Calculation staff scheduling method and system and electronic equipment

Abstract

The invention discloses a schedule arrangement method and system for trainees and electronic equipment, and is applied to the technical field of teaching management, basic information of the trainees and basic information of tutors are acquired, and trainees requirements are formulated according to a trainee manual; acquiring operation room information, operation arrangement information, operation planning requirements and basic information, matching the operation planning personnel with the tutor, and issuing a matching result to the operation planning personnel and the tutor; and performing expert scoring on the performance of the trainee by adopting a field recording and video recording mode in an operation, and adjusting a matching result according to an evaluation result. The operation process of the trainee is evaluated more comprehensively from subjective and objective aspects, so that whether the trainee meets the trainee requirement or not is determined, and a subsequent scheduling scheme is guided; and the tutor can perform grading according to the scoring result and feed back the grading result to the tutor, so that the tutor can know the proficiency degree of the trainee according to the grading result, and the trainee can practice unfamiliar items in a targeted manner according to the scoring result.

Description

Technical Field

[0001] The present invention relates to the field of teaching management technology, and more specifically to a method, system and electronic equipment for scheduling regular training personnel. Background Art

[0002] The problems and defects of the existing technology are: the existing hospitals do not have strict control over the entry into the operating room, and there is no entry qualification training or examination, and the qualifications of the personnel entering the operating room cannot be guaranteed.

[0003] The difficulty in solving the above problems and defects is as follows: In the current teaching and management practices of major hospitals, the operating room has limitations due to the complexity of the number of people entering, and the qualification access training is constrained by personnel, venues, and time, making it difficult to achieve good results. There are generally lagging teaching, shortage of teaching manpower, and violations of hospital management regulations and work processes in actual work, resulting in chaotic management of the operating room.

[0004] In order to solve the above problems, the hospital usually needs to manually set up training courses based on the trainees' majors, rotation departments, rotation time and other information, and arrange a training plan for each trainee, and then let the trainees receive training according to the training plan. Due to the large amount of information about trainees, the hospital needs to spend a lot of energy to organize and summarize the information and arrange a training plan for each trainee. At the same time, when arranging the training plan, it is necessary to manually check whether the time and number of training rotation arrangements in each department are reasonable to ensure that each trainee can receive training and study in an orderly manner according to the training plan. The workload is very heavy and time-consuming, and it is easy to have omissions and duplications.

[0005] Therefore, how to provide a method, system and electronic equipment for scheduling regular training personnel that automatically matches regular training personnel and tutors and can score regular training results with high pertinence is an issue that technical personnel in this field urgently need to solve. Summary of the invention

[0006] In view of this, the present invention provides a method, system and electronic device for scheduling regular training personnel to solve the problems in the background technology. According to the tutor evaluation and comprehensive evaluation, the operation process of the regular training personnel is more comprehensively evaluated from both subjective and objective aspects, so as to determine whether the regular training personnel meet the regular training requirements and guide the subsequent scheduling plan.

[0007] In order to achieve the above object, the present invention adopts the following technical solution:

[0008] A method for scheduling regular training personnel, comprising:

[0009] Obtain basic information of the trainees and tutors. The basic information of the trainees includes: basic personal information, professional background, training stage, and learning needs; the basic information of tutors includes: name, professional field, clinical experience, teaching experience, personality characteristics, and teaching style;

[0010] Formulate training requirements according to the training manual, and formulate clear training goals and requirements based on the training standards and the actual situation of the hospital, including theoretical examinations, skills assessments, and clinical case analysis;

[0011] Obtain operating room information and surgical arrangement information, training requirements and basic information to match training personnel and mentors, and publish the matching results to training personnel and mentors;

[0012] During the operation, on-site recording and video recording are used to conduct expert evaluation of the training personnel's performance, and the matching result is adjusted according to the evaluation result, wherein the expert evaluation includes: instructor evaluation and comprehensive evaluation.

[0013] Preferably, in the above-mentioned method for scheduling regular trainees, the specific steps for matching regular trainees and tutors are as follows:

[0014] Quantify the basic information of training personnel and tutors;

[0015] Constructing a training matching model, the training matching model includes an input layer, a hidden layer and an output layer, the number of nodes in the input layer is determined according to the number of preprocessed data features; the number of nodes in the hidden layer can be determined by experience or experiment; the output layer represents the matching score between the training personnel and the tutor;

[0016] Combine the collected data of trainees and mentors into multiple sample pairs, each of which has a known matching label determined by manual evaluation or historical data;

[0017] Select optimization algorithm and loss function: Select stochastic gradient descent and Adam as optimization algorithm; select mean square error as loss function, the goal is to make the output of the neural network close to the actual matching label;

[0018] When the training of the regular training matching model is completed, the data of the new regular training personnel and the tutors are input into the regular training matching model to obtain the matching score. The scoring standard is a five-point system, where 1 point means no complementarity at all and 5 points means high complementarity.

