resistant bacteria of the third-generation cephalosporin antibiotic in the hospital where the target patient is treated within the predetermined time, the maximum value of body temperature of the target patient before the onset of the disease, the incidence of drug-resistant bacteria of the quinolone antibiotic in the hospital where the target patient is treated within the predetermined time, the use intensity of the carbapenem antibiotic of the target patient before the onset of the disease, the use intensity of antibacterial drugs of the target patient before the onset of the disease, the maximum value of white blood cell number of the target patient before the onset of the disease, the number of strains of carbapenem antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, the number of strains of third-generation cephalosporin antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, the number of strains of quinolone antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, the number of days of use of the quinolone antibiotic of the target patient before the onset of the disease, the number of categories of antibiotics used by the target patient before the onset of the disease, the number of days of use of the third-generation cephalosporin antibiotic of the target patient before the onset of the disease, the use intensity of the third-generation cephalosporin antibiotic of the target patient before the onset of the disease, and the use intensity of the quinolone antibiotic of the target patient before the onset of the disease.

[0033] In a possible implementation, the target antibiotic is a carbapenem antibiotic, and the first plurality of index data includes at least five of the following index data: a number of days of use of the carbapenem antibiotic by the target patient before the onset of the disease, a length of time of use of mechanical ventilation by the target patient before the onset of the disease, a minimum value of C-reactive protein (CRP) of the target patient before the onset of the disease, a value of the last procalcitonin (PCT) of the target patient before the onset of the disease, a length of hospitalization of the target patient before the onset of the disease, a length of time in an intensive care unit (ICU) of the target patient before the onset of the disease, a number of days of use of antibacterial drugs by the target patient before the onset of the disease, an incidence of drug-resistant bacteria to the carbapenem antibiotic in a hospital where the target patient is treated within the predetermined time, a maximum value of procalcitonin (PCT) of the target patient before the onset of the disease, an age of the target patient, a last serum albumin value of the target patient before the onset of the disease, a minimum value of serum albumin of the target patient before the onset of the disease, a maximum value of body temperature of the target patient before the onset of the disease, an incidence of drug-resistant bacteria to the quinolone antibiotic in the hospital where the target patient is treated within the predetermined time, a use intensity of the carbapenem antibiotic by the target patient before the onset of the disease, a use intensity of antibacterial drugs by the target patient before the onset of the disease, and a maximum value of white blood cell count of the target patient before the onset of the disease.

[0034] In a possible implementation, the target antibiotic is a carbapenem antibiotic, and the first plurality of index data includes at least five of the following index data: a number of days of use of the carbapenem antibiotic by the target patient before the onset of the disease, a length of time of use of mechanical ventilation by the target patient before the onset of the disease, a minimum value of C-reactive protein (CRP) of the target patient before the onset of the disease, a value of the last procalcitonin (PCT) of the target patient before the onset of the disease, a length of hospitalization of the target patient before the onset of the disease, a length of time in an intensive care unit (ICU) of the target patient before the onset of the disease, a number of days of use of antibacterial drugs by the target patient before the onset of the disease, an incidence of drug-resistant bacteria to the carbapenem antibiotic in a hospital where the target patient is treated within the predetermined time, a maximum value of procalcitonin (PCT) of the target patient before the onset of the disease, a minimum value of serum albumin of the target patient before the onset of the disease, and an incidence of drug-resistant bacteria to the quinolone antibiotic in the hospital where the target patient is treated within the predetermined time.

[0035] In a possible implementation, the first plurality of index data further includes at least one of the following index data: an age of the target patient, a last serum albumin value of the target patient before the onset of the disease, a maximum value of body temperature of the target patient before the onset of the disease, a use intensity of the carbapenem antibiotic by the target patient before the onset of the disease, a use intensity of antibacterial drugs by the target patient before the onset of the disease, and a maximum value of white blood cell count of the target patient before the onset of the disease.

[0036] In a possible implementation, the first plurality of index data further comprises at least one of the following index data: an incidence of drug-resistant bacteria to the third-generation cephalosporin antibiotic in a hospital where the target patient is treated within the predetermined time, a strain number of the carbapenem antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a strain number of the third-generation cephalosporin antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a strain number of the quinolone antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a number of days of the quinolone antibiotic used by the target patient before the onset of the disease, a number of categories of antibiotics used by the target patient before the onset of the disease, a number of days of the third-generation cephalosporin antibiotic used by the target patient before the onset of the disease, a use intensity of the third-generation cephalosporin antibiotic used by the target patient before the onset of the disease, and a use intensity of the quinolone antibiotic used by the target patient before the onset of the disease.

[0037] In a possible implementation, the target antibiotic is a quinolone antibiotic, and the second plurality of index data includes at least five of the following index data: a length of time in an intensive care unit (ICU) before the onset of the target patient, an intensity of use of an antibacterial drug before the onset of the target patient, an incidence of a drug-resistant bacterium of the quinolone antibiotic in a hospital where the target patient is treated within a predetermined time, a value of a last procalcitonin (PCT) before the onset of the target patient, a strain number of the drug-resistant bacterium of the quinolone antibiotic in the hospital where the target patient is treated within the predetermined time, a maximum value of the PCT before the onset of the target patient, a length of hospital stay before the onset of the target patient, a number of days of use of the carbapenem antibiotic before the onset of the target patient, a minimum value of serum albumin before the onset of the target patient, an age of the target patient, a minimum value of C-reactive protein (CRP) before the onset of the target patient, an incidence of a drug-resistant bacterium of the carbapenem antibiotic in the hospital where the target patient is treated within the predetermined time, a number of days of use of the quinolone antibiotic before the onset of the target patient, a last serum albumin value before the onset of the target patient, a maximum value of body temperature before the onset of the target patient, a length of time of use of mechanical ventilation before the onset of the target patient, a number of days of use of an antibacterial drug before the onset of the target patient, an incidence of a drug-resistant bacterium of a third-generation cephalosporin in the hospital where the target patient is treated within the predetermined time, a maximum value of white blood cell count before the onset of the target patient, a strain number of the drug-resistant bacterium of the third-generation cephalosporin in the hospital where the target patient is treated within the predetermined time, a strain number of the drug-resistant bacterium of the carbapenem antibiotic in the hospital where the target patient is treated within the predetermined time, an intensity of use of the carbapenem antibiotic before the onset of the target patient, a number of categories of antibiotics used before the onset of the target patient, an intensity of use of the quinolone antibiotic before the onset of the target patient, a number of days of use of the third-generation cephalosporin before the onset of the target patient, and an intensity of use of the third-generation cephalosporin before the onset of the target patient.

[0038] In a possible implementation, the target antibiotic is a quinolone antibiotic, and the second multiple index data includes at least five of the following index data: a length of time in an intensive care unit (ICU) before the onset of the target patient, an intensity of use of an antibacterial drug before the onset of the target patient, an incidence of drug-resistant bacteria to the quinolone antibiotic in a hospital where the target patient is treated within a predetermined time, a value of a last procalcitonin (PCT) before the onset of the target patient, a strain number of the quinolone antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a maximum value of the PCT before the onset of the target patient, a length of hospitalization before the onset of the target patient, a number of days of use of the carbapenem antibiotic before the onset of the target patient, a minimum value of serum albumin before the onset of the target patient, an age of the target patient, a minimum value of C-reactive protein (CRP) before the onset of the target patient, an incidence of drug-resistant bacteria to the carbapenem antibiotic in the hospital where the target patient is treated within the predetermined time, a last serum albumin value before the onset of the target patient, a maximum value of body temperature before the onset of the target patient, a length of time of use of mechanical ventilation before the onset of the target patient, a number of days of use of the antibacterial drug before the onset of the target patient, an incidence of drug-resistant bacteria to the third-generation cephalosporin antibiotic in the hospital where the target patient is treated within the predetermined time, a maximum value of white blood cell count before the onset of the target patient, and a strain number of the third-generation cephalosporin antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time.

[0039] In a possible implementation, the target antibiotic is a quinolone antibiotic, and the second multiple index data includes at least five of the following index data: a length of time in an intensive care unit (ICU) before the onset of the target patient, an intensity of use of an antibacterial drug before the onset of the target patient, an incidence of drug-resistant bacteria to the quinolone antibiotic in a hospital where the target patient is treated within a predetermined time, a value of a last procalcitonin (PCT) before the onset of the target patient, a strain number of the quinolone antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a maximum value of the PCT before the onset of the target patient, a length of hospitalization before the onset of the target patient, a minimum value of serum albumin before the onset of the target patient, an age of the target patient, a minimum value of C-reactive protein (CRP) before the onset of the target patient, an incidence of drug-resistant bacteria to the carbapenem antibiotic in the hospital where the target patient is treated within the predetermined time, a length of time of use of mechanical ventilation before the onset of the target patient, a number of days of use of the antibacterial drug before the onset of the target patient, a maximum value of white blood cell count before the onset of the target patient.

[0040] In a possible implementation, the second plurality of index data further comprises at least one of the following index data: a number of days of use of the carbapenem antibiotic before the onset of the disease of the target patient, a last serum albumin value before the onset of the disease of the target patient, a maximum body temperature before the onset of the disease of the target patient, an incidence of the third-generation cephalosporin-resistant bacteria in a hospital where the target patient is treated within the predetermined time, and a number of strains of the third-generation cephalosporin-resistant bacteria in the hospital where the target patient is treated within the predetermined time.

[0041] In a possible implementation, the second plurality of index data further comprises at least one of the following index data: a number of days of use of the quinolone antibiotic before the onset of the disease of the target patient, a number of strains of the carbapenem antibiotic-resistant bacteria in a hospital where the target patient is treated within the predetermined time, a use intensity of the carbapenem antibiotic before the onset of the disease of the target patient, a number of categories of antibiotics used before the onset of the disease of the target patient, a use intensity of the quinolone antibiotic before the onset of the disease of the target patient, a number of days of use of the third-generation cephalosporin before the onset of the disease of the target patient, and a use intensity of the third-generation cephalosporin before the onset of the disease of the target patient.

[0042] In a possible implementation, the target antibiotic is a third-generation cephalosporin antibiotic, and the third plurality of index data further includes at least five of the following index data: a number of days of use of an antibacterial drug before the onset of the target patient, a length of time in an intensive care unit (ICU) before the onset of the target patient, a minimum value of serum albumin before the onset of the target patient, an age of the target patient, a number of days of use of the carbapenem antibiotic before the onset of the target patient, an incidence of drug-resistant bacteria to the quinolone antibiotic in a hospital where the target patient is treated within the predetermined time, a maximum value of procalcitonin (PCT) before the onset of the target patient, a minimum value of C-reactive protein (CRP) before the onset of the target patient, a length of hospital stay before the onset of the target patient, an incidence of drug-resistant bacteria to the carbapenem antibiotic in a hospital where the target patient is treated within the predetermined time, an intensity of use of an antibacterial drug before the onset of the target patient, a number of strains of carbapenem antibiotic-resistant bacteria in a hospital where the target patient is treated within the predetermined time, a value of procalcitonin (PCT) at the last time before the onset of the target patient, a maximum value of white blood cell count before the onset of the target patient, a number of strains of third-generation cephalosporin antibiotic-resistant bacteria in a hospital where the target patient is treated within the predetermined time, a last value of serum albumin before the onset of the target patient, a number of strains of quinolone antibiotic-resistant bacteria in a hospital where the target patient is treated within the predetermined time, a length of time of use of mechanical ventilation before the onset of the target patient, an incidence of drug-resistant bacteria to the third-generation cephalosporin antibiotic in a hospital where the target patient is treated within the predetermined time, an intensity of use of the carbapenem antibiotic before the onset of the target patient, a maximum value of body temperature before the onset of the target patient, a number of days of use of the quinolone antibiotic before the onset of the target patient, a number of categories of antibiotics used before the onset of the target patient, a number of days of use of the third-generation cephalosporin antibiotic before the onset of the target patient, an intensity of use of the third-generation cephalosporin antibiotic before the onset of the target patient, and an intensity of use of the quinolone antibiotic before the onset of the target patient.

[0043] In a possible implementation, the target antibiotic is a third-generation cephalosporin antibiotic, and the third multi-index data further include at least five of the following index data: the number of days of use of antibacterial drugs before the onset of the target patient, the length of time in an intensive care unit (ICU) before the onset of the target patient, the minimum value of serum albumin before the onset of the target patient, the age of the target patient, the number of days of use of the carbapenem antibiotic before the onset of the target patient, the incidence of drug-resistant bacteria to the quinolone antibiotic in the hospital where the target patient is treated within the predetermined time, the maximum value of procalcitonin (PCT) before the onset of the target patient, the minimum value of C-reactive protein (CRP) before the onset of the target patient, the length of hospital stay before the onset of the target patient, the incidence of drug-resistant bacteria to the carbapenem antibiotic in the hospital where the target patient is treated within the predetermined time, the intensity of use of antibacterial drugs before the onset of the target patient, the number of strains of carbapenem antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, the value of the last procalcitonin (PCT) before the onset of the target patient, the maximum value of white blood cell count before the onset of the target patient, the last serum albumin before the onset of the target patient, the number of strains of quinolone antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, the length of time of use of mechanical ventilation before the onset of the target patient, and the intensity of use of the carbapenem antibiotic before the onset of the target patient.

[0044] In a possible implementation, the target antibiotic is a third-generation cephalosporin antibiotic, and the third multi-index data further include at least five of the following index data: the number of days of use of antibacterial drugs before the onset of the target patient, the length of time in an intensive care unit (ICU) before the onset of the target patient, the minimum value of serum albumin before the onset of the target patient, the age of the target patient, the number of days of use of the carbapenem antibiotic before the onset of the target patient, the maximum value of procalcitonin (PCT) before the onset of the target patient, the minimum value of C-reactive protein (CRP) before the onset of the target patient, the length of hospital stay before the onset of the target patient, the incidence of drug-resistant bacteria to the carbapenem antibiotic in the hospital where the target patient is treated within the predetermined time, the intensity of use of antibacterial drugs before the onset of the target patient, the number of strains of carbapenem antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, the value of the last procalcitonin (PCT) before the onset of the target patient, the maximum value of white blood cell count before the onset of the target patient, the length of time of use of mechanical ventilation before the onset of the target patient.

[0045] In a possible implementation, the third plurality of index data further comprises at least one of the following index data: an incidence of drug-resistant bacteria to the quinolone antibiotic in a hospital where the target patient is treated within the predetermined time, a number of strains of the carbapenem antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a last serum white blood cell count of the target patient before the onset of the disease, a number of strains of the quinolone antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a use intensity of the carbapenem antibiotic of the target patient before the onset of the disease.

[0046] In a possible implementation, the third plurality of index data further comprises at least one of the following index data: a number of strains of the third-generation cephalosporin antibiotic-resistant bacteria in a hospital where the target patient is treated within the predetermined time, an incidence of drug-resistant bacteria to the third-generation cephalosporin antibiotic in the hospital where the target patient is treated within the predetermined time, a maximum body temperature of the target patient before the onset of the disease, a number of days of the quinolone antibiotic used by the target patient before the onset of the disease, a number of categories of antibiotics used by the target patient before the onset of the disease, a number of days of the third-generation cephalosporin antibiotic used by the target patient before the onset of the disease, a use intensity of the third-generation cephalosporin antibiotic of the target patient before the onset of the disease, a use intensity of the quinolone antibiotic of the target patient before the onset of the disease.

