CN108461154B - A hospital infection monitoring and management device and monitoring and management method - Google Patents

Abstract

The invention discloses a kind of Hospital Infection managing device and method for managing and monitoring, including Hospital Infection management server, information for hospital integrated platform, hospital information system, laboratory information system, Picture Archiving and Communication System, electronic medical record system, nosocomial infection detection service device, wechat enterprise number server, smart phone；It is pushed by new media information, picture and text propaganda and education, the various aspects interaction function such as questionnaire survey realizes plan (Plan), executes (Do), check (Check), adjusts the nosocomial infection overall flow management of (Action) closed loop.The present invention improves doubtful institute's sense detection accuracy, reduces doctor's workload, optimizes Hospital Infection management process, can be applied to Hospital Infection.

Description

A hospital infection monitoring and management device and monitoring and management method

technical field

The invention belongs to the technical field of medical and health care, and in particular relates to a hospital infection monitoring and management device and a monitoring and management method.

Background technique

Nosocomial infection refers to the infection acquired by inpatients in the hospital, including the infection that occurs during hospitalization and the infection that occurs after discharge in the hospital. Many hospitals have implemented information-based hospital infection monitoring and management. With the development of machine learning, WeChat enterprise account and hospital information platform, hospital infection information management has ushered in opportunities for further improvement. The traditional nosocomial infection information system realizes the basic monitoring and management functions of nosocomial infection. Publication No. CN201410658773 (Authorization Announcement Date: 2017.05.17) discloses a remote monitoring and management system and monitoring and management method for hospital-sensing equipment. The invention monitors the operation data of the hospital-sensing equipment and the data of the completion of the hospital-sensing equipment work through the front-end device , solves the need to manually collect data to obtain the operation status of hospital infection equipment and work completion status, reduces monitoring time, and improves monitoring efficiency, but lacks the process management of hospital infection events. Publication No. CN201410636914 (application date: 2014.11.13) discloses a hospital infection prevention and control whole-process management system and a method thereof, which adopts the design of hardware structure and conducts comprehensive, rapid, Accurate retrieval greatly reduces the scope of cases that the infection control staff need to consult, and improves the efficiency of infection control work. Interpretation is mainly based on the basic method of collecting data, and lacks the analysis and judgment method of machine learning. Publication No. CN 201611214892 (application date: 2016.12.26) discloses an intelligent decision-making assistance system for predicting, diagnosing, treating and controlling nosocomial infections, which can be effectively combined with epidemiological models according to patients' medical records and measures taken by hospital management staff. Predict the probability of infection in patients, and make corresponding diagnosis and treatment recommendations, but it also lacks the interactive functions of information push, graphic education, questionnaire survey and other interactive functions that new media have. The method has certain errors, and it does not classify and deal with different infection categories, and does not calculate and analyze the principal component factors that may cause various types of nosocomial infections, which will affect the calculation of the probability of predicting the occurrence of infection in patients.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a nosocomial infection monitoring and management device and a monitoring and management method, so as to further advance the technology of nosocomial infection detection and realize the function of nosocomial infection monitoring and management.

In order to solve the above technical problems, the present invention uses the existing mature machine learning, WeChat enterprise account related functions and hospital information platform technology, etc., through the design of information technology and hardware equipment, integrated information push, graphic propaganda and education, questionnaire survey and other functions to achieve the purpose of the invention. The specific technical solutions adopted are as follows:

A hospital infection monitoring and management device, comprising a hospital infection monitoring and management server, a hospital information integration platform, a hospital information system, a laboratory information system, a medical image archiving and communication system, an electronic medical record system, a hospital infection detection server, a WeChat enterprise account server, smart phone;

The hospital infection monitoring and management server is connected with the hospital information integration platform through the hospital LAN;

The hospital information integration platform is connected with the hospital information system through the hospital LAN;

The hospital information integration platform is connected with the laboratory information system through the hospital LAN;

The hospital information integration platform is connected with the medical image archiving and communication system through the hospital LAN;

The hospital information integration platform is connected with the electronic medical record system through the hospital LAN;

The hospital infection monitoring and management server is connected with the hospital infection detection server through the hospital LAN;

The hospital infection monitoring and management server is connected with the WeChat enterprise account server through the hospital LAN;

The WeChat enterprise number server is connected with the smart phone through the hospital wireless network;

The hospital information system is used to collect basic hospitalization information such as the patient's name, hospitalization number, etc., as well as medical information such as antibiotic drug orders.

The laboratory information system is used for collecting information such as patient blood test and drug resistance bacteria detection.

The medical image archiving and communication system is used for collecting the result information of the patient's chest X-ray examination and CT examination report.

The electronic medical record system is used to collect the patient's body temperature state information.

The hospital information integration platform is used for managing and connecting hospital information systems, laboratory information systems, medical image archiving and communication systems, and electronic medical record systems.

The nosocomial infection detection server is used for analyzing and processing the basic medical data of the patients, and detecting the patients suspected of nosocomial infection.

The WeChat enterprise account server is used to provide various interactive functions such as information push, graphic propaganda and education, and questionnaire survey.

The smart phone is used to receive and display various service information provided by the WeChat enterprise account server, and send relevant feedback information to the WeChat enterprise account server.

The hospital infection monitoring and management server is used to manage and connect the hospital information integration platform, the hospital infection detection server, and the WeChat enterprise account server, and to manage the overall process of hospital infection.