[0019] Preferably, in the above-mentioned method for scheduling regular training personnel, the steps for generating regular training requirements are as follows:

[0020] Collect the content of the training manual, training standard documents and the actual situation of the hospital, including but not limited to the following aspects: the scope of professional knowledge, skill operation items, and clinical case types in the training manual, the specific requirements for knowledge mastery, skill proficiency, and case analysis ability in the training standards, the hospital's disciplinary characteristics, common case types, equipment resources, and teaching staff; pre-process the collected data and convert text information into quantifiable feature vectors;

[0021] Constructing a training plan model, the training plan model includes an input layer, a hidden layer and an output layer, the number of nodes in the input layer is determined according to the number of preprocessed data features; the number of nodes in the hidden layer can be determined through experience or experiments; the output layer represents the output quantitative table of training requirements;

[0022] The feature vector is input into the regular training planning model, and a preliminary quantitative table is determined as a label according to the regular training manual, standards and actual conditions of the hospital. For a certain regular training specialty, the weight of the theoretical examination on different subject knowledge, the scoring standard of each operation item in the skill assessment, and the specific scoring points of the clinical case analysis are determined, and the quantitative table of regular training requirements is output.

[0023] Preferably, in the above-mentioned method for scheduling regular training personnel, the expert scoring of the regular training personnel's performance includes tutor evaluation and comprehensive evaluation, and the tutor evaluation is subjectively scored based on the quantitative table of regular training requirements and on-site records; the comprehensive evaluation is objectively scored based on video records.

[0024] Preferably, in the above-mentioned method for scheduling regular training personnel, the specific steps of comprehensive evaluation are as follows:

[0025] Shoot a frontal action video and several side action videos of the training personnel, where each side action video is shot at a different angle;

[0026] For each side action video, input it together with the front action video into the 3D reconstruction algorithm model to construct a number of 3D action videos corresponding to each side action video;

[0027] Inputting the plurality of three-dimensional action videos into a video action recognition algorithm model to perform specific action recognition, obtaining all continuous frame images between a start frame image and an end frame image where the specific action appears in each three-dimensional action video, and identifying a key frame image that best reflects the specific action among all continuous frame images;

[0028] Acquire a preset standard action library, wherein the standard action library contains all standard action continuous frame images of each specific action;

[0029] Input each key frame image into a human key point regression algorithm to obtain a human key point feature map corresponding to a specific action in each key frame image, then compare the human key point feature map of each key frame image with all standard action continuous frame images of each specific action in the standard action library, select one of the key frame images that best meets the standard action and its corresponding three-dimensional action video according to the comparison result, and identify the action category, specification coefficient and difficulty coefficient to which the specific action in the selected key frame image belongs;

[0030] According to the action category to which the specific action in the selected key frame image belongs, the number of all standard action continuous frame images corresponding to the action category is obtained from the standard action library, the number of all continuous frame images in which the specific action appears in the three-dimensional action video corresponding to the selected key frame image is compared with the number of all obtained standard action continuous frame images, and then the adjustment coefficient is set in combination with the degree of the patient;

[0031] The operation coefficient is calculated according to the action category and adjustment coefficient of the specific action in the selected key frame image, and the evaluation score is then calculated according to the specification coefficient, difficulty coefficient, adjustment coefficient and operation coefficient of the specific action in the selected key frame image.

[0032] Preferably, in the above-mentioned method for scheduling regular training personnel, an adjustment coefficient is set in combination with the severity of the patient:

[0033]

[0034] Among them, ω2 represents the adjustment coefficient of the specific action in the selected key frame image, M i represents the number of all continuous frame images of a specific action in the 3D action video corresponding to the selected key frame image, M0 represents the number of all continuous frame images of standard actions obtained from the standard action library, and A represents the patient coefficient fitted with different degrees of illness;

[0035] Where A=u0x0+u1x1+…+u n x n ; n represents the number of factors that affect surgical operations, including at least: factors of the training personnel themselves, surgical environment factors, surgical instrument factors, and patient factors; u0 to u n Represents the preset fitting constants; x0 to x n It represents the scores of the training personnel's own factors, surgical environment factors, surgical instrument factors, and patient factors.

