Adaptive test paper generation method, device and equipment for pathology teaching

Abstract

The application provides a pathology teaching-oriented adaptive test paper generation method, device and equipment, and relates to the technical field of educational information processing, and comprises the following steps: determining evaluation index values corresponding to pathology teaching test questions contained in a pathology teaching question bank; performing data integrity inspection on the evaluation index values; in the case of passing the data integrity inspection, establishing examinee characteristic groupings, and associating the pathology teaching test questions to their matching examinee characteristic groupings; performing quality evaluation on the evaluation index values corresponding to the pathology teaching test questions associated to the examinee characteristic groupings, and then determining the evaluation results of the pathology teaching test questions in their associated examinee characteristic groupings; in the case of receiving a test paper grouping request uploaded by a user, generating a pathology teaching test paper corresponding to the test paper grouping request based on the pathology teaching test questions and their evaluation results. The application can significantly improve the efficiency of case screening, grading and test paper grouping processes, so as to meet the increasingly enhanced pathology examination requirements.

Description

Technical Field

[0001] This invention relates to the field of educational information processing technology, and in particular to an adaptive test paper generation method, apparatus and equipment for pathology teaching. Background Technology

[0002] With the increasing importance of pathology in medical education and clinical practice, the assessment of pathological diagnostic skills has become a crucial aspect of medical talent training. Pathology examinations typically require real cases, evaluating candidates' diagnostic abilities through comprehensive analysis of tissue sections, imaging data, and clinical history. In traditional pathology teaching systems, the collection and organization of case resources mainly rely on the accumulated clinical experience of pathologists; these cases need to undergo professional evaluation to determine their suitability as examination questions.

[0003] The current process of constructing pathology exam questions involves multiple stages: First, typical cases with teaching value are selected from a large number of clinical cases. Then, senior pathology experts assess the difficulty, typicality, and teaching applicability of the cases. Qualified cases are included in the question bank system and categorized into different levels (beginner, intermediate, and advanced) based on their complexity for use in physician assessments at different stages. However, existing technology lacks a test paper generation system specifically designed for the characteristics of pathology, resulting in inefficient case selection, grading, and test paper assembly processes, making it difficult to meet the growing demand for pathology assessments. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide an adaptive test paper generation method, apparatus and equipment for pathology teaching, which can significantly improve the efficiency of case screening, grading and test paper generation processes to meet the increasing demands of pathology assessment.

[0005] In a first aspect, the present invention provides an adaptive test paper generation method for pathology teaching, comprising: Determine the evaluation index values ​​corresponding to the pathology teaching test questions included in the pathology teaching test question bank in at least one historical examination. The pathology teaching test questions are constructed based on case data. The same case data corresponds to one or more pathology teaching test questions. The evaluation index values ​​are determined based on the examinee's examination results for the pathology teaching test questions in that historical examination. Perform data integrity checks on the assessment indicator values ​​corresponding to the pathology teaching test questions; After passing the data integrity check, establish candidate characteristic groups and associate pathology teaching test questions with their matching candidate characteristic groups; The evaluation index values ​​corresponding to the pathology teaching test items associated with the candidate characteristic groups are evaluated for quality. Based on the evaluation index values ​​that pass the quality evaluation, the evaluation results of the pathology teaching test items in their associated candidate characteristic groups are determined. Upon receiving a user's request to create a test paper, the system generates a pathology teaching test paper corresponding to the request, based on the pathology teaching test questions and their evaluation results.

[0006] In one implementation, before establishing candidate feature groups and associating pathology teaching questions with their matching candidate feature groups, the method further includes: When pathology teaching exam questions fall under the multi-stage diagnostic pathway analysis type, the following steps should be taken: To obtain the candidates' operational behavior regarding pathology teaching test questions; The diagnostic path constraint model corresponding to the candidate's seniority level is invoked, and the operational behavior is matched with the diagnostic path constraint model to obtain the compliance judgment result corresponding to the operational behavior. If the compliance assessment result is compliant, the evaluation index values ​​corresponding to the pathology teaching test questions will be retained.

