Digitalized diagnosis and evaluation method for infection risk integrated in medical information system

Abstract

The invention relates to an infection risk digital diagnosis and evaluation method integrated in a medical information system, and belongs to the technical field of risk evaluation. The method comprises the following steps: acquiring three-tube risk screening data, and automatically generating a risk score through a medical information system according to the three-tube risk screening data; a three-management early warning threshold value is preset, and the medical information system obtains continuous prediction data through a monitoring early warning model according to the risk score and the three-management early warning threshold value; obtaining patient infection risk digital diagnosis data through a risk prediction model according to the continuous prediction data; and obtaining a personnel optimization scheduling result through a nursing personnel scheduling model according to the patient infection risk digital diagnosis data. The infection risk assessment integrated in the medical information system is realized, the accuracy of infection risk assessment is improved, and the nursing service is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of risk assessment, and in particular relates to a digital diagnosis and assessment method for infection risk integrated into a medical information system. Background Art

[0002] Since the 1960s and 1970s, the construction of medical information technology has been continuously promoted, and the progress of Internet information technology has prompted medical institutions to enter the information age. As the core medical data resource, the medical information system not only records the patient's hospitalization information, nursing level information, the total number of hospitalized patients, the total number of graded nursing, etc., but also effectively reduces the problem of information islands, simplifies the nursing process, and improves medical efficiency. In view of the high infection risk of "three-tube" patients, with the construction of the medical information system, it is possible to actively monitor the patient's infection during the process, reduce the infection rate, and improve the quality of nursing.

[0003] With the development of artificial intelligence technology, we are no longer satisfied with timely discovering the infection risk of patients during the process. We hope to be able to make timely predictions of the infection risk of patients, make timely adjustments to nursing methods, and make corresponding countermeasures. Traditional prediction methods based on random forests cannot fully utilize the contextual relationship of data in medical information systems. At the same time, when faced with unstructured data, a large amount of data preprocessing is required, which is not suitable for medical data processing with large data volumes. Using methods such as long short-term memory networks, gated recurrent units, and bidirectional long short-term memory networks to learn patient data for infection risk prediction can help reduce the infection rate, but it often uses a single data in the patient data and cannot fully learn the complex relationships in the patient data. Therefore, the above methods still have limitations in processing unstructured multimodal patient data to predict and assess infection risks. Summary of the invention

[0004] In order to solve the above problems existing in the prior art, the present invention provides a digital diagnosis and assessment method for infection risk integrated into a medical information system.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A digital diagnosis and assessment method for infection risk integrated into a medical information system, comprising:

[0007] Acquiring three-tube risk screening data, and automatically generating a risk score through a medical information system based on the three-tube risk screening data;

[0008] A three-channel early warning threshold is preset, and the medical information system obtains continuous prediction data through a monitoring early warning model according to the risk score and the three-channel early warning threshold;

[0009] Obtaining digital diagnostic data of the patient's infection risk through a risk prediction model based on the continuous prediction data;

[0010] According to the digital diagnosis data of the patient's infection risk, a nursing staff scheduling model is used to obtain a personnel optimization scheduling result.

[0011] Preferably, the medical information system obtains continuous prediction data through monitoring and early warning models according to the risk score and the three-pipe early warning thresholds, including:

[0012] The three-control warning thresholds include low risk threshold, medium risk threshold, and high risk threshold; the risk warning levels include low risk, medium risk, and high risk;

[0013] The risk warning level is obtained by risk level judgment according to the risk score and the three-control warning threshold;

[0014] Performing early warning judgment on the risk early warning level to obtain the continuous prediction data.

[0015] Preferably, the risk level judgment includes judging whether the risk score is not greater than the low risk threshold value, if so, the risk warning level is the low risk; if not, judging whether the risk score is not greater than the medium risk threshold value, if so, the risk warning level is the medium risk; if not, the risk warning level is the high risk.

[0016] Preferably, the early warning judgment includes judging whether the risk warning level is the high risk. If so, the medical information system transmits the three-tube risk screening data and the risk warning level to the smart terminal for early warning; if not, the continuous prediction data is obtained, and the continuous prediction data is the three-tube risk screening data.

