CN115545580B

CN115545580B - Medical training process standardization verification method and system

Info

Publication number: CN115545580B
Application number: CN202211524414.5A
Authority: CN
Inventors: 龚姝; 李进; 徐禹; 魏申毅; 王磊
Original assignee: West China Hospital of Sichuan University
Current assignee: West China Hospital of Sichuan University
Priority date: 2022-12-01
Filing date: 2022-12-01
Publication date: 2023-04-07
Anticipated expiration: 2042-12-01
Also published as: CN115545580A

Abstract

The embodiment of the invention provides a medical training process standardization verification method and system, wherein in the method, when a verification instruction is received, a medical training activity log to be verified is obtained, then training process description results corresponding to each training node data set in the medical training activity log to be verified are obtained first, j training process description results are obtained, then j training process description result weights are obtained based on the j training process description results, j description result knowledge fields are obtained based on the j training process description result weights, and finally j training node compliance evaluation information corresponding to the medical training activity log to be verified is obtained by adopting an evaluation model according to the j description result knowledge fields. According to the embodiment of the application, the compliance evaluation information of each training node is efficiently acquired, manpower is saved, the fairness and the justice of verification are guaranteed, processing and mining of related data are not needed by means of extra computing resources, and the computing resources are saved.

Description

Medical training process standardization verification method and system

Technical Field

The application relates to the field of medical training, in particular to a medical training process standardization verification method and system.

Background

The medical education training is different from the traditional examination taking education, and the judgment of the learning result is carried out only through a single theoretical examination. The method has strict requirements on multiple links such as course setting, training range, training process, assessment process, training objects and the like. In the process of medical graduation back education and training, trainees can generate training data in all links and all aspects, for example, when the trainees are managed through an online training system, a series of data such as attendance conditions of the trainees, training duration, rotation department sequence and duration, signers information, employment assessment data, assessment approval results, matching conditions of talent with teachers and trainees and learning of the trainees can be recorded. However, for medical care training requiring strong standardized training, training nodes involved in the training process are often checked one by one through manpower and office software, the checking efficiency is low, standardized checking cannot be guaranteed, and erroneous judgment or subjective erroneous judgment is often generated in the middle. If the data records generated by the online training system can be analyzed efficiently and objectively, the obtained verification result can be relatively objective and fair, and the standardization of training is facilitated.

Disclosure of Invention

The invention aims to provide a medical training process standardization verification method and a medical training process standardization verification system so as to solve the problems.

The embodiment of the application is realized as follows:

in a first aspect, an embodiment of the present application provides a medical training procedure standardization verification method, which is applied to a verification system, and the method includes:

responding to the received checking instruction, and acquiring a medical training activity log to be checked;

acquiring training process description results corresponding to each training node data set in the medical training activity log to be verified to obtain j training process description results; the medical training activity log to be verified comprises j training node data sets, and the training node data sets and the training process description results are mapped one by one;

obtaining j training process description result weights based on the j training process description results; the training process description result weight indicates the examination importance ratio of the training node data set to the medical training activity log to be verified in the training process, and the training process description result weight and the training node data set are mapped one by one;

obtaining j description result knowledge fields based on the j training process description result weights; wherein the description result knowledge fields are mapped one-to-one with the training node data set;

acquiring the compliance evaluation information of j training nodes corresponding to the medical training activity log to be verified by adopting an evaluation model according to the j description result knowledge fields; wherein the training node compliance evaluation information is mapped one-to-one with the training node data set.

Optionally, the obtaining of the training process description result corresponding to each training node data set in the medical training activity log to be verified to obtain j training process description results includes:

acquiring a characteristic value matched with the ith training node data set aiming at the ith training node data set in the medical training activity log to be verified; wherein the ith training node data set is any one of the j training node data sets, i < j;

aiming at the ith training node data set in the medical training activity log to be verified, acquiring training process description result probability corresponding to the ith training node data set;

and aiming at the ith training node data set in the medical training activity log to be verified, determining a training process description result corresponding to the ith training node data set based on the training process description result probability corresponding to the ith training node data set and the characteristic value matched with the ith training node data set.

Optionally, the obtaining of the weights of the j training process description results based on the j training process description results includes:

aiming at the ith training node data set in the medical training activity log to be verified, determining a characteristic analysis result corresponding to the ith training node data set according to a training process description result of the ith training node data set; wherein the ith training node data set is any one of the j training node data sets, and i < j;

aiming at the ith training node data set in the medical training activity log to be verified, obtaining training process description result probability corresponding to the ith training node data set;

and aiming at the ith training node data set in the medical training activity log to be verified, obtaining the training process description result weight corresponding to the ith training node data set based on the training process description result probability corresponding to the ith training node data set and the characteristic analysis result corresponding to the ith training node data set.

Optionally, the determining j description result knowledge fields based on the j training process description result weights includes:

aiming at the ith training node data set in the medical training activity log to be verified, acquiring a pooling strategy corresponding to the ith training node data set; wherein the ith training node data set is any one of the j training node data sets, and i is less than j;

and aiming at the ith training node data set in the medical training activity log to be verified, obtaining a description result knowledge field corresponding to the ith training node data set based on the pooling strategy corresponding to the ith training node data set and the training process description result weight corresponding to the ith training node data set.