[0007] In one implementation, the seniority hierarchy includes at least three levels: low seniority, middle seniority, and high seniority. The diagnostic path constraint model includes at least one or a combination of the following parameters: the maximum number of allowed special examination orders, the maximum number of allowed rounds of special examination orders, and verification conditions for operational accuracy. The operational behavior is matched with the diagnostic path constraint model to obtain a compliance judgment result corresponding to the operational behavior, including: For junior staff, if the operation meets the verification conditions for operational accuracy, the corresponding compliance judgment result of the operation is determined to be compliant. For mid-level staff, if the operation meets the verification conditions for operational accuracy, and the number of special medical orders corresponding to the operation does not exceed the maximum number of special medical orders allowed for mid-level staff, and the maximum number of rounds of special medical orders corresponding to the operation does not exceed the maximum number of rounds of special medical orders allowed for mid-level staff, the compliance judgment result corresponding to the operation is determined to be compliant. For senior staff, if the operation meets the verification conditions for operational accuracy, and the number of special examination orders corresponding to the operation does not exceed the maximum number of special examination orders allowed for senior staff, and the maximum number of rounds of special examination orders corresponding to the operation does not exceed the maximum number of rounds of special examination orders allowed for senior staff, the compliance judgment result corresponding to the operation is determined to be compliant.

[0008] In one implementation, candidate characteristic groups are established, and pathology teaching test questions are associated with their matching candidate characteristic groups, including: Extract the candidate characteristics corresponding to each candidate in historical exams; Based on the candidate characteristics corresponding to each candidate, the dominant characteristics corresponding to the historical examination are determined. The dominant feature is grouped as a candidate feature group, and the pathology teaching questions included in the historical exam are associated with the candidate feature group.

[0009] In one implementation, a quality assessment is performed on the evaluation index values ​​corresponding to pathology teaching test items associated with candidate characteristic groups, including: For pathology teaching questions included in the same candidate's characteristic group, the following operations shall be performed: If the number of times a pathology teaching test item is used is greater than or equal to a preset threshold, the corresponding evaluation index value of the pathology teaching test item is determined to pass the quality assessment. And / or, if the fluctuation of the examination results of the pathology teaching test items is less than or equal to the preset fluctuation threshold, the evaluation index value corresponding to the pathology teaching test items is determined to pass the quality assessment.

[0010] In one implementation, the assessment results of pathology teaching test items in their associated candidate characteristic groups are determined based on assessment index values ​​obtained through quality assessment, including: For the same candidate characteristic group, when comparing the evaluation index value of the same pathology teaching test item with the same indicator dimension, the following operations are performed: A time decay weight is assigned to the pathology teaching test item relative to the multiple evaluation index values ​​of this indicator dimension. The multiple evaluation index values ​​of this indicator dimension are then weighted according to the time decay weight to obtain the evaluation result of the pathology teaching test item relative to this indicator dimension.

[0011] In one implementation, the evaluation index values ​​corresponding to pathology teaching test items are subjected to data integrity checks, including: In cases where the number of candidates in the historical exam is lower than the preset threshold, or the results of the historical exam are abnormal, or there is a validity failure event in the historical exam, the determination of the evaluation index value corresponding to the pathology teaching test questions included in the historical exam fails the data integrity check.

[0012] Secondly, the present invention also provides an adaptive test paper generation device for pathology teaching, comprising: The indicator determination module is used to determine the evaluation indicator values ​​corresponding to the pathology teaching test questions contained in the pathology teaching test question bank in at least one historical examination. The pathology teaching test questions are constructed based on case data. The same case data corresponds to one or more pathology teaching test questions. The evaluation indicator values ​​are determined based on the examinee's examination results for the pathology teaching test questions in the historical examination. The integrity check module is used to check the data integrity of the evaluation index values ​​corresponding to pathology teaching test questions. The feature grouping module is used to create candidate feature groups after passing the data integrity check, and to associate pathology teaching test questions with their matching candidate feature groups; The quality assessment module is used to assess the quality of the assessment index values ​​corresponding to the pathology teaching test items associated with the candidate characteristic groups, and to determine the assessment results of the pathology teaching test items in their associated candidate characteristic groups based on the assessment index values ​​that have passed the quality assessment. The test paper generation module is used to generate a pathology teaching test paper corresponding to the test paper generation request based on the pathology teaching test questions and their evaluation results, upon receiving a user's uploaded test paper generation request.

[0013] Thirdly, the present invention also provides an electronic device including a processor and a memory, the memory storing computer-executable instructions executable by the processor, the processor executing the computer-executable instructions to implement any of the methods provided in the first aspect.