The described a hospital infection monitoring and management method is as follows:

Step A1, the hospital infection staff organizes the detection methods and policies related to hospital infection, and sends them to the doctor's smartphone through the image and text education function provided by the WeChat enterprise account server;

Step A2, the hospital infection monitoring and management server uses the hospital infection model establishment method to establish a hospital infection model on the hospital infection detection server, the hospital infection monitoring and management server uses the hospital infection detection method to detect the suspected hospital infection event, and pushes it to the doctor, and the doctor reports the hospital infection using the hospital infection detection method. Methods to report nosocomial infection events;

Step A3, the hospital infection staff receives and handles the nosocomial infection event;

Step A4, the hospital infection staff organizes the relevant infection cases in the hospital infection monitoring and management server, generates a questionnaire, and sends it to the doctor's smartphone through the questionnaire function provided by the WeChat enterprise account server. After the doctor completes the filling, it is sent back to the hospital infection monitoring and management server , the hospital infection staff analyzes the data reported by doctors and the questionnaire survey data, forms rectification opinions, and readjusts the detection methods and policies related to hospital infection.

The methods for reporting nosocomial infections are as follows:

Step B1, set the infection category set in the hospital infection monitoring and management server as {C ₁ , C ₂ ... C _j ... C _J }, where C _j is the j-th infection category, 1≤j≤J, J is the total number of infection categories, C _j = {D ₁ , D ₂ ... D _k ... D _K }, where D _k is the kth case of the jth infection category, 1≤k≤K, K is the kth case The total number of cases in j infection categories;

Step B2, the doctor checks the case information of the patient in the hospital by himself, or checks the information of the suspected hospital infection case detected by the hospital infection detection method pushed by the hospital infection monitoring and management server;

Step B3, the doctor judges whether the case is a nosocomial infection case, if it is transferred to B4, otherwise transferred to B8;

Step B4, generating infection case D _find ;

Step B5, push D _find to the hospital staff through the information push function provided by the WeChat enterprise account server;

Step B6, the hospital infection staff analyzes the infection case D _find , if it is confirmed to be the infection category C _j , go to B7, otherwise go to B8;

Step B7, store the infection case D _find in the corresponding infection category C _j ;

Step B8, the reporting ends.

The nosocomial infection model establishment method is as follows:

Step C1, the hospital infection monitoring and management server loads all infection categories as the training set of principal component analysis, and initializes j=0;

Step C2, j=j+1;

Step C3, the hospital infection monitoring and management server applies to the hospital information integration platform to extract the medical index information of all cases in C _j = {D ₁ , D ₂ ... D _k ... D _K };

Step C4, the hospital information integration platform extracts antibiotic drug information {drug _{k, 1} , drug _{k, 2} , ... drug _{k, a} , ... drug _{k, A} } from the hospital information system, where drug _{k, a} is The a-th antibiotic dose of the k-th case, 1≤a≤A, A is the number of antibiotics in the k-th case;

Step C5, the hospital information integration platform extracts blood test information {blood _{k, 1} , blood _{k, 2} , ... blood _{k, b} , ... blood _{k, B} } from the laboratory information system, where blood _{k, b} is The bth blood test value of the kth case, 1≤b≤B, B is the number of blood test types of the kth case, the hospital information integration platform extracts the drug-resistant bacteria detection information from the laboratory information system {cell _{k, 1} , cell _{k, 2} ,...cell _k,c ,...cell _k,C }, where cell _k,c is the detection value of the c-th drug-resistant bacteria in the k-th case, 1≤c≤C,C is the number of drug-resistant bacteria detected in the kth case;

Step C6, the hospital information integration platform extracts the chest X-ray examination report result information xray _k and the CT examination report result information CT _k from the medical image archiving and communication system;

Step C7, the hospital information integration platform extracts recent body temperature information {tem _{k, 1} , tem _{k, 2} ,...tem _k,d ,...tem _k,D } from the electronic medical record system, where tem _k,d is the first The value of the d-th body temperature measurement of the k cases, 1≤d≤D, D is the number of body temperature measurements of the k-th case;

Step C8, after standardizing all medical index information, arrange in columns to form a matrix of all indexes of all cases

where { _xk,1 , _xk,2 ,... _xk,p ,... _xk,P } is given by {drugk _,1 ,drugk _,2 ,...drugk _,a ,.. .drug _k,A },{blood _k,1 ,blood _k,2 ,...blood _k,b ,...blood _k,B },{cell _k,1 ,cell _k,2 ,...cell _k,c ,...cell _{k ,C} },xrayk, _CTk ,{temk _,1 ,temk _,2 ,...temk _,d ,...temk _,D } from left to right Integrated into the same set, P=A+B+C+1+1+D;

Step C9, the hospital infection detection server calculates the correlation coefficient matrix of X

in

in

Step C10, the hospital infection detection server finds the eigenvalues {λ ₁ ,λ ₂ ...λ _p ...λ _P } of the matrix R, and arranges them in order of magnitude, λ ₁ ≥λ ₂ ≥...λ _p ...≥λ _P ≥0 ;

Step C11, the hospital infection detection server generates principal components corresponding to {λ ₁ ,λ ₂ ...λ _p ...λ _P }

Step C12, set the cumulative proportion threshold STEP, i=0, i is the cumulative variable;

Step C13, i=i+1;

Step C14, the hospital infection detection server calculates the cumulative contribution rate of the principal components

Step C15, if setp<STEP, go to C13, otherwise go to C16;

Step C16, the nosocomial infection detection server determines that the feature vector corresponding to the principal component is Detection threshold Test _j =setp of infection category C _j ;

Step C17, the hospital infection detection server generates a detection model of infection category C _j

Step C18, if j<J, go to C2, otherwise go to C19;

Step C19, the establishment of the hospital infection model is completed.