[0036] Preferably, in the above-mentioned method for scheduling regular training personnel, the operating coefficient formula is: ω3=ω2×A×g;

[0037] Among them, ω3 represents the operation coefficient of the specific action in the selected key frame image, A represents the adjustment coefficient of the specific action in the selected key frame image, which is fitted based on the patient's degree, and g is the operation constant preset for the action category to which each specific action belongs.

[0038] Preferably, in the above-mentioned method for scheduling regular training personnel, the calculation formula for the evaluation score is: y=ω0+ω1+ω2+ω3;

[0039] Among them, y represents the evaluation score, ω0 represents the specification coefficient of the specific action in the selected key frame image, ω1 represents the difficulty coefficient of the specific action in the selected key frame image, ω2 represents the adjustment coefficient of the specific action in the selected key frame image, and ω3 represents the operation coefficient of the specific action in the selected key frame image.

[0040] A scheduling system for regular training personnel, using the above-mentioned scheduling method for regular training personnel, comprises:

[0041] The data acquisition module acquires basic information of the training personnel and the basic information of the tutor. The basic information of the training personnel includes: basic personal information, professional background, training stage, and learning needs; the basic information of the tutor includes: name, professional field, clinical experience, teaching experience, personality characteristics, and teaching style;

[0042] The training requirement generation module formulates the training requirements according to the training manual, and formulates clear training goals and requirements according to the training standards and the actual situation of the hospital, including theoretical examinations, skills assessments, and clinical case analysis;

[0043] The matching module obtains the operating room information and surgical arrangement information, training requirements and basic information to match the training personnel and the tutor, and publishes the matching results to the training personnel and the tutor;

[0044] The evaluation module uses on-site and video recording methods to conduct expert scoring of the training personnel's performance during the operation, and adjusts the matching results based on the evaluation results, wherein the expert scoring includes: tutor evaluation and comprehensive evaluation.

[0045] An electronic device includes a memory and a processor, wherein the memory stores a computer program, and is characterized in that when the processor executes the computer program, it implements the steps of a method for scheduling regular training personnel.

[0046] It can be seen from the above technical solutions that, compared with the prior art, the present invention discloses a method, system and electronic device for scheduling regular training personnel, which matches regular training personnel and tutors according to the acquired data, and more comprehensively evaluates the operation process of regular training personnel from both subjective and objective aspects, so as to determine whether the regular training personnel meet the regular training requirements and guide subsequent scheduling plans; tutors can grade the regular training personnel according to the scoring results and feed back the grading results to tutors, so that tutors can understand the proficiency of regular training personnel according to the grading results, and regular training personnel can also practice unfamiliar projects in a targeted manner according to the scoring results. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0048] Figure 1 A flow chart of the method provided by the present invention;

[0049] Figure 2 A flow chart of the comprehensive evaluation method provided by the present invention;

[0050] Figure 3 A system block diagram provided by the present invention. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] The embodiment of the present invention discloses a method for scheduling training personnel, such as Figure 1 As shown, including:

[0053] Obtain basic information of the trainees and tutors. The basic information of the trainees includes: basic personal information, professional background, training stage, and learning needs; the basic information of tutors includes: name, professional field, clinical experience, teaching experience, personality characteristics, and teaching style;

[0054] Formulate training requirements according to the training manual, and formulate clear training goals and requirements based on the training standards and the actual situation of the hospital, including theoretical examinations, skills assessments, and clinical case analysis;

[0055] Obtain operating room information and surgical arrangement information, training requirements and basic information to match training personnel and mentors, and publish the matching results to training personnel and mentors;

[0056] During the operation, on-site recording and video recording are used to conduct expert evaluation of the training personnel's performance, and the matching result is adjusted according to the evaluation result, wherein the expert evaluation includes: instructor evaluation and comprehensive evaluation.

[0057] Specifically, obtain the basic information of the training personnel and the basic information of the tutor:

[0058] Information entry for training personnel:

[0059] Professional background: Fill in the details of the majors studied and research directions at undergraduate and graduate levels.

[0060] Training needs: clarify your specific needs in terms of professional knowledge, clinical skills, scientific research capabilities, etc. Personality characteristics: summarize your personality characteristics through some personality test questionnaires (such as the simplified version of the Big Five Personality Test, etc.) or self-description, such as cheerfulness, introversion, rigor, easy-going, etc.