[0014] Fourthly, the present invention also provides a computer-readable storage medium storing computer-executable instructions, which, when invoked and executed by a processor, cause the processor to implement any of the methods provided in the first aspect.

[0015] This invention provides an adaptive test paper generation method, apparatus, and device for pathology teaching. First, it determines the evaluation index values ​​corresponding to pathology teaching test questions in at least one historical examination, as contained in a pathology teaching question bank. These pathology teaching test questions are constructed based on case data, with one or more pathology teaching test questions corresponding to the same case data. The evaluation index values ​​are determined based on the examinee's examination results for these pathology teaching test questions in that historical examination. Next, it performs a data integrity check on the evaluation index values ​​corresponding to the pathology teaching test questions. If the data integrity check passes, it establishes examinee feature groups and associates the pathology teaching test questions with their matching examinee feature groups. Then, it performs a quality assessment on the evaluation index values ​​corresponding to the pathology teaching test questions associated with the examinee feature groups, and based on the evaluation index values ​​that pass the quality assessment, it determines the evaluation result of the pathology teaching test questions in their associated examinee feature groups. Finally, upon receiving a user's uploaded test paper generation request, it generates a pathology teaching test paper corresponding to the test paper generation request based on the pathology teaching test questions and their evaluation results. The above method integrates evaluation index analysis of historical examination data, data integrity verification, candidate characteristic grouping modeling, and dynamic evaluation of test quality. It realizes personalized, hierarchical, and data-driven adaptive test paper generation for pathology teaching scenarios, which significantly improves the scientific nature of case test paper selection, the accuracy of hierarchical matching, and the automation and teaching adaptability of the test paper generation process.

[0016] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention are realized and obtained through the structures particularly pointed out in the description and the drawings.

[0017] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0018] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0019] Figure 1 A flowchart illustrating an adaptive test paper generation method for pathology teaching provided in an embodiment of the present invention; Figure 2 This invention provides a technical framework for an adaptive test paper generation method for pathology teaching. Figure 3 A schematic diagram of the structure of an adaptive test paper generation device for pathology teaching provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Currently, existing technologies lack test paper generation systems specifically designed for the characteristics of pathology, resulting in low efficiency in case screening, grading, and test paper assembly processes, making it difficult to meet the growing demand for pathology assessments. Based on this, the present invention provides an adaptive test paper generation method, device, and equipment for pathology teaching, which can significantly improve the efficiency of case screening, grading, and test paper assembly processes to meet the increasing demand for pathology assessments.

[0022] To facilitate understanding of this embodiment, a detailed description of an adaptive test paper generation method for pathology teaching disclosed in this embodiment of the invention will be provided first. (See [link to relevant documentation]). Figure 1 The diagram shows a flowchart of an adaptive test paper generation method for pathology teaching. The method mainly includes the following steps S102 to S110: Step S102: Determine the evaluation index values ​​corresponding to the pathology teaching test questions included in the pathology teaching question bank in at least one historical examination.

[0023] Among them, the pathology teaching test questions are constructed based on case data. The same case data corresponds to one or more pathology teaching test questions. The evaluation index value is determined based on the examinee's examination results on the pathology teaching test questions in the past examination. The evaluation index value specifically includes the values ​​corresponding to the index dimensions such as difficulty dimension, discrimination dimension, reliability dimension and validity dimension.

[0024] Step S104: Perform a data integrity check on the evaluation index values ​​corresponding to the pathology teaching test questions.

[0025] In one example, the following exclusion conditions can be preset: the number of candidates in the historical exam is lower than a preset threshold, the results of the historical exam are abnormal, or there is a validity failure event in the historical exam. If any of the above exclusion conditions are met in the historical exam to which the pathology teaching test question belongs, it can be determined that the evaluation index value of the pathology teaching test question in that historical exam has not passed the data integrity test, and the evaluation index value of the pathology teaching test question in that historical exam is removed.

[0026] Step S106: If the data integrity check is passed, establish candidate feature groups and associate the pathology teaching test questions with their matching candidate feature groups.

[0027] In one example, feature extraction is performed on the relevant information of all candidates in a history exam to determine the individual candidate characteristics of each candidate. These characteristics include at least seniority distribution (e.g., junior, intermediate, senior), education level distribution (e.g., bachelor's / master's and above), job / professional distribution (e.g., doctor / technician), and historical performance level. The combination of these features constitutes the candidate feature group. For this particular history exam, the dominant feature is determined based on the individual candidate characteristics of each candidate and their combinations. This dominant feature is then used as the final candidate feature group for the history exam, and the pathology teaching questions included in the exam are associated with this candidate feature group.