[0047] In a possible implementation, the prediction data is data of the target patient acquired 2 days before the onset of the disease.

[0048] According to a second aspect of the present disclosure, a drug resistance prediction apparatus is provided for predicting drug resistance data of bacteria infected by a target patient to a target antibiotic, the drug resistance prediction apparatus comprising: an acquisition module configured to acquire prediction data of the target patient; and a determination module configured to determine, according to the prediction data and a prediction model pre-constructed, the drug resistance data of the bacteria infected by the target patient to the target antibiotic, wherein the prediction model is pre-constructed according to historical data of sample patients, the prediction data comprises a plurality of index data of the target patient as model features of the prediction model, and the drug resistance data of the bacteria infected by the target patient to the target antibiotic is determined according to a drug resistance value corresponding to each of the plurality of index data output by the prediction model and a threshold value corresponding to the index data.

[0049] According to a third aspect of the present disclosure, an electronic device is provided, comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to implement the above drug resistance prediction method when executing the instructions stored in the memory.

[0050] According to a fourth aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium having stored thereon computer program instructions which, when executed by a processor, implement the drug resistance prediction method described above.

[0051] According to a fifth aspect of the present disclosure, there is provided a computer program product comprising computer readable code, or a non-transitory computer-readable storage medium carrying computer readable code, which when run in a processor of an electronic device, the processor in the electronic device implements the drug resistance prediction method described above.

[0052] According to the drug resistance prediction method and device of the present disclosure, a prediction model for predicting drug resistance data of bacteria infected by a target patient to a target antibiotic is constructed in advance according to historical data of the sample patient, so that when an empirical anti-infection treatment is to be performed on the target patient, only prediction data of the target patient needs to be obtained, and the obtained prediction data is input into the prediction model constructed in advance, so that the drug resistance data of the bacteria infected by the target patient to the target antibiotic can be quickly and accurately predicted. Therefore, based on the quickly predicted drug resistance data, the antibacterial drug to be used by the target patient can be accurately selected, so as to improve the accuracy of drug selection for the empirical anti-infection treatment performed on the target patient.

[0053] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, but not limiting the present disclosure.

[0054] Other features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments with reference to the drawings. BRIEF DESCRIPTION OF DRAWINGS

[0055] The drawings herein are incorporated into the specification and constitute a part of the specification, which illustrate embodiments consistent with the present disclosure, and together with the specification serve to explain the technical solutions of the present disclosure.

[0056] FIG. 1 shows a flowchart of a drug resistance prediction method according to an exemplary embodiment.

[0057] FIG. 2 shows a flowchart of a drug resistance prediction method according to an exemplary embodiment.

[0058] FIG. 3 shows a schematic diagram of a prediction model according to an exemplary embodiment.

[0059] FIG. 4 shows a schematic diagram of a prediction model according to an exemplary embodiment.

[0060] FIG. 5 shows a schematic diagram of a prediction model according to an exemplary embodiment.

[0061] FIG. 6 shows a schematic diagram of a prediction model according to an exemplary embodiment.

[0062] Figure 7 shows a schematic diagram of a prediction model according to an exemplary embodiment.

[0063] Figure 8 shows a schematic diagram of a prediction model according to an exemplary embodiment.

[0064] Figure 9 shows a block diagram of an antimicrobial resistance prediction device according to an exemplary embodiment.

[0065] Figure 10 shows a block diagram of an electronic device according to an exemplary embodiment. Detailed Implementation

[0066] Various exemplary embodiments, features, and aspects of this disclosure will now be described in detail with reference to the accompanying drawings. The same reference numerals in the drawings denote elements that have the same or similar functions. Although various aspects of the embodiments are shown in the drawings, they are not necessarily drawn to scale unless specifically indicated otherwise.

[0067] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments.

[0068] In this document, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. Furthermore, the term "at least one" in this document means any combination of at least two of any one or more elements. For example, including at least one of A, B, and C can mean including any one or more elements selected from the set consisting of A, B, and C.

[0069] Furthermore, to better illustrate this disclosure, numerous specific details are set forth in the following detailed description. Those skilled in the art will understand that this disclosure can be practiced without certain specific details. In some instances, methods, means, components, and circuits well known to those skilled in the art have not been described in detail in order to highlight the main points of this disclosure.

[0070] Figure 1 shows a flowchart of a drug resistance prediction method according to an embodiment of the present disclosure, which is used to predict the drug resistance data of bacteria infecting a target patient to a target antibiotic. In one possible implementation, the target patient includes, but is not limited to, a hospital-acquired pneumonia patient, and the target antibiotic includes, but is not limited to, carbapenems, quinolones, and third-generation cephalosporins. As shown in Figure 1, the drug resistance prediction method may include the following steps.

[0071] In step S110 (acquisition step), predictive data for the target patient is acquired.

[0072] In this embodiment, the prediction data of the target patient can be any data of the target patient that can be used to predict the drug resistance data of the bacteria infected by the target patient to the target antibiotic. In one possible implementation, the prediction data can include, but is not limited to, patient information data and / or hospital environment information data.

[0073] It should be understood that the patient information data can represent data related to patient information, such as the patient information data can include, but is not limited to, the following data: age, name, gender, body temperature, length of hospital stay, white blood cell count, serum albumin, antibiotic use intensity, antibiotic use days, length of stay in ICU, length of mechanical ventilation, PCT value, antibiotic class used, and the like. The hospital environment information data can represent data related to the environment information in the hospital where the patient is treated, such as the hospital environment information data can include, but is not limited to, the following data: the incidence of drug-resistant bacteria to the corresponding antibiotic in the hospital where the patient is treated, the number of strains of the corresponding antibiotic-resistant bacteria in the hospital where the patient is treated, and the like.

[0074] When the target patient is to be treated empirically, the target patient's corresponding index data can be collected as the prediction data, and the collected index data can include, but is not limited to, the above exemplary data.

[0075] It should be understood that any suitable method in the prior art can be used to collect the corresponding index data of the target patient, and the collection methods of different types of index data can be the same or different, and the embodiment does not make specific limitations on the collection method of the index data.

[0076] For example, assuming that the index data of the target patient includes the age of the target patient, the age of the target patient can be obtained by scanning the target patient's ID card, or the age of the target patient can be obtained by retrieving the medical record of the target patient from the server of the hospital where the target patient is treated, or the age of the target patient can be obtained by scanning the medical insurance card used by the target patient when checking into the hospital, and the like.

[0077] In one possible implementation, the prediction data is the data of the target patient obtained 2 days before the onset. In this way, the accuracy of the predicted drug resistance data of the bacteria infected by the target patient to the target antibiotic can be improved.

[0078] In step S130 (determination step), the drug resistance data of the bacteria infected by the target patient to the target antibiotic is determined according to the prediction data and a prediction model constructed in advance, wherein the prediction model is constructed in advance according to historical data of sample patients, the prediction data includes multiple index data of the target patient as model features of the prediction model, and the drug resistance data of the bacteria infected by the target patient to the target antibiotic is determined according to the drug resistance value corresponding to each index data in the multiple index data output by the prediction model and the threshold value corresponding to the index data.

[0079] In this embodiment, a prediction model for predicting the drug resistance data of the bacteria infected by the target patient to the target antibiotic can be constructed in advance according to historical data of multiple sample patients, the input data of the prediction model is multiple index data, and the output data of the prediction model is a drug resistance value corresponding to each index data in the multiple index data. The construction method of the prediction model will be described below. It should be understood that the historical data is medical history data, which can include patient information data and hospital environment information data. The types of index data included in the historical data of multiple sample patients can be the same as and / or different from the types of prediction data of the target patient.

[0080] After the prediction model is constructed in advance, when the drug resistance data of the bacteria infected by the target patient to the target antibiotic is to be predicted, after the multiple index data of the target patient is obtained to be used as the prediction data of the target patient, the multiple index data and the prediction model constructed in advance are used to determine the drug resistance data of the bacteria infected by the target patient to the target antibiotic. The drug resistance data of the bacteria infected by the target patient to the target antibiotic can include, but is not limited to, whether the bacteria infected by the target patient is resistant to the target antibiotic and / or the probability that the bacteria infected by the target patient is resistant to the target antibiotic, etc.

[0081] Specifically, the obtained multiple index data can be input into the prediction model, i.e., the multiple index data is used as the input data of the prediction model, in response to inputting the multiple index data, the prediction model outputs a drug resistance value corresponding to each index data in the multiple index data, so that the output data of the prediction model is the drug resistance value, and the drug resistance data of the bacteria infected by the target patient to the target antibiotic can be determined according to the drug resistance value corresponding to each index data and the threshold value corresponding to the index data.

[0082] For example, assuming that five pieces of index data of a target patient are obtained, i.e., first index data, second index data, third index data, fourth index data and fifth index data, and the threshold values corresponding to the five pieces of index data are first threshold value, second threshold value, third threshold value, fourth threshold value and fifth threshold value respectively, the five pieces of index data can be input into the prediction model, and thus first drug resistance value, second drug resistance value, third drug resistance value, fourth drug resistance value and fifth drug resistance value corresponding to the five pieces of index data can be obtained. The drug resistance data of the bacteria infected by the target patient to the target antibiotic can be determined according to the first drug resistance value and the first threshold value, the second drug resistance value and the second threshold value, the third drug resistance value and the third threshold value, the fourth drug resistance value and the fourth threshold value, and the fifth drug resistance value and the fifth threshold value.

[0083] In an application scenario, the prediction model can be constructed using the medical history data of a plurality of hospital patients after anonymization processing. For a target hospital-acquired pneumonia patient A, the historical data of the patient A can be input into the prediction model, and the prediction model can output the prediction result of the patient A, i.e., the bacteria infected by the patient A, whether the bacteria is resistant to the corresponding antibiotic and / or the probability that the bacteria is resistant to the corresponding antibiotic. In this way, the doctor can be assisted to select a more appropriate antibiotic for the target hospital-acquired pneumonia patient A more quickly, so as to improve the treatment effect on the target hospital-acquired pneumonia patient A.

[0084] In a possible implementation, the pre-constructed prediction model is a model that is pre-trained and tested. As shown in FIG. 2, the pre-construction, training and testing of the prediction model can include the following steps.

[0085] In step S210, the historical data of a sample patient is obtained.

[0086] In this embodiment, the historical data of the sample patient represents any data of the sample patient that can be used to predict the drug resistance data of the bacteria infected by the sample patient to the target antibiotic. In a possible implementation, the historical data of the sample patient can include but is not limited to patient information data and / or hospital environment information data. For the patient information data and the hospital environment information data, refer to the foregoing description of step S110, which will not be described here.

[0087] In step S220, a feature data set is constructed.

[0088] In this embodiment, after obtaining the historical data of the sample patient, the historical data can be filtered to filter out interference information, and then the historical data can be processed to convert the historical data into a format consistent with the standard data of the prediction model. The data after format conversion constitutes the feature data set.

[0089] In step S230, the training data set and the test data set are divided.

[0090] In this embodiment, the constructed feature data set can be divided into the training data set and the test data set according to a certain proportion, for example, the training data set accounts for 80%, and the test data set accounts for 20%. It should be understood that the proportions of the training data set and the test data set can be dynamically adjusted according to actual application requirements to obtain a prediction model meeting the actual application requirements.

[0091] In step S240, the training data set is used to train the prediction model.

[0092] In this embodiment, the training data set divided can be used to train the prediction model. As shown in FIG. 3, the training data is data X, and the data X includes data X t-2 , data X t-1 , data X t , and data X t-2 represent data corresponding to time t-2, data X t-1 represent data corresponding to time t-1, and data X t represent data corresponding to time t. These data are sequentially input into the prediction model, and the prediction model outputs the drug resistance data of the bacteria infecting the target patient to the first target antibiotic, the drug resistance data of the bacteria infecting the target patient to the second target antibiotic, and the drug resistance data of the bacteria infecting the target patient to the third target antibiotic.

[0093] In a possible implementation manner, the prediction model includes a machine learning model.

[0094] In a possible implementation manner, the machine learning model includes a neural network model, a logistic regression model, a decision tree model, or a naive Bayes model.

[0095] In this embodiment, the prediction model can select a machine learning model, such as a neural network model, a logistic regression model, a decision tree model, a naive Bayes model, etc., and the model outputs multiple binary classification labels or probabilities of multiple labels, which are used to predict whether or the probability of the type of bacteria (bacterial class / bacterial family / bacterial genus / bacterial species can be predicted) currently infected by the target patient (for example, b1 is the prediction result of enterobacteriaceae / enterobacteriaceae / escherichia coli, b2 is the prediction result of pseudomonas / pseudomonas / pseudomonas aeruginosa, and b3 is the prediction result of pseudomonas / moraxellaceae / acinetobacter baumannii), and can also predict whether or the probability of the drug resistance of the bacteria currently infected by the target patient to each class / each species of antibiotic (for example, i1 is the prediction result of carbapenem antibiotic / meropenem drug resistance, i2 is the prediction result of third-generation cephalosporin antibiotic / ceftazidime drug resistance, and i3 is the prediction result of fluoroquinolone / levofloxacin antibiotic drug resistance).

[0096] The index data input into the model is the model feature, and 85 model features are constructed by collecting sample data during modeling. The method for selecting features can use the feature importance provided by the model, and the importance of each model feature to the model and the importance type of the model parameter are obtained during the model training process, and important features are automatically selected.

[0097] Of course, the Shap value can also be used to represent the influence of each model feature on the model and the positive and negative nature of the feature influence, and important features are selected accordingly.

[0098] It should be understood that the model can not only predict the probability of the type of bacteria currently infected by the target patient and the drug resistance of the bacteria currently infected by the target patient to each class / each species of antibiotic, but also can first predict the type of bacteria currently infected by the target patient, and use this result as one of the input data (i.e., one of the input Y described below) to output the corresponding prediction result, which is actually the prediction result of the drug resistance of the bacteria currently infected by the target patient to each class / each species of antibiotic.

[0099] In step S250, the trained prediction model is tested using the test data set.

[0100] In this embodiment, the test data set can be used to test the trained prediction model. In one possible implementation, each test data in the test data set can be input into the trained prediction model in sequence, and the prediction model outputs the drug resistance data of the bacteria infected by the target patient to the first target antibiotic, the drug resistance data of the bacteria infected by the target patient to the second target antibiotic, and the drug resistance data of the bacteria infected by the target patient to the third target antibiotic. The drug resistance data can be compared with the corresponding drug resistance results. If the deviation is within the allowable range, it indicates that the parameters of the prediction model are appropriate, and the prediction model can be used to predict the drug resistance data of the target antibiotic to the bacteria infected by the target patient. Otherwise, the parameters of the prediction model are adjusted, and the training and testing are performed again until the deviation is within the allowable range.

[0101] After the prediction model is completed, trained and tested, the prediction model can be used to predict the drug resistance data of the target antibiotic (including the first target antibiotic, the second target antibiotic and the third target antibiotic) to the bacteria infected by the target patient.

[0102] In step S260, the prediction data of the target patient is obtained. The prediction data is processed into index data conforming to the format of the prediction model.

[0103] In step S270, the prediction data of the target patient and the trained and tested prediction model are used to determine the drug resistance data of the target antibiotic to the bacteria infected by the target patient.

[0104] In this embodiment, the prediction data of the target patient is input into the trained and tested prediction model to determine the drug resistance data of the target antibiotic to the bacteria infected by the target patient.