Optionally, the obtaining, by using an evaluation model according to the j description result knowledge fields, the j training node compliance evaluation information corresponding to the medical training activity log to be verified includes:

acquiring j target description result knowledge fields by adopting a plurality of groups of feature extraction modules in the evaluation model according to the j description result knowledge fields;

and acquiring the compliance evaluation information of j training nodes by adopting a multilayer deep neural network in the evaluation model according to the j target description result knowledge fields.

Optionally, before the obtaining, by using an evaluation model, the compliance evaluation information of j training nodes corresponding to the medical training activity log to be verified according to the j description result knowledge fields, the method further includes:

acquiring a training-simulated training process description result corresponding to each training-simulated data set in a training-simulated medical training activity log to obtain j training-simulated training process description results; the log of activity of the quasi-training medical training comprises j quasi-training data sets, and the quasi-training data sets are mapped with the description results of the quasi-training process one by one;

determining weights of j training procedure description results to be trained based on the j training procedure description results to be trained; wherein the quasi-training process describes one-to-one mapping of result weights and the quasi-training data set;

determining j training simulation description result knowledge fields based on the j training simulation training process description result weights; wherein, the quasi-training description result knowledge field is mapped with the quasi-training data set one by one;

according to the j quasi-training description result knowledge fields, adopting a quasi-training evaluation model to obtain j quasi-training node compliance evaluation information corresponding to the quasi-training medical training activity log; the compliance evaluation information of the training nodes to be trained is mapped with the data sets to be trained one by one;

acquiring j real-value training node compliance evaluation information corresponding to a real-value medical training activity log;

and training the quasi-training evaluation model through the j truth-value training node compliance evaluation information and the j quasi-training node compliance evaluation information until a training cut-off condition is met, and obtaining the evaluation model.

Optionally, the obtaining of the compliance evaluation information of j true-value training nodes corresponding to the true-value medical training activity log includes:

acquiring truth value training process description results corresponding to each truth value data set in a truth value medical training activity log to obtain j truth value training process description results; wherein the truth medical training activity log comprises j truth value data sets, and the truth value data sets are mapped with the truth value training process description results one by one;

determining weights of the j truth training process description results based on the j truth training process description results; the weight of the real-value training process description result is mapped with the real-value training process description result one by one;

determining j truth value description result knowledge fields based on the j truth value training process description result weights; wherein the truth value description result knowledge field is mapped with the truth value data set one by one;

acquiring the compliance evaluation information of j true-value training nodes corresponding to the true-value medical training activity log through a quasi-training evaluation model according to the j true-value description result knowledge fields; the real-value training node compliance evaluation information is mapped with the real-value data set one by one;

the training the quasi-training evaluation model through the j true value training node compliance evaluation information and the j quasi-training node compliance evaluation information until a training cutoff condition is met to obtain the evaluation model, and the method comprises the following steps:

aiming at each quasi-training data set in the quasi-training medical training activity log and a truth value data set corresponding to each quasi-training data set, acquiring the fitting evaluation information of the quasi-training nodes and the cost values of the truth value training node fitting evaluation information through a metric learning objective function, and acquiring j cost values;

adjusting parameters of the quasi-training evaluation model according to the j cost values;

and when the training cutoff condition is met, acquiring the evaluation model according to the adjusted parameters.

Optionally, before the j training node compliance evaluation information corresponding to the medical training activity log to be verified is obtained by using an evaluation model based on the j description result knowledge fields, the method further includes:

acquiring a training-simulated training process description result corresponding to each training-simulated data set in a training-simulated medical training activity log to obtain j training-simulated training process description results; the medical training activity log to be trained comprises j training data sets, and the training data sets and the training process description results to be trained are mapped one by one;

obtaining j weights of the description results of the training process to be trained based on the j description results of the training process to be trained; wherein the quasi-training process description result weights are mapped with the quasi-training data set one by one;

determining j training-to-be-trained description result knowledge fields based on the j training-to-be-trained training process description result weights; mapping the quasi-training description result knowledge fields and the quasi-training data sets one by one;

acquiring the compliance evaluation information of j training-to-be-trained training nodes corresponding to the medical training activity log to be trained through a training-to-be-trained evaluation model according to the j training-to-be-trained description result knowledge fields; the compliance evaluation information of the training nodes to be trained is mapped with the data sets to be trained one by one;

acquiring the compliance evaluation information of j false value training nodes corresponding to the false value medical training activity log;

and training the quasi-training evaluation model according to the j true value training node compliance evaluation information, the j false value training node compliance evaluation information and the j quasi-training node compliance evaluation information until a training cut-off condition is met, and obtaining the evaluation model.