The nosocomial infection detection methods described are as follows:

Step D1, select the cases of patients in the hospital as the detection set of the hospital infection model, the case set of all patients in the hospital is {D ₁ , D ₂ ... D _f ... D _F }, and D _f is the fth inpatient Cases, 1≤f≤F, F is the total number of patients currently in the hospital, initialized f=0;

Step D2, f=f+1;

Step D3, judge whether the patient's case has been judged as a nosocomial infection case; if it is, transfer to D12, otherwise transfer to D4;

Step D4, initialize j=0;

Step D5, j=j+1;

Step D6, the hospital infection monitoring and management server applies to the hospital information integration platform to extract relevant information from the D _f cases, and the hospital information integration platform extracts antibiotic drug information {drug _{f, 1} , drug _{f, 2} , ... drug _{f, a} ,...drug _{f, A} }, the hospital information integration platform draws blood test information from the laboratory information system {blood _{f, 1} , blood _{f, 2} ,...blood _f,b ,...blood _{f, B} }, drug-resistant bacteria detection information {cell _{f, 1} , cell _{f, 2} , ... cell _{f, c} , ... cell _{f, C} }, the hospital information integration platform extracts from the medical image archiving and communication system Chest X-ray examination report result information xray _f , CT examination report result information CT _f , the hospital information integration platform extracts recent body temperature information from the electronic medical record system {tem _{f, 1} , tem _{f, 2} ,...tem _{f, d} ,. ..tem _{f, D} };

Step D7, after standardizing all medical index information, arrange them in sequence as {x _{f, 1} , x _{f, 2} ,...x _{f, p} ,...x _{f, P} };

Step D8, input the medical index information into the detection model in the hospital infection detection server corresponding to the infection category C _j for calculation, current=e ₁ , 1 ×x _f,1 +...e _p,1 ×x _f,p . ..+e _P,1 × x _f,P ;

Step D9, judging whether current is greater than Test _j , it is transferred to D10, otherwise it is transferred to D11;

Step D10, the hospital infection monitoring and management server pushes the detected suspected hospital infection cases to the relevant doctors through the information push function provided by the WeChat enterprise account server, j=J;

Step D11, if j<J, go to D5, otherwise go to D12;

Step D12, if f<F, go to D2, otherwise go to D13;

In step D13, the nosocomial infection detection is completed.

The present invention has beneficial effects. The present invention integrates the whole process of hospital infection monitoring and management through the hospital infection monitoring and management server, and realizes the planning (Plan), execution (Do), inspection ( Check), adjust (Action) closed-loop hospital infection overall process management. The hospital infection monitoring and management server uses the hospital information integration platform to connect other hospital information system components to standardize the data collection method, so that when other information systems are upgraded or adjusted, they only need to change the interface with the hospital information integration platform. Does not affect the use of hospital infection monitoring and management server. The machine learning method is used to establish a nosocomial infection model, and for different infection categories, different weights of different categories are calculated respectively, and the principal component factors of each category of nosocomial infection can be analyzed, which can accurately calculate the probability of a patient's infection. Using the nosocomial infection cases reported by relevant doctors, iteratively calculates and updates the nosocomial infection model to obtain more accurate calculation results. The invention improves the detection accuracy of suspected hospital infection, reduces the workload of doctors, and optimizes the monitoring and management process of hospital infection.

Description of drawings

FIG. 1 is a schematic diagram of the overall structure of a hospital infection monitoring and management device.

In the picture: 1-Hospital infection monitoring and management server, 2-Hospital information integration platform, 3-Hospital information system, 4-Laboratory information system, 5-Medical image archiving and communication system, 6-Electronic medical record system, 7-Hospital infection Detection server, 8-WeChat enterprise number server, 9-Smart phone.

Fig. 2 is a flow chart of a hospital infection monitoring and management method.

FIG. 3 is a flow chart of a method for reporting nosocomial infections.

Figure 4 is a flow chart of a method for establishing a hospital infection model.

Figure 5 is a flow chart of a nosocomial infection detection method.

FIG. 6 is a schematic diagram of a closed-loop operation of a hospital infection monitoring and management method.

Detailed ways

The present invention will be described in further detail below with reference to the accompanying drawings and specific embodiments.

It can be seen from the schematic diagram of the overall structure of a hospital infection monitoring and management device shown in Figure 1 that it includes a hospital infection monitoring and management server 1, a hospital information integration platform 2, a hospital information system 3, a laboratory information system 4, medical image archiving and communication. System 5, electronic medical record system 6, hospital infection detection server 7, WeChat enterprise account server 8, smart phone 9;

The hospital infection monitoring and management server 1 is connected with the hospital information integration platform 2 through the hospital LAN;

The hospital information integration platform 2 is connected with the hospital information system 3 through the hospital LAN;

The hospital information integration platform 2 is connected with the laboratory information system 4 through the hospital LAN;

The hospital information integration platform 2 is connected with the medical image archiving and communication system 5 through the hospital LAN;

The hospital information integration platform 2 is connected with the electronic medical record system 6 through the hospital LAN;

The hospital infection monitoring and management server 1 is connected with the hospital infection detection server 7 through the hospital LAN;

The hospital infection monitoring and management server 1 is connected with the WeChat enterprise account server 8 through the hospital LAN;

The WeChat enterprise number server 8 is connected with the smart phone 9 through the hospital wireless network;

When the present invention is in use, the functions of the various components are described below.

The hospital information system 3 is used to collect basic hospitalization information such as the patient's name, hospitalization number, etc., and medical information such as antibiotic drug orders.

The laboratory information system 4 is used for collecting information such as patient blood test and drug-resistant bacteria detection.

The medical image archiving and communication system 5 is used to collect the result information of the patient's chest X-ray examination and CT examination report.

The electronic medical record system 6 is used to collect the patient's body temperature state information.

The hospital information integration platform 2 is used to manage and connect the hospital information system, the laboratory information system, the medical image archiving and communication system, and the electronic medical record system.

The nosocomial infection detection server 7 is used for analyzing and processing the basic medical data of the patient, and detecting the patient suspected of nosocomial infection.

The WeChat enterprise account server 8 is used to provide various interactive functions such as information push, graphic propaganda and education, and questionnaire survey.

The smart phone 9 is used for receiving and displaying various service information provided by the WeChat enterprise account server 8, and sending relevant feedback information to the WeChat enterprise account server.