[0105] According to this embodiment, the prediction model for predicting the drug resistance data of the target antibiotic to the bacteria infected by the target patient is constructed in advance according to the historical data of the sample patient. When the empirical anti-infective treatment is to be performed on the target patient, only the prediction data of the target patient needs to be obtained, and the obtained prediction data is input into the prediction model constructed in advance, so that the drug resistance data of the target antibiotic to the bacteria infected by the target patient can be quickly and accurately predicted. Therefore, the antibacterial drug used by the target patient can be accurately selected based on the quickly predicted drug resistance data, so that the accuracy of the drug selection for the empirical anti-infective treatment of the target patient can be improved.

[0106] Thus, compared with the result of the drug resistance of the bacteria infecting the target patient to the target antibiotic obtained after several days of diagnosis by using the antibacterial drug sensitivity test in the prior art, the present disclosure obtains the prediction data of the target patient at the time of diagnosis, inputs the prediction data into the prediction model constructed, trained and tested in advance, and quickly obtains the drug resistance data of the bacteria infecting the target patient to the target antibiotic, in other words, the present disclosure can obtain the drug resistance result of the bacteria infecting the target patient to the target antibiotic at the time of diagnosis.

[0107] For example, in actual application, the prediction data of the target patient can be collected, an interface can be called to predict the drug resistance data of the bacteria infecting the target patient to the target antibiotic using the prediction data, the corresponding model can be called in the background to predict the drug resistance data of the bacteria infecting the target patient to the corresponding target antibiotic, and the predicted drug resistance data of the bacteria infecting the target patient to the corresponding target antibiotic can be output via the interface. The target antibiotic can include one or more antibiotics, so that the drug resistance result of the bacteria infecting the target patient to the target antibiotic can be obtained at the time of diagnosis.

[0108] It should be understood that the prediction of the drug resistance data of the bacteria infecting the target patient to the target antibiotic can be customized according to the actual application requirements. For example, if it is obtained that the doctor needs to predict the drug resistance data of the bacteria infecting the target patient to three types of antibiotics, all the index data of the target patient can be input into the model, and the drug resistance data of the bacteria infecting the target patient to the three types of antibiotics can be output, so that the operation convenience can be improved.

[0109] For another example, if it is obtained that the doctor needs to predict the drug resistance data of the bacteria infecting the target patient to the three types of antibiotics respectively and the doctor needs to give the interpretation of which type of antibiotic each index data belongs to, the corresponding index data of the target patient belonging to the three types of antibiotics can be displayed, and the index data can be input into the model to output the drug resistance data of the bacteria infecting the target patient to the antibiotic, so that more accurate prediction result can be provided.

[0110] In a possible implementation, the drug resistance prediction method is used to predict the drug resistance data of the bacteria currently infecting the target patient to the target antibiotic.

[0111] Although the above-mentioned drug resistance prediction method can be used to predict the drug resistance data of the bacteria past infected by the target patient to the target antibiotic, in the embodiment, the drug resistance prediction method is used to predict the drug resistance data of the bacteria currently infected by the target patient to the target antibiotic, so that the drug resistance data of the bacteria currently infected by the target patient to the target antibiotic can be quickly and accurately predicted, and based on the quickly predicted drug resistance data, the antibacterial drug currently used by the target patient can be accurately selected, thereby improving the accuracy of drug selection for the empirical anti-infective treatment of the target patient.

[0112] In a possible implementation, the multiple index data is index data meeting a first condition selected from the prediction data, and the first condition includes that the importance of the multiple index data output by the prediction model is higher than the importance of other index data of the prediction data.

[0113] In the embodiment, considering that a large amount of index data of the patient can be obtained, if all the index data is input into the prediction model, the construction of the model can become complex and the time required for prediction using the model can increase, therefore, in order to more quickly and accurately determine the drug resistance data of the bacteria infected by the target patient to the target antibiotic, the index data can be sorted according to the feature importance of the model, and index data meeting a first condition in the sorting is selected as the prediction data.

[0114] In a possible implementation, the prediction model includes a first prediction model and a second prediction model, the first prediction model is used to make a prediction related to the bacteria infected by the patient, and the second prediction model is used to make a prediction related to the drug resistance of the bacteria infected by the patient to the antibiotic, accordingly, step S130 can include: inputting the prediction data into the first prediction model, the first prediction model outputs a first prediction result, the first prediction result includes whether the target patient is infected with each of the bacteria in the list of bacteria and / or the infection probability of the target patient to the each of the bacteria; inputting the prediction data and the first prediction result into the second prediction model, the second prediction model outputs a second prediction result, the second prediction result includes whether the each of the bacteria is resistant to the target antibiotic and / or the resistance probability of the each of the bacteria to the target antibiotic.

[0115] In the embodiment, as shown in FIG. 4, input X represents prediction data, and the prediction data can include data X t-2 , X t-1 , X t, the output X represents the first prediction result, which for example represents whether the target patient is infected with the corresponding bacteria, which can be refined to class, family, genus, species, etc. In this way, the prediction data X t-2 , X t-1 , X t is input into the first prediction model, so as to obtain the first prediction result such as whether the target patient is infected with Escherichia, and even more, the first prediction result such as whether the target patient is infected with Escherichia coli.

[0116] Correspondingly, as shown in FIG. 5, the input X represents the prediction data, which can include data X t-2 , X t-1 , X t , the output X represents the first prediction result, which for example represents the probability of the target patient being infected with the corresponding bacteria, which can be refined to class, family, genus, species, etc. In this way, the prediction data X t-2 , X t-1 , X t is input into the first prediction model, so as to obtain the first prediction result such as the probability of the target patient being infected with Escherichia being 89%, and even more, the first prediction result such as the probability of the target patient being infected with Escherichia coli being 89%.

[0117] Correspondingly, as shown in FIG. 6, the input X represents the prediction data, which can include data X t-2 , X t-1 , X t , the output X represents the first prediction result, which for example represents whether the bacteria infected by the target patient is resistant to the corresponding antibiotic, which can be all categories of antibiotics or common categories of antibiotics. In this way, the prediction data X t-2 , X t-1 , X t is input into the second prediction model, so as to obtain the prediction result such as whether the bacteria infected by the target patient is resistant to carbapenem antibiotics, and even more, the prediction result such as whether the bacteria infected by the target patient is resistant to meropenem.

[0118] Correspondingly, as shown in FIG. 7, the input X represents the prediction data, which can include data X t-2 , X t-1 , X t , the output X represents the first prediction result, which for example represents whether the bacteria infected by the target patient is resistant to the corresponding antibiotic, which can be all categories of antibiotics or common categories of antibiotics or the antibiotic drug name and the resistance probability of the antibiotic to which the bacteria infected by the target patient can be resistant. In this way, the prediction data X t-2 , X t-1, X t The input is inputted into the third prediction model, so that a prediction result such as a probability that the bacteria infected in the target patient is resistant to carbapenem antibiotics is 89% is obtained, and a prediction result such as a probability that the bacteria infected in the target patient is resistant to meropenem is 76% is further obtained.

[0119] That is, the prediction model can include the first prediction model for predicting the bacteria infected in the target patient, and can further include the third prediction model for predicting the resistance data of the bacteria infected in the target patient to the corresponding antibiotic, so that a prediction result indicating whether the target patient is infected with the corresponding bacteria and / or a probability that the target patient is infected with the corresponding bacteria is obtained using the first prediction model, and a prediction result indicating whether the bacteria infected in the target patient is resistant to the corresponding antibiotic and / or a probability that the bacteria infected in the target patient is resistant to the corresponding antibiotic is further obtained using the third prediction model.

[0120] Thus, as shown in FIG. 8, input X indicates prediction data, the prediction model includes the first prediction model and the second prediction model, the second prediction model is used to make a prediction related to the resistance of the bacteria infected in the patient to the antibiotic, and output X and output Y respectively indicate the first prediction result and the second prediction result, output X is, for example, a prediction result composed of "bacterial species + prediction result", and output Y is, for example, a prediction result indicating whether the bacteria infected in the target patient is resistant to the corresponding antibiotic or a probability that the bacteria infected in the target patient is resistant to the corresponding antibiotic. The prediction data can include data X t-2 , X t-1 , X t The prediction data X t-2 , X t-1 , X t may be inputted into the first prediction model first, so that a prediction result of output X is obtained, and then, the prediction result of output X and input X are taken as input Y, and input Y is inputted into the second prediction model, so that a prediction result of output Y is obtained.

[0121] Although the second prediction model and the third prediction model are both used to make a prediction related to the resistance of the bacteria infected in the patient to the antibiotic, the inputs thereof are different, the input of the former is the prediction data and the first prediction result outputted by the first prediction model, and the input of the latter is only the prediction data.

[0122] In one possible implementation, the prediction model includes a third prediction model for making predictions related to the bacteria infecting the patient and the antibiotic resistance of the bacteria infecting the patient. Accordingly, step S130 may include: inputting the prediction data into the prediction model, the prediction model outputting a third prediction result, the third prediction result including whether the target patient is infected with each bacteria in the bacterial list and / or the probability of the target patient being infected with each bacteria, and whether each bacteria is resistant to the target antibiotic and / or the probability of each bacteria being resistant to the target antibiotic.

[0123] In this embodiment, as shown in Figures 6 and 7, the input X represents the prediction data, which may include data X. t-2 X t-1 X t The prediction data X can be used. t-2 X t-1 X t The data is input into the third prediction model, which directly yields the output prediction results. For example, Figure 6 shows whether the bacteria infecting the target patient are resistant to the corresponding antibiotic, or Figure 7 shows the probability that the bacteria infecting the target patient are resistant to the corresponding antibiotic. Thus, resistance data to the target antibiotic can be directly obtained, and the accuracy of this resistance data is higher than that shown in Figure 8.

[0124] As can be seen, the prediction model in this embodiment may include, but is not limited to, the first prediction model, the second prediction model and the third prediction model described above. In actual use, the model features of the corresponding prediction model can be adjusted according to the application requirements.

[0125] It should be understood that if models with different features and corresponding inputs are called different models, then each antibiotic corresponds to a different model. Accordingly, for each model, the methods described above can be used to predict the antibiotic resistance data of the bacteria infecting the target patient. For example, for carbapenems, quinolones, and third-generation cephalosporins, there can be corresponding models, and the corresponding models and corresponding index data can be used to predict the antibiotic resistance data of the bacteria infecting the target patient. This embodiment will not elaborate on this further.

[0126] In one possible implementation, the bacteria list includes a list of bacteria that the target patient may be infected with or a list of bacteria that the sample patient is infected with.

[0127] In a possible implementation, whether the target patient is infected with each bacterium in the list of bacteria includes at least one of whether the target patient is infected with a bacterium of a corresponding bacterium class, whether the target patient is infected with a bacterium of a corresponding bacterium family, whether the target patient is infected with a bacterium of a corresponding bacterium genus, and whether the target patient is infected with a bacterium of a corresponding bacterium species.

[0128] In this embodiment, the prediction model can be used to obtain only the bacterium that the target patient is likely to be infected with, or the prediction model can be used to obtain all the list of bacteria or the list of common bacteria, and the probability that the target patient is infected with each bacterium in the list of bacteria and / or the target patient is infected with each bacterium in the list of bacteria can be obtained. Of course, the bacterium can be refined to a class, a family, a genus, and a species.

[0129] In a possible implementation, the plurality of index data includes a use intensity of the target antibiotic of the target patient within a predetermined time before the onset of the disease.

[0130] In this embodiment, the predetermined time is, for example, 3 months. Of course, the predetermined time can be dynamically adjusted according to actual application needs.

[0131] In a possible implementation, the use intensity of the target antibiotic is calculated according to the number of days in which the target antibiotic is used by the target patient within the predetermined time, the dose of the target antibiotic used at a time, the frequency of using the target antibiotic per day, and the total number of days in which antibiotics are used by the target patient within the predetermined time.

[0132] In this embodiment, the use intensity of the target antibiotic can be calculated by using the following formula:

[0133] DAUDP is the use intensity of the target antibiotic of the target patient within the predetermined time before the onset of the disease, Drug days before onset is the total number of days in which antibiotics are used by the target patient before the onset of the disease (the day of the onset of the disease can be set as the day of sampling of bacterial culture, the day of ordering of drug sensitivity test, and the like), i is the target antibiotic, Dose is the dose of the target antibiotic used at a time, Frequency is the frequency of using the target antibiotic per day, Drug days before onset is the number of days in which the target antibiotic is used by the target patient before the onset of the disease, and WHO DDD of drug is the DDD value of the target antibiotic specified by the WHO.

[0134] In a possible implementation, the target antibiotics include carbapenems, quinolones and third-generation cephalosporins, and step S130 can include: inputting the prediction data into the prediction model, determining the drug resistance data of the bacteria currently infected by the target patient to the carbapenems according to the drug resistance value corresponding to each index data in the first plurality of index data corresponding to the carbapenems output by the prediction model and the threshold value corresponding to the index data, determining the drug resistance data of the bacteria currently infected by the target patient to the quinolones according to the drug resistance value corresponding to each index data in the second plurality of index data corresponding to the quinolones output by the prediction model and the threshold value corresponding to the index data, and determining the drug resistance data of the bacteria currently infected by the target patient to the third-generation cephalosporins according to the drug resistance value corresponding to each index data in the third plurality of index data corresponding to the third-generation cephalosporins output by the prediction model and the threshold value corresponding to the index data.

[0135] In this embodiment, the prediction model can be used to predict the drug resistance data of the bacteria currently infected by the target patient to the carbapenems, the quinolones and the third-generation cephalosporins. The carbapenems, the quinolones and the third-generation cephalosporins can correspond to the first target antibiotic, the second target antibiotic and the third target antibiotic in FIG. 3, respectively.

[0136] In a possible implementation, the target antibiotics further include at least one of the first-generation cephalosporins, the second-generation cephalosporins, the fourth-generation cephalosporins, the penicillins, the penams, the monobactams and the beta-lactam-beta-lactamase inhibitor combination preparations.

[0137] Various types of relevant data of the target patient can be obtained, such as the basic information of the target patient, the medication record, the detection report and the hospital information, etc. The basic information includes the age, the gender, the disease history, the allergy history, etc. of the patient. The medication record includes the history of the use of the antibacterial drugs, etc. The detection report includes the blood detection report, the ultrasonic detection report, the CT detection report, etc. during the previous hospitalization of the patient. The hospital information includes the frequency of the occurrence of the hospital infection bacteria and the frequency of the use of the antibacterial drugs, etc. during the previous hospitalization of the patient in the hospital.

[0138] The inventors have extracted some index data with higher influence degree on the prediction of drug resistance data from various types of related data of the target patient through a large amount of research and analysis. It should be noted that the extraction of some index data with higher influence degree from a large number of index data for the prediction of drug resistance data is the result of the inventors' creative labor. Moreover, the extracted index data is different for different types of target antibiotics.

[0139] In a possible implementation, the target antibiotic is a carbapenem antibiotic, and the first plurality of index data includes a minimum value of C-reactive protein (CRP) of the target patient before the onset of the disease.

[0140] In a possible implementation, the first plurality of index data further includes a length of time of the target patient in an intensive care unit (ICU) before the onset of the disease.

[0141] In a possible implementation, the first plurality of index data further includes a length of hospitalization of the target patient before the onset of the disease and a length of time of the target patient using mechanical ventilation before the onset of the disease.

[0142] In a possible implementation, the first plurality of index data further includes a number of days of the carbapenem antibiotic used by the target patient before the onset of the disease.