[0061] Career planning: Briefly describe your future career development direction and goals.

[0062] Tutor information entry:

[0063] Professional field: Describe in detail your professional field, sub-professional areas, etc.

[0064] Clinical experience: Introduce your years of clinical work, types of surgeries you are good at, difficult cases you have handled, etc.

[0065] Teaching experience: Describe your experience and achievements in teaching trainee students, graduate students, etc.

[0066] Personality traits: Also presented through personality tests or self-descriptions.

[0067] Teaching style: such as strict requirements, inspiring guidance, encouraging autonomy, etc.

[0068] In order to further optimize the above technical solutions, the specific steps for matching training personnel and tutors are as follows:

[0069] Quantify the basic information of training personnel and tutors;

[0070] Constructing a training matching model, the training matching model includes an input layer, a hidden layer and an output layer, the number of nodes in the input layer is determined according to the number of preprocessed data features; the number of nodes in the hidden layer can be determined by experience or experiment; the output layer represents the matching score between the training personnel and the tutor;

[0071] Combine the collected data of trainees and mentors into multiple sample pairs, each of which has a known matching label determined by manual evaluation or historical data;

[0072] Select optimization algorithm and loss function: Select stochastic gradient descent and Adam as optimization algorithm; select mean square error as loss function, the goal is to make the output of the neural network close to the actual matching label;

[0073] When the training of the regular training matching model is completed, the data of the new regular training personnel and the tutors are input into the regular training matching model to obtain the matching score. The scoring standard is a five-point system, where 1 point means no complementarity at all and 5 points means high complementarity.

[0074] Specifically, the scoring criteria can be divided into five levels: very unmatched, not very matched, average, relatively matched, and very matched. The matching results can be presented to hospital management departments, training personnel, and mentors in the form of charts, reports, etc. For example, a radar chart can be used to show the scores in three aspects: professional matching, ability complementarity, and personality matching, and the overall matching degree and advantages and disadvantages can be described in words.

[0075] Going further,

[0076] Explain the matching results: The contribution of each input feature in the neural network to the matching degree can be analyzed, for example, by calculating the gradient of the input features to understand which features have a greater impact on the matching degree. This can help understand why certain trainees and mentors are considered to be a good match, and in which aspects the matching degree can be further improved.

[0077] Application matching results: Recommend matching of trainees and mentors based on matching scores. For example, pairs with high matching scores can be recommended to management departments for arrangement. Update the neural network regularly, and as new data accumulates, retrain the network to improve matching accuracy.

[0078] It is important to understand that the basic principles for formulating training requirements include: following training standards, strictly following national and local training standards, and ensuring the legality and standardization of training requirements. In combination with the actual situation of the hospital, full consideration should be given to the actual situation of the hospital's medical resources, teaching staff, discipline characteristics, etc., to formulate practical and feasible training requirements. With the cultivation of clinical ability as the core, the cultivation of clinical practice ability should be highlighted, and the combination of theory and practice should be emphasized to improve the clinical diagnosis, treatment and operation skills of training personnel. Focus on the cultivation of comprehensive qualities, not only focusing on the cultivation of professional knowledge and skills, but also on the cultivation of comprehensive qualities such as professional ethics, communication skills, and teamwork.

[0079] In order to further optimize the above technical solution, the steps for generating the regulatory training requirements are as follows:

[0080] Collect the content of the training manual, training standard documents and the actual situation of the hospital, including but not limited to the following aspects: the scope of professional knowledge, skill operation items, and clinical case types in the training manual, the specific requirements for knowledge mastery, skill proficiency, and case analysis ability in the training standards, the hospital's disciplinary characteristics, common case types, equipment resources, and teaching staff; pre-process the collected data and convert text information into quantifiable feature vectors;

[0081] Constructing a training plan model, the training plan model includes an input layer, a hidden layer and an output layer, the number of nodes in the input layer is determined according to the number of preprocessed data features; the number of nodes in the hidden layer can be determined through experience or experiments; the output layer represents the output quantitative table of training requirements;

[0082] The feature vector is input into the regular training planning model, and a preliminary quantitative table is determined as a label according to the regular training manual, standards and actual conditions of the hospital. For a certain regular training specialty, the weight of the theoretical examination on different subject knowledge, the scoring standard of each operation item in the skill assessment, and the specific scoring points of the clinical case analysis are determined, and the quantitative table of regular training requirements is output.