[0017] Preferably, obtaining digital diagnostic data of the patient's infection risk through a risk prediction model based on the continuous prediction data includes:

[0018] Obtaining preprocessed data to be predicted through a preprocessing model according to the continuous prediction data;

[0019] According to the preprocessed data to be predicted, a word vector is obtained by mapping the word vector using the CBOW model;

[0020] According to the word vector, feature extraction is performed through LSTM to obtain local attention features;

[0021] The digital diagnostic data of the patient's infection risk is obtained based on the local attention features through a Transformer encoder model.

[0022] Preferably, obtaining the patient's infection risk digital diagnostic data through a Transformer encoder model according to the local attention feature includes:

[0023] Obtaining a projection value by inter-vector projection calculation according to the local attention feature;

[0024] Obtaining a normalized projection value by attention normalization calculation according to the projection value;

[0025] Obtaining the global attention feature by weighted summing the normalized projection value and the local attention feature;

[0026] Obtaining a hidden state through a Transformer encoder according to the global attention feature;

[0027] The digital diagnosis data of the patient's infection risk is obtained by calculating the hidden state through a fully connected layer.

[0028] Preferably, obtaining the personnel optimization scheduling result through the nursing staff scheduling model according to the patient's infection risk digital diagnosis data includes:

[0029] Acquire patient information data and nursing resource data, use the patient information data, the nursing resource data, and the patient infection risk digital diagnosis data as a state space, and assign nursing staff as an action space;

[0030] A reward function is allocated according to the state space and the action space through a reward function definition;

[0031] The personnel optimization scheduling result is obtained according to the state space, the action space, and the scheduling reward function through the deep Q network algorithm DQN.

[0032] A digital infection risk diagnosis and assessment system integrated into a medical information system, comprising a data processing module, a risk monitoring module, a risk prediction module, and a personnel scheduling module;

[0033] The data processing module is used to obtain three-tube risk screening data and automatically generate a risk score through a medical information system based on the three-tube risk screening data;

[0034] The risk monitoring module is used to preset the three-pipe warning thresholds, and the medical information system obtains continuous prediction data through the monitoring and early warning model according to the risk score and the three-pipe warning thresholds;

[0035] The risk prediction module is used to obtain digital diagnostic data of the patient's infection risk through a risk prediction model based on the continuous prediction data;

[0036] The personnel scheduling module is used to obtain the personnel optimization scheduling result through the nursing staff scheduling model according to the patient's infection risk digital diagnosis data.

[0037] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the digital diagnosis and assessment method for infection risk integrated into a medical information system is implemented.

[0038] A storage medium containing computer executable instructions, which, when executed by a computer processor, are used to execute the above-mentioned digital diagnosis and assessment method for infection risk integrated into a medical information system.

[0039] The beneficial effects of the present invention are:

[0040] (1) Through the monitoring and early warning model, patients with high infection risk can be screened out in time and timely intervention can be carried out; at the same time, through continuous data prediction, continuous infection risk assessment can be formed for medium-risk and low-risk patients, potential risks can be identified in advance, and digital diagnosis of patient infection risks can be achieved, thereby reducing high-risk infection rates and improving service quality.

[0041] (2) The continuous prediction data is predicted by a two-layer attention neural network model based on LSTM and Transformer encoder, which can well learn the multimodal information of the continuous prediction data and the complex relationship between the data, improve the accuracy of risk infection prediction, and lay the foundation for subsequent nursing staff scheduling.

[0042] (3) Through the deep Q network, nursing staff are allocated and scheduled according to the digital diagnostic data of patients' infection risk. Nursing staff can be allocated according to the patient's infection risk level, avoiding the waste of medical resources and improving nursing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0044] Figure 1 The figure is a flow chart of a digital diagnosis and assessment method for infection risk integrated into a medical information system of the present invention. DETAILED DESCRIPTION

[0045] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0046] See also Figure 1, a digital diagnosis and assessment method for infection risk integrated into a medical information system, comprising:

[0047] S1: Acquire three-tube risk screening data, and automatically generate a risk score through a medical information system based on the three-tube risk screening data;

[0048] S2: presetting the three-channel warning threshold, the medical information system obtains continuous prediction data through the monitoring and warning model according to the risk score and the three-channel warning threshold;

[0049] S3: Obtaining digital diagnostic data of the patient's infection risk through a risk prediction model based on the continuous prediction data;

[0050] S4: Obtaining personnel optimization scheduling results through a nursing staff scheduling model according to the patient's digital diagnosis data on infection risk.