[0143] In a possible implementation, the target antibiotic is a carbapenem antibiotic, and the first plurality of index data includes at least five of the following index data: a number of days of use of the carbapenem antibiotic by the target patient before the onset of the disease, a length of time of use of mechanical ventilation by the target patient before the onset of the disease, a minimum value of C-reactive protein (CRP) of the target patient before the onset of the disease, a value of the last procalcitonin (PCT) of the target patient before the onset of the disease, a length of hospitalization of the target patient before the onset of the disease, a length of time in an intensive care unit (ICU) of the target patient before the onset of the disease, a number of days of use of antibacterial drugs by the target patient before the onset of the disease, an incidence of drug-resistant bacteria to the carbapenem antibiotic in a hospital where the target patient is treated within the predetermined time, a maximum value of procalcitonin (PCT) of the target patient before the onset of the disease, an age of the target patient, a value of the last serum albumin of the target patient before the onset of the disease, a minimum value of serum albumin of the target patient before the onset of the disease, an incidence of drug-resistant bacteria to the third-generation cephalosporin antibiotic in the hospital where the target patient is treated within the predetermined time, a maximum value of body temperature of the target patient before the onset of the disease, an incidence of drug-resistant bacteria to the quinolone antibiotic in the hospital where the target patient is treated within the predetermined time, a use intensity of the carbapenem antibiotic by the target patient before the onset of the disease, a use intensity of antibacterial drugs by the target patient before the onset of the disease, a maximum value of white blood cell count of the target patient before the onset of the disease, a number of strains of the carbapenem antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a number of strains of the third-generation cephalosporin antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a number of strains of the quinolone antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a number of days of use of the quinolone antibiotic by the target patient before the onset of the disease, a number of categories of antibiotics used by the target patient before the onset of the disease, a number of days of use of the third-generation cephalosporin antibiotic by the target patient before the onset of the disease, a use intensity of the third-generation cephalosporin antibiotic by the target patient before the onset of the disease, and a use intensity of the quinolone antibiotic by the target patient before the onset of the disease.

[0144] In a possible implementation, the target antibiotic is a carbapenem antibiotic, and the first plurality of index data includes at least five of the following index data: a number of days of use of the carbapenem antibiotic by the target patient before the onset of the disease, a length of time of use of mechanical ventilation by the target patient before the onset of the disease, a minimum value of C-reactive protein (CRP) of the target patient before the onset of the disease, a value of the last procalcitonin (PCT) of the target patient before the onset of the disease, a length of hospitalization of the target patient before the onset of the disease, a length of time in an intensive care unit (ICU) of the target patient before the onset of the disease, a number of days of use of antibacterial drugs by the target patient before the onset of the disease, an incidence of drug-resistant bacteria to the carbapenem antibiotic in a hospital where the target patient is treated within the predetermined time, a maximum value of procalcitonin (PCT) of the target patient before the onset of the disease, an age of the target patient, a last serum albumin value of the target patient before the onset of the disease, a minimum value of serum albumin of the target patient before the onset of the disease, a maximum value of body temperature of the target patient before the onset of the disease, an incidence of drug-resistant bacteria to the quinolone antibiotic in the hospital where the target patient is treated within the predetermined time, a use intensity of the carbapenem antibiotic by the target patient before the onset of the disease, a use intensity of antibacterial drugs by the target patient before the onset of the disease, and a maximum value of white blood cell count of the target patient before the onset of the disease.

[0145] In a possible implementation, the target antibiotic is a carbapenem antibiotic, and the first plurality of index data includes at least five of the following index data: a number of days of use of the carbapenem antibiotic by the target patient before the onset of the disease, a length of time of use of mechanical ventilation by the target patient before the onset of the disease, a minimum value of C-reactive protein (CRP) of the target patient before the onset of the disease, a value of the last procalcitonin (PCT) of the target patient before the onset of the disease, a length of hospitalization of the target patient before the onset of the disease, a length of time in an intensive care unit (ICU) of the target patient before the onset of the disease, a number of days of use of antibacterial drugs by the target patient before the onset of the disease, an incidence of drug-resistant bacteria to the carbapenem antibiotic in a hospital where the target patient is treated within the predetermined time, a maximum value of procalcitonin (PCT) of the target patient before the onset of the disease, a minimum value of serum albumin of the target patient before the onset of the disease, and an incidence of drug-resistant bacteria to the quinolone antibiotic in the hospital where the target patient is treated within the predetermined time.

[0146] In a possible implementation, the first plurality of index data further includes at least one of the following index data: an age of the target patient, a last serum albumin value of the target patient before the onset of the disease, a maximum value of body temperature of the target patient before the onset of the disease, a use intensity of the carbapenem antibiotic by the target patient before the onset of the disease, a use intensity of antibacterial drugs by the target patient before the onset of the disease, and a maximum value of white blood cell count of the target patient before the onset of the disease.

[0147] In a possible implementation, the first plurality of index data further includes at least one of the following index data: an incidence of drug-resistant bacteria of the third-generation cephalosporin antibiotic in a hospital where the target patient is treated within the predetermined time, a strain number of the carbapenem antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a strain number of the third-generation cephalosporin antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a strain number of the quinolone antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a number of days of use of the quinolone antibiotic before the onset of the target patient, a number of categories of antibiotics used before the onset of the target patient, a number of days of use of the third-generation cephalosporin antibiotic before the onset of the target patient, a use intensity of the third-generation cephalosporin antibiotic before the onset of the target patient, and a use intensity of the quinolone antibiotic before the onset of the target patient.

[0148] In this embodiment, the carbapenem antibiotic can include the plurality of index data, and optionally, the plurality of index data is selected from the plurality of index data according to the feature importance of the model. The index data can be sorted in descending order according to the feature importance, and any suitable index data can be selected according to the sorting result. In a possible implementation, the plurality of index data can be selected according to the result of single sorting according to the feature importance. In another possible implementation, the ranking of each index data in multiple cases can be counted to determine the number of times of ranking of the index data in a predetermined number of ranks, such as 5 or 10 ranks. Of course, other predetermined numbers, such as 6, 7, or 8, can also be used. Any suitable value can be selected as the predetermined number according to the actual application scenario and requirements. The importance of the index data can be comprehensively evaluated according to the number of times to select the plurality of index data. The corresponding index data for the carbapenem antibiotic can be appropriately increased or decreased according to actual application requirements.

[0149] In a possible implementation, the target antibiotic is a quinolone antibiotic, and the second plurality of index data includes a length of time of the target patient in an intensive care unit (ICU) before the onset and a length of hospital stay of the target patient before the onset.

[0150] In a possible implementation, the second plurality of index data further includes an incidence of drug-resistant bacteria of the quinolone antibiotic in a hospital where the target patient is treated within the predetermined time.

[0151] In a possible implementation, the second plurality of index data further includes a minimum value of serum albumin of the target patient before the onset.

[0152] In a possible implementation, the second plurality of index data further comprises a length of time that the target patient uses mechanical ventilation before the onset of the disease.

[0153] In a possible implementation, the target antibiotic is a quinolone antibiotic, and the second plurality of index data comprises at least five of the following index data: a length of time that the target patient stays in an intensive care unit (ICU) before the onset of the disease, a use intensity of antibacterial drugs of the target patient before the onset of the disease, an incidence of drug-resistant bacteria of the quinolone antibiotic in a hospital where the target patient is treated within the predetermined time, a value of the last procalcitonin (PCT) of the target patient before the onset of the disease, a strain number of the drug-resistant bacteria of the quinolone antibiotic in the hospital where the target patient is treated within the predetermined time, a maximum value of the PCT of the target patient before the onset of the disease, a length of hospital stay of the target patient before the onset of the disease, a number of days of the carbapenem antibiotic used by the target patient before the onset of the disease, a minimum value of serum albumin of the target patient before the onset of the disease, an age of the target patient, a minimum value of C-reactive protein (CRP) of the target patient before the onset of the disease, an incidence of drug-resistant bacteria of the carbapenem antibiotic in the hospital where the target patient is treated within the predetermined time, a number of days of the quinolone antibiotic used by the target patient before the onset of the disease, the last serum albumin value of the target patient before the onset of the disease, a maximum value of body temperature of the target patient before the onset of the disease, a length of time that the target patient uses mechanical ventilation before the onset of the disease, a number of days of antibacterial drugs used by the target patient before the onset of the disease, an incidence of drug-resistant bacteria of the third-generation cephalosporin antibiotic in the hospital where the target patient is treated within the predetermined time, a maximum value of white blood cell count of the target patient before the onset of the disease, a strain number of the drug-resistant bacteria of the third-generation cephalosporin antibiotic in the hospital where the target patient is treated within the predetermined time, a strain number of the drug-resistant bacteria of the carbapenem antibiotic in the hospital where the target patient is treated within the predetermined time, a use intensity of the carbapenem antibiotic of the target patient before the onset of the disease, a number of categories of antibiotics used by the target patient before the onset of the disease, a use intensity of the quinolone antibiotic of the target patient before the onset of the disease, a number of days of the third-generation cephalosporin antibiotic used by the target patient before the onset of the disease, a use intensity of the third-generation cephalosporin antibiotic of the target patient before the onset of the disease.

[0154] In a possible implementation, the target antibiotic is a quinolone antibiotic, and the second multiple index data includes at least five of the following index data: a length of time in an intensive care unit (ICU) before the onset of the target patient, an intensity of use of an antibacterial drug before the onset of the target patient, an incidence of drug-resistant bacteria to the quinolone antibiotic in a hospital where the target patient is treated within a predetermined time, a value of a last procalcitonin (PCT) before the onset of the target patient, a strain number of the quinolone antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a maximum value of the PCT before the onset of the target patient, a length of hospitalization before the onset of the target patient, a number of days of use of the carbapenem antibiotic before the onset of the target patient, a minimum value of serum albumin before the onset of the target patient, an age of the target patient, a minimum value of C-reactive protein (CRP) before the onset of the target patient, an incidence of drug-resistant bacteria to the carbapenem antibiotic in the hospital where the target patient is treated within the predetermined time, a last serum albumin value before the onset of the target patient, a maximum value of body temperature before the onset of the target patient, a length of time of use of mechanical ventilation before the onset of the target patient, a number of days of use of the antibacterial drug before the onset of the target patient, an incidence of drug-resistant bacteria to the third-generation cephalosporin antibiotic in the hospital where the target patient is treated within the predetermined time, a maximum value of white blood cell count before the onset of the target patient, and a strain number of the third-generation cephalosporin antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time.

[0155] In a possible implementation, the target antibiotic is a quinolone antibiotic, and the second multiple index data includes at least five of the following index data: a length of time in an intensive care unit (ICU) before the onset of the target patient, an intensity of use of an antibacterial drug before the onset of the target patient, an incidence of drug-resistant bacteria to the quinolone antibiotic in a hospital where the target patient is treated within a predetermined time, a value of a last procalcitonin (PCT) before the onset of the target patient, a strain number of the quinolone antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a maximum value of the PCT before the onset of the target patient, a length of hospitalization before the onset of the target patient, a minimum value of serum albumin before the onset of the target patient, an age of the target patient, a minimum value of C-reactive protein (CRP) before the onset of the target patient, an incidence of drug-resistant bacteria to the carbapenem antibiotic in the hospital where the target patient is treated within the predetermined time, a length of time of use of mechanical ventilation before the onset of the target patient, a number of days of use of the antibacterial drug before the onset of the target patient, a maximum value of white blood cell count before the onset of the target patient.

[0156] In a possible implementation, the second plurality of index data further comprises at least one of the following index data: a number of days of use of the carbapenem antibiotic before the onset of the disease of the target patient, a last serum albumin value before the onset of the disease of the target patient, a maximum body temperature before the onset of the disease of the target patient, an incidence of the third-generation cephalosporin-resistant bacteria in a hospital where the target patient is treated within the predetermined time, and a number of strains of the third-generation cephalosporin-resistant bacteria in the hospital where the target patient is treated within the predetermined time.

[0157] In a possible implementation, the second plurality of index data further comprises at least one of the following index data: a number of days of use of the quinolone antibiotic before the onset of the disease of the target patient, a number of strains of the carbapenem antibiotic-resistant bacteria in a hospital where the target patient is treated within the predetermined time, a use intensity of the carbapenem antibiotic before the onset of the disease of the target patient, a number of categories of antibiotics used before the onset of the disease of the target patient, a use intensity of the quinolone antibiotic before the onset of the disease of the target patient, a number of days of use of the third-generation cephalosporin before the onset of the disease of the target patient, and a use intensity of the third-generation cephalosporin before the onset of the disease of the target patient.

[0158] In this embodiment, the quinolone antibiotic can include the plurality of index data, and optionally, the plurality of index data is selected from a plurality of index data according to feature importance of a model. The index data can be sorted in descending order according to the feature importance, and any suitable index data can be selected according to the sorting result. In a possible implementation, the plurality of index data can be selected according to a single sorting result according to the feature importance. In another possible implementation, the ranking of each index data in multiple cases can be counted to determine the number of times of ranking in a predetermined number, such as 5 or 10, and of course, other predetermined numbers, such as 6, 7, or 8, can also be used. Any suitable value can be selected as the predetermined number according to the actual application scenario and requirements. The importance of the index data can be comprehensively evaluated according to the number of times to select the plurality of index data. The corresponding index data of the quinolone antibiotic can be appropriately increased or reduced according to actual application requirements.

[0159] In a possible implementation, the target antibiotic is a third-generation cephalosporin, and the third plurality of index data comprises a length of time of the target patient in an intensive care unit (ICU) before the onset of the disease and a length of hospital stay of the target patient before the onset of the disease.

[0160] In a possible implementation, the third plurality of index data further comprises a number of days of use of an antibacterial drug before the onset of the disease of the target patient.

[0161] In a possible implementation, the third plurality of index data further comprises an incidence of bacteria resistant to the carbapenem antibiotic in the hospital where the target patient is treated within the predetermined time.

[0162] In a possible implementation, the third plurality of index data further comprises a maximum value of white blood cell count of the target patient before the onset of the disease.

[0163] In a possible implementation, the target antibiotic is a third-generation cephalosporin antibiotic, and the third plurality of index data further comprises at least five of the following index data: a duration of antibacterial drug use of the target patient before the onset of the disease, a length of time in an intensive care unit (ICU) of the target patient before the onset of the disease, a minimum value of serum albumin of the target patient before the onset of the disease, an age of the target patient, a duration of the carbapenem antibiotic use of the target patient before the onset of the disease, an incidence of bacteria resistant to the quinolone antibiotic in the hospital where the target patient is treated within the predetermined time, a maximum value of procalcitonin (PCT) of the target patient before the onset of the disease, a minimum value of C-reactive protein (CRP) of the target patient before the onset of the disease, a length of hospitalization of the target patient before the onset of the disease, an incidence of bacteria resistant to the carbapenem antibiotic in the hospital where the target patient is treated within the predetermined time, an intensity of antibacterial drug use of the target patient before the onset of the disease, a number of strains of bacteria resistant to the carbapenem antibiotic in the hospital where the target patient is treated within the predetermined time, a value of procalcitonin (PCT) of the target patient at the last time before the onset of the disease, a maximum value of white blood cell count of the target patient before the onset of the disease, a number of strains of bacteria resistant to the third-generation cephalosporin antibiotic in the hospital where the target patient is treated within the predetermined time, a last value of serum albumin of the target patient before the onset of the disease, a number of strains of bacteria resistant to the quinolone antibiotic in the hospital where the target patient is treated within the predetermined time, a length of time of mechanical ventilation use of the target patient before the onset of the disease, an incidence of bacteria resistant to the third-generation cephalosporin antibiotic in the hospital where the target patient is treated within the predetermined time, an intensity of the carbapenem antibiotic use of the target patient before the onset of the disease, a maximum value of body temperature of the target patient before the onset of the disease, a duration of the quinolone antibiotic use of the target patient before the onset of the disease, a number of categories of antibiotics used by the target patient before the onset of the disease, a duration of the third-generation cephalosporin antibiotic use of the target patient before the onset of the disease, an intensity of the third-generation cephalosporin antibiotic use of the target patient before the onset of the disease, and an intensity of the quinolone antibiotic use of the target patient before the onset of the disease.