[0083] In order to further optimize the above technical solutions, the performance of the trainees is evaluated by experts, including tutor evaluation and comprehensive evaluation. The tutor evaluation is subjectively scored according to the quantitative table of the training requirements and the on-site records; the comprehensive evaluation is objectively scored according to the video records.

[0084] In order to further optimize the above technical solution, Figure 2 The specific steps of comprehensive evaluation are as follows:

[0085] Shoot a frontal action video and several side action videos of the training personnel, where each side action video is shot at a different angle;

[0086] For each side action video, input it together with the front action video into the 3D reconstruction algorithm model to construct a number of 3D action videos corresponding to each side action video;

[0087] Inputting the plurality of three-dimensional action videos into a video action recognition algorithm model to perform specific action recognition, obtaining all continuous frame images between a start frame image and an end frame image where the specific action appears in each three-dimensional action video, and identifying a key frame image that best reflects the specific action among all continuous frame images;

[0088] Acquire a preset standard action library, wherein the standard action library contains all standard action continuous frame images of each specific action;

[0089] Input each key frame image into a human key point regression algorithm to obtain a human key point feature map corresponding to a specific action in each key frame image, then compare the human key point feature map of each key frame image with all standard action continuous frame images of each specific action in the standard action library, select one of the key frame images that best meets the standard action and its corresponding three-dimensional action video according to the comparison result, and identify the action category, specification coefficient and difficulty coefficient to which the specific action in the selected key frame image belongs;

[0090] According to the action category to which the specific action in the selected key frame image belongs, the number of all standard action continuous frame images corresponding to the action category is obtained from the standard action library, the number of all continuous frame images in which the specific action appears in the three-dimensional action video corresponding to the selected key frame image is compared with the number of all obtained standard action continuous frame images, and then the adjustment coefficient is set in combination with the degree of the patient;

[0091] The operation coefficient is calculated according to the action category and adjustment coefficient of the specific action in the selected key frame image, and the evaluation score is then calculated according to the specification coefficient, difficulty coefficient, adjustment coefficient and operation coefficient of the specific action in the selected key frame image.

[0092] In order to further optimize the above technical solution, the adjustment coefficient is set according to the severity of the patient:

[0093]

[0094] Among them, ω2 represents the adjustment coefficient of the specific action in the selected key frame image, M i represents the number of all continuous frame images of a specific action in the 3D action video corresponding to the selected key frame image, M0 represents the number of all continuous frame images of standard actions obtained from the standard action library, and A represents the patient coefficient fitted with different degrees of illness;

[0095] Where A=u0x0+u1x1+…+u n x n ; n represents the number of factors that affect surgical operations, including at least: factors of the training personnel themselves, surgical environment factors, surgical instrument factors, and patient factors; u0 to u n Represents the preset fitting constants; x0 to x n It represents the scores of the training personnel's own factors, surgical environment factors, surgical instrument factors, and patient factors.

[0096] Furthermore, factors affecting surgical procedures include at least:

[0097] Factors related to the training staff themselves:

[0098] Surgical experience (Number of operating tables: The number of operating tables completed by the doctor in a specific time period can be counted, such as the total number of various types of operations completed in the past year. Surgical success rate: By calculating the success rate of the operations performed by the doctor, such as the percentage of operations in which the patient recovered well and had no major complications. Standard deviation of surgical time: The time of the same type of operation is counted and its standard deviation is calculated to reflect the stability of the doctor's operation time for a specific operation.) Physical condition (Fatigue index: It can be obtained by combining continuous working time, heart rate changes, subjective fatigue scores (such as 1-10 points), etc. For example, the fatigue score for working continuously for more than 4 hours is more than 6 points. Hand stability index: Use professional equipment to measure the tremor amplitude of the hand in a specific state, and the unit can be millimeter / second. Psychological state stress test score: The stress score is obtained through a psychological test scale (such as the anxiety self-rating scale, etc.). The higher the score, the greater the stress. Emotional stability index: It can be quantified by the number and degree of emotional fluctuations over a continuous period of time (such as a week). For example, emotional fluctuations greater than 5 times / week and affecting the work status are high emotional instability.)

[0099] Surgical environment factors:

[0100] Operating room layout (Instrument retrieval time: Count the time it takes for the doctor to get the instrument from the time he issues the instruction to get the instrument, in seconds. Activity space area: Use a measuring tool to measure the area of ​​the doctor's main operating area in the operating room, in square meters. Lighting condition illuminance value: Use a illuminance meter to measure the light intensity in the operating area, in lux. Shadowless effect index: You can place objects in different positions in the operating area, observe the size and clarity of their shadows, and make subjective scores (1-10 points). Temperature value: Use a thermometer to measure the temperature of the operating room, in degrees Celsius (℃). Humidity value: Use a hygrometer to measure the humidity in the operating room, in percentage (%).)

