Intelligent whole-course management system and method for repair personnel

Abstract

The invention provides an intelligent whole-course management system and method for repair personnel. Classification is carried out based on multiple dimensions of hospitals, ages, educational background and the like where the repair personnel are located, the multi-dimensional classification mode can comprehensively cover various characteristics of the repair personnel, and the universality of the system on management of different types of repair personnel is ensured. According to the method, different repair requirements are met, and corresponding management and teaching schemes can be customized according to the requirements of various personnel through accurate classification, so that effective management of the adaptive type of repair personnel is realized. According to the system, automation and intellectualization of repair personnel management are realized, and a series of management processes including registration, classification, repair plan making, teacher matching, return visit, later teaching data analysis and the like of repair personnel can be automatically operated under the framework of the model. The automation reduces the manual intervention, reduces the management cost, and improves the management efficiency at the same time.

Description

Technical Field

[0001] The present application relates to an intelligent full-process management system and method for trainees, and belongs to the field of data processing and artificial intelligence technology. Background Art

[0002] In today's era, new knowledge and technologies in the medical field are constantly emerging. In order for medical personnel to keep up with the pace of these rapid developments, continuous learning and further training have become essential. There are differences in medical levels between hospitals of different regions and levels. Medical staff in grassroots hospitals hope to go to tertiary hospitals for further training to learn advanced clinical experience, management models, etc. At present, the management process of trainees is cumbersome and inefficient, information sharing and communication are not smooth, and data management and analysis are difficult.

[0003] The current drawbacks are mainly reflected in the following aspects: 1. The management process is cumbersome and inefficient. Under the traditional model, the registration, review, course arrangement, and assessment of trainees often rely on manual operations. 2. Information sharing and communication are not smooth. There are many departments involved in the management of trainees. Under the traditional management model, information sharing between departments is difficult. 3. Data management and analysis are difficult. The data of trainees, such as academic performance, attendance records, clinical practice evaluation, and teaching effect follow-up, are all stored in a decentralized manner. 4. The teaching plan and teaching content are fixed, lacking in dynamism and individualization. The teaching plan and teaching content are not dynamically adjusted according to the basic knowledge and skills of trainees.

[0004] To sum up, in order to adapt to the development needs of the medical industry, overcome the drawbacks of traditional hospital trainee management, and meet the trend and needs of information management, it is of great significance to develop a hospital trainee management information system. Summary of the invention

[0005] The present application provides an intelligent full-process management system and method for trainees to solve the many drawbacks of the management of trainees in traditional hospitals.

[0006] In a first aspect, the present application provides an intelligent full-process management system for trainees, including: An information acquisition and classification module is used to acquire relevant information of trainees and classify trainees based on the relevant information; wherein the relevant information includes basic personal information, training major, and hospital where the trainees are located; when classifying trainees based on the relevant information, a preliminary classification is made according to the hospital grade where the trainees are located, and the age range is set as the classification standard, and the trainees are divided into multiple categories according to the educational level; The course customization module is used to customize individual courses for trainees and assign corresponding teaching teachers according to different classification results and training purposes; The online teaching video library module is used to store various teaching videos, including teaching videos of various lectures, surgical demonstrations, and case analysis explanations, providing trainees with resources for after-class review, in-depth learning, and gap-filling. The teaching effect feedback module is used to obtain the practice status of trainees after they return to their original hospitals at set time intervals and evaluate the teaching effect. The feedback method is online questionnaire survey and online skill test. Through the end time point of the trainees' training in the system, the relevant questionnaire survey and test are automatically pushed to the corresponding trainees to complete the information acquisition, and the teaching effect is evaluated based on the acquired information. The course customization module is configured with a pre-trained neural network model, which is used to generate matching courses to be learned according to the trainees' mastery of the corresponding training projects; the neural network model has multiple configurations according to different training projects, and the neural network model includes an input layer, a hidden layer and an output layer connected in sequence; the input variable x=[x1,x2,x3,x4,…,xi] of the input layer is each knowledge point of the corresponding training project, where xi represents the i-th knowledge point; the hidden layer is represented by hi=σ(Wixi+bi), which is used to calculate the test score corresponding to the input variable, where hi represents the output of the i-th hidden layer, Wi represents the weight matrix from the previous layer of the i-th hidden layer to the i-th hidden layer, bi represents the bias vector of the i-th hidden layer, and σ represents the activation function; the output layer is represented by y=σ(Whi+b), which is used to determine the content of the chapter to be learned according to the test score output by the hidden layer, where y represents the output of the output layer, W represents the weight matrix from the last hidden layer to the output layer, and b represents the bias vector of the output layer; In addition, the model is trained based on the trainees' mastery of different knowledge points and learning test results in the original data. Before the model training begins, decimals are randomly selected from the uniform distribution or normal distribution as initialization data for the weight matrix and bias vector. During the training process, the model parameters are automatically adjusted through the back propagation algorithm and the optimization algorithm to minimize the loss function.