[0051] Specifically, "three tubes" refer to central catheter-related bloodstream infection, catheter-related urinary tract infection, and ventilator-associated pneumonia.

[0052] Specifically, the three-tube risk screening data include patient conditions and catheter usage; the patient conditions include patient basic information, underlying medical conditions, infection history, immune status, surgical information, patient body temperature, and patient blood indicators; the catheter usage includes catheter type, catheter placement location, insertion time and duration, and catheter care records.

[0053] In this embodiment, the medical information system process specifically includes first displaying patient information data, and the patient information data includes patient hospitalization information and nursing level information. The medical information system automatically captures and displays key data of hospitalized patients based on the patient information data, including the total number of hospitalized patients, the number of nursing levels, and the number of high-risk patients in the "three-control" category. Then, the information of new patients is obtained, and the information of new patients is entered into a risk screening form. The medical information system automatically generates a risk score based on the risk screening form. A risk warning threshold is preset, and based on the risk score and the risk warning threshold, the medical information system automatically starts a three-level warning to obtain a warning result. The three-level warning includes a green light for low risk, a yellow light for medium risk, and a red light for high risk. The medical information system pushes the warning result to the nurse and manager interface.

[0054] In step S2, the monitoring and early warning model includes:

[0055] S201: The three-control warning thresholds include a low risk threshold, a medium risk threshold, and a high risk threshold; the risk warning levels include low risk, medium risk, and high risk;

[0056] S202: Obtaining the risk warning level through risk level judgment according to the risk score and the three-control warning thresholds;

[0057] The risk level judgment includes judging whether the risk score is not greater than the low risk threshold, if yes, the risk warning level is the low risk; if no, judging whether the risk score is not greater than the medium risk threshold, if yes, the risk warning level is the medium risk; if no, the risk warning level is the high risk;

[0058] S203: Performing early warning judgment on the risk early warning level to obtain the continuous prediction data;

[0059] The early warning judgment includes judging whether the risk warning level is the high risk. If yes, the medical information system transmits the three-tube risk screening data and the risk warning level to the smart terminal for early warning; if not, the continuous prediction data is obtained, and the continuous prediction data is the three-tube risk screening data.

[0060] In step S3, the risk prediction model includes:

[0061] S301: obtaining preprocessed data to be predicted through a preprocessing model according to the continuous prediction data;

[0062] S302: Obtain a word vector by mapping the preprocessed data to be predicted through a CBOW model;

[0063] S303: Extracting features through LSTM according to the word vector to obtain local attention features;

[0064] S304: Obtain the digital diagnosis data of the patient's infection risk through a Transformer encoder model according to the local attention features.

[0065] Specifically, the preprocessing model includes linear interpolation to fill missing values, exponential smoothing, and normalization processing.

[0066] Specifically, the CBOW model is a natural language processing model for word vector training, which embeds words into a vector space so that similar words are close to each other in the vector space.

[0067] In step S304, the Transformer encoder encoder model includes:

[0068] S304-1: Obtaining a projection value by inter-vector projection calculation according to the local attention feature;

[0069] The inter-vector projection calculation expression is:

[0070]

[0071] in, is the projection value, X represents the local attention feature, and T represents the transposition operation;

[0072] S304-2: Obtaining a normalized projection value by performing attention normalization calculation according to the projection value;

[0073] S304-3: Obtaining the global attention feature by weighted summing the normalized projection value and the local attention feature;

[0074] The attention weighted sum calculation expression is:

[0075]

[0076] Where Z represents the global attention feature, is the projection value, is the normalized projection value, X represents the local attention feature, and softmax is the normalization function;

[0077] S304-4: Obtaining a hidden state through a Transformer encoder according to the global attention feature;

[0078] S304-5: Obtain digital diagnostic data of the patient's infection risk through calculation using a fully connected layer according to the hidden state.