Optionally, the obtaining of the compliance evaluation information of j truth training nodes corresponding to the truth medical training activity log includes:

acquiring truth value training process description results corresponding to each truth value data set in a truth value medical training activity log to obtain j truth value training process description results; wherein the truth medical training activity log comprises j truth value data sets, and the truth value data sets are mapped with the truth training process description results one by one;

determining weights of j truth training process description results based on the j truth training process description results; the weight of the real-value training process description result is mapped with the real-value training process description result one by one;

determining j truth describing result knowledge fields based on the j truth training process describing result weights; the truth value description field and the truth value data set are mapped one by one;

the acquiring of the compliance evaluation information of j false value training nodes corresponding to the false value medical training activity log comprises the following steps:

acquiring a false value training process description result corresponding to each false value data set in a false value medical training activity log to obtain j false value training process description results; wherein the log of the medical training activities of the false values comprises j data sets of the false values, and the data sets of the false values are mapped with the description results of the medical training processes of the false values one by one;

determining j false value training process description result weights based on the j false value training process description results; wherein the weight of the false value training process description result is mapped with the false value training process description result one by one;

determining j false value description result knowledge fields based on the j false value training process description result weights; wherein the false value description result knowledge field is mapped with the false value data set one by one;

acquiring the compliance evaluation information of j false value training nodes corresponding to the false value medical training activity log through a quasi-training evaluation model according to the j false value description result knowledge fields; the compliance evaluation information of the false value training nodes is mapped with the false value data sets one by one;

the training of the quasi-training evaluation model based on the j true value training node compliance evaluation information, the j false value training node compliance evaluation information and the j quasi-training node compliance evaluation information until a training cutoff condition is met to obtain the evaluation model comprises:

aiming at a truth value data set corresponding to each quasi-training data set in the quasi-training medical training activity log, acquiring the quasi-training node compliance evaluation information and a first generation value of the truth value training node compliance evaluation information through a first metric learning objective function to obtain j first generation values;

aiming at each quasi-training data set and a false value data set in the quasi-training medical training activity log, acquiring the compliance evaluation information of the quasi-training nodes and a second generation value which is the compliance evaluation information of the false value training nodes through a second metric learning objective function to obtain j second generation values;

adjusting parameters of the quasi-training evaluation model according to the j first generation values and the j second generation values;

and when the training cutoff condition is met, acquiring the evaluation model through the adjusted parameters.

In a second aspect, an embodiment of the present application provides a verification system, where the system includes a processor and a memory, where the memory stores a program, and the processor is configured to retrieve a computer program from the memory and implement the method provided in the first aspect of the present application by executing the computer program.

According to the medical training process standardization verification method and system, when a verification instruction is received, a medical training activity log to be verified is obtained, then training process description results corresponding to each training node data set in the medical training activity log to be verified are obtained first, j training process description results are obtained, then j training process description result weights are obtained based on the j training process description results, j description result knowledge fields are obtained based on the j training process description result weights, and finally j training node compliance evaluation information corresponding to the medical training activity log to be verified is obtained through an evaluation model according to the j description result knowledge fields. The embodiment of the application efficiently acquires the compliance evaluation information of each training node based on the artificial intelligence model, saves manpower, ensures fairness and fairness, and saves computing resources by processing and mining related data without the help of extra computing resources at the aspect of computer resource consumption.

In the description that follows, additional features will be set forth, in part, in the description. These features will be in part apparent to those of ordinary skill in the art upon examination of the following and the accompanying drawings or may be learned by production or use. The features of the present application may be realized and attained by practice or use of various aspects of the methodologies, instrumentalities and combinations particularly pointed out in the detailed examples which follow.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained from the drawings without inventive effort.

The methods, systems, and/or processes of the figures are further described in accordance with the exemplary embodiments. These exemplary embodiments will be described in detail with reference to the drawings. These exemplary embodiments are non-limiting exemplary embodiments in which reference numerals represent similar mechanisms throughout the various views of the drawings.

FIG. 1 is a schematic diagram illustrating the hardware and software components of a verification system according to some embodiments of the present application.

Fig. 2 is a flow chart of a medical training procedure normalization check method according to some embodiments of the present application.

Fig. 3 is a schematic diagram of a functional module architecture of a verification apparatus according to an embodiment of the present application.

Detailed Description

In order to better understand the technical solutions, the technical solutions of the present application are described in detail below with reference to the drawings and specific embodiments, and it should be understood that the specific features in the embodiments and examples of the present application are detailed descriptions of the technical solutions of the present application, and are not limitations of the technical solutions of the present application, and the technical features in the embodiments and examples of the present application may be combined with each other without conflict.

In the following detailed description, numerous specific details are set forth by way of examples in order to provide a thorough understanding of the relevant teachings. It will be apparent, however, to one skilled in the art that the present application may be practiced without these specific details. In other instances, well-known methods, procedures, systems, components, and/or circuits have been described at a relatively high-level, without detail, in order to avoid unnecessarily obscuring aspects of the present application.

These and other features, functions, methods of execution, and combination of functions and elements of related elements in the structure and economies of manufacture disclosed in the present application may become more apparent upon consideration of the following description with reference to the accompanying drawings, all of which form a part of this application. It is to be expressly understood, however, that the drawings are for the purpose of illustration and description only and are not intended as a definition of the limits of the application. It should be understood that the drawings are not to scale. It is to be expressly understood, however, that the drawings are for the purpose of illustration and description only and are not intended as a definition of the limits of the application. It should be understood that the drawings are not to scale.