[0028] Step S108: Perform quality assessment on the evaluation index values ​​corresponding to the pathology teaching test items associated with the candidate feature groups, and determine the evaluation results of the pathology teaching test items in their associated candidate feature groups based on the evaluation index values ​​that have passed the quality assessment.

[0029] In one example, the quality assessment of pathology teaching test items associated with a certain candidate characteristic group can be conducted using test item stability scores and indicator consistency scores. For example, if the test item stability score or indicator consistency score is high, it will be confirmed that it has passed the quality assessment. The assessment indicator values ​​that have passed the quality assessment will be assigned time decay weights. Then, the assessment indicator values ​​that have passed the quality assessment will be weighted according to the time decay weights to obtain the assessment result of the pathology teaching test item in its associated candidate characteristic group.

[0030] Step S110: Upon receiving a user's uploaded test paper creation request, a pathology teaching test paper corresponding to the test paper creation request is generated based on the pathology teaching test questions and their evaluation results.

[0031] In one example, if the test paper creation request carries the target group for this test paper creation (e.g., the target group is primary level), then according to the screening criteria corresponding to primary level, pathology teaching test questions that meet the above screening criteria are selected from the candidate characteristic groups matched to primary level. The pathology teaching test questions obtained from the screening are used to create the test paper to obtain the pathology teaching test paper.

[0032] The adaptive test paper generation method for pathology teaching provided in this invention achieves personalized, hierarchical, and data-driven adaptive test paper generation for pathology teaching scenarios by integrating evaluation index analysis of historical examination data, data integrity verification, student characteristic grouping modeling, and dynamic evaluation of test paper quality. This significantly improves the scientific nature of case test paper selection, the accuracy of hierarchical matching, and the automation and teaching adaptability of the test paper generation process.

[0033] For ease of understanding, this invention provides a specific implementation of an adaptive test paper generation method for pathology teaching. See [link to relevant documentation]. Figure 2 The technical framework of an adaptive test paper generation method for pathology teaching is shown, including: (1) difficulty, discrimination, reliability and validity algorithms; (2) a three-level exclusion logic system, including: data integrity check, candidate feature matching degree screening and quality assessment; (3) adaptive test paper generation.

[0034] (1) Algorithms for difficulty, discrimination, reliability, and validity: (1.1) Difficulty: The difficulty of the test paper is a quantitative indicator reflecting the overall difficulty level of the test paper, and its value range is [0,1]. A higher value indicates that the test paper is easier.

[0035] Difficulty calculation for single multiple-choice questions: The difficulty of a single multiple-choice question is usually expressed as the pass rate, which is the percentage of test takers who answered the question correctly or passed. The calculation formula is as follows: ,in The number of people who answered the question correctly or passed the test. The total number of candidates; Difficulty calculation for individual non-multiple-choice questions: These questions do not have only two possible outcomes—correct or incorrect—but rather multiple possible results ranging from full marks to zero. The calculation formula is as follows: ,in The average score of the test takers on a particular question. This question deserves full marks; The difficulty of the entire exam paper is calculated as the average score rate, using the following formula: ,in The average score of all candidates' exam papers. This is the total score for the exam.

[0036] For example, the evaluation criteria for the difficulty of the test paper are as follows: P ≥ 0.8: Too easy, the discrimination should be improved; 0.6≤P<0.8: The difficulty is moderate and can be maintained; 0.4≤P<0.6: Slightly difficult but reasonable; needs to be checked to see if it focuses on the core test points. P<0.4: Too difficult; the question design should be re-examined.

[0037] (1.2) Discrimination analysis: Discrimination This refers to the ability of exam questions to differentiate between candidates' psychological characteristics. Highly discriminating questions can separate candidates of different levels, allowing high-achieving candidates to score higher and low-achieving candidates to score lower. In a highly discriminating exam, there is a certain proportion of students at the excellent, average, and poor levels. However, if students are relatively concentrated in a certain score range, with too many high scores or too many failing scores, the discriminating power is low.