[0164] In a possible implementation, the target antibiotic is a third-generation cephalosporin antibiotic, and the third multi-index data further include at least five of the following index data: the number of days of use of antibacterial drugs before the onset of the target patient, the length of time in an intensive care unit (ICU) before the onset of the target patient, the minimum value of serum albumin before the onset of the target patient, the age of the target patient, the number of days of use of the carbapenem antibiotic before the onset of the target patient, the incidence of drug-resistant bacteria to the quinolone antibiotic in the hospital where the target patient is treated within the predetermined time, the maximum value of procalcitonin (PCT) before the onset of the target patient, the minimum value of C-reactive protein (CRP) before the onset of the target patient, the length of hospital stay before the onset of the target patient, the incidence of drug-resistant bacteria to the carbapenem antibiotic in the hospital where the target patient is treated within the predetermined time, the intensity of use of antibacterial drugs before the onset of the target patient, the number of strains of carbapenem antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, the value of the last procalcitonin (PCT) before the onset of the target patient, the maximum value of white blood cell count before the onset of the target patient, the last serum albumin before the onset of the target patient, the number of strains of quinolone antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, the length of time of use of mechanical ventilation before the onset of the target patient, and the intensity of use of the carbapenem antibiotic before the onset of the target patient.

[0165] In a possible implementation, the target antibiotic is a third-generation cephalosporin antibiotic, and the third multi-index data further include at least five of the following index data: the number of days of use of antibacterial drugs before the onset of the target patient, the length of time in an intensive care unit (ICU) before the onset of the target patient, the minimum value of serum albumin before the onset of the target patient, the age of the target patient, the number of days of use of the carbapenem antibiotic before the onset of the target patient, the maximum value of procalcitonin (PCT) before the onset of the target patient, the minimum value of C-reactive protein (CRP) before the onset of the target patient, the length of hospital stay before the onset of the target patient, the incidence of drug-resistant bacteria to the carbapenem antibiotic in the hospital where the target patient is treated within the predetermined time, the intensity of use of antibacterial drugs before the onset of the target patient, the number of strains of carbapenem antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, the value of the last procalcitonin (PCT) before the onset of the target patient, the maximum value of white blood cell count before the onset of the target patient, the length of time of use of mechanical ventilation before the onset of the target patient.

[0166] In a possible implementation, the third plurality of index data further comprises at least one of the following index data: an incidence of drug-resistant bacteria to the quinolone antibiotic in the hospital where the target patient is treated within the predetermined time, a number of strains of the carbapenem antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a last serum albumin level before the onset of the target patient, a number of strains of the quinolone antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a use intensity of the quinolone antibiotic before the onset of the target patient.

[0167] In a possible implementation, the third plurality of index data further comprises at least one of the following index data: a number of strains of the third-generation cephalosporin antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, an incidence of drug-resistant bacteria to the third-generation cephalosporin antibiotic in the hospital where the target patient is treated within the predetermined time, a maximum body temperature before the onset of the target patient, a number of days of the quinolone antibiotic before the onset of the target patient, a number of categories of antibiotics used before the onset of the target patient, a number of days of the third-generation cephalosporin antibiotic before the onset of the target patient, a use intensity of the third-generation cephalosporin antibiotic before the onset of the target patient, a use intensity of the quinolone antibiotic before the onset of the target patient.

[0168] In this embodiment, the third-generation cephalosporin antibiotic can comprise the plurality of index data, which is selected from the plurality of index data according to the feature importance of the model. The index data can be sorted in descending order of the feature importance, and any suitable index data can be selected according to the sorting result. In a possible implementation, the plurality of index data can be selected according to the result of single sorting of the feature importance. In another possible implementation, the ranking of each index data in multiple cases can be counted to determine the number of times of ranking in a predetermined number, such as 5 or 10, and of course, other predetermined numbers, such as 6, 7, or 8, can also be used. Any suitable value can be selected as the predetermined number according to the actual application scenario and requirements. The importance of the index data can be comprehensively evaluated according to the number of times to select the plurality of index data. The corresponding index data for the third-generation cephalosporin antibiotic can be appropriately increased or reduced according to actual application requirements.

[0169] In a possible implementation, the incidence of drug-resistant bacteria is calculated according to the number of drug-resistant bacteria and the number of non-drug-resistant bacteria, wherein the number of drug-resistant bacteria represents the number of all drug-resistant bacteria monitored by drug sensitivity tests of a plurality of patients in a hospital during the predetermined time when the target patient visits the hospital, and the number of non-drug-resistant bacteria represents the number of all non-drug-resistant bacteria monitored by drug sensitivity tests of the plurality of patients in the hospital during the predetermined time when the target patient visits the hospital.

[0170] In this embodiment, the predetermined time can include but is not limited to 1 month, 2 months, 3 months, or half a year, etc. The incidence of drug-resistant bacteria = the number of drug-resistant bacteria / (the number of drug-resistant bacteria + the number of non-drug-resistant bacteria).

[0171] It should be understood that the extracted index data generally includes demographic data, diagnosis data, medication history data, time series biochemical test results, environmental factor data, etc. The biochemical test results can use the latest detection data and the maximum and minimum values. The environmental factor data can be calculated by sampling from the hospital pneumonia related patients to calculate the hospital antibiotic resistance rate, and the drug resistance rate of different types of antibiotics per day can be calculated in a sliding window manner.

[0172] It can be understood that the above-mentioned various method embodiments mentioned in the present disclosure can be combined with each other to form combined embodiments without deviating from the principle logic. Due to the limited space, the present disclosure will not be repeated.

[0173] In addition, the present disclosure also provides a drug resistance prediction device, an electronic device, a computer readable storage medium, and a program, which can be used to implement any one of the drug resistance prediction methods provided by the present disclosure. The corresponding technical solutions and descriptions are referred to the corresponding description in the method part, and will not be repeated.

[0174] Those skilled in the art can understand that in the above method of the specific embodiment, the writing order of each step does not mean a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of each step should be determined by its function and possible internal logic.

[0175] FIG. 9 shows a block diagram of a drug resistance prediction device for predicting the drug resistance data of bacteria infected in a target patient to a target antibiotic according to an exemplary embodiment. In a possible implementation, the target patient includes but is not limited to a hospital pneumonia infected person, and the target antibiotic includes but is not limited to carbapenem antibiotics, quinolone antibiotics, and third-generation cephalosporin antibiotics.

[0176] As shown in FIG. 9, the drug resistance prediction apparatus can include an acquisition module 410 and a determination module 430. The acquisition module 410 is configured to acquire prediction data of the target patient. The determination module 430 is connected with the acquisition module 410 and configured to determine drug resistance data of a bacterium infected by the target patient to a target antibiotic according to the prediction data and a prediction model pre-constructed according to historical data of sample patients, wherein the prediction data includes multiple index data of the target patient as model features of the prediction model, and the drug resistance data of the bacterium infected by the target patient to the target antibiotic is determined according to a drug resistance value corresponding to each index data in the multiple index data output by the prediction model and a threshold value corresponding to the index data.

[0177] In a possible implementation, the prediction model includes a first prediction model and a second prediction model, the first prediction model is configured to make a prediction related to a bacterium infected by a patient, and the second prediction model is configured to make a prediction related to drug resistance of a bacterium infected by a patient to an antibiotic. Accordingly, the determination module 430 is configured to: input the prediction data into the first prediction model, the first prediction model outputs a first prediction result, the first prediction result includes whether the target patient is infected with each bacterium in a bacterium list and / or an infection probability of the target patient to the each bacterium; and input the prediction data and the first prediction result into the second prediction model, the second prediction model outputs a second prediction result, the second prediction result includes whether the each bacterium is resistant to the target antibiotic and / or a resistance probability of the each bacterium to the target antibiotic.

[0178] In a possible implementation, the prediction model includes a third prediction model, the third prediction model is configured to make a prediction related to a bacterium infected by a patient and drug resistance of the bacterium infected by the patient to an antibiotic. Accordingly, the determination module 430 is configured to: input the prediction data into the third prediction model, the third prediction model outputs a third prediction result, the third prediction result includes whether the target patient is infected with each bacterium in a bacterium list and / or an infection probability of the target patient to the each bacterium, and whether the each bacterium is resistant to the target antibiotic and / or a resistance probability of the each bacterium to the target antibiotic.

[0179] In a possible implementation, the bacterium list includes a list of bacteria that can be infected by the target patient or a list of bacteria infected by sample patients.

[0180] In one possible implementation, each item in the list of bacteria in which the target patient is infected includes at least one of the following: bacteria in which the target patient is infected with a corresponding bacterial class, bacteria in which the target patient is infected with a corresponding bacterial family, bacteria in which the target patient is infected with a corresponding bacterial genus, and bacteria in which the target patient is infected with a corresponding bacterial species.

[0181] In one possible implementation, the multiple indicator data includes the intensity of the target antibiotic used by the target patient within a predetermined time period prior to the onset of illness.

[0182] In one possible implementation, the intensity of the target antibiotic use is calculated based on the number of days the target patient uses the target antibiotic within the predetermined time period, the dose of a single dose of the target antibiotic, the frequency of daily use of the target antibiotic, and the total number of days the target patient uses the antibiotic within the predetermined time period.

[0183] In one possible implementation, the target antibiotic includes carbapenem antibiotics, quinolone antibiotics, and third-generation cephalosporin antibiotics. The determination step includes: inputting the prediction data into the prediction model; determining the resistance data of the bacteria currently infecting the target patient to the carbapenem antibiotics based on the resistance value corresponding to each of the first polynomial index data corresponding to the carbapenem antibiotics output by the prediction model and the threshold corresponding to that index data; determining the resistance data of the bacteria currently infecting the target patient to the quinolone antibiotics based on the resistance value corresponding to each of the second polynomial index data corresponding to the quinolone antibiotics output by the prediction model and the threshold corresponding to that index data; and determining the resistance data of the bacteria currently infecting the target patient to the third-generation cephalosporin antibiotics based on the resistance value corresponding to each of the third polynomial index data corresponding to the third-generation cephalosporin antibiotics output by the prediction model and the threshold corresponding to that index data.

[0184] In one possible implementation, the target antibiotic is a carbapenem antibiotic, and the first polynomial data includes the minimum C-reactive protein (CRP) value of the target patient prior to the onset of the disease.

[0185] In one possible implementation, the first plurality of indicator data also includes the duration of time the target patient spent in the intensive care unit (ICU) before the onset of illness.

[0186] In one possible implementation, the first plurality of indicator data further includes the length of hospital stay of the target patient before the onset of illness and the length of time the target patient used mechanical ventilation before the onset of illness.

[0187] In a possible implementation, the first plurality of index data further comprises a number of days of use of the carbapenem antibiotic by the target patient before the onset of the disease.

[0188] In a possible implementation, the target antibiotic is a quinolone antibiotic, and the second plurality of index data comprises a length of stay in an intensive care unit (ICU) by the target patient before the onset of the disease and a length of hospital stay by the target patient before the onset of the disease.

[0189] In a possible implementation, the second plurality of index data further comprises an incidence of drug-resistant bacteria to the quinolone antibiotic in a hospital where the target patient is treated within the predetermined time.

[0190] In a possible implementation, the second plurality of index data further comprises a minimum value of serum albumin of the target patient before the onset of the disease.

[0191] In a possible implementation, the second plurality of index data further comprises a length of use of mechanical ventilation by the target patient before the onset of the disease.

[0192] In a possible implementation, the target antibiotic is a third-generation cephalosporin antibiotic, and the third plurality of index data comprises a length of stay in an intensive care unit (ICU) by the target patient before the onset of the disease and a length of hospital stay by the target patient before the onset of the disease.

[0193] In a possible implementation, the third plurality of index data further comprises a number of days of use of an antibacterial drug by the target patient before the onset of the disease.

[0194] In a possible implementation, the third plurality of index data further comprises an incidence of drug-resistant bacteria to the carbapenem antibiotic in a hospital where the target patient is treated within the predetermined time.

[0195] In a possible implementation, the third plurality of index data further comprises a maximum value of white blood cell count of the target patient before the onset of the disease.

[0196] In a possible implementation, the incidence of drug-resistant bacteria is calculated according to a number of drug-resistant bacteria and a number of non-drug-resistant bacteria, wherein the number of drug-resistant bacteria represents a total number of drug-resistant bacteria monitored by drug sensitivity tests on a plurality of patients in a hospital where the target patient is treated within the predetermined time, and the number of non-drug-resistant bacteria represents a total number of non-drug-resistant bacteria monitored by drug sensitivity tests on the plurality of patients in the hospital where the target patient is treated within the predetermined time.

[0197] In a possible implementation, the target antibiotic further includes at least one of a first generation cephalosporin, a second generation cephalosporin, a fourth generation cephalosporin, a penicillin antibiotic, a penam antibiotic, a monobactam antibiotic, or a beta-lactam-beta lactamase inhibitor combination.

[0198] In a possible implementation, the model includes a machine learning model.

[0199] In a possible implementation, the machine learning model includes a neural network model, a logistic regression model, a decision tree model, or a Naive Bayes model.

[0200] In a possible implementation, the drug resistance prediction device is configured to predict drug resistance data of a bacterium currently infecting the target patient to the target antibiotic.

[0201] In a possible implementation, the multiple pieces of index data are index data that meet a first condition selected from the prediction data, and the first condition includes that an importance of the multiple pieces of index data output by the prediction model is higher than an importance of other index data of the prediction data.

[0202] In a possible implementation, the target antibiotic is a carbapenem antibiotic, and the first plurality of index data includes at least five of the following index data: a number of days of use of the carbapenem antibiotic by the target patient before the onset of the disease, a length of time of use of mechanical ventilation by the target patient before the onset of the disease, a minimum value of C-reactive protein (CRP) of the target patient before the onset of the disease, a value of the last procalcitonin (PCT) of the target patient before the onset of the disease, a length of hospitalization of the target patient before the onset of the disease, a length of time in an intensive care unit (ICU) of the target patient before the onset of the disease, a number of days of use of antibacterial drugs by the target patient before the onset of the disease, an incidence of drug-resistant bacteria to the carbapenem antibiotic in a hospital where the target patient is treated within the predetermined time, a maximum value of procalcitonin (PCT) of the target patient before the onset of the disease, an age of the target patient, a value of the last serum albumin of the target patient before the onset of the disease, a minimum value of serum albumin of the target patient before the onset of the disease, an incidence of drug-resistant bacteria to the third-generation cephalosporin antibiotic in the hospital where the target patient is treated within the predetermined time, a maximum value of body temperature of the target patient before the onset of the disease, an incidence of drug-resistant bacteria to the quinolone antibiotic in the hospital where the target patient is treated within the predetermined time, a use intensity of the carbapenem antibiotic by the target patient before the onset of the disease, a use intensity of antibacterial drugs by the target patient before the onset of the disease, a maximum value of white blood cell count of the target patient before the onset of the disease, a number of strains of the carbapenem antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a number of strains of the third-generation cephalosporin antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a number of strains of the quinolone antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a number of days of use of the quinolone antibiotic by the target patient before the onset of the disease, a number of categories of antibiotics used by the target patient before the onset of the disease, a number of days of use of the third-generation cephalosporin antibiotic by the target patient before the onset of the disease, a use intensity of the third-generation cephalosporin antibiotic by the target patient before the onset of the disease, and a use intensity of the quinolone antibiotic by the target patient before the onset of the disease.