The hospital infection monitoring and management server 1 is used to manage and connect the hospital information integration platform 2, the hospital infection detection server 7, and the WeChat enterprise account server 8, and manage the overall process of hospital infection.

The various components in the schematic diagram of the overall structure of a hospital infection monitoring and management device shown in FIG. 1 are described as follows in conjunction with the flowchart of the hospital infection monitoring and management method shown in FIG. 2 :

In step A1, the hospital infection staff organizes the detection methods and policies related to hospital infection, and sends them to the doctor's smartphone 9 through the image and text propaganda and education function provided by the WeChat enterprise account server 8;

Step A2, the hospital infection monitoring and management server 1 uses the hospital infection model establishment method to establish a hospital infection model on the hospital infection detection server 7, the hospital infection monitoring and management server 1 uses the hospital infection detection method to detect the suspected hospital infection event, and pushes it to the doctor, the doctor uses Nosocomial infection reporting method to report nosocomial infection incidents;

Step A3, the hospital infection staff receives and handles the nosocomial infection event;

In step A4, the hospital infection staff organizes the relevant infection cases in the hospital infection monitoring and management server 1, generates a questionnaire, and sends it to the doctor's smartphone 9 through the questionnaire survey function provided by the WeChat enterprise account server 8, and the doctor sends it back to the hospital after completing the filling. Monitoring and management server 1, the hospital infection staff analyzes the data reported by the doctor and the questionnaire survey data, forms rectification opinions, and readjusts the detection methods and policies related to hospital infection.

The various components in the schematic diagram of the overall structure of a hospital infection monitoring and management device shown in FIG. 1 are described as follows in conjunction with the flowchart of the hospital infection reporting method shown in FIG. 3 :

Step B1, set the infection category set in the hospital infection monitoring and management server 1 as {C ₁ , C ₂ ... C _j ... C _J }, where C _j is the j-th infection category, 1≤j≤J, J is the total number of infection categories, C _j = {D ₁ , D ₂ ... D _k ... D _K }, where D _k is the k-th case of the j-th infection category, 1≤k≤K, K is The total number of cases in the jth infection category;

In step B2, the doctor checks the case information of the patient in the hospital by himself, or checks the information of the suspected hospital infection case detected by the hospital infection detection method pushed by the hospital infection monitoring and management server 1;

Step B3, the doctor judges whether the case is a nosocomial infection case, if it is transferred to B4, otherwise transferred to B8;

Step B4, generating infection case D _find ;

Step B5, through the information push function provided by the WeChat enterprise number server 8, push D _find to the hospital staff;

Step B6, the hospital infection staff analyzes the infection case D _find , if it is confirmed to be the infection category C _j , go to B7, otherwise go to B8;

Step B7, store the infection case D _find in the corresponding infection category C _j ;

Step B8, the reporting ends.

The various components in the schematic diagram of the overall structure of a hospital infection monitoring and management device shown in FIG. 1 are described as follows in conjunction with the flowchart of the method for establishing a hospital infection model shown in FIG. 4 :

Step C1, the hospital infection monitoring and management server 1 loads all infection categories as the training set of principal component analysis, and initializes j=0;

Step C2, j=j+1;

Step C3, the hospital infection monitoring and management server 1 applies to the hospital information integration platform 2 to extract the medical index information of all cases in C _j = {D ₁ , D ₂ ... D _k ... D _K };

Step C4, the hospital information integration platform 2 extracts antibiotic drug information {drug _{k, 1} , drug _{k, 2} , ... drug _{k, a} , ... drug _{k, A} } from the hospital information system 3, where drug _{k, a} is the dose of antibiotic used in the kth case, 1≤a≤A, and A is the number of antibiotics in the kth case;

Step C5, the hospital information integration platform 2 extracts blood test information {blood _{k, 1} , blood _{k, 2} , ... blood _{k, b} , ... blood _{k, B} } from the laboratory information system 4, where blood _{k, b} is the value of the bth blood test of the kth case, 1≤b≤B, B is the number of blood test types of the kth case, the hospital information integration platform 2 extracts the drug-resistant bacteria detection information from the laboratory information system 4 { cell _{k, 1} , cell _{k, 2} , ... cell _{k, c} , ... cell _{k, C} }, where cell _{k, c} is the c-th drug-resistant bacteria detection value of the k-th case, 1≤c ≤C, C is the number of drug-resistant bacteria detected in the kth case;

Step C6, the hospital information integration platform 2 extracts the chest X-ray examination report result information xray _k and the CT examination report result information CT _k from the medical image archiving and communication system 5;

Step C7, the hospital information integration platform 2 extracts recent body temperature information {tem _{k, 1} , tem _{k, 2} , ... tem _{k, d} , ... tem _{k, D} } from the electronic medical record system 6, where tem _{k, d} is the d-th body temperature measurement value of the kth case, 1≤d≤D, and D is the number of body temperature measurements of the kth case;

Step C8, after standardizing all medical index information, arrange in columns to form a matrix of all indexes of all cases

where { _xk,1 , _xk,2 ,... _xk,p ,... _xk,P } is given by {drugk _,1 ,drugk _,2 ,...drugk _,a ,.. .drug _k,A },{blood _k,1 ,blood _k,2 ,...blood _k,b ,...blood _k,B },{cell _k,1 ,cell _k,2 ,...cell _k,c ,...cell _{k ,C} },xrayk, _CTk ,{temk _,1 ,temk _,2 ,...temk _,d ,...temk _,D } from left to right Integrated into the same set, P=A+B+C+1+1+D;

Step C9, the hospital infection detection server calculates the correlation coefficient matrix of X

in

in

Step C10, the hospital infection detection server finds the eigenvalues {λ ₁ ,λ ₂ ...λ _p ...λ _P } of the matrix R, and arranges them in order of magnitude, λ ₁ ≥λ ₂ ≥...λ _p ...≥λ _P ≥0 ;

Step C11, the hospital infection detection server generates principal components corresponding to {λ ₁ ,λ ₂ ...λ _p ...λ _P }

Step C12, set the cumulative proportion threshold STEP, i=0, i is the cumulative variable;

Step C13, i=i+1;

Step C14, the hospital infection detection server calculates the cumulative contribution rate of the principal components

Step C15, if setp<STEP, go to C13, otherwise go to C16;

Step C16, the nosocomial infection detection server determines that the feature vector corresponding to the principal component is Detection threshold Test _j =setp of infection category C _j ;

Step C17, the hospital infection detection server generates a detection model of infection category C _j

Step C18, if j<J, go to C2, otherwise go to C19;

Step C19, the establishment of the hospital infection model is completed.