The present application uses flowcharts to illustrate the implementations performed by a system according to embodiments of the present application. It should be expressly understood that the processes performed by the flowcharts may be performed out of order. Rather, these implementations may be performed in the reverse order or simultaneously. In addition, at least one other implementation may be added to the flowchart. One or more implementations may be deleted from the flowchart.

Referring to fig. 1, a schematic diagram of a verification system 100 is shown, in which the verification system 100 includes a verification device 110, a memory 120, a processor 130 and a communication unit 140. The memory 120, processor 130, and communication unit 140 are electrically connected to each other, directly or indirectly, to enable data transmission or interaction. For example, the components may be electrically connected to each other via one or more communication buses or signal lines. The verification apparatus 110 includes at least one software function module that may be stored in the memory 120 in the form of software or firmware (firmware) or solidified in an Operating System (OS) of the verification system 100. The processor 130 is used for executing executable modules stored in the memory 120, such as software functional modules and computer programs included in the verification apparatus 110.

The Memory 120 may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Read-Only Memory (EPROM), an electrically Erasable Read-Only Memory (EEPROM), and the like. The memory 120 is used for storing a program, and the processor 130 executes the program after receiving the execution instruction. The communication unit 140 is used for establishing a communication connection between the verification system 100 and the service interaction device through a network, and for transceiving data through the network.

The processor may be an integrated circuit chip having signal processing capabilities. The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but may also be a Digital Signal Processor (DSP)), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components. The various methods, steps, and logic blocks disclosed in the embodiments of the present invention may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

It will be appreciated that the configuration shown in FIG. 1 is merely illustrative, and that verification system 100 may also include more or fewer components than shown in FIG. 1, or have a different configuration than shown in FIG. 1. The components shown in fig. 1 may be implemented in hardware, software, or a combination thereof.

Fig. 2 is a flowchart of a medical training procedure standardized verification method according to some embodiments of the present application, where the method is applied to the verification system 100 in fig. 1, and may specifically include the following steps S1 to S5. On the basis of the following steps S1 to S5, some alternative embodiments will be described, which should be understood as examples and should not be understood as technical features essential for implementing the present solution.

And S1, responding to the received checking instruction, and acquiring a medical training activity log to be checked.

The medical training activity log to be verified is a data set generated by the trainee in the process of medical training, and the data set corresponding to each training node is included in the log. The data can be, for example, training data common to multiple trainees, which are classified according to the same training node, or data generated by a single trainee in the training process, and the data can include a series of data of the record of attendance of the trainee, the record of training duration, the record of sequence and duration of rotary departments, the record of information of signers of each node, the record of performance assessment data, the record of assessment approval results, the record of matching conditions of the seniors with education and the study of the trainee, and the like.

And S2, acquiring training process description results corresponding to each training node data set in the medical training activity log to be verified, and acquiring j training process description results.

The medical training activity log to be verified comprises j training node data sets, and the training node data sets and the training process description results are mapped one by one. The one-to-one mapping described herein represents a one-to-one correspondence of training node data sets to training procedure description results. It is readily understood that j is an integer greater than or equal to 1. The training node data set may be a collection of training data generated by individual training nodes. For example, a training node data set is the learning data during a medical care training process for training trainees in the order of rounds for each department, and for the corresponding duration and duration of each department round. The training process description result can be used for describing the degree of conformity of the corresponding training node data set to the normalized process or the degree of non-conformity to the normalized process, and can be presented in the form of a feature vector.

As an embodiment, the process of obtaining the training process description result corresponding to each training node data set in the medical training activity log to be verified to obtain j training process description results may include the following steps: acquiring a characteristic value matched with an ith training node data set, such as a quantization result, aiming at the ith training node data set in a medical training activity log to be verified, wherein the ith training node data set is any one of j training node data sets, i is less than j, and i is more than or equal to 0; the method comprises the steps that training process description result probability corresponding to an ith training node data set in a medical training activity log to be verified is obtained, and is used for representing the possibility of tending to a training process description result; and aiming at the ith training node data set in the medical training activity log to be verified, determining a training process description result corresponding to the ith training node data set based on the training process description result probability corresponding to the ith training node data set and the characteristic value matched with the ith training node data set. Therefore, the training process description result of the training node data set is obtained by combining the characteristic value and the training process description result probability, and the result is more complete and accurate.

And S3, obtaining the weights of the j training process description results based on the j training process description results.

The training process description result weight indicates the examination importance ratio of the training node data set to the medical training activity log to be verified in the training process, and the training process description result weight and the training node data set are mapped one by one. The assessment importance ratio can exist in a proportion mode and reflects the importance degree.