[0203] In a possible implementation, the target antibiotic is a carbapenem antibiotic, and the first plurality of index data includes at least five of the following index data: a number of days of use of the carbapenem antibiotic by the target patient before the onset of the disease, a length of time of use of mechanical ventilation by the target patient before the onset of the disease, a minimum value of C-reactive protein (CRP) of the target patient before the onset of the disease, a value of the last procalcitonin (PCT) of the target patient before the onset of the disease, a length of hospitalization of the target patient before the onset of the disease, a length of time in an intensive care unit (ICU) of the target patient before the onset of the disease, a number of days of use of antibacterial drugs by the target patient before the onset of the disease, an incidence of drug-resistant bacteria to the carbapenem antibiotic in a hospital where the target patient is treated within the predetermined time, a maximum value of procalcitonin (PCT) of the target patient before the onset of the disease, an age of the target patient, a last serum albumin value of the target patient before the onset of the disease, a minimum value of serum albumin of the target patient before the onset of the disease, a maximum value of body temperature of the target patient before the onset of the disease, an incidence of drug-resistant bacteria to the quinolone antibiotic in the hospital where the target patient is treated within the predetermined time, a use intensity of the carbapenem antibiotic by the target patient before the onset of the disease, a use intensity of antibacterial drugs by the target patient before the onset of the disease, and a maximum value of white blood cell count of the target patient before the onset of the disease.

[0204] In a possible implementation, the target antibiotic is a carbapenem antibiotic, and the first plurality of index data includes at least five of the following index data: a number of days of use of the carbapenem antibiotic by the target patient before the onset of the disease, a length of time of use of mechanical ventilation by the target patient before the onset of the disease, a minimum value of C-reactive protein (CRP) of the target patient before the onset of the disease, a value of the last procalcitonin (PCT) of the target patient before the onset of the disease, a length of hospitalization of the target patient before the onset of the disease, a length of time in an intensive care unit (ICU) of the target patient before the onset of the disease, a number of days of use of antibacterial drugs by the target patient before the onset of the disease, an incidence of drug-resistant bacteria to the carbapenem antibiotic in a hospital where the target patient is treated within the predetermined time, a maximum value of procalcitonin (PCT) of the target patient before the onset of the disease, a minimum value of serum albumin of the target patient before the onset of the disease, and an incidence of drug-resistant bacteria to the quinolone antibiotic in the hospital where the target patient is treated within the predetermined time.

[0205] In a possible implementation, the first plurality of index data further includes at least one of the following index data: an age of the target patient, a last serum albumin value of the target patient before the onset of the disease, a maximum value of body temperature of the target patient before the onset of the disease, a use intensity of the carbapenem antibiotic by the target patient before the onset of the disease, a use intensity of antibacterial drugs by the target patient before the onset of the disease, and a maximum value of white blood cell count of the target patient before the onset of the disease.

[0206] In a possible implementation, the first plurality of index data further comprises at least one of the following index data: an incidence of drug-resistant bacteria to the third-generation cephalosporin antibiotic in a hospital where the target patient is treated within the predetermined time, a strain number of the carbapenem antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a strain number of the third-generation cephalosporin antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a strain number of the quinolone antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a number of days of the quinolone antibiotic used by the target patient before the onset of the disease, a number of categories of antibiotics used by the target patient before the onset of the disease, a number of days of the third-generation cephalosporin antibiotic used by the target patient before the onset of the disease, a use intensity of the third-generation cephalosporin antibiotic used by the target patient before the onset of the disease, and a use intensity of the quinolone antibiotic used by the target patient before the onset of the disease.

[0207] In a possible implementation, the target antibiotic is a quinolone antibiotic, and the second plurality of index data includes at least five of the following index data: a length of time in an intensive care unit (ICU) before the onset of the target patient, an intensity of use of an antibacterial drug before the onset of the target patient, an incidence of a drug-resistant bacterium of the quinolone antibiotic in a hospital where the target patient is treated within a predetermined time, a value of a last procalcitonin (PCT) before the onset of the target patient, a strain number of the drug-resistant bacterium of the quinolone antibiotic in the hospital where the target patient is treated within the predetermined time, a maximum value of the PCT before the onset of the target patient, a length of hospital stay before the onset of the target patient, a number of days of use of the carbapenem antibiotic before the onset of the target patient, a minimum value of serum albumin before the onset of the target patient, an age of the target patient, a minimum value of C-reactive protein (CRP) before the onset of the target patient, an incidence of a drug-resistant bacterium of the carbapenem antibiotic in the hospital where the target patient is treated within the predetermined time, a number of days of use of the quinolone antibiotic before the onset of the target patient, a last serum albumin value before the onset of the target patient, a maximum value of body temperature before the onset of the target patient, a length of time of use of mechanical ventilation before the onset of the target patient, a number of days of use of an antibacterial drug before the onset of the target patient, an incidence of a drug-resistant bacterium of a third-generation cephalosporin in the hospital where the target patient is treated within the predetermined time, a maximum value of white blood cell count before the onset of the target patient, a strain number of the drug-resistant bacterium of the third-generation cephalosporin in the hospital where the target patient is treated within the predetermined time, a strain number of the drug-resistant bacterium of the carbapenem antibiotic in the hospital where the target patient is treated within the predetermined time, an intensity of use of the carbapenem antibiotic before the onset of the target patient, a number of categories of antibiotics used before the onset of the target patient, an intensity of use of the quinolone antibiotic before the onset of the target patient, a number of days of use of the third-generation cephalosporin before the onset of the target patient, and an intensity of use of the third-generation cephalosporin before the onset of the target patient.

[0208] In a possible implementation, the target antibiotic is a quinolone antibiotic, and the second multiple index data includes at least five of the following index data: a length of time in an intensive care unit (ICU) before the onset of the target patient, an intensity of use of an antibacterial drug before the onset of the target patient, an incidence of drug-resistant bacteria to the quinolone antibiotic in a hospital where the target patient is treated within a predetermined time, a value of a last procalcitonin (PCT) before the onset of the target patient, a strain number of the quinolone antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a maximum value of the PCT before the onset of the target patient, a length of hospitalization before the onset of the target patient, a number of days of use of the carbapenem antibiotic before the onset of the target patient, a minimum value of serum albumin before the onset of the target patient, an age of the target patient, a minimum value of C-reactive protein (CRP) before the onset of the target patient, an incidence of drug-resistant bacteria to the carbapenem antibiotic in the hospital where the target patient is treated within the predetermined time, a last serum albumin value before the onset of the target patient, a maximum value of body temperature before the onset of the target patient, a length of time of use of mechanical ventilation before the onset of the target patient, a number of days of use of the antibacterial drug before the onset of the target patient, an incidence of drug-resistant bacteria to the third-generation cephalosporin antibiotic in the hospital where the target patient is treated within the predetermined time, a maximum value of white blood cell count before the onset of the target patient, and a strain number of the third-generation cephalosporin antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time.

[0209] In a possible implementation, the target antibiotic is a quinolone antibiotic, and the second multiple index data includes at least five of the following index data: a length of time in an intensive care unit (ICU) before the onset of the target patient, an intensity of use of an antibacterial drug before the onset of the target patient, an incidence of drug-resistant bacteria to the quinolone antibiotic in a hospital where the target patient is treated within a predetermined time, a value of a last procalcitonin (PCT) before the onset of the target patient, a strain number of the quinolone antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a maximum value of the PCT before the onset of the target patient, a length of hospitalization before the onset of the target patient, a minimum value of serum albumin before the onset of the target patient, an age of the target patient, a minimum value of C-reactive protein (CRP) before the onset of the target patient, an incidence of drug-resistant bacteria to the carbapenem antibiotic in the hospital where the target patient is treated within the predetermined time, a length of time of use of mechanical ventilation before the onset of the target patient, a number of days of use of the antibacterial drug before the onset of the target patient, a maximum value of white blood cell count before the onset of the target patient.

[0210] In a possible implementation, the second plurality of index data further comprises at least one of the following index data: a number of days of use of the carbapenem antibiotic before the onset of the disease of the target patient, a last serum albumin value before the onset of the disease of the target patient, a maximum body temperature before the onset of the disease of the target patient, an incidence of the third-generation cephalosporin-resistant bacteria in a hospital where the target patient is treated within the predetermined time, and a number of strains of the third-generation cephalosporin-resistant bacteria in the hospital where the target patient is treated within the predetermined time.

[0211] In a possible implementation, the second plurality of index data further comprises at least one of the following index data: a number of days of use of the quinolone antibiotic before the onset of the disease of the target patient, a number of strains of the carbapenem antibiotic-resistant bacteria in a hospital where the target patient is treated within the predetermined time, a use intensity of the carbapenem antibiotic before the onset of the disease of the target patient, a number of categories of antibiotics used before the onset of the disease of the target patient, a use intensity of the quinolone antibiotic before the onset of the disease of the target patient, a number of days of use of the third-generation cephalosporin before the onset of the disease of the target patient, and a use intensity of the third-generation cephalosporin before the onset of the disease of the target patient.

[0212] In a possible implementation, the target antibiotic is a third-generation cephalosporin antibiotic, and the third plurality of index data further includes at least five of the following index data: a number of days of use of an antibacterial drug before the onset of the target patient, a length of time in an intensive care unit (ICU) before the onset of the target patient, a minimum value of serum albumin before the onset of the target patient, an age of the target patient, a number of days of use of the carbapenem antibiotic before the onset of the target patient, an incidence of drug-resistant bacteria to the quinolone antibiotic in a hospital where the target patient is treated within the predetermined time, a maximum value of procalcitonin (PCT) before the onset of the target patient, a minimum value of C-reactive protein (CRP) before the onset of the target patient, a length of hospital stay before the onset of the target patient, an incidence of drug-resistant bacteria to the carbapenem antibiotic in a hospital where the target patient is treated within the predetermined time, an intensity of use of an antibacterial drug before the onset of the target patient, a number of strains of carbapenem-resistant bacteria in a hospital where the target patient is treated within the predetermined time, a value of procalcitonin (PCT) at the last time before the onset of the target patient, a maximum value of white blood cell count before the onset of the target patient, a number of strains of third-generation cephalosporin-resistant bacteria in a hospital where the target patient is treated within the predetermined time, a last value of serum albumin before the onset of the target patient, a number of strains of quinolone-resistant bacteria in a hospital where the target patient is treated within the predetermined time, a length of time of use of mechanical ventilation before the onset of the target patient, an incidence of drug-resistant bacteria to the third-generation cephalosporin in a hospital where the target patient is treated within the predetermined time, an intensity of use of the carbapenem antibiotic before the onset of the target patient, a maximum value of body temperature before the onset of the target patient, a number of days of use of the quinolone antibiotic before the onset of the target patient, a number of categories of antibiotics used before the onset of the target patient, a number of days of use of the third-generation cephalosporin before the onset of the target patient, an intensity of use of the third-generation cephalosporin before the onset of the target patient, and an intensity of use of the quinolone antibiotic before the onset of the target patient.

[0213] In a possible implementation, the target antibiotic is a third-generation cephalosporin antibiotic, and the third multi-index data further include at least five of the following index data: the number of days of use of antibacterial drugs before the onset of the target patient, the length of time in an intensive care unit (ICU) before the onset of the target patient, the minimum value of serum albumin before the onset of the target patient, the age of the target patient, the number of days of use of the carbapenem antibiotic before the onset of the target patient, the incidence of drug-resistant bacteria to the quinolone antibiotic in the hospital where the target patient is treated within the predetermined time, the maximum value of procalcitonin (PCT) before the onset of the target patient, the minimum value of C-reactive protein (CRP) before the onset of the target patient, the length of hospital stay before the onset of the target patient, the incidence of drug-resistant bacteria to the carbapenem antibiotic in the hospital where the target patient is treated within the predetermined time, the intensity of use of antibacterial drugs before the onset of the target patient, the number of strains of carbapenem antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, the value of the last procalcitonin (PCT) before the onset of the target patient, the maximum value of white blood cell count before the onset of the target patient, the last serum albumin before the onset of the target patient, the number of strains of quinolone antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, the length of time of use of mechanical ventilation before the onset of the target patient, and the intensity of use of the carbapenem antibiotic before the onset of the target patient.

[0214] In a possible implementation, the target antibiotic is a third-generation cephalosporin antibiotic, and the third multi-index data further include at least five of the following index data: the number of days of use of antibacterial drugs before the onset of the target patient, the length of time in an intensive care unit (ICU) before the onset of the target patient, the minimum value of serum albumin before the onset of the target patient, the age of the target patient, the number of days of use of the carbapenem antibiotic before the onset of the target patient, the maximum value of procalcitonin (PCT) before the onset of the target patient, the minimum value of C-reactive protein (CRP) before the onset of the target patient, the length of hospital stay before the onset of the target patient, the incidence of drug-resistant bacteria to the carbapenem antibiotic in the hospital where the target patient is treated within the predetermined time, the intensity of use of antibacterial drugs before the onset of the target patient, the number of strains of carbapenem antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, the value of the last procalcitonin (PCT) before the onset of the target patient, the maximum value of white blood cell count before the onset of the target patient, the length of time of use of mechanical ventilation before the onset of the target patient.

[0215] In a possible implementation, the third plurality of index data further comprises at least one of the following index data: an incidence of drug-resistant bacteria to the quinolone antibiotic in the hospital where the target patient is treated within the predetermined time, a number of strains of the carbapenem antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a last serum albumin value before the onset of the target patient, a number of strains of the quinolone antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a use intensity of the carbapenem antibiotic before the onset of the target patient.

[0216] In a possible implementation, the third plurality of index data further comprises at least one of the following index data: a number of strains of the third-generation cephalosporin antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, an incidence of drug-resistant bacteria to the third-generation cephalosporin antibiotic in the hospital where the target patient is treated within the predetermined time, a maximum body temperature before the onset of the target patient, a number of days of the quinolone antibiotic before the onset of the target patient, a number of categories of antibiotics used before the onset of the target patient, a number of days of the third-generation cephalosporin antibiotic before the onset of the target patient, a use intensity of the third-generation cephalosporin antibiotic before the onset of the target patient, a use intensity of the quinolone antibiotic before the onset of the target patient.

[0217] In a possible implementation, the prediction data is data of the target patient acquired 2 days before the onset.

[0218] In some embodiments, the apparatus provided by the embodiments of the present disclosure has functions or includes modules that can be used to perform the methods described in the above method embodiments, and the specific implementation can be referred to the description of the above method embodiments. For briefness, details are not described here.