[0007] Based on the above system, optionally, the training program includes intravenous catheterization, and the input variables of the input layer of the corresponding neural network model include the development of intravenous infusion therapy, the clinical application of PICC, the insertion and management of pediatric vascular channels, other vascular channel technologies, intravenous infusion therapy response, infusion therapy and management, PICC training and teaching, nursing research and paper writing.

[0008] Based on the above system, optionally, the learning test results when training the model are comprehensive results of the mastery of each learning chapter.

[0009] Based on the above system, optionally, a binary cross entropy loss function is used to evaluate the performance of the model.

[0010] Based on the above system, optionally, the hidden layer uses a ReLU activation function and the output layer uses a Sigmoid activation function.

[0011] In a second aspect, the present application also provides an intelligent full-process management method for trainees, which includes: Obtaining relevant information of trainees and classifying the trainees based on the relevant information; wherein the relevant information includes basic personal information, training major, and hospital where the trainees are located; when classifying the trainees based on the relevant information, a preliminary classification is made according to the hospital grade where the trainees are located, an age range is set as a classification standard, and the trainees are divided into multiple categories according to the educational level; According to different classification results and training purposes, we customize individual courses for trainees and assign corresponding teaching teachers; Utilize the stored teaching videos, including lectures, surgical demonstrations, and case analysis videos, to provide trainees with resources for after-class review, in-depth learning, and gap-filling. At set time intervals, the practical situation of trainees after they return to their original hospitals is obtained to evaluate the teaching effect; the return visit method is online questionnaire survey and online skill test. Through the end time point of the trainees' training in the system, the relevant questionnaire survey and test are automatically pushed to the corresponding trainees to complete the information acquisition, and the teaching effect is evaluated based on the information obtained; Among them, the course customization module is configured with a pre-trained neural network model, which is used to generate matching courses to be learned according to the trainees' mastery of the corresponding training projects; the neural network model has multiple configurations according to different training projects, and the neural network model includes an input layer, a hidden layer and an output layer connected in sequence; the input variable x=[x1,x2,x3,x4,…,xi] of the input layer is each knowledge point of the corresponding training project, where xi represents the i-th knowledge point; the hidden layer is represented by hi=σ(Wixi+bi), which is used to calculate the test score corresponding to the input variable, where hi represents the output of the i-th hidden layer, Wi represents the weight matrix from the previous layer of the i-th hidden layer to the i-th hidden layer, bi represents the bias vector of the i-th hidden layer, and σ represents the activation function; the output layer is represented by y=σ(Whi+b), which is used to determine the content of the chapter to be learned according to the test score output by the hidden layer, where y represents the output of the output layer, W represents the weight matrix from the last hidden layer to the output layer, and b represents the bias vector of the output layer; In addition, the model is trained based on the trainees' mastery of different knowledge points and learning test results in the original data. Before the model training begins, decimals are randomly selected from the uniform distribution or normal distribution as initialization data for the weight matrix and bias vector. During the training process, the model parameters are automatically adjusted through the back propagation algorithm and the optimization algorithm to minimize the loss function.