The various components in the schematic diagram of the overall structure of a hospital infection monitoring and management device shown in FIG. 1 are described as follows in conjunction with the flowchart of the hospital infection detection method shown in FIG. 5 :

Step D1, select the cases of patients in the hospital as the detection set of the hospital infection model, the case set of all patients in the hospital is {D ₁ , D ₂ ... D _f ... D _F }, and D _f is the fth inpatient Cases, 1≤f≤F, F is the total number of patients currently in the hospital, initialized f=0;

Step D2, f=f+1;

Step D3, judge whether the patient's case has been judged as a nosocomial infection case; if it is, transfer to D12, otherwise transfer to D4;

Step D4, initialize j=0;

Step D5, j=j+1;

In step D6, the hospital infection monitoring and management server 1 applies to the hospital information integration platform 2 to extract relevant information from the case D _f , and the hospital information integration platform 2 extracts antibiotic drug information {drug _{f, 1} , drug _f, 2 from the hospital information system 3 . , ...drug _{f, a} , ...drug _{f, A} }, the hospital information integration platform 2 draws blood test information from the laboratory information system 4 {blood _{f, 1} , blood _{f, 2} , ...blood _{f, b} ,...blood _f,B }, drug-resistant bacteria detection information {cell _f,1 ,cell _f,2 ,...cell _f,c ,...cell _f,C }, hospital information integration platform 2 from The medical image archiving and communication system 5 extracts the result information xray _{f of the chest X-ray examination report and the result information CT f of} the CT examination report, and the hospital information integration platform 2 extracts the recent body temperature information {tem _{f, 1} , tem _f, 2 from the electronic medical record system 6 . ,...tem _f,d ,...tem _f,D };

Step D7, after standardizing all medical index information, arrange them in sequence as {x _{f, 1} , x _{f, 2} ,...x _{f, p} ,...x _{f, P} };

Step D8, input the medical index information into the detection model in the hospital infection detection server 7 corresponding to the infection category C _j for calculation, current=e ₁ , 1 ×x _f,1 +...e _p,1 ×x _f,p ...+e _P,1 × x _f,P ;

Step D9, judging whether current is greater than Test _j , it is transferred to D10, otherwise it is transferred to D11;

Step D10, the hospital infection monitoring and management server 1 pushes the detected suspected hospital infection cases to the relevant doctors through the information push function provided by the WeChat enterprise account server 8, j=J;

Step D11, if j<J, go to D5, otherwise go to D12;

Step D12, if f<F, go to D2, otherwise go to D13;

Step D13, the nosocomial infection detection is completed.

The following are examples given by the inventors

Example 1

As shown in Figure 6, the lower respiratory tract infection category management process is used as an example, which reflects the overall process of planning (Plan), executing (Do), checking (Check), and adjusting (Action). The specific process is as follows:

Step A1, the hospital infection staff organize the detection methods and policies related to lower respiratory tract infection, and send it to the smart phone 9 of the respiratory doctor through the graphic and text education function as shown in the plan (Plan) in Figure 6 provided by the WeChat enterprise account server 8;

Step A2, as shown in Figure 6 Execute (Do), the hospital infection monitoring and management server 1 uses the hospital infection model establishment method to establish a hospital infection model on the hospital infection detection server 7, and the hospital infection monitoring and management server 1 uses the hospital infection detection method to detect suspected cases. Respiratory tract infection events, and push them to respiratory doctors, who use the hospital infection reporting method to report lower respiratory tract infection events;

Step A3, as shown in the check (Check) in Figure 6, the hospital infection staff receives and handles lower respiratory tract infection events;

Step A4, as shown in Figure 6 Adjustment (Action), the hospital infection staff organizes the relevant lower respiratory tract infection cases in the hospital infection monitoring and management server 1, generates a questionnaire, and sends it to the respiratory department through the questionnaire function provided by the WeChat enterprise account server 8. The doctor's smartphone 1, the respiratory doctor sends it back to the hospital infection monitoring and management server 1 after filling it out, and the hospital infection staff analyzes the data reported by the respiratory doctor and the questionnaire survey data, forms rectification opinions, and readjusts the detection methods and policies related to lower respiratory tract infection .

Example 2

An example of the process of receiving, confirming and reporting a suspected case of lower respiratory tract infection by a respiratory doctor is as follows:

Step B1, set the infection category set in the hospital infection monitoring and management server 1 as {C ₁ , C ₂ ... C _j ... C _J }, where C _j is the lower respiratory tract infection category, 1≤j≤J, J is the total number of infection categories, C _j = {D ₁ , D ₂ ... D _k ... D _K }, where D _k is the kth case of the lower respiratory tract infection category, 1≤k≤K, K is the lower respiratory tract infection the total number of cases in the infection category;

Step B2, the respiratory doctor checks the information of suspected lower respiratory tract infection cases detected by the hospital infection detection method pushed by the hospital infection monitoring and management server 1;

Step B3, the respiratory doctor judges that the case is a nosocomial infection case, and transfers to B4;

Step B4, generating infection case D _find ;

Step B5, through the information push function provided by the WeChat enterprise number server 8, push D _find to the hospital staff;

Step B6, the hospital infection staff analyzes the infection case D _find , confirms it is a lower respiratory tract infection category, and transfers to B7;

Step B7, store the infection case D _find in the lower respiratory tract infection category C _j ;

Step B8, the reporting ends.