In one embodiment, the deriving the weights of the j training process description results based on the j training process description results may include the following steps: determining a characteristic analysis result, such as a characteristic analysis value, corresponding to an ith training node data set according to a training process description result of the ith training node data set aiming at the ith training node data set in the medical training activity log to be verified, wherein the ith training node data set is any one of j training node data sets, i is less than j, and i is more than or equal to 0; aiming at the ith training node data set in the medical training activity log to be verified, acquiring the training process description result probability corresponding to the ith training node data set; and aiming at the ith training node data set in the medical training activity log to be verified, obtaining the training process description result weight corresponding to the ith training node data set based on the training process description result probability corresponding to the ith training node data set and the characteristic analysis result corresponding to the ith training node data set. In this way, the training node data sets and training process description result weights are more correlated.

And S4, obtaining j description result knowledge fields based on the j training process description result weights.

The mapping of the description result knowledge field and the training node data set one by one, wherein the description result knowledge field may be a description result feature vector, and the process of obtaining j description result knowledge fields based on j training process description result weights may include: aiming at the ith training node data set in the medical training activity log to be verified, acquiring a pooling strategy corresponding to the ith training node data set, wherein the pooling strategy is a down-sampling mode, such as common maximal pooling, the ith training node data set is any one of j training node data sets, i is less than j, and i is more than or equal to 0; and aiming at the ith training node data set in the medical training activity log to be verified, obtaining a description result knowledge field corresponding to the ith training node data set based on the pooling strategy corresponding to the ith training node data set and the training process description result weight corresponding to the ith training node data set. In the process, the general characteristics and the data consistency of the description result knowledge fields corresponding to different training node data sets are enhanced through the pooling strategy, and the deviation probability of subsequent evaluation analysis is reduced.

And S5, acquiring the compliance evaluation information of j training nodes corresponding to the medical training activity log to be verified by adopting an evaluation model according to j description result knowledge fields.

The training node compliance evaluation information and the training node data sets are mapped one by one, the j training node compliance evaluation information corresponds to the j training node data sets, and the evaluation whether the training nodes corresponding to the j training node data sets are in compliance or not is represented. The training nodes correspond to compliance evaluation, and compared with the traditional single assessment evaluation, the method has higher fairness and rigor. The evaluation model is a machine learning model obtained by pre-training, wherein the process of obtaining the compliance evaluation information of j training nodes corresponding to the medical training activity log to be verified by adopting the evaluation model according to j description result knowledge fields may include: acquiring j target description result knowledge fields by adopting a plurality of groups of feature extraction modules in the evaluation model according to the j description result knowledge fields, wherein the plurality of groups of feature extraction modules can be multilayer convolutional neural networks; and acquiring the compliance evaluation information of j training nodes by adopting a multilayer deep neural network in the evaluation model according to j target description result knowledge fields. The multi-layer deep neural network may be a multi-layer fully-connected neural network.

As an embodiment, before step S5, the method may further include a training step of the evaluation model: acquiring a training-to-be-trained training process description result corresponding to each training-to-be-trained data set in a training-to-be-trained medical training activity log to obtain j training-to-be-trained training process description results, wherein the training-to-be-trained medical training activity log comprises j training-to-be-trained data sets, and the training-to-be-trained data sets and the training-to-be-trained training process description results are mapped one by one; determining the weights of the j training procedure description results to be trained based on the j training procedure description results to be trained, wherein the weights of the training procedure description results to be trained are mapped with the training data sets one by one; determining j training description result knowledge fields based on the j training procedure description result weights, wherein the training description result knowledge fields are mapped with a training data set one by one; acquiring j pieces of compliance evaluation information of the to-be-trained training nodes corresponding to the activity logs of the to-be-trained medical training by adopting a to-be-trained evaluation model according to j pieces of knowledge fields of the to-be-trained description results, wherein the compliance evaluation information of the to-be-trained training nodes is mapped with a to-be-trained data set one by one; acquiring j real-value training node compliance evaluation information corresponding to a real-value medical training activity log; and training the quasi-training evaluation model through the j truth-value training node compliance evaluation information and the j quasi-training node compliance evaluation information until the training cutoff condition is met, and obtaining the evaluation model. The true medical training activity log and the true training node compliance evaluation information belong to positive samples.

As an embodiment, the process of obtaining the j truth training node compliance evaluation information corresponding to the truth medical training activity log may include: acquiring truth value training process description results corresponding to each truth value data set in a truth value medical training activity log to obtain j truth value training process description results, wherein the truth value medical training activity log comprises j truth value data sets, and the truth value data sets and the truth value training process description results are mapped one by one and belong to positive samples; determining j truth-value training process description result weights based on j truth-value training process description results, wherein the truth-value training process description result weights are mapped with the truth-value training process description results one by one; determining j truth-value description result knowledge fields based on j truth-value training process description result weights, wherein the truth-value description result knowledge fields are mapped with truth-value data sets one by one; and acquiring j truth-value training node compliance evaluation information corresponding to the truth-value medical training activity log through a quasi-training evaluation model according to j truth-value description result knowledge fields, wherein the truth-value training node compliance evaluation information is mapped with a truth-value data set one by one.