[0012] The technical solution provided by this application has the following beneficial effects: In the intelligent whole-process management system and method for trainees provided by the present application, classification is carried out based on the hospital where the trainees are located, age, education and other dimensions. This multi-dimensional classification method can comprehensively cover the various characteristics of trainees, ensuring the versatility of the system in the management of different types of trainees. Adapting to different training requirements, through accurate classification, corresponding management and teaching plans can be customized according to the needs of various types of personnel, so as to achieve effective management of trainees of the adapted type. The system realizes the automation and intelligence of trainee management, and a series of management processes such as registration, classification, formulation of training plans, matching of teachers, return visits and later teaching data analysis of trainees can be automatically operated under the framework of the model. This automation reduces manual intervention, reduces management costs, and improves management efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The accompanying drawings herein are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present application, and together with the specification are used to explain the principles of the present application. In addition, these drawings and text descriptions are not intended to limit the scope of the concept of the present application in any way, but to illustrate the concept of the present application for those skilled in the art by referring to specific embodiments.

[0014] Figure 1 A schematic diagram of the structure of an intelligent full-process management system for trainees provided in one embodiment of the present application; Figure 2 A schematic diagram of the structure of a god-level network model provided for one embodiment of the present application; Figure 3 A flowchart of an intelligent full-process management method for trainees provided in one embodiment of the present application. DETAILED DESCRIPTION

[0015] In order to make the purpose, technical solutions and advantages of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in combination with the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0016] In order to solve the problems existing in the management of medical personnel's continuing education, this program proposes an intelligent full-process management program for trainees. In this program, a series of management processes from registration, classification, training plan formulation, teacher matching to follow-up visits and post-teaching data analysis of trainees can be automated and intelligent, which reduces manual intervention, reduces management costs, and improves management efficiency.

[0017] Firstly, the embodiment of the present application provides an intelligent full-process management system for trainees. The full-process intelligent management of trainees can be achieved through the management system.

[0018] Specifically, refer to Figure 1 As shown, an intelligent whole-process management system for trainees includes: an information acquisition and classification module, a course customization module, an online teaching video library module and a teaching effect feedback module.

[0019] Among them, the information acquisition and classification module is used to obtain relevant information of trainees and classify trainees based on the relevant information; wherein, the relevant information includes personal basic information, training major, and the hospital where they are located; when classifying trainees based on the relevant information, a preliminary division is made according to the hospital grade of the hospital where they are located, age range is set as the classification standard, and they are divided into multiple categories according to the educational level.

[0020] Specifically, a comprehensive registration information form is designed in the information acquisition and classification module. In addition to the training major and personal basic information (such as name, gender, age, etc.), the fields about the hospital, age, and education level are set. A data entry interface is established, and trainees can fill in the registration information themselves through the online platform, or the staff of the hospital receiving the training can assist in entering the information. The classification standards are refined, and a preliminary division is made according to the hospital level. The age range is set as the classification standard, and the education level is divided into college, undergraduate, master, doctoral and other categories. Different training plans can be automatically set for different classification results.

[0021] This multi-dimensional classification method can comprehensively cover various characteristics of trainees, ensuring the versatility of the system in managing different types of trainees.

[0022] The course customization module is used to customize individual courses for trainees and assign corresponding teaching teachers according to different classification results and training purposes.

[0023] Specifically, through the automated classification of trainees, courses that meet individual needs can be customized for trainees with different classification results and training purposes, and suitable teaching teachers can be assigned to improve training results and adapt to different training requirements. Through accurate classification, corresponding management and teaching plans can be customized according to the needs of each type of personnel, thereby achieving effective management of trainees of appropriate types.

[0024] The online teaching video library module is used to store various teaching videos, including teaching videos of various lectures, surgical demonstrations and case analysis explanations, providing trainees with resources for after-class review, in-depth learning and filling in the gaps.

[0025] Specifically, in order to facilitate the learning of trainees, the system also uses an online teaching video library module to store various teaching videos. These teaching videos can be videos recorded by the hospital itself, or videos obtained from other hospitals or the Internet. In this way, in addition to learning from the instructor, trainees can also learn through videos to improve the effect of their training.

[0026] The teaching effect follow-up module is used to obtain the practical situation of trainees after they return to their original hospitals at set time intervals and evaluate the teaching effect. The follow-up method is online questionnaire survey and online skill test. Through the trainees' training completion time point in the system, the relevant questionnaire survey and test are automatically pushed to the corresponding trainees to complete the information acquisition, and the teaching effect is evaluated based on the information obtained.