Example 3

The upper respiratory tract infection, the lower respiratory tract infection category model generation process, and the detection process using the infection model are examples. The detection model generation process is as follows:

Step C1, the hospital infection monitoring and management server 1 loads the upper respiratory tract infection and the lower respiratory tract infection category as the training set of the principal component analysis, and initializes j=0;

Step C2, j=j+1;

Step C3, the hospital infection monitoring and management server 1 applies to the hospital information integration platform 2 to extract the medical index information of all cases in upper respiratory tract infection;

Step C4, the hospital information integration platform 2 extracts antibiotic drug information {100, 200, ... 200, ... 300} from the hospital information system 3;

Step C5, the hospital information integration platform 2 extracts blood test information {1, 0, ... 1, ... 1} from the laboratory information system 4, and the hospital information integration platform 2 extracts the drug-resistant bacteria detection from the laboratory information system 4 info{0,0,...0,...0};

Step C6, the hospital information integration platform 2 extracts the chest X-ray examination report result information xray _k =0 and the CT examination report result information CT _k =0 from the medical image archiving and communication system 5;

Step C7, the hospital information integration platform 2 extracts recent body temperature information {37.2, 36.5, ... 36.5, ... 38.2} from the electronic medical record system 6, where tem _{k, d} is the d-th body temperature measurement value of the k-th case , 1≤d≤D, D is the number of body temperature measurements of the kth case;

Step C8, after standardizing all medical index information, arrange in columns to form a matrix of all indexes of all cases for

Step C9, the hospital infection detection server 7 calculates the correlation coefficient matrix of X

in

in

Step C10, the hospital infection detection server 7 finds the characteristic root {3.56, 2.32...1.22...0.23} of the matrix R, and arranges it in order of magnitude, λ ₁ ≥λ ₂ ≥...λ _p ...≥λ _P ≥0;

Step C11, the hospital infection detection server 7 generates principal components corresponding to {λ ₁ , λ ₂ ... λ _p ... λ _P }

Step C12, set the cumulative proportion threshold STEP=0.85, i=0, i is the cumulative variable;

Step C13, i=i+1;

Step C14, the hospital infection detection server 7 calculates the cumulative contribution rate of the principal components

Step C15, setp>STEP, transfer to C16;

In step C16, the nosocomial infection detection server 7 determines that the feature vector corresponding to the principal component is: The detection threshold of upper respiratory tract infection Test _j = 0.87;

Step C17, the hospital infection detection server 7 generates a detection model of the infection category C _j

Step C18, j<2, transfer to C2;

Step C2, j=j+1;

Step C3, the hospital infection monitoring and management server 1 applies to the hospital information integration platform 2 to extract the medical index information of all cases in lower respiratory tract infection;

Step C4, the hospital information integration platform 2 extracts antibiotic drug information {300, 400, ... 500, ... 800} from the hospital information system 3;

Step C5, the hospital information integration platform 2 extracts blood test information {1, 1, ... 1, ... 1} from the laboratory information system 4, and the hospital information integration platform 2 extracts the drug-resistant bacteria detection from the laboratory information system 4 info{0,1,...1,...0};

Step C6, the hospital information integration platform 2 extracts the chest X-ray examination report result information xray _k =1, and the CT examination report result information CT _k =1 from the medical image archiving and communication system 5;

Step C7, the hospital information integration platform 2 extracts recent body temperature information {38.2, 38.5, ... 38.5, ... 38.2} from the electronic medical record system 6, where tem _{k, d} is the d-th body temperature measurement value of the k-th case , 1≤d≤D, D is the number of body temperature measurements of the kth case;

Step C8, after standardizing all medical index information, arrange in columns to form a matrix of all indexes of all cases for

Step C9, the hospital infection detection server 7 calculates the correlation coefficient matrix of X

in

in

Step C10, the hospital infection detection server 7 finds the characteristic root {3.96, 3.12...2.22...0.63} of the matrix R, and arranges it in order of magnitude, λ ₁ ≥λ ₂ ≥...λ _p ...≥λ _P ≥0;

Step C11, the hospital infection detection server 7 generates principal components corresponding to {λ ₁ , λ ₂ ... λ _p ... λ _P }

Step C12, set the cumulative proportion threshold STEP=0.85, i=0, i is the cumulative variable;

Step C13, i=i+1;

Step C14, the hospital infection detection server 7 calculates the cumulative contribution rate of the principal components

Step C15, setp>STEP, transfer to C16;

In step C16, the nosocomial infection detection server 7 determines that the feature vector corresponding to the principal component is: The detection threshold of lower respiratory tract infection Test _j = 0.87;

Step C17, the hospital infection detection server 7 generates a detection model of the infection category C _j

Step C18, j==J, go to C19;

In step C19, the upper respiratory tract infection and lower respiratory tract infection category models are established.