[0038] Calculation of the discrimination index for a single multiple-choice question: Arrange all candidates' total scores from highest to lowest. Define the top 27% of candidates as the high-scoring group and the bottom 27% as the low-scoring group. Calculate the pass rate for each group on a given question. The difference between these two pass rates is the discrimination index (also called the identification index) for that question. The formula is: ;in, and The pass rates are for the high-performing group and the low-performing group, respectively.

[0039] The discrimination index of the entire test (commonly used): The calculation formula is as follows: ;in For differentiation, The average score of the 27% high-achieving group was [missing information]. The average score for the 27% low group was [missing information]. This is the total score for the exam.

[0040] Generally, all candidates are ranked from highest to lowest total score, with the top 27% of candidates in the high group and the bottom 27% in the low group, and then the average score of each group is calculated.

[0041] For example, in a test paper with a total score of 100, the average score of the high-scoring group is 90 and the average score of the low-scoring group is 60, then the discrimination index is 2(90-60) / 100=0.6; in a test paper with a question worth 2 points, the average score of the high-scoring group is 1.5 and the average score of the low-scoring group is 0.5, then the discrimination index is 2(1.5-0.5) / 2=1.

[0042] In one implementation, the discrimination index (D) ranges from -1.00 to +1.00. A positive D value is typically called positive discrimination; a negative D value is called negative discrimination; and a D value of 0 indicates no discrimination. For items with positive discrimination, the higher the D value, the better the discrimination effect. For example, a discrimination index above 0.4 indicates very good discrimination, 0.3–0.39 indicates good discrimination, 0.2–0.29 indicates poor discrimination requiring modification, and below 0.19 indicates poor discrimination and should be eliminated.

[0043] (1.3) Reliability Analysis (Overall Test Paper): Reliability refers to the consistency of results obtained when the same method is used to repeatedly measure the same subject. The Cronbach's α reliability coefficient is the most commonly used reliability coefficient. It is a commonly used method for assessing the reliability of educational tests, estimating the internal consistency of a test using a specific formula, and serving as an indicator of reliability. The formula is: ;in, This represents the total number of items in the scale. For the first The in-question variance of the scores. This represents the variance of the total score for all items. As can be seen from the formula, The coefficient evaluates the consistency between the scores of each item in the scale and belongs to the internal consistency coefficient.

[0044] For example, the evaluation criteria for test paper reliability are as follows: ≥0.9 indicates very high reliability of the test paper; 0.8≤ <0.9 indicates good reliability of the test paper; 0.7≤ <0.8 indicates the exam paper is acceptable; A score <0.7 indicates that the reliability of the test paper needs improvement.

[0045] (1.4) Validity Analysis (Overall Test Paper): Validity refers to the degree to which a measurement tool can accurately measure what it is meant to measure. Content Validity specifically refers to the degree to which the test paper content covers and matches the importance of the teaching syllabus (examination objectives).

[0046] The embodiments of this invention use the Weighted Content Validity Index (WCVI) to measure the content validity of the test paper. Its core is to evaluate the consistency between the test paper content and the teaching syllabus in terms of knowledge point coverage and teaching weight distribution.

[0047] The system automatically extracts the following data: a list of knowledge points ( ); weight of teaching time for each knowledge point ( The relative importance weight of each knowledge point (); At the same time, the system also needs to obtain test paper data from historical exams, including: a list of questions; the knowledge points tested in each question (there may be multiple knowledge points, which are obtained by the system automatically extracting the question content and using AI analysis); and the score for each question.

[0048] The calculation formula is as follows: ; Content Coverage Ratio, Formula: = Number of core knowledge points actually covered in the exam / Total number of core knowledge points specified in the teaching syllabus This is used to measure whether an exam paper covers the core content that should be tested. The specific algorithm is as follows: Extract all core knowledge points from the teaching syllabus to form a set. Extract all the knowledge points tested from the exam paper to form a set. Content Coverage Ratio (CCR) is calculated as follows: ;in Represents a set and The number of elements in the intersection. Furthermore, the relative importance weights of knowledge points can be introduced (…). ): .

[0049] Weight distribution consistency, formula: ;in, Knowledge points in the teaching syllabus The weight of teaching time is calculated using the following formula: Weight of teaching time = Knowledge Points Teaching time / Total teaching time. Knowledge points in the exam paper The actual score weight is calculated using the following formula: Exam Paper Score Weight = Knowledge Points Total score in the exam / Total score of the exam. Principle: Use cosine similarity to calculate the consistency between the distribution vectors of teaching weights and exam score weights. The closer the value is to 1, the higher the match between the exam score distribution and the teaching focus.