On the basis, optionally, training the pseudo-training evaluation model through the j true-value training node compliance evaluation information and the j pseudo-training node compliance evaluation information until the training cutoff condition is met, and the process of obtaining the evaluation model may include: aiming at each quasi-training data set in a quasi-training medical training activity log and a true value data set corresponding to each quasi-training data set, obtaining cost values (or loss values) of the quasi-training node compliance evaluation information and the true value training node compliance evaluation information through a metric learning (metric learning) objective function, and obtaining j generation values; adjusting parameters of the simulated training evaluation model according to the j cost values; and when the training cutoff condition is met, acquiring an evaluation model according to the adjusted parameters. The loss method of the metric learning objective function, that is, the loss function for training, is not limited in the embodiments of the present application.

As an embodiment, in combination with the training process, negative samples may be added for training to improve the performance of the model, and the training process may include: acquiring a training-simulated training process description result corresponding to each training-simulated data set in a training-simulated medical training activity log to obtain j training-simulated training process description results; the activity log of the quasi-training medical training comprises j quasi-training data sets, and the quasi-training data sets and the description results of the quasi-training process are mapped one by one; obtaining j weights of the description results of the training process to be trained based on the j description results of the training process to be trained; wherein, the weight of the description result of the training process to be trained is mapped with the data set to be trained one by one; determining j training-to-be-trained description result knowledge fields based on j training-to-be-trained process description result weights; wherein, the quasi-training description result knowledge field and the quasi-training data set are mapped one by one; acquiring the compliance evaluation information of j training-to-be-trained training nodes corresponding to the activity logs of the medical training to be trained through a training-to-be-trained evaluation model according to the j training-to-be-trained description result knowledge fields; the method comprises the following steps that compliance evaluation information of nodes to be trained is mapped with a data set to be trained one by one; acquiring j real-value training node compliance evaluation information corresponding to a real-value medical training activity log; acquiring the compliance evaluation information of j false value training nodes corresponding to the false value medical training activity log; and training the quasi-training evaluation model according to the j true value training node compliance evaluation information, the j false value training node compliance evaluation information and the j quasi-training node compliance evaluation information until a training cut-off condition is met to obtain the evaluation model. The related false value medical training activity logs and the false value training node compliance evaluation information are negative samples.

As one embodiment, the process of obtaining the j true value training node compliance evaluation information corresponding to the true value medical training activity log may include: acquiring truth value training process description results corresponding to each truth value data set in a truth value medical training activity log to obtain j truth value training process description results; the truth value medical training activity log comprises j truth value data sets, and the truth value data sets are mapped with truth value training process description results one by one; determining j truth-value training process description result weights based on j truth-value training process description results, wherein the truth-value training process description result weights are mapped with the truth-value training process description results one by one; determining j truth value description result knowledge fields based on j truth value training process description result weights, wherein the truth value description result knowledge fields are mapped with truth value data sets one by one; and acquiring j truth-value training node compliance evaluation information corresponding to the truth-value medical training activity log through a quasi-training evaluation model according to j truth-value description result knowledge fields, wherein the truth-value training node compliance evaluation information is mapped with a truth-value data set one by one.

Optionally, the obtaining of the compliance evaluation information of j false value training nodes corresponding to the false value medical training activity log includes: acquiring a false value training process description result corresponding to each false value data set in a false value medical training activity log to obtain j false value training process description results, wherein the false value medical training activity log comprises j false value data sets, and the false value data sets and the false value training process description results are mapped one by one; determining the weights of the description results of the j false value training processes based on the description results of the j false value training processes, wherein the weights of the description results of the false value training processes are mapped with the description results of the false value training processes one by one; determining j false value description result knowledge fields based on j false value training process description result weights, wherein the false value description result knowledge fields are mapped with a false value data set one by one; and acquiring the compliance evaluation information of j false value training nodes corresponding to the false value medical training activity log through a quasi-training evaluation model according to the knowledge field of the j false value description results, wherein the compliance evaluation information of the false value training nodes is mapped with the false value data set one by one.

Optionally, training the quasi-training evaluation model based on the j true-value training node compliance evaluation information, the j false-value training node compliance evaluation information, and the j quasi-training node compliance evaluation information until a training cutoff condition is met, and obtaining the evaluation model may include: aiming at a truth value data set corresponding to each quasi-training data set in a quasi-training medical training activity log, acquiring first generation values of quasi-training node compliance evaluation information and truth value training node compliance evaluation information through a first metric learning objective function to obtain j first generation values; aiming at each quasi-training data set and a false value data set in the quasi-training medical training activity log, acquiring the compliance evaluation information of quasi-training nodes and the second generation value of the compliance evaluation information of the false value training nodes through a second metric learning objective function to obtain j second generation values; adjusting parameters of the to-be-trained evaluation model according to the j first generation values and the j second generation values; and when the training cutoff condition is met, acquiring an evaluation model through the adjusted parameters.

Referring to fig. 3, which is a schematic structural diagram of a verification apparatus 110 according to an embodiment of the present invention, the verification apparatus 110 may be used for performing a medical training procedure standardized verification method, wherein the verification apparatus 110 includes:

and the receiving module 111 is used for responding to the received verification instruction and acquiring the medical training activity log to be verified.