[0101] Surgical instrument factors: (Instrument quality sharpness test: The sharpness can be quantified by the amount of force required to cut a specific material, in Newtons (N). Durability: Record the number of times the same instrument can be used for surgery under normal use. Instrument operation difficulty score: The doctor subjectively scores the difficulty of operating a specific instrument in surgery, 1-10 points, with higher scores indicating greater difficulty. Instrument and surgery matching: Based on the type of surgery and the functional characteristics of the instrument, a subjective matching score is given, 1-10 points, with higher scores indicating higher matching.)

[0102] Patient factors: (Complexity of the disease: Score of anatomical abnormality: Through imaging examinations and other means, multiple experts subjectively score the degree of anatomical abnormality of the patient, with a score of 1-10, and the higher the score, the greater the degree of abnormality. Number of comorbidities: Count the number of other diseases that the patient suffers from at the same time. Physical condition: Body mass index (BMI): Weight (kg) divided by height (m) squared, which can reflect the patient's obesity. Age: The specific age is used as a quantitative indicator.)

[0103] Specifically, the variable definition is: let the dependent variable Y be the quantitative value of surgical action performance. The independent variables are the training personnel's own factor variable X1, the surgical environment factor variable X2, the surgical instrument factor variable X3, and the patient factor variable X4.

[0104] A linear regression model is established. The linear regression model is Y=a+b1X1+b2X2+b3X3+b4X4, where a is the intercept, and b1, b2, b3, and b4 are the regression coefficients of the corresponding independent variables.

[0105] Parameter estimation, use the least squares method and other methods to perform regression analysis on the collected data and estimate the values ​​of parameters a, b1, b2, b3, and b4.

[0106] Calculate the adjustment factor. The adjustment factor can be defined as C = e Y (This is just a definition method, and it can also be adjusted according to actual conditions.) Where Y is the quantitative value of surgical action performance calculated by the linear regression model.

[0107] In order to further optimize the above technical solution, the operating coefficient formula is: ω3=ω 2×A×g ;

[0108] Among them, ω3 represents the operation coefficient of the specific action in the selected key frame image, A represents the adjustment coefficient of the specific action in the selected key frame image, which is fitted based on the patient's degree, and g is the operation constant preset for the action category to which each specific action belongs.

[0109] In order to further optimize the above technical solution, the calculation formula of the evaluation score is: y = ω0 + ω1 + ω2 + ω3;

[0110] Among them, y represents the evaluation score, ω0 represents the specification coefficient of the specific action in the selected key frame image, ω1 represents the difficulty coefficient of the specific action in the selected key frame image, ω2 represents the adjustment coefficient of the specific action in the selected key frame image, and ω3 represents the operation coefficient of the specific action in the selected key frame image.

[0111] When conducting a comprehensive evaluation, several three-dimensional action videos of the training personnel are obtained, and a video action recognition algorithm model is used to identify all continuous frame images of specific actions in each three-dimensional action video and the key frame images that best reflect the specific action. Then, one that best meets the standard action and its corresponding three-dimensional action video are selected from each key frame image. Then, according to a preset standard action library, the action category, specification coefficient and difficulty coefficient of the specific action in the selected key frame image are identified. Then, according to the action category to which the specific action in the selected key frame image belongs, the corresponding number of all standard action continuous frame images is obtained from the standard action library. Then, the three-dimensional action video corresponding to the selected key frame image is selected. The number of all continuous frame images in which specific actions appear in the action video is compared with the number of all continuous frame images of standard actions obtained, and then the adjustment coefficient is obtained in combination with the patient's severity. Then, the operation coefficient is obtained according to the action category and adjustment coefficient of the specific action in the selected key frame image. Then, the evaluation score is obtained according to the specification coefficient, difficulty coefficient, adjustment coefficient and operation coefficient of the specific action in the selected key frame image. The final evaluation score comprehensively considers the action category to which the specific action of the training personnel belongs, the action specification reflected by the specification coefficient, the action difficulty reflected by the difficulty coefficient, and the operation coefficient obtained according to the action category and adjustment coefficient, so that the scoring result is no longer one-sided.