The lower respiratory tract infection detection process is as follows:

Step D1, select the cases of patients in the hospital as the detection set of the hospital infection model, the case set of all patients in the hospital is {D ₁ , D ₂ ... D _f ... D _F }, and D _f is the fth inpatient Cases, 1≤f≤F, F is the total number of patients currently in the hospital, initialized f=0;

Step D2, f=f+1;

Step D3, judging that the patient's case is not judged to be a nosocomial case, transfer to D4;

Step D4, initialize j=0;

Step D5, j=j+1;

In step D6, the hospital infection monitoring and management server 1 applies to the hospital information integration platform 2 to extract relevant information from the D _f cases, and the hospital information integration platform 2 extracts antibiotic drug information from the hospital information system 3 {300, 300, ... 400, ...600}, hospital information integration platform 2 extracts blood test information {1, 1, ... 1, ... 1} from laboratory information system 4, drug resistance detection information {0, 0, ... _{1 ,} . The electronic medical record system 6 extracts recent body temperature information {37.5, 38.2, ... 38.5, ... 38.6};

Step D7, after standardizing all medical index information, arrange them in sequence as {1, 0.6, ... 0.5, ... 1};

Step D8, input the medical index information into the lower respiratory tract infection detection model in the hospital infection detection server 7 for calculation, current=0.67×1+...0.22×0.5...+0.02×1=0.91;

Step D9, judge that 0.91 is greater than 0.87, and transfer to D10;

Step D10, the hospital infection monitoring and management server 1 pushes the detected suspected lower respiratory tract infection case to the respiratory doctor through the information push function provided by the WeChat enterprise account server 8, j=J;

Step D11, j==J, go to D12;

Step D12, f==F, go to D13;

In step D13, the nosocomial infection detection is completed.

Claims (5)