A description result obtaining module 112, configured to obtain a training process description result corresponding to each training node data set in the medical training activity log to be verified, so as to obtain j training process description results; the medical training activity log to be verified comprises j training node data sets, and the training node data sets are mapped with the training process description results one by one.

A description result weight obtaining module 113, configured to obtain weights of j training process description results based on the j training process description results; the training process description result weight indicates the assessment importance ratio of the training node data set to the medical training activity log to be verified in the training process, and the training process description result weight and the training node data set are mapped one by one.

A description result knowledge field acquisition module 114, configured to obtain j description result knowledge fields based on j training process description result weights; wherein the description result knowledge fields are mapped one-to-one with the training node data set.

The training node compliance evaluation information generation module 115 acquires j pieces of training node compliance evaluation information corresponding to the medical training activity log to be verified by adopting an evaluation model according to the j description result knowledge fields; and mapping the training node compliance evaluation information and the training node data sets one by one.

The receiving module 111 may be configured to perform step S1, the description result obtaining module 112 may be configured to perform step S2, the description result weight obtaining module 113 may be configured to perform step S3, the description result knowledge field obtaining module 114 may be configured to perform step S4, and the training node compliance evaluation information generating module 115 may be configured to perform step S5.

Since the medical training procedure standardized verification method provided by the embodiment of the present invention has been described in the foregoing embodiment, and the principle of the verification apparatus 110 is the same as that of the method, the implementation principle of each module of the verification apparatus 110 will not be described again.

To sum up, according to the medical training process standardization verification method and system provided by the embodiment of the application, when a verification instruction is received, a to-be-verified medical training activity log is obtained, then training process description results corresponding to each training node data set in the to-be-verified medical training activity log are obtained first, j training process description results are obtained, then j training process description result weights are obtained based on the j training process description results, j description result knowledge fields are obtained based on the j training process description result weights, and finally j node compliance evaluation information corresponding to the to-be-verified medical training activity log is obtained by adopting an evaluation model according to the j description result knowledge fields. The embodiment of the application efficiently acquires the compliance evaluation information of each training node based on the artificial intelligence model, saves manpower, ensures fairness and fairness, and saves computing resources by processing and mining related data without the help of extra computing resources at the aspect of computer resource consumption.

In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented in other ways. The apparatus embodiments described above are merely illustrative, and for example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

In addition, the functional modules in the embodiments of the present invention may be integrated together to form an independent part, or each module may exist alone, or two or more modules may be integrated to form an independent part.

The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.

It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrases "comprising a," "8230," "8230," or "comprising" does not exclude the presence of additional like elements in a process, method, article, or apparatus that comprises the element.

In addition, those skilled in the art will recognize that the various aspects of the application may be illustrated and described in terms of several patentable species or contexts, including any new and useful combination of procedures, machines, articles, or materials, or any new and useful modifications thereof. Accordingly, various aspects of the present application may be embodied entirely in hardware, entirely in software (including firmware, resident software, micro-code, etc.) or in a combination of hardware and software. The above hardware or software may be referred to as a "unit", "component", or "system". Furthermore, aspects of the present application may be represented as a computer product, including computer readable program code, embodied in at least one computer readable medium.

A computer readable signal medium may comprise a propagated data signal with computer program code embodied therein, for example, on a baseband or as part of a carrier wave. The propagated signal may take any of a variety of forms, including electromagnetic, optical, and the like, or any suitable combination. A computer readable signal medium may be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code on a computer readable signal medium may be propagated over any suitable medium, including radio, electrical cable, fiber optic cable, RF, or the like, or any combination of the preceding.

Additionally, unless explicitly stated in the claims, the order of processing elements and sequences, use of numerical letters, or use of other designations in this application is not intended to limit the order of the processes and methods in this application. While various presently contemplated embodiments of the invention have been discussed in the foregoing disclosure by way of example, it should be understood that such detail is solely for that purpose and that the appended claims are not limited to the disclosed embodiments, but, on the contrary, are intended to cover all modifications and equivalent arrangements that are within the spirit and scope of the embodiments herein. For example, although the system components described above may be implemented by hardware means, they may also be implemented by software-only solutions, such as installing the described system on an existing server or mobile device.

It should also be appreciated that in the foregoing description of embodiments of the present application, various features are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure aiding in the understanding of at least one embodiment of the invention. However, this method of disclosure is not intended to require more features than are expressly recited in the claims. Indeed, the embodiments may be characterized as having less than all of the features of a single disclosed embodiment.