[0027] Specifically, a teaching effect feedback window is designed in the teaching effect feedback module. The time can be divided into half-year short-term effect evaluation and one-year long-term outcome test. The feedback method is online questionnaire survey and online skill test. Through this module, the system can automatically push relevant questionnaire surveys and tests, investigate the practical situation of trainees after returning to the hospital, and evaluate the teaching effect.

[0028] Among them, the course customization module is the core module of this system, which is configured with a pre-trained neural network model to generate matching courses to be learned according to the trainees' mastery of the corresponding training projects. In addition, considering the diversity of training projects, there are multiple neural network models configured according to different training projects. Each neural network model corresponds to a training project, and can match the corresponding trainees with appropriate courses to be learned according to the training project.

[0029] like Figure 2As shown, the neural network model as a whole includes an input layer, a hidden layer, and an output layer connected in sequence. The input variables x=[x1,x2,x3,x4,…,xi] of the input layer are the knowledge points of the corresponding training project, where xi represents the i-th knowledge point; the hidden layer is represented by hi=σ(Wixi+bi), which is used to calculate the test score corresponding to the input variable, where hi represents the output of the i-th hidden layer, Wi represents the weight matrix from the previous layer of the i-th hidden layer to the i-th hidden layer, bi represents the bias vector of the i-th hidden layer, and σ represents the activation function; the output layer is represented by y=σ(Whi+b), which is used to determine the content of the chapter to be learned according to the test score output by the hidden layer, where y represents the output of the output layer, W represents the weight matrix from the last hidden layer to the output layer, and b represents the bias vector of the output layer.

[0030] In addition, the model is trained based on the trainees' mastery of different knowledge points and learning test results in the original data; before the model training begins, decimals are randomly selected from the uniform distribution or normal distribution as the initialization data of the weight matrix and bias vector; during the training process, the model parameters are automatically adjusted through the back propagation algorithm and the optimization algorithm to minimize the loss function. The gradient of the loss function relative to the weight and bias is calculated through the back propagation algorithm. This process uses the chain rule to propagate the error back from the output layer to the input layer. The weight and bias update uses the gradient descent method or its variants (such as Adam, RMSprop, etc.). This training and learning process enables the neural network model to continuously optimize its own parameters based on the input data and target output, so as to achieve better performance on specific tasks.

[0031] In this neural network model, the input variables of the input layer represent all the knowledge points of the corresponding training program, the hidden layer is the chapter test scores corresponding to these variables of the trainees, and the output layer is the chapter content that needs to be learned. In this way, according to the trainees' mastery of different knowledge points, the chapter content that needs to be learned can be obtained, so that individualized teaching can be implemented.

[0032] For better understanding, a specific example is provided for illustration.

[0033] In some embodiments, the training project includes intravenous catheterization, and the input variables of the input layer of the corresponding neural network model include the development of intravenous infusion therapy, clinical application of PICC, placement and management of pediatric vascular access, other vascular access technologies, intravenous infusion therapy response, infusion therapy and management, PICC training and teaching, nursing research and paper writing. In addition, there are two hidden layers. The output layer is one or more of the three chapters that can be learned.

[0034] The code is as follows: import numpy as np from keras.models import Sequential from keras.layers import Dense # Example of hypothetical input data (development of IV therapy, etc.) # Each row represents the input features of a student, 1 means mastered, 0 means not mastered X = np.array([[1, 1, 0, 0, 1, 0, 1, 0], # Student 1 [0, 1, 1, 0, 0, 1, 0, 1], # Student 2 [1, 0, 0, 1, 1, 0, 1, 1]]) # Student 3 # Example of assuming output data (section content to be learned) # Each row represents the content of the chapter that students need to learn. 1 means they need to learn, and 0 means they don't need to learn. y = np.array([[0, 1, 1], # Chapters that student 1 needs to learn [1, 0, 1], # The chapter that student 2 needs to learn [0, 1, 0]]) # The chapters that student 3 needs to learn # Create a neural network model model = Sequential() model.add(Dense(8, input_dim=8, activation='relu'))# Input layer and the first hidden layer model.add(Dense(4, activation='relu'))# The second hidden layer model.add(Dense(3, activation='sigmoid'))# Output layer (section to be learned) # Compile the model model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy']) # Train the model model.fit(X, y, epochs=100, batch_size=1) # Test the model test_data = np.array([[1, 0, 1, 0, 0, 1, 1, 0]])# Input data for new students predictions = model.predict(test_data) # Output prediction results print("Chapter content to be learned (prediction results):", predictions) The code description is as follows: Input data (X): Each row represents a student's mastery of each chapter, 1 means mastery, and 0 means no mastery.