[0050] For example, the weighted content validity index (WCVI) evaluation criteria are shown below: ≥ 0.9: Excellent validity (comprehensive content coverage and excellent weighting). 0.8 ≤ <0.9: Good (good coverage and weight matching); 0.7 ≤ <0.8: Acceptable (basically covers the key points, but with some deviation). <0.7: Needs improvement (insufficient content coverage or score distribution is out of sync with teaching focus).

[0051] This invention provides a specific application example. The teaching syllabus contains 10 core knowledge points, with a total teaching time of 100 hours. The exam paper has a full score of 100 points, actually covering 8 of the knowledge points. Based on this, the Content Coverage Ratio (CCR) calculation is divided into a basic version and a weighted version: the basic CCR is equal to the number of covered knowledge points (8) divided by the total number of knowledge points in the syllabus (10), resulting in 0.80; the weighted CCR considers the importance weight of each knowledge point. Assuming the sum of the importance weights of the covered knowledge points is 0.85 and the sum of the importance weights of all knowledge points is 1.00, then the weighted CCR is 0.85. Weight Distribution Consistency (WDC) measures the degree of matching between the weight distribution of the exam paper score and the weight distribution of the teaching time through cosine similarity. For example, the teaching time for knowledge point 1 accounts for 10% of the total class time (i.e., 0.10), and its score in the test paper accounts for 8% (i.e., 0.08); the teaching time for knowledge point 2 accounts for 15% (0.15), and its score accounts for 20% (0.20); and so on. A teaching weight vector W and a test paper score weight vector P are constructed, and the cosine similarity is used to calculate WDC as 0.82. The comprehensive validity is calculated using the weighted average formula: the sum of 0.55 times the weighted CCR (0.85) and 0.45 times the WDC (0.82) is 0.8365, which is equivalent to 83.65 points on a percentage scale, corresponding to a validity level of "good".

[0052] (2) The three-level elimination logic system has the core elimination principle of "comparison within a homogeneous group is meaningful". Its basic concept is that the difficulty and discrimination of a question are not inherent absolute attributes of the question, but relative attributes dependent on a specific group of candidates; the difficulty and discrimination effect of the same question may vary significantly when facing candidates of different ability levels (such as novices and experts); therefore, when conducting statistical analysis, it is necessary to strictly limit it to within a comparable group of candidates (that is, belonging to the same group of candidate characteristics) to avoid cross-group data mixing that would lead to indicator distortion. This elimination logic system includes: data integrity verification, candidate characteristic matching screening, and quality assessment.

[0053] (2.1) Data integrity check: If the number of candidates in the historical exam is lower than the preset threshold, or the results of the historical exam are abnormal, or there is a validity failure event in the historical exam, the evaluation index value corresponding to the pathology teaching test questions included in the historical exam is determined, and the data integrity check is not passed.

[0054] Specifically, the exclusion criteria include: Insufficient number of examinees: If fewer than 30 examinees provide valid responses to a pathology teaching question in a single examination, the evaluation index value for that pathology teaching question will be excluded from that examination. The reason given is that the sample size is too small, resulting in poor stability and low reliability of statistical indicators such as difficulty and discrimination. Exceptions may be made for highly specialized certification examinations, where the minimum number of valid responses may be lowered to 10.

[0055] Abnormal scoring distribution (i.e., abnormal exam results in historical exams): If a question is answered correctly by all candidates in a particular exam (100% pass rate) or answered incorrectly by all candidates (0% pass rate), then the evaluation index value of that pathology teaching question in that exam will be excluded. The reason is that such extreme distributions cannot support the calculation of discrimination, and often indicate that the question difficulty is seriously inappropriate or that there are design flaws in terms of expression, technology, etc.

[0056] Anomalies in the examination environment (i.e., validity failure events in historical examinations): When there are obvious anomalies in the examination process, the relevant question data will not be used. Specifically, this includes: the actual average examination time being significantly shorter than the prescribed time (e.g., in a 60-minute exam, candidates complete it in an average of only 20 minutes); or the abnormal interruption rate of candidates exceeding 10% (e.g., a large number of candidates dropping out midway due to system failure, misunderstanding of rules, etc.). Such situations indicate that the examination implementation deviated from normal conditions, and the obtained answer data lacks ecological validity and a reliable basis.