[0112] Furthermore, combined with the instructor's evaluation, we can have a more comprehensive understanding of the entire operation process.

[0113] A scheduling system for regular training personnel, such as Figure 3 As shown, the above-mentioned method of scheduling regular training personnel is applied, including:

[0114] The data acquisition module acquires basic information of the training personnel and the basic information of the tutor. The basic information of the training personnel includes: basic personal information, professional background, training stage, and learning needs; the basic information of the tutor includes: name, professional field, clinical experience, teaching experience, personality characteristics, and teaching style;

[0115] The training requirement generation module formulates the training requirements according to the training manual, and formulates clear training goals and requirements according to the training standards and the actual situation of the hospital, including theoretical examinations, skills assessments, and clinical case analysis;

[0116] The matching module obtains the operating room information and surgical arrangement information, training requirements and basic information to match the training personnel and the tutor, and publishes the matching results to the training personnel and the tutor;

[0117] The evaluation module uses on-site and video recording methods to conduct expert scoring of the training personnel's performance during the operation, and adjusts the matching results based on the evaluation results, wherein the expert scoring includes: tutor evaluation and comprehensive evaluation.

[0118] An electronic device includes a memory and a processor, wherein the memory stores a computer program, and is characterized in that when the processor executes the computer program, it implements the steps of a method for scheduling regular training personnel.

[0119] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0120] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for scheduling regular training personnel, characterized in that: include: Obtain basic information of the trainees and tutors. The basic information of the trainees includes: basic personal information, professional background, training stage, and learning needs; the basic information of tutors includes: name, professional field, clinical experience, teaching experience, personality characteristics, and teaching style; Formulate training requirements according to the training manual, and formulate clear training goals and requirements based on the training standards and the actual situation of the hospital, including theoretical examinations, skills assessments, and clinical case analysis; Obtain operating room information and surgical arrangement information, training requirements and basic information to match training personnel and mentors, and publish the matching results to training personnel and mentors; During the operation, on-site recording and video recording are used to conduct expert evaluation of the training personnel's performance, and the matching result is adjusted according to the evaluation result, wherein the expert evaluation includes: instructor evaluation and comprehensive evaluation.

2. A method for scheduling regular training personnel according to claim 1, characterized in that: The specific steps for matching trainees and mentors are as follows: Quantify the basic information of training personnel and tutors; Constructing a training matching model, the training matching model includes an input layer, a hidden layer and an output layer, the number of nodes in the input layer is determined according to the number of preprocessed data features; the number of nodes in the hidden layer can be determined by experience or experiment; the output layer represents the matching score between the training personnel and the tutor; Combine the collected data of trainees and mentors into multiple sample pairs, each of which has a known matching label determined by manual evaluation or historical data; Select optimization algorithm and loss function: Select stochastic gradient descent or Adam as the optimization algorithm; The loss function selects the mean square error, and the goal is to make the output of the neural network close to the actual matching label; When the training of the regular training matching model is completed, the data of the new regular training personnel and the tutors are input into the regular training matching model to obtain the matching score. The scoring standard is a five-point system, where 1 point means no complementarity at all and 5 points means high complementarity.

3. A method for scheduling regular training personnel according to claim 1, characterized in that: The steps for generating the regulatory training requirements are as follows: Collect the content of the training manual, training standard documents and the actual situation of the hospital, including but not limited to the following aspects: the scope of professional knowledge, skill operation items, and clinical case types in the training manual, the specific requirements for knowledge mastery, skill proficiency, and case analysis ability in the training standards, the hospital's disciplinary characteristics, common case types, equipment resources, and teaching staff; pre-process the collected data and convert text information into quantifiable feature vectors; Constructing a training plan model, the training plan model includes an input layer, a hidden layer and an output layer, the number of nodes in the input layer is determined according to the number of preprocessed data features; the number of nodes in the hidden layer can be determined through experience or experiments; the output layer represents the output quantitative table of training requirements; The feature vector is input into the regular training planning model, and a preliminary quantitative table is determined as a label according to the regular training manual, standards and actual conditions of the hospital. For a certain regular training specialty, the weight of the theoretical examination on different subject knowledge, the scoring standard of each operation item in the skill assessment, and the specific scoring points of the clinical case analysis are determined, and the quantitative table of regular training requirements is output.

4. A method for scheduling regular training personnel according to claim 1, characterized in that: The expert evaluation of the training personnel's performance includes tutor evaluation and comprehensive evaluation. The tutor evaluation is subjectively scored according to the quantitative table of training requirements and on-site records; the comprehensive evaluation is objectively scored according to video records.