[0035] Output data (y): Each row represents the content of the chapter that the students need to learn, 1 means it needs to be learned, and 0 means it does not need to be learned.

[0036] Model structure: The input layer and the first hidden layer have 8 nodes (corresponding to 8 input variables).

[0037] The second hidden layer has 4 nodes.

[0038] The output layer has 3 nodes (assuming there are 3 chapters).

[0039] Activation function: The hidden layer uses the ReLU activation function, and the output layer uses the Sigmoid activation function.

[0040] Loss function: Use binary cross entropy loss function, which is suitable for binary classification problems.

[0041] Training model: Use the fit method to train, setting the number of training rounds and batch size.

[0042] Prediction: Use the predict method to predict the data of new students.

[0043] The relevant formula is as follows: Input Layer: Suppose the feature vector of the input layer is x=[x1,x2,x3,x4,x5,x6,x7,x8], where x1~x8 represents each input variable (such as the development of intravenous infusion therapy, etc.).

[0044] Hidden Layer: The output h1 of the first hidden layer is expressed as: h1=σ(W1x+b1) Where: W1 is the weight matrix from the input layer to the first hidden layer. b1 is the bias vector of the first hidden layer.

[0045] The output h2 of the second hidden layer is expressed as: h2=σ(W2h1+b2) Where: W2 is the weight matrix from the first hidden layer to the second hidden layer. b2 is the bias vector of the second hidden layer.

[0046] Output layer: The output y of the output layer can be expressed as: y=σ(W3h2+b3) Where: W3 is the weight matrix from the second hidden layer to the output layer. b3 is the bias vector of the output layer.

[0047] The activation function of the output layer uses the Sigmoid function, which is defined as:

[0048] Loss function: The performance of the model is evaluated using the binary cross entropy loss function, defined as: L(Y, )=- log( i)+(1- )(1-log( i)) in is the predicted output of the model and N is the number of samples.

[0049] Training process: Update the weights W and bias b through the back-propagation algorithm, using the gradient descent method: W←W−η b←b−η

[0050] where η is the learning rate.

[0051] When applying the model, the model is trained based on the students' mastery of different knowledge and learning test results in the original data. The test results are divided into three levels: high, medium, and low. For example, if student A gets a high score on chapter one, a medium score on chapter two, and a low score on chapter three, the output layer is the student's overall test result of medium. The model is trained based on such results, so that after knowing the student's mastery of each chapter, the overall test result can be known, so that targeted training can be provided to the students, giving them individualized teaching.

[0052] Through the above scheme, the trainees are classified based on multiple dimensions such as the hospital where they are located, age, and education. This multi-dimensional classification method can comprehensively cover the various characteristics of trainees, ensuring the versatility of the system in the management of different types of trainees. To adapt to different training requirements, through accurate classification, corresponding management and teaching plans can be customized according to the needs of various types of personnel, thereby achieving effective management of trainees of the appropriate type. The system realizes the automation and intelligence of trainee management. A series of management processes from trainee registration, classification, training plan formulation, teacher matching to return visits and later teaching data analysis can be automatically run under the framework of the model. This automation reduces manual intervention, reduces management costs, and improves management efficiency.