[0057] Furthermore, this embodiment of the invention also further filters pathology teaching test questions of the multi-stage diagnostic pathway analysis type, and the filtering process is as follows: Step 1: Obtain the candidate's operational behavior in response to pathology teaching test questions.

[0058] For example, during the examination system's operation, the system automatically and continuously collects the candidate's complete operational behavior in real case diagnostic questions, forming structured diagnostic behavior sequence data. The collected data includes at least the following four categories: first, HE slide reading completion markers, used to confirm whether the candidate has completed basic morphological observation; second, records of medical history and case information review behavior, reflecting the candidate's access to clinical background information; third, operational data related to special examination orders, specifically covering the type of special examination (such as immunohistochemistry, molecular detection, or special staining), the actual number of special examination items requested, and the time sequence of each special examination item (used to estimate the number of special examination rounds in the diagnostic pathway); and fourth, the candidate's final submitted pathological diagnosis conclusion. The above multi-dimensional behavioral data is organized in an orderly manner according to timestamps, jointly constituting a traceable and resolvable individualized diagnostic behavior sequence, providing the original basis for subsequent pathway modeling and statistical stability control.

[0059] Step 2: Invoke the diagnostic path constraint model corresponding to the candidate's seniority level, match the operational behavior with the diagnostic path constraint model, and obtain the compliance judgment result corresponding to the operational behavior. The seniority level includes at least low seniority level, mid-senior seniority level, and high seniority level. The diagnostic path constraint model includes at least one of the following parameters or a combination thereof: maximum number of allowed special examination orders, maximum number of allowed rounds of special examination orders, and verification conditions for operational accuracy.

[0060] In one implementation, the system automatically identifies and maps a candidate's clinical practice seniority level based on their registered and verified professional qualification information (including but not limited to physician qualification certificate number, professional title appointment document, standardized residency training completion certificate, and synchronized data from the personnel system of their medical institution). The seniority level adopts a three-tiered structure, including: Lower seniority level: refers to licensed physicians who have completed standardized residency training but have not yet obtained the attending physician title, typically represented by resident physicians; Middle seniority level: refers to licensed physicians who have been appointed as attending physicians and possess independent consultation and preliminary diagnostic decision-making abilities; Senior seniority level: refers to senior physicians who have been appointed as associate chief physicians or chief physicians and are responsible for difficult case consultations, teaching guidance, and quality control. This seniority level serves as a key parameter input for subsequent diagnostic pathway analysis and statistical control.

[0061] Furthermore, the system configures and activates the corresponding Diagnostic Pathway Constraint Model (DPCM) for each seniority level. This model is used to determine in real time during the examination whether the diagnostic behavior sequence generated by the examinee conforms to the reasonable clinical decision-making paradigm corresponding to their seniority. DPCM should include at least one or more of the following quantifiable and verifiable technical constraint parameters: (a) Maximum threshold for the number of special examination orders: limiting the total number of special examination items allowed to be issued during a single diagnosis; (b) Maximum threshold for the number of special examination rounds: dividing rounds according to the time stamp sequence of the special examination initiation, limiting the number of diagnostic iterations allowed (e.g., first round of basic screening, second round of differential diagnosis, third round of targeted verification); (c) Verification conditions for operational accuracy: defining structured verification rules that the diagnostic conclusions must meet, including the standardization of diagnostic terminology (matching ICD-10 / WHO classification codes), the completeness of key criteria (e.g., the diagnosis of "adenocarcinoma" must be associated with histological grade and description of invasion depth), and logical consistency with the ability to observe HE slides (e.g., if a diagnosis of high-grade intraepithelial neoplasia is given directly without identifying typical dysplastic areas, verification failure is triggered).

[0062] The DPCM corresponding to different seniority levels exhibits a significant gradient design, for example: For junior staff, if the operational behavior meets the verification criteria for accuracy, the corresponding compliance judgment result is determined to be compliant. In practical applications, the model prioritizes the correctness of diagnostic results, allowing for a relatively lenient number and rounds of special examinations, focusing on assessing the level of mastery of basic pathological identification and standard diagnostic procedures.

[0063] For mid-level staff, the compliance assessment result is determined to be compliant if the operational