Claims

1. A medical training flow standardization verification method is applied to a verification system and comprises the following steps:

acquiring training process description results corresponding to each training node data set in the medical training activity log to be verified to obtain j training process description results; the medical training activity log to be verified comprises j training node data sets, the training node data sets and the training process description results are mapped one by one, and the training process description results are used for describing the degree of the corresponding training node data sets conforming to the standardized process or the degree of the corresponding training node data sets not conforming to the standardized process;

acquiring j training node compliance evaluation information corresponding to the medical training activity log to be verified by adopting an evaluation model according to the j description result knowledge fields; wherein the training node compliance evaluation information is mapped one-to-one with the training node data sets;

wherein the obtaining j description result knowledge fields based on the j training process description result weights comprises: aiming at the ith training node data set in the medical training activity log to be verified, acquiring a pooling strategy corresponding to the ith training node data set; wherein the ith training node data set is any one of the j training node data sets, and i is less than j; and aiming at the ith training node data set in the medical training activity log to be verified, obtaining a description result knowledge field corresponding to the ith training node data set based on the pooling strategy corresponding to the ith training node data set and the training process description result weight corresponding to the ith training node data set.

2. The method according to claim 1, wherein the obtaining training process description results corresponding to each training node data set in the medical training activity log to be verified to obtain j training process description results comprises:

acquiring a characteristic value matched with the ith training node data set aiming at the ith training node data set in the medical training activity log to be verified; wherein the ith training node data set is any one of the j training node data sets, and i < j;

3. The method of claim 1, wherein deriving j training procedure description result weights based on the j training procedure description results comprises:

determining a characteristic analysis result corresponding to an ith training node data set according to a training process description result of the ith training node data set aiming at the ith training node data set in the medical training activity log to be verified; wherein the ith training node data set is any one of the j training node data sets, and i < j;

4. The method according to claim 1, wherein the obtaining, by using an evaluation model, the j training node compliance evaluation information corresponding to the medical training activity log to be verified according to the j description result knowledge fields comprises:

5. The method according to any one of claims 1 to 4, wherein before obtaining the j pieces of training node compliance evaluation information corresponding to the medical training activity log to be verified by using an evaluation model according to the j pieces of description result knowledge fields, the method further comprises:

acquiring a training process description result corresponding to each training data set in a training activity log of the quasi-training medical training to obtain j training process description results; the log of activity of the quasi-training medical training comprises j quasi-training data sets, and the quasi-training data sets are mapped with the description results of the quasi-training process one by one;

determining j training simulation description result knowledge fields based on the j training simulation training process description result weights; mapping the quasi-training description result knowledge fields and the quasi-training data sets one by one;

acquiring the compliance evaluation information of j training-to-be-trained training nodes corresponding to the medical training activity log to be trained by adopting a training-to-be-trained evaluation model according to the j training-to-be-trained description result knowledge fields; the compliance evaluation information of the training nodes to be trained is mapped with the data sets to be trained one by one;

acquiring j truth value training node compliance evaluation information corresponding to a truth value medical training activity log;

and training the quasi-training evaluation model through the j true value training node compliance evaluation information and the j quasi-training node compliance evaluation information until a training cut-off condition is met, and obtaining the evaluation model.

6. The method of claim 5, wherein obtaining j true value training node compliance evaluation information for a log of true value medical training activities comprises:

determining j truth describing result knowledge fields based on the j truth training process describing result weights; wherein the truth value description result knowledge field is mapped with the truth value data set one by one;

aiming at each quasi-training data set in the quasi-training medical training activity log and a truth value data set corresponding to each quasi-training data set, acquiring the cost values of the quasi-training node compliance evaluation information and the truth value training node compliance evaluation information through a metric learning objective function, and acquiring j cost values;

adjusting parameters of the model to be trained and evaluated according to the j cost values;

7. The method according to any one of claims 1 to 4, wherein before the obtaining of the j pieces of training node compliance evaluation information corresponding to the medical training activity log to be verified by using an evaluation model based on the j pieces of description result knowledge fields, the method further comprises:

determining j training-to-be-trained description result knowledge fields based on the j training-to-be-trained training process description result weights; wherein, the quasi-training description result knowledge field is mapped with the quasi-training data set one by one;

8. The method of claim 7, wherein obtaining j true value training node compliance evaluation information for a log of true medical training activities comprises:

according to the j truth value description result knowledge fields, acquiring j truth value training node compliance evaluation information corresponding to the truth value medical training activity logs through a simulated training evaluation model; the real-value training node compliance evaluation information is mapped with the real-value data set one by one;

determining j false value training process description result weights based on the j false value training process description results; the weight of the description result of the false-value training process is mapped to the description result of the false-value training process one by one;

the training of the quasi-training evaluation model based on the j true value training node compliance evaluation information, the j false value training node compliance evaluation information and the j quasi-training node compliance evaluation information until a training cut-off condition is met to obtain the evaluation model comprises the following steps:

aiming at a truth value data set corresponding to each quasi-training data set in the quasi-training medical training activity log, acquiring the compliance evaluation information of quasi-training nodes and a first generation value of the compliance evaluation information of the truth training nodes through a first metric learning objective function to obtain j first generation values;

adjusting parameters of the model to be trained and evaluated according to the j first generation values and the j second generation values;

9. A checking system, characterized in that the system comprises a processor and a memory communicating with each other, the memory storing a program, the processor being configured to retrieve a computer program from the memory and to implement the method of any of claims 1-8 by running the computer program.