[0053] In addition, the embodiment of the present application provides an intelligent full-process management method for trainees, which can be implemented through the above-mentioned management system, and includes: S101: Obtaining relevant information of trainees and classifying the trainees based on the relevant information; wherein the relevant information includes basic personal information, training major, and hospital where the trainees are located; when classifying the trainees based on the relevant information, a preliminary classification is performed according to the hospital grade where the trainees are located, an age range is set as a classification standard, and the trainees are divided into multiple categories according to the educational level; S102: Customize individual courses for trainees and assign corresponding teachers according to different classification results and training purposes; S103: Utilize various stored teaching videos, including teaching videos of lectures, surgical demonstrations and case analysis, to provide trainees with resources for after-class review, in-depth learning and gap-filling; S104: According to the set time interval, the practice situation of the trainees after returning to their original hospitals is obtained to evaluate the teaching effect; the return visit method is online questionnaire survey and online skill test. According to the end time point of the trainees' training in the system, the relevant questionnaire survey and test are automatically pushed to the corresponding trainees to complete the information acquisition, and the teaching effect is evaluated based on the acquired information; Among them, the course customization module is configured with a pre-trained neural network model, which is used to generate matching courses to be learned according to the trainees' mastery of the corresponding training projects; the neural network model has multiple configurations according to different training projects, and the neural network model includes an input layer, a hidden layer and an output layer connected in sequence; the input variable x=[x1,x2,x3,x4,…,xi] of the input layer is each knowledge point of the corresponding training project, where xi represents the i-th knowledge point; the hidden layer is represented by hi=σ(Wixi+bi), which is used to calculate the test score corresponding to the input variable, where hi represents the output of the i-th hidden layer, Wi represents the weight matrix from the previous layer of the i-th hidden layer to the i-th hidden layer, bi represents the bias vector of the i-th hidden layer, and σ represents the activation function; the output layer is represented by y=σ(Whi+b), which is used to determine the content of the chapter to be learned according to the test score output by the hidden layer, where y represents the output of the output layer, W represents the weight matrix from the last hidden layer to the output layer, and b represents the bias vector of the output layer; In addition, the model is trained based on the trainees' mastery of different knowledge points and learning test results in the original data. Before the model training begins, decimals are randomly selected from the uniform distribution or normal distribution as initialization data for the weight matrix and bias vector. During the training process, the model parameters are automatically adjusted through the back propagation algorithm and the optimization algorithm to minimize the loss function.

[0054] Among them, regarding the specific implementation method of each step of the above method, reference can be made to the corresponding content in the aforementioned method embodiment, which will not be repeated here.

[0055] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0056] It should be noted that, in the description of the present invention, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, the meaning of "plurality" refers to at least two.

[0057] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention belong.

[0058] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0059] A person skilled in the art may understand that all or part of the steps of the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0060] In addition, each functional unit in each embodiment of the present invention may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium. The above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0061] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0062] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. An intelligent full-process management system for trainees, characterized in that: include: An information acquisition and classification module is used to acquire relevant information of trainees and classify trainees based on the relevant information; wherein the relevant information includes basic personal information, training major, and hospital where the trainees are located; when classifying trainees based on the relevant information, a preliminary classification is made according to the hospital grade where the trainees are located, and the age range is set as the classification standard, and the trainees are divided into multiple categories according to the educational level; The course customization module is used to customize individual courses for trainees and assign corresponding teaching teachers according to different classification results and training purposes; The online teaching video library module is used to store various teaching videos, including teaching videos of various lectures, surgical demonstrations, and case analysis explanations, providing trainees with resources for after-class review, in-depth learning, and gap-filling. The teaching effect feedback module is used to obtain the practice status of trainees after they return to their original hospitals at set time intervals and evaluate the teaching effect. The feedback method is online questionnaire survey and online skill test. Through the end time point of the trainees' training in the system, the relevant questionnaire survey and test are automatically pushed to the corresponding trainees to complete the information acquisition, and the teaching effect is evaluated based on the acquired information. The course customization module is configured with a pre-trained neural network model, which is used to generate matching courses to be learned according to the trainees' mastery of the corresponding training projects; the neural network model has multiple configurations according to different training projects, and the neural network model includes an input layer, a hidden layer and an output layer connected in sequence; the input variable x=[x1,x2,x3,x4,…,xi] of the input layer is each knowledge point of the corresponding training project, where xi represents the i-th knowledge point; the hidden layer is represented by hi=σ(Wixi+bi), which is used to calculate the test score corresponding to the input variable, where hi represents the output of the i-th hidden layer, Wi represents the weight matrix from the previous layer of the i-th hidden layer to the i-th hidden layer, bi represents the bias vector of the i-th hidden layer, and σ represents the activation function; the output layer is represented by y=σ(Whi+b), which is used to determine the content of the chapter to be learned according to the test score output by the hidden layer, where y represents the output of the output layer, W represents the weight matrix from the last hidden layer to the output layer, and b represents the bias vector of the output layer; In addition, the model is trained based on the trainees' mastery of different knowledge points and learning test results in the original data. Before the model training begins, decimals are randomly selected from the uniform distribution or normal distribution as initialization data for the weight matrix and bias vector. During the training process, the model parameters are automatically adjusted through the back propagation algorithm and the optimization algorithm to minimize the loss function.