[0079] In this embodiment, the Transformer encoder model is based on a multi-head attention mechanism, and the multi-head attention mechanism has 8 self-attention heads.

[0080] In step S4, the nursing staff scheduling model includes:

[0081] S401: Acquire patient information data and nursing resource data, use the patient information data, the nursing resource data, and the digital diagnosis data of the patient's infection risk as a state space, and assign nursing staff as an action space;

[0082] S402: Obtaining a distribution reward function according to the state space and the action space through a reward function definition;

[0083] The reward function definition expression is:

[0084]

[0085] Wherein, R represents the allocation reward function, ω1, ω2, ω3 are weight coefficients, s represents the state space, a represents the action space, L(s,a) represents the coverage rate of high-risk patients who perform action a in state s, H(s,a) represents the workload balance of the nursing staff who perform action a in state s, W(s,a) represents the waste of nursing resources when performing action a in state s, N r (s,a) represents the number of high-risk patients with assigned caregivers who perform action a in state s, N patient represents the total number of high-risk patients, i represents the i-th nurse, L i represents the actual workload of the i-th nurse, represents the average workload of all nursing staff, N represents the total number of nursing staff, M represents the total number of patients, j represents the jth patient, and R j represents the nursing resources allocated to the jth patient, P j is the risk warning level of the jth patient;

[0086] S403: Obtain the personnel optimization scheduling result according to the state space, the action space, and the scheduling reward function through a deep Q network algorithm DQN.

[0087] Specifically, the patient information data includes the patient condition, the catheter usage, and the risk warning level; the nursing resource data includes the number of nursing staff, patient distribution, and the total number of high-risk patients.

[0088] A digital infection risk diagnosis and assessment system integrated into a medical information system, comprising a data processing module, a risk monitoring module, a risk prediction module, and a personnel scheduling module;

[0089] The data processing module is used to obtain three-tube risk screening data and automatically generate a risk score through a medical information system based on the three-tube risk screening data;

[0090] The risk monitoring module is used to preset the three-pipe warning thresholds, and the medical information system obtains continuous prediction data through the monitoring and early warning model according to the risk score and the three-pipe warning thresholds;

[0091] The risk prediction module is used to obtain digital diagnostic data of the patient's infection risk through a risk prediction model based on the continuous prediction data;

[0092] The personnel scheduling module is used to obtain the personnel optimization scheduling result through the nursing staff scheduling model according to the patient's infection risk digital diagnosis data.

[0093] The computer storage medium of the embodiment of the present invention may adopt any combination of one or more computer-readable media. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device.

[0094] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0095] The program code included on the computer readable medium can be transmitted with any appropriate medium, including but not limited to wireless, electric wire, optical cable, RF, etc., or any suitable combination of the above. The computer program code for performing the operation of the present invention can be written in one or more programming languages ​​or their combinations, and the programming language includes object-oriented programming languages-such as Java, Smalltalk, C++, and also includes conventional procedural programming languages-such as "C" language or similar programming languages. The program code can be executed completely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on the remote computer, or completely on the remote computer or server. In the case of a remote computer, 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 it can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0096] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. A digital diagnosis and assessment method for infection risk integrated into a medical information system, characterized in that: The following steps are involved: Acquiring three-tube risk screening data, and automatically generating a risk score through a medical information system based on the three-tube risk screening data; A three-channel early warning threshold is preset, and the medical information system obtains continuous prediction data through a monitoring early warning model according to the risk score and the three-channel early warning threshold; Obtaining digital diagnostic data of the patient's infection risk through a risk prediction model based on the continuous prediction data; According to the digital diagnosis data of the patient's infection risk, a nursing staff scheduling model is used to obtain a personnel optimization scheduling result.