[0219] The embodiments of the present disclosure also provide a computer-readable storage medium having stored thereon computer program instructions, and the computer program instructions, when executed by a processor, implement the above method. The computer-readable storage medium can be a non-volatile computer-readable storage medium.

[0220] The embodiments of the present disclosure also provide an electronic device, including: a processor; a memory for storing processor-executable instructions; and wherein the processor is configured to perform the above method.

[0221] The electronic device can be provided as a terminal, a server, or other forms of devices.

[0222] FIG. 10 is a block diagram of an electronic device according to an embodiment of the present disclosure. For example, the electronic device 1900 can be provided as a terminal, a server, or other forms of device. Referring to FIG. 10, the electronic device 1900 includes a processing component 1922, further including one or more processors, and a memory resource represented by a memory 1932, for storing instructions executable by the processing component 1922, such as an application program. The application program stored in the memory 1932 can include one or more than one module each corresponding to a set of instructions. In addition, the processing component 1922 is configured to execute the instructions to perform the above-described method.

[0223] The electronic device 1900 can further include a power supply component 1926 configured to perform power management of the electronic device 1900, a wired or wireless network interface 1950 configured to connect the electronic device 1900 to a network, and an input / output interface (I / O interface) 1958. The electronic device 1900 can operate based on an operating system stored in the memory 1932, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, or the like.

[0224] In an exemplary embodiment, a non-transitory computer readable storage medium, such as the memory 1932 including computer program instructions, is also provided, which can be executed by the processing component 1922 of the electronic device 1900 to complete the above-described method.

[0225] The present disclosure can be a system, a method, and / or a computer program product. The computer program product can include a computer readable storage medium (or media) having computer readable program instructions thereon for causing a processor to carry out aspects of the present disclosure.

[0226] Computer readable storage media can be tangible storage media which can retain and store instructions for use by an instruction execution device. Computer readable storage media can be, for example, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of computer readable storage media include the following: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanically encoded device such as punch-cards or raised structures in a groove having instructions recorded thereon, and any suitable combination of the foregoing. A computer readable storage medium, as used herein, is not to be construed as being transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission media (e.g., light pulses passing through a fiber-optic cable), or electrical signals transmitted through a wire.

[0227] Computer readable program instructions described herein can be downloaded to respective computing / processing devices from a computer readable storage medium or to an external computer or external storage device via a network, for example, the Internet, a local area network, a wide area network and / or a wireless network. The network can comprise copper transmission cables, optical transmission fibers, wireless transmission, routers, firewalls, switches, gateway computers and / or edge servers. A network adapter card or network interface in each computing / processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium within the respective computing / processing device.

[0228] Computer readable program instructions for carrying out operations of the present disclosure can be assembler instructions, instruction-set-architecture (ISA) instructions, machine instructions, machine dependent instructions, microcode, firmware instructions, state-setting data, or either source code or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer readable program instructions can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider). In some embodiments, electronic circuitry including, for example, programmable logic circuitry, field-programmable gate array (FPGA), or programmable logic array (PLA) can execute the computer readable program instructions by utilizing state information of the computer readable program instructions to personalize the electronic circuitry, in order to perform aspects of the present disclosure.

[0229] The computer readable program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0230] These computer readable program instructions can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks. These computer readable program instructions can also be stored in a computer readable storage medium that can include random access memory (RAM), read only memory (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or nonvolatile memory, or a suitable combination of the different types of computer readable storage media. The computer readable program instructions can also be downloaded to a computer, other programmable data processing apparatus, or other device from a computer readable storage medium or to an external computer or external storage device via a data signal that can be transmitted for example via a wired medium or a wireless medium such as the Internet or Wireless Application Protocol (WAP) signaling.

[0231] Having described above several embodiments of the disclosure, any modifications and variations that fall within the scope of the described embodiments are also intended to be within the scope of the disclosure. As will be apparent to those skilled in the art, some modifications and variations to the embodiments described above can be practiced while staying within the scope and spirit of the described embodiments. The foregoing description of the described embodiments has been presented for the purposes of illustration and description. It is not intended to be exhaustive or to limit the described embodiments to the precise form disclosed. Many modifications and variations are possible in light of the above teachings. It is intended that the disclosed embodiments be limited only by the claims.

Claims

1. A drug resistance prediction method for predicting drug resistance data of a bacterium infected in a target patient against a target antibiotic, characterized by, The method comprises: an acquisition step of acquiring prediction data of the target patient; a determination step of determining, according to the prediction data and a pre-constructed prediction model, the drug resistance data of the bacteria infected by the target patient to the target antibiotic, wherein the prediction model is pre-constructed according to historical data of sample patients, the prediction data comprises multiple index data of the target patient as model features of the prediction model, and the drug resistance data of the bacteria infected by the target patient to the target antibiotic is determined according to a drug resistance value corresponding to each index data in the multiple index data output by the prediction model and a threshold value corresponding to the index data.

2. The drug resistance prediction method according to claim 1, wherein the prediction model comprises a first prediction model and a second prediction model, the first prediction model is used for making a prediction related to bacteria infected by a patient, and the second prediction model is used for making a prediction related to drug resistance of bacteria infected by a patient to an antibiotic, and accordingly the determination step comprises: inputting the prediction data into the first prediction model, the first prediction model outputting a first prediction result, the first prediction result comprising whether each bacterium in a bacterium list is infected by the target patient and / or an infection probability of the target patient to the each bacterium; inputting the prediction data and the first prediction result into the second prediction model, the second prediction model outputting a second prediction result, the second prediction result comprising whether the each bacterium is resistant to the target antibiotic and / or a drug resistance probability of the each bacterium to the target antibiotic.

3. The drug resistance prediction method according to claim 1, wherein the prediction model comprises a third prediction model, the third prediction model is used for making a prediction related to bacteria infected by a patient and drug resistance of bacteria infected by a patient to an antibiotic, and accordingly the determination step comprises: inputting the prediction data into the third prediction model, the third prediction model outputting a third prediction result, the third prediction result comprising whether each bacterium in a bacterium list is infected by the target patient and / or an infection probability of the target patient to the each bacterium, and whether the each bacterium is resistant to the target antibiotic and / or a drug resistance probability of the each bacterium to the target antibiotic.

4. The drug resistance prediction method according to claim 2 or 3, characterized by, The bacterium list comprises a list of bacteria that can be infected by the target patient or a list of bacteria infected by sample patients.

5. The drug resistance prediction method according to claim 2 or 3, characterized by, Whether each bacterium in the bacterium list is infected by the target patient comprises at least one of whether a bacterium of a corresponding bacterium class is infected by the target patient, whether a bacterium of a corresponding bacterium family is infected by the target patient, whether a bacterium of a corresponding bacterium genus is infected by the target patient, and whether a bacterium of a corresponding bacterium species is infected by the target patient.

6. The drug resistance prediction method according to claim 1, wherein The multiple index data comprises a use intensity of the target antibiotic within a predetermined time before the onset of the target patient.

7. The drug resistance prediction method according to claim 6, wherein The use intensity of the target antibiotic is calculated according to the number of days of using the target antibiotic, the dose of a single target antibiotic, the frequency of using the target antibiotic per day, the total number of days of using antibiotics of the target patient within the predetermined time.

8. The drug resistance prediction method according to any one of claims 1 to 7, characterized by, The target antibiotic includes carbapenem antibiotics, quinolone antibiotics, and third-generation cephalosporin antibiotics, The determining step includes: The predicted data is input into the prediction model, and the drug resistance data of the bacteria currently infected by the target patient to the carbapenem antibiotic is determined according to the drug resistance value corresponding to each index data in the first plurality of index data corresponding to the carbapenem antibiotic output by the prediction model and the threshold value corresponding to the index data, the drug resistance data of the bacteria currently infected by the target patient to the quinolone antibiotic is determined according to the drug resistance value corresponding to each index data in the second plurality of index data corresponding to the quinolone antibiotic output by the prediction model and the threshold value corresponding to the index data, and the drug resistance data of the bacteria currently infected by the target patient to the third-generation cephalosporin antibiotic is determined according to the drug resistance value corresponding to each index data in the third plurality of index data corresponding to the third-generation cephalosporin antibiotic output by the prediction model and the threshold value corresponding to the index data.

9. The drug resistance prediction method according to claim 8, wherein, The target antibiotic is carbapenem antibiotic, and the first plurality of index data includes the minimum value of C-reactive protein (CRP) of the target patient before the onset.

10. The drug resistance prediction method according to claim 9, wherein, The first plurality of index data further includes the length of time of the target patient in an intensive care unit (ICU) before the onset.

11. The drug resistance prediction method according to claim 10, wherein, The first plurality of index data further includes the length of hospitalization of the target patient before the onset and the length of time of the target patient using mechanical ventilation before the onset.

12. The drug resistance prediction method according to claim 11, wherein, The first plurality of index data further includes the number of days of using the carbapenem antibiotic of the target patient before the onset.

13. The drug resistance prediction method according to claim 8, wherein, The target antibiotic is quinolone antibiotic, and the second plurality of index data includes the length of time of the target patient in an intensive care unit (ICU) before the onset and the length of hospitalization of the target patient before the onset.

14. The drug resistance prediction method according to claim 13, wherein, The second plurality of index data further includes the incidence of drug-resistant bacteria to the quinolone antibiotic in the hospital where the target patient was treated within a predetermined time before the onset of the target patient.

15. The drug resistance prediction method according to claim 14, wherein, The second plurality of index data further includes the minimum value of serum albumin of the target patient before the onset.

16. The drug resistance prediction method according to claim 15, wherein, The second plurality of index data further includes the length of time of the target patient using mechanical ventilation before the onset.

17. The drug resistance prediction method according to claim 8, wherein, The target antibiotic is third-generation cephalosporin antibiotic, and the third plurality of index data includes the length of time of the target patient in an intensive care unit (ICU) before the onset and the length of hospitalization of the target patient before the onset.

18. The drug resistance prediction method according to claim 17, wherein, The third plurality of index data further includes the number of days of using the antibacterial drug of the target patient before the onset.

19. The drug resistance prediction method according to claim 18, wherein, The third plurality of index data further includes an incidence of the drug-resistant bacteria in the hospital where the target patient is treated within a predetermined time before the onset of the disease.

20. The drug resistance prediction method according to claim 19, wherein, The third plurality of index data further includes a maximum value of white blood cell count of the target patient before the onset of the disease.

21. The drug resistance prediction method according to claim 14 or 19, wherein, The incidence of the drug-resistant bacteria is calculated based on a number of drug-resistant bacteria and a number of non-drug-resistant bacteria, wherein the number of drug-resistant bacteria represents a total number of drug-resistant bacteria monitored by drug sensitivity tests of a plurality of patients in the hospital where the target patient is treated within the predetermined time, and the number of non-drug-resistant bacteria represents a total number of non-drug-resistant bacteria monitored by drug sensitivity tests of the plurality of patients in the hospital where the target patient is treated within the predetermined time.

22. The drug resistance prediction method according to claim 8, wherein, The target antibiotic further includes at least one of a first-generation cephalosporin, a second-generation cephalosporin, a fourth-generation cephalosporin, a penicillin antibiotic, a penam antibiotic, a monobactam antibiotic, and a beta-lactam-beta lactamase inhibitor combination.

23. The drug resistance prediction method according to any one of claims 1-22, wherein, The prediction model includes a machine learning model.

24. The drug resistance prediction method according to claim 23, wherein, The machine learning model includes a neural network model, a logistic regression model, a decision tree model, or a naive Bayes model.

25. The drug resistance prediction method according to any one of claims 1-24, wherein, The drug resistance prediction method is used to predict drug resistance data of bacteria currently infected in the target patient to the target antibiotic.

26. The drug resistance prediction method of any one of claims 1-25, wherein the plurality of index data is index data that meets a first condition selected from the prediction data, and the first condition includes that an importance of the plurality of index data output by the prediction model is higher than an importance of other index data of the prediction data.

27. The drug resistance prediction method according to claim 8, wherein, The target antibiotic is a carbapenem antibiotic, and the first plurality of index data includes at least five of the following index data: the number of days of use of the carbapenem antibiotic by the target patient before the onset of the disease, the length of time of use of mechanical ventilation by the target patient before the onset of the disease, the minimum value of C-reactive protein (CRP) of the target patient before the onset of the disease, the value of the last procalcitonin (PCT) of the target patient before the onset of the disease, the length of hospitalization of the target patient before the onset of the disease, the length of time in an intensive care unit (ICU) of the target patient before the onset of the disease, the number of days of use of an antibacterial drug by the target patient before the onset of the disease, the incidence of drug-resistant bacteria to the carbapenem antibiotic in a hospital visited by the target patient within a predetermined time before the onset of the disease, the maximum value of procalcitonin (PCT) of the target patient before the onset of the disease, the age of the target patient, the last serum albumin value of the target patient before the onset of the disease, the minimum value of serum albumin of the target patient before the onset of the disease, the incidence of drug-resistant bacteria to the third-generation cephalosporin antibiotic in a hospital visited by the target patient within the predetermined time, the maximum body temperature of the target patient before the onset of the disease, the incidence of drug-resistant bacteria to the quinolone antibiotic in a hospital visited by the target patient within the predetermined time, the use intensity of the carbapenem antibiotic by the target patient before the onset of the disease, the use intensity of an antibacterial drug by the target patient before the onset of the disease, the maximum value of white blood cell count of the target patient before the onset of the disease, the number of strains of carbapenem antibiotic-resistant bacteria in a hospital visited by the target patient within the predetermined time, the number of strains of third-generation cephalosporin antibiotic-resistant bacteria in a hospital visited by the target patient within the predetermined time, the number of strains of quinolone antibiotic-resistant bacteria in a hospital visited by the target patient within the predetermined time, the number of days of use of the quinolone antibiotic by the target patient before the onset of the disease, the number of categories of antibiotics used by the target patient before the onset of the disease, the number of days of use of the third-generation cephalosporin antibiotic by the target patient before the onset of the disease, the use intensity of the third-generation cephalosporin antibiotic by the target patient before the onset of the disease, and the use intensity of the quinolone antibiotic by the target patient before the onset of the disease.

28. The drug resistance prediction method according to claim 8, wherein, The target antibiotic is a carbapenem antibiotic, and the first plurality of index data includes at least five of the following index data: a number of days of use of the carbapenem antibiotic before the onset of the target patient, a length of time of use of mechanical ventilation before the onset of the target patient, a minimum value of C-reactive protein (CRP) before the onset of the target patient, a value of the last procalcitonin (PCT) before the onset of the target patient, a length of hospitalization before the onset of the target patient, a length of time in an intensive care unit (ICU) before the onset of the target patient, a number of days of use of an antibacterial drug before the onset of the target patient, an incidence of a drug-resistant bacterium to the carbapenem antibiotic in a hospital visited by the target patient within a predetermined time before the onset of the target patient, a maximum value of procalcitonin (PCT) before the onset of the target patient, an age of the target patient, a last serum albumin value before the onset of the target patient, a minimum serum albumin value before the onset of the target patient, a maximum body temperature before the onset of the target patient, an incidence of a drug-resistant bacterium to the quinolone antibiotic in a hospital visited by the target patient within the predetermined time, a use intensity of the carbapenem antibiotic before the onset of the target patient, a use intensity of an antibacterial drug before the onset of the target patient, a maximum white blood cell count before the onset of the target patient.