2. The intelligent full-process management system for trainees according to claim 1 is characterized in that: The training program includes intravenous catheterization, and the input variables of the input layer of the corresponding neural network model include the development of intravenous infusion therapy, the clinical application of PICC, the insertion and management of pediatric vascular channels, other vascular channel technologies, intravenous infusion therapy response, infusion therapy and management, PICC training and teaching, nursing research and paper writing.

3. The intelligent full-process management system for trainees according to claim 1 is characterized in that: The learning test results when training the model are comprehensive results of the mastery of each learning chapter.

4. The intelligent full-process management system for trainees according to any one of claims 1 to 3, characterized in that: The binary cross entropy loss function is used to evaluate the performance of the model.

5. The intelligent full-process management system for trainees according to any one of claims 1 to 3, characterized in that: The hidden layer uses the ReLU activation function, and the output layer uses the Sigmoid activation function.

6. An intelligent full-process management method for trainees, characterized in that: include: Obtaining relevant information of trainees and classifying the trainees based on the relevant information; wherein the relevant information includes basic personal information, training major, and hospital where the trainees are located; when classifying the trainees based on the relevant information, a preliminary classification is made according to the hospital grade where the trainees are located, an age range is set as a classification standard, and the trainees are divided into multiple categories according to the educational level; According to different classification results and training purposes, we customize individual courses for trainees and assign corresponding teaching teachers; Utilize the stored teaching videos, including lectures, surgical demonstrations, and case analysis videos, to provide trainees with resources for after-class review, in-depth learning, and gap-filling. At set time intervals, the practical situation of trainees after they return to their original hospitals is obtained to evaluate the teaching effect; the return visit method is online questionnaire survey and online skill test. Through the end time point of the trainees' training in the system, the relevant questionnaire survey and test are automatically pushed to the corresponding trainees to complete the information acquisition, and the teaching effect is evaluated based on the information obtained; Among them, the course customization module is configured with a pre-trained neural network model, which is used to generate matching courses to be learned according to the trainees' mastery of the corresponding training projects; the neural network model has multiple configurations according to different training projects, and the neural network model includes an input layer, a hidden layer and an output layer connected in sequence; the input variable x=[x1,x2,x3,x4,…,xi] of the input layer is each knowledge point of the corresponding training project, where xi represents the i-th knowledge point; the hidden layer is represented by hi=σ(Wixi+bi), which is used to calculate the test score corresponding to the input variable, where hi represents the output of the i-th hidden layer, Wi represents the weight matrix from the previous layer of the i-th hidden layer to the i-th hidden layer, bi represents the bias vector of the i-th hidden layer, and σ represents the activation function; the output layer is represented by y=σ(Whi+b), which is used to determine the content of the chapter to be learned according to the test score output by the hidden layer, where y represents the output of the output layer, W represents the weight matrix from the last hidden layer to the output layer, and b represents the bias vector of the output layer; In addition, the model is trained based on the trainees' mastery of different knowledge points and learning test results in the original data. Before the model training begins, decimals are randomly selected from the uniform distribution or normal distribution as initialization data for the weight matrix and bias vector. During the training process, the model parameters are automatically adjusted through the back propagation algorithm and the optimization algorithm to minimize the loss function.

Images