5. A method for scheduling regular training personnel according to claim 4, characterized in that: The specific steps of comprehensive evaluation are as follows: Shoot a frontal action video and several side action videos of the training personnel, where each side action video is shot at a different angle; For each side action video, input it together with the front action video into the 3D reconstruction algorithm model to construct a number of 3D action videos corresponding to each side action video; Inputting the plurality of three-dimensional action videos into a video action recognition algorithm model to perform specific action recognition, obtaining all continuous frame images between a start frame image and an end frame image where the specific action appears in each three-dimensional action video, and identifying a key frame image that best reflects the specific action among all continuous frame images; Acquire a preset standard action library, wherein the standard action library contains all standard action continuous frame images of each specific action; Input each key frame image into a human key point regression algorithm to obtain a human key point feature map corresponding to a specific action in each key frame image, then compare the human key point feature map of each key frame image with all standard action continuous frame images of each specific action in the standard action library, select one of the key frame images that best meets the standard action and its corresponding three-dimensional action video according to the comparison result, and identify the action category, specification coefficient and difficulty coefficient to which the specific action in the selected key frame image belongs; According to the action category to which the specific action in the selected key frame image belongs, the number of all standard action continuous frame images corresponding to the action category is obtained from the standard action library, the number of all continuous frame images in which the specific action appears in the three-dimensional action video corresponding to the selected key frame image is compared with the number of all obtained standard action continuous frame images, and then the adjustment coefficient is set in combination with the degree of the patient; The operation coefficient is calculated according to the action category and adjustment coefficient of the specific action in the selected key frame image, and the evaluation score is then calculated according to the specification coefficient, difficulty coefficient, adjustment coefficient and operation coefficient of the specific action in the selected key frame image.

6. A method for scheduling regular training personnel according to claim 5, characterized in that: Set adjustment coefficient based on patient severity Among them, ω2 represents the adjustment coefficient of the specific action in the selected key frame image, M i represents the number of all continuous frame images of a specific action in the 3D action video corresponding to the selected key frame image, M0 represents the number of all continuous frame images of standard actions obtained from the standard action library, and A represents the patient coefficient fitted with different degrees of illness; Where A=u0x0+u1x1+…+u n x n ; n represents the number of factors that affect surgical operations, including at least: factors of the training personnel themselves, surgical environment factors, surgical instrument factors, and patient factors; u0 to u n Represents the preset fitting constants; x0 to x n It represents the scores of the training personnel's own factors, surgical environment factors, surgical instrument factors, and patient factors.

7. A method for scheduling regular training personnel according to claim 5, characterized in that: The operating coefficient formula is: ω3=ω2×A×g; Among them, ω3 represents the operation coefficient of the specific action in the selected key frame image, A represents the adjustment coefficient of the specific action in the selected key frame image, which is fitted based on the patient's degree, and g is the operation constant preset for the action category to which each specific action belongs.

8. A method for scheduling regular training personnel according to claim 5, characterized in that: The calculation formula for the evaluation score is: y=ω0+ω1+ω2+ω3; Among them, y represents the evaluation score, ω0 represents the specification coefficient of the specific action in the selected key frame image, ω1 represents the difficulty coefficient of the specific action in the selected key frame image, ω2 represents the adjustment coefficient of the specific action in the selected key frame image, and ω3 represents the operation coefficient of the specific action in the selected key frame image.

9. A training staff scheduling system, using a training staff scheduling method according to any one of claims 1 to 8, characterized in that: include: The data acquisition module acquires basic information of the training personnel and the basic information of the tutor. The basic information of the training personnel includes: basic personal information, professional background, training stage, and learning needs; the basic information of the tutor includes: name, professional field, clinical experience, teaching experience, personality characteristics, and teaching style; The training requirement generation module formulates the training requirements according to the training manual, and formulates clear training goals and requirements according to the training standards and the actual situation of the hospital, including theoretical examinations, skills assessments, and clinical case analysis; The matching module obtains the operating room information and surgical arrangement information, training requirements and basic information to match the training personnel and the tutor, and publishes the matching results to the training personnel and the tutor; The evaluation module uses on-site and video recording methods to conduct expert scoring of the training personnel's performance during the operation, and adjusts the matching results based on the evaluation results, wherein the expert scoring includes: tutor evaluation and comprehensive evaluation.

10. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of a method for scheduling training personnel described in any one of claims 1 to 8 are implemented.

Images