29. The drug resistance prediction method according to claim 8, wherein, The target antibiotic is a carbapenem antibiotic, and the first plurality of index data includes at least five of the following index data: a number of days of use of the carbapenem antibiotic before the onset of the target patient, a length of time of use of mechanical ventilation before the onset of the target patient, a minimum value of C-reactive protein (CRP) before the onset of the target patient, a value of the last procalcitonin (PCT) before the onset of the target patient, a length of hospitalization before the onset of the target patient, a length of time in an intensive care unit (ICU) before the onset of the target patient, a number of days of use of an antibacterial drug before the onset of the target patient, an incidence of a drug-resistant bacterium to the carbapenem antibiotic in a hospital visited by the target patient within a predetermined time before the onset of the target patient, a maximum value of procalcitonin (PCT) before the onset of the target patient, a minimum serum albumin value before the onset of the target patient, an incidence of a drug-resistant bacterium to the quinolone antibiotic in a hospital visited by the target patient within the predetermined time.

30. The drug resistance prediction method according to claim 29, wherein, The first plurality of index data further includes at least one of the following index data: an age of the target patient, a last serum albumin value before the onset of the target patient, a maximum body temperature before the onset of the target patient, a use intensity of the carbapenem antibiotic before the onset of the target patient, a use intensity of an antibacterial drug before the onset of the target patient, a maximum white blood cell count before the onset of the target patient.

31. The drug resistance prediction method according to claim 30, wherein, The first plurality of index data further comprises at least one of the following index data: an incidence of drug-resistant bacteria to the third generation cephalosporin antibiotics in a hospital where the target patient is treated within the predetermined time, a number of strains of the carbapenem antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a number of strains of the third generation cephalosporin antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a number of strains of the quinolone antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a number of days of the quinolone antibiotics used by the target patient before the onset of the disease, a number of categories of the antibiotics used by the target patient before the onset of the disease, a number of days of the third generation cephalosporin antibiotics used by the target patient before the onset of the disease, a use intensity of the third generation cephalosporin antibiotics used by the target patient before the onset of the disease, and a use intensity of the quinolone antibiotics used by the target patient before the onset of the disease.

32. The drug resistance prediction method according to claim 8, wherein, The target antibiotic is a quinolone antibiotic, and the second plurality of index data includes at least five of the following index data: a length of time in an intensive care unit (ICU) before the onset of the target patient, an intensity of use of an antibacterial drug before the onset of the target patient, an incidence of a drug-resistant bacterium of the quinolone antibiotic in a hospital visited by the target patient within a predetermined time before the onset of the target patient, a value of a last procalcitonin (PCT) before the onset of the target patient, a number of strains of the drug-resistant bacterium of the quinolone antibiotic in the hospital visited by the target patient within the predetermined time, a maximum value of PCT before the onset of the target patient, a length of hospital stay before the onset of the target patient, a number of days of use of the carbapenem antibiotic before the onset of the target patient, a minimum value of serum albumin before the onset of the target patient, an age of the target patient, a minimum value of C-reactive protein (CRP) before the onset of the target patient, an incidence of a drug-resistant bacterium of the carbapenem antibiotic in the hospital visited by the target patient within the predetermined time, a number of days of use of the quinolone antibiotic before the onset of the target patient, a last value of serum albumin before the onset of the target patient, a maximum value of body temperature before the onset of the target patient, a length of time of use of mechanical ventilation before the onset of the target patient, a number of days of use of an antibacterial drug before the onset of the target patient, an incidence of a drug-resistant bacterium of a third-generation cephalosporin in the hospital visited by the target patient within the predetermined time, a maximum value of white blood cell count before the onset of the target patient, a number of strains of the drug-resistant bacterium of the third-generation cephalosporin in the hospital visited by the target patient within the predetermined time, a number of strains of the drug-resistant bacterium of the carbapenem antibiotic in the hospital visited by the target patient within the predetermined time, an intensity of use of the carbapenem antibiotic before the onset of the target patient, a number of categories of antibiotics used before the onset of the target patient, an intensity of use of the quinolone antibiotic before the onset of the target patient, a number of days of use of the third-generation cephalosporin before the onset of the target patient, and an intensity of use of the third-generation cephalosporin before the onset of the target patient.

33. The drug resistance prediction method according to claim 8, wherein, The target antibiotic is a quinolone antibiotic, and the second multiple index data includes at least five of the following index data: the length of time in an intensive care unit (ICU) before the onset of the disease in the target patient, the use intensity of antibacterial drugs before the onset of the disease in the target patient, the incidence of drug-resistant bacteria to the quinolone antibiotic in a hospital visited by the target patient within a predetermined time before the onset of the disease, the value of the last procalcitonin (PCT) before the onset of the disease in the target patient, the strain number of the quinolone antibiotic-resistant bacteria in the hospital visited by the target patient within the predetermined time, the maximum value of the procalcitonin (PCT) before the onset of the disease in the target patient, the length of hospital stay before the onset of the disease in the target patient, the number of days of use of the carbapenem antibiotic before the onset of the disease in the target patient, the minimum value of serum albumin before the onset of the disease in the target patient, the age of the target patient, the minimum value of C-reactive protein (CRP) before the onset of the disease in the target patient, the incidence of drug-resistant bacteria to the carbapenem antibiotic in a hospital visited by the target patient within the predetermined time, the last serum albumin value before the onset of the disease in the target patient, the maximum value of body temperature before the onset of the disease in the target patient, the length of time of using mechanical ventilation before the onset of the disease in the target patient, the number of days of use of antibacterial drugs before the onset of the disease in the target patient, the incidence of drug-resistant bacteria to the third-generation cephalosporin antibiotic in a hospital visited by the target patient within the predetermined time, the maximum value of white blood cell count before the onset of the disease in the target patient, and the strain number of the third-generation cephalosporin antibiotic-resistant bacteria in the hospital visited by the target patient within the predetermined time.

34. The drug resistance prediction method according to claim 8, wherein, The target antibiotic is a quinolone antibiotic, and the second multiple index data includes at least five of the following index data: the length of time in an intensive care unit (ICU) before the onset of the disease in the target patient, the use intensity of antibacterial drugs before the onset of the disease in the target patient, the incidence of drug-resistant bacteria to the quinolone antibiotic in a hospital visited by the target patient within a predetermined time before the onset of the disease, the value of the last procalcitonin (PCT) before the onset of the disease in the target patient, the strain number of the quinolone antibiotic-resistant bacteria in the hospital visited by the target patient within the predetermined time, the maximum value of the procalcitonin (PCT) before the onset of the disease in the target patient, the length of hospital stay before the onset of the disease in the target patient, the minimum value of serum albumin before the onset of the disease in the target patient, the age of the target patient, the minimum value of C-reactive protein (CRP) before the onset of the disease in the target patient, the incidence of drug-resistant bacteria to the carbapenem antibiotic in a hospital visited by the target patient within the predetermined time, the last serum albumin value before the onset of the disease in the target patient, the maximum value of body temperature before the onset of the disease in the target patient, the length of time of using mechanical ventilation before the onset of the disease in the target patient, the number of days of use of antibacterial drugs before the onset of the disease in the target patient, the incidence of drug-resistant bacteria to the third-generation cephalosporin antibiotic in a hospital visited by the target patient within the predetermined time, the maximum value of white blood cell count before the onset of the disease in the target patient, and the strain number of the third-generation cephalosporin antibiotic-resistant bacteria in the hospital visited by the target patient within the predetermined time.

35. The drug resistance prediction method according to claim 34, wherein, The second multiple index data further comprises at least one of the following index data: the number of days of the carbapenem antibiotic used by the target patient before the onset of the disease, the last serum albumin value before the onset of the disease by the target patient, the maximum body temperature before the onset of the disease by the target patient, the incidence of the third-generation cephalosporin-resistant bacteria in the hospital where the target patient is treated within the predetermined time, and the number of strains of the third-generation cephalosporin-resistant bacteria in the hospital where the target patient is treated within the predetermined time.

36. The drug resistance prediction method according to claim 35, wherein, The second multiple index data further comprises at least one of the following index data: the number of days of the quinolone antibiotic used by the target patient before the onset of the disease, the number of strains of the carbapenem antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, the intensity of the carbapenem antibiotic used by the target patient before the onset of the disease, the number of categories of antibiotics used by the target patient before the onset of the disease, the intensity of the quinolone antibiotic used by the target patient before the onset of the disease, the number of days of the third-generation cephalosporin used by the target patient before the onset of the disease, and the intensity of the third-generation cephalosporin used by the target patient before the onset of the disease.

37. The drug resistance prediction method according to claim 8, wherein, The target antibiotic is a third-generation cephalosporin antibiotic, and the third plurality of index data further includes at least five of the following index data: the number of days of use of an antibacterial drug before the onset of the target patient, the length of time in an intensive care unit (ICU) before the onset of the target patient, the minimum value of serum albumin before the onset of the target patient, the age of the target patient, the number of days of use of the carbapenem antibiotic before the onset of the target patient, the incidence of drug-resistant bacteria to the quinolone antibiotic in the hospital where the target patient was treated within a predetermined time before the onset of the target patient, the maximum value of procalcitonin (PCT) before the onset of the target patient, the minimum value of C-reactive protein (CRP) before the onset of the target patient, the length of hospital stay before the onset of the target patient, the incidence of drug-resistant bacteria to the carbapenem antibiotic in the hospital where the target patient was treated within the predetermined time, the intensity of use of an antibacterial drug before the onset of the target patient, the number of strains of carbapenem antibiotic-resistant bacteria in the hospital where the target patient was treated within the predetermined time, the last value of procalcitonin (PCT) before the onset of the target patient, the maximum value of white blood cell count before the onset of the target patient, the number of strains of third-generation cephalosporin antibiotic-resistant bacteria in the hospital where the target patient was treated within the predetermined time, the last serum albumin value before the onset of the target patient, the number of strains of quinolone antibiotic-resistant bacteria in the hospital where the target patient was treated within the predetermined time, the length of time of mechanical ventilation before the onset of the target patient, the incidence of drug-resistant bacteria to the third-generation cephalosporin antibiotic in the hospital where the target patient was treated within the predetermined time, the intensity of use of the carbapenem antibiotic before the onset of the target patient, the maximum value of body temperature before the onset of the target patient, the number of days of use of the quinolone antibiotic before the onset of the target patient, the number of categories of antibiotics used before the onset of the target patient, the number of days of use of the third-generation cephalosporin antibiotic before the onset of the target patient, the intensity of use of the third-generation cephalosporin antibiotic before the onset of the target patient, and the intensity of use of the quinolone antibiotic before the onset of the target patient.

38. The drug resistance prediction method of claim 8, wherein, The target antibiotic is a third-generation cephalosporin antibiotic, and the third multi-index data further includes at least five of the following index data: the number of days of use of antibacterial drugs before the onset of the target patient, the length of time in an intensive care unit (ICU) before the onset of the target patient, the minimum value of serum albumin before the onset of the target patient, the age of the target patient, the number of days of use of the carbapenem antibiotic before the onset of the target patient, the incidence of drug-resistant bacteria to the quinolone antibiotic in the hospital where the target patient was treated within a predetermined time before the onset of the target patient, the maximum value of procalcitonin (PCT) before the onset of the target patient, the minimum value of C-reactive protein (CRP) before the onset of the target patient, the length of hospital stay before the onset of the target patient, the incidence of drug-resistant bacteria to the carbapenem antibiotic in the hospital where the target patient was treated within the predetermined time, the intensity of use of antibacterial drugs before the onset of the target patient, the number of strains of carbapenem antibiotic-resistant bacteria in the hospital where the target patient was treated within the predetermined time, the value of the last procalcitonin (PCT) before the onset of the target patient, the maximum value of white blood cell count before the onset of the target patient, the last serum albumin value before the onset of the target patient, the number of strains of quinolone antibiotic-resistant bacteria in the hospital where the target patient was treated within the predetermined time, the length of time of mechanical ventilation before the onset of the target patient, and the intensity of use of the carbapenem antibiotic before the onset of the target patient.

39. The drug resistance prediction method according to claim 8, wherein, The target antibiotic is a third-generation cephalosporin antibiotic, and the third multi-index data further includes at least five of the following index data: the number of days of use of antibacterial drugs before the onset of the target patient, the length of time in an intensive care unit (ICU) before the onset of the target patient, the minimum value of serum albumin before the onset of the target patient, the age of the target patient, the number of days of use of the carbapenem antibiotic before the onset of the target patient, the maximum value of procalcitonin (PCT) before the onset of the target patient, the minimum value of C-reactive protein (CRP) before the onset of the target patient, the length of hospital stay before the onset of the target patient, the incidence of drug-resistant bacteria to the carbapenem antibiotic in the hospital where the target patient was treated within a predetermined time before the onset of the target patient, the intensity of use of antibacterial drugs before the onset of the target patient, the number of strains of carbapenem antibiotic-resistant bacteria in the hospital where the target patient was treated within the predetermined time, the value of the last procalcitonin (PCT) before the onset of the target patient, the maximum value of white blood cell count before the onset of the target patient, the length of time of mechanical ventilation before the onset of the target patient.

40. The drug resistance prediction method according to claim 39, wherein, The third plurality of index data further comprises at least one of the following index data: an incidence of drug-resistant bacteria to the quinolone antibiotic in the hospital where the target patient is treated within the predetermined time, a number of strains of the carbapenem antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a last serum white blood cell count of the target patient before the onset of the disease, a number of strains of the quinolone antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, a use intensity of the carbapenem antibiotic of the target patient before the onset of the disease.

41. The drug resistance prediction method according to claim 40, wherein, The third plurality of index data further comprises at least one of the following index data: a number of strains of the third-generation cephalosporin antibiotic-resistant bacteria in the hospital where the target patient is treated within the predetermined time, an incidence of drug-resistant bacteria to the third-generation cephalosporin antibiotic in the hospital where the target patient is treated within the predetermined time, a maximum body temperature of the target patient before the onset of the disease, a number of days of the quinolone antibiotic use of the target patient before the onset of the disease, a number of categories of antibiotics used by the target patient before the onset of the disease, a number of days of the third-generation cephalosporin antibiotic use of the target patient before the onset of the disease, a use intensity of the third-generation cephalosporin antibiotic of the target patient before the onset of the disease, a use intensity of the quinolone antibiotic of the target patient before the onset of the disease.

42. The drug resistance prediction method of any one of claims 1-41, wherein the prediction data of the target patient is data of the target patient acquired 2 days before the onset of the disease.

43. A drug resistance prediction device for predicting drug resistance data of a bacterium infected in a target patient against a target antibiotic, characterized by, comprising: an acquisition module configured to acquire prediction data of the target patient; a determination module configured to determine, according to the prediction data and a prediction model pre-constructed, drug resistance data of a bacterium infected by the target patient to the target antibiotic, wherein the prediction model is pre-constructed according to historical data of sample patients, the prediction data comprises a plurality of index data of the target patient as model features of the prediction model, and the drug resistance data of the bacterium infected by the target patient to the target antibiotic is determined according to a drug resistance value corresponding to each index data in the plurality of index data and a threshold value corresponding to the index data according to an output of the prediction model.

44. An electronic device, comprising: comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to implement the drug resistance prediction method of any one of claims 1-42 when executing the instructions stored in the memory.

45. A non-transitory computer readable storage medium having stored thereon computer program instructions, wherein, The computer program instructions, when executed by the processor, implement the drug resistance prediction method of any one of claims 1-42. The computer program instructions, when executed by the processor, implement the drug resistance prediction method of any one of claims 1-42.

Images