1. a hospital infection monitoring and management system is characterized in that comprising: hospital infection monitoring and management server, hospital information integration platform, hospital information system, laboratory information system, medical image archive and communication system, electronic medical record system, hospital infection detection server , WeChat enterprise number server, smart phone; The hospital infection monitoring and management server is connected with the hospital information integration platform through the hospital LAN; The hospital information integration platform is connected with the hospital information system through the hospital LAN; The hospital information integration platform is connected with the laboratory information system through the hospital LAN; The hospital information integration platform is connected with the medical image archiving and communication system through the hospital LAN; The hospital information integration platform is connected with the electronic medical record system through the hospital LAN; The hospital infection monitoring and management server is connected with the hospital infection detection server through the hospital LAN; The hospital infection monitoring and management server is connected with the WeChat enterprise account server through the hospital LAN; The WeChat enterprise number server is connected with the smart phone through the hospital wireless network; The hospital information system is used to collect the patient's name, hospital number, and antibiotic drug order information; The laboratory information system is used for collecting information on blood tests and drug-resistant bacteria detection of patients; The medical image archiving and communication system is used for collecting patient chest X-ray examination and CT examination report result information; The electronic medical record system is used for collecting patient temperature state information; The hospital information integration platform is used for managing and connecting hospital information systems, laboratory information systems, medical image archiving and communication systems, and electronic medical record systems; The nosocomial infection detection server is used for analyzing and processing the basic medical data of patients, and detecting patients suspected of nosocomial infection; The WeChat enterprise account server is used to provide information push, graphic propaganda and education, and questionnaire survey functions; The smart phone is used to receive and display various service information provided by the WeChat enterprise account server, and send relevant feedback information to the WeChat enterprise account server; The hospital infection monitoring and management server is used to manage and connect the hospital information integration platform, the hospital infection detection server, and the WeChat enterprise account server, and to manage the overall process of hospital infection. 2. the monitoring and management method of a kind of hospital infection monitoring and management system according to claim 1, is characterized in that comprising the following steps: Step A1, the hospital infection staff organizes the detection methods and policies related to hospital infection, and sends them to the doctor's smartphone through the image and text education function provided by the WeChat enterprise account server; Step A2, the hospital infection monitoring and management server uses the hospital infection model establishment method to establish a hospital infection model on the hospital infection detection server, the hospital infection monitoring and management server uses the hospital infection detection method to detect the suspected hospital infection event, and pushes it to the doctor, and the doctor reports the hospital infection using the hospital infection detection method. Methods to report nosocomial infection events; Step A3, the hospital infection staff receives and handles the nosocomial infection event; Step A4, the hospital infection staff organizes the relevant infection cases in the hospital infection monitoring and management server, generates a questionnaire, and sends it to the doctor's smartphone through the questionnaire function provided by the WeChat enterprise account server. After the doctor completes the filling, it is sent back to the hospital infection monitoring and management server , the hospital infection staff analyzes the data reported by doctors and the questionnaire survey data, forms rectification opinions, and readjusts the detection methods and policies related to hospital infection. 3. a kind of nosocomial infection monitoring and management method according to claim 2, is characterized in that described nosocomial infection reporting method specifically comprises the following steps: Step B1, set the infection category set in the hospital infection monitoring and management server as {C ₁ , C ₂ ... C _j ... C _J }, where C _j is the j-th infection category, 1≤j≤J, J is the total number of infection categories, C _j = {D ₁ , D ₂ ... D _k ... D _K }, where D _k is the kth case of the jth infection category, 1≤k≤K, K is the kth case The total number of cases in j infection categories; Step B2, the doctor checks the case information of the patient in the hospital by himself, or checks the information of the suspected hospital infection case detected by the hospital infection detection method pushed by the hospital infection monitoring and management server; Step B3, the doctor judges whether the case is a nosocomial infection case, if it is transferred to B4, otherwise transferred to B8; Step B4, generating infection case D _find ; Step B5, push D _find to the hospital staff through the information push function provided by the WeChat enterprise account server; Step B6, the hospital infection staff analyzes the infection case D _find , if it is confirmed to be the infection category C _j , go to B7, otherwise go to B8; Step B7, store the infection case D _find in the corresponding infection category C _j ; Step B8, the reporting ends. 4. a kind of hospital infection monitoring and management method according to claim 2 is characterized in that described hospital infection model establishment method is as follows: Step C1, the hospital infection monitoring and management server loads all infection categories as the training set of principal component analysis, and initializes j=0; Step C2, j=j+1; Step C3, the hospital infection monitoring and management server applies to the hospital information integration platform to extract the medical index information of all cases in C _j = {D ₁ , D ₂ ... D _k ... D _K }; Step C4, the hospital information integration platform extracts antibiotic drug information {drug _{k, 1} , drug _{k, 2} , ... drug _{k, a} , ... drug _{k, A} } from the hospital information system, where drug _{k, a} is The a-th antibiotic dose of the k-th case, 1≤a≤A, A is the number of antibiotics in the k-th case; Step C5, the hospital information integration platform extracts blood test information {blood _{k, 1} , blood _{k, 2} , ... blood _{k, b} , ... blood _{k, B} } from the laboratory information system, where blood _{k, b} is The bth blood test value of the kth case, 1≤b≤B, B is the number of blood test types of the kth case, the hospital information integration platform extracts the drug-resistant bacteria detection information from the laboratory information system {cell _{k, 1} , cell _{k, 2} ,...cell _k,c ,...cell _k,C }, where cell _k,c is the detection value of the c-th drug-resistant bacteria in the k-th case, 1≤c≤C,C is the number of drug-resistant bacteria detected in the kth case; Step C6, the hospital information integration platform extracts the chest X-ray examination report result information xray _k and the CT examination report result information CT _k from the medical image archiving and communication system; Step C7, the hospital information integration platform extracts recent body temperature information {tem _{k, 1} , tem _{k, 2} ,...tem _k,d ,...tem _k,D } from the electronic medical record system, where tem _k,d is the first The value of the d-th body temperature measurement of the k cases, 1≤d≤D, D is the number of body temperature measurements of the k-th case; Step C8, after standardizing all medical index information, arrange in columns to form a matrix of all indexes of all cases where { _xk,1 , _xk,2 ,... _xk,p ,... _xk,P } is given by {drugk _,1 ,drugk _,2 ,...drugk _,a ,.. .drug _k,A },{blood _k,1 ,blood _k,2 ,...blood _k,b ,...blood _k,B },{cell _k,1 ,cell _k,2 ,...cell _k,c ,...cell _{k ,C} },xrayk, _CTk ,{temk _,1 ,temk _,2 ,...temk _,d ,...temk _,D } from left to right Integrated into the same set, P=A+B+C+1+1+D, 1≤p≤P, p is the serial number of medical indicators, and P is the total number of medical indicators; Step C9, the hospital infection detection server calculates the correlation coefficient matrix of X in in Step C10, the hospital infection detection server finds the eigenvalues {λ ₁ ,λ ₂ ...λ _p ...λ _P } of the matrix R, and arranges them in order of magnitude, λ ₁ ≥λ ₂ ≥...λ _p ...≥λ _P ≥0 ; Step C11, the hospital infection detection server generates principal components corresponding to {λ ₁ ,λ ₂ ...λ _p ...λ _P } Step C12, set the cumulative proportion threshold STEP, i=0, i is the cumulative variable; Step C13, i=i+1; Step C14, the hospital infection detection server calculates the cumulative contribution rate of the principal components 1≤i≤P; Step C15, if setp<STEP, go to C13, otherwise go to C16; Step C16, the nosocomial infection detection server determines that the feature vector corresponding to the principal component is Detection threshold Test _j =setp of infection category C _j ; Step C17, the hospital infection detection server generates a detection model of infection category C _j Step C18, if j<J, go to C2, otherwise go to C19; Step C19, the establishment of the nosocomial infection model is completed; where C _j is the j-th infection category, 1≤j≤J, J is the total number of infection categories, C _j ={D ₁ ,D ₂ ... D _k ... D _K }, where D _k is the j-th The kth case of the infection category, 1≤k≤K, where K is the total number of cases in the jth infection category. 5. a kind of nosocomial infection monitoring and management method according to claim 2, is characterized in that described nosocomial infection detection method is as follows: Step D1, select the cases of patients in the hospital as the detection set of the hospital infection model, the case set of all patients in the hospital is {D ₁ , D ₂ ... D _f ... D _F }, and D _f is the fth inpatient Cases, 1≤f≤F, F is the total number of patients currently in the hospital, initialized f=0; Step D2, f=f+1; Step D3, judge whether the patient's case has been judged as a nosocomial infection case; if it is, transfer to D12, otherwise transfer to D4; Step D4, initialize j=0; Step D5, j=j+1; Step D6, the hospital infection monitoring and management server applies to the hospital information integration platform to extract relevant information from the D _f cases, and the hospital information integration platform extracts antibiotic drug information {drug _{f, 1} , drug _{f, 2} , ... drug _{f, a} ,...drug _{f, A} }, the hospital information integration platform draws blood test information from the laboratory information system {blood _{f, 1} , blood _{f, 2} ,...blood _f,b ,...blood _{f, B} }, drug-resistant bacteria detection information {cell _{f, 1} , cell _{f, 2} , ... cell _{f, c} , ... cell _{f, C} }, the hospital information integration platform extracts from the medical image archiving and communication system Chest X-ray examination report result information xray _f , CT examination report result information CT _f , the hospital information integration platform extracts recent body temperature information from the electronic medical record system {tem _{f, 1} , tem _{f, 2} ,...tem _{f, d} ,. ..tem _{f, D} }; Step D7, after standardizing all medical index information, arrange them in sequence as {x _{f, 1} , x _{f, 2} ,...x _{f, p} ,...x _{f, P} }; Step D8, input the medical index information into the detection model in the hospital infection detection server corresponding to the infection category C _j for calculation, current=e ₁ , 1 ×x _f,1 +...e _p,1 ×x _f,p . ..+e _P,1 × x _f,P ; Step D9, judging whether current is greater than Test _j , it is transferred to D10, otherwise it is transferred to D11; Step D10, the hospital infection monitoring and management server pushes the detected suspected hospital infection cases to the relevant doctors through the information push function provided by the WeChat enterprise account server, j=J; Step D11, if j<J, go to D5, otherwise go to D12; Step D12, if f<F, go to D2, otherwise go to D13; Step D13, the nosocomial infection detection is completed; J is the total number of infection categories; A is the number of antibiotic types in the kth case; B is the number of blood test types in the kth case; C is the number of drug-resistant bacteria detected in the kth case; D is the number of drug-resistant bacteria in the kth case Number of body temperature measurements; P is the total number of medical indicators.