2. The digital diagnosis and assessment method for infection risk integrated into a medical information system according to claim 1, characterized in that: The medical information system obtains continuous prediction data through monitoring and early warning models according to the risk score and the three-pipe early warning thresholds, including: The three-control warning thresholds include low risk threshold, medium risk threshold, and high risk threshold; the risk warning levels include low risk, medium risk, and high risk; The risk warning level is obtained by risk level judgment according to the risk score and the three-control warning threshold; Performing early warning judgment on the risk early warning level to obtain the continuous prediction data.

3. The digital diagnosis and assessment method for infection risk integrated into a medical information system according to claim 2, characterized in that: The risk level judgment includes judging whether the risk score is not greater than the low risk threshold, and if so, the risk warning level is the low risk; If not, it is determined whether the risk score is not greater than the medium risk threshold; if yes, the risk warning level is the medium risk; No, the risk warning level is the high risk.

4. The digital diagnosis and assessment method for infection risk integrated into a medical information system according to claim 2, characterized in that: The early warning judgment includes judging whether the risk warning level is the high risk. If yes, the medical information system transmits the three-tube risk screening data and the risk warning level to the smart terminal for early warning; If not, the continuous prediction data is obtained, and the continuous prediction data is the three-tube risk screening data.

5. The digital diagnosis and assessment method for infection risk integrated into a medical information system according to claim 1, characterized in that: The step of obtaining the patient's infection risk digital diagnostic data through a risk prediction model according to the continuous prediction data includes: Obtaining preprocessed data to be predicted through a preprocessing model according to the continuous prediction data; According to the preprocessed data to be predicted, a word vector is obtained by mapping the word vector using the CBOW model; Perform feature extraction through LSTM according to the word vector to obtain local attention features; The digital diagnostic data of the patient's infection risk is obtained based on the local attention features through a Transformer encoder model.

6. The digital diagnosis and assessment method for infection risk integrated into a medical information system according to claim 1, characterized in that: The step of obtaining the patient's infection risk digital diagnostic data through a Transformer encoder model according to the local attention feature includes: Obtaining a projection value by inter-vector projection calculation according to the local attention feature; Obtaining a normalized projection value by attention normalization calculation according to the projection value; Obtaining the global attention feature by weighted summing the normalized projection value and the local attention feature; Obtaining a hidden state through a Transformer encoder according to the global attention feature; The digital diagnosis data of the patient's infection risk is obtained by calculating the hidden state through a fully connected layer.

7. The digital diagnosis and assessment method for infection risk integrated into a medical information system according to claim 1, characterized in that: The optimizing scheduling result of the personnel is obtained by using the nursing staff scheduling model according to the digital diagnosis data of the patient's infection risk, including: Acquire patient information data and nursing resource data, use the patient information data, the nursing resource data, and the patient infection risk digital diagnosis data as a state space, and assign nursing staff as an action space; A reward function is allocated according to the state space and the action space through a reward function definition; The personnel optimization scheduling result is obtained according to the state space, the action space, and the scheduling reward function through the deep Q network algorithm DQN.

8. A digital infection risk diagnosis and assessment system integrated into a medical information system, characterized in that: Including data processing module, risk monitoring module, risk prediction module, and personnel scheduling module; The data processing module is used to obtain three-tube risk screening data and automatically generate a risk score through a medical information system based on the three-tube risk screening data; The risk monitoring module is used to preset the three-pipe warning thresholds, and the medical information system obtains continuous prediction data through the monitoring and early warning model according to the risk score and the three-pipe warning thresholds; The risk prediction module is used to obtain digital diagnostic data of the patient's infection risk through a risk prediction model based on the continuous prediction data; The personnel scheduling module is used to obtain the personnel optimization scheduling result through the nursing staff scheduling model according to the patient's infection risk digital diagnosis data.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the digital diagnosis and assessment method for infection risk integrated into a medical information system as described in any one of claims 1 to 7 is implemented.

10. A storage medium containing computer executable instructions, characterized in that: The computer executable instructions, when executed by a computer processor, are used to execute the digital diagnosis and assessment method for infection risk integrated into a medical information system as described in any one of claims 1 to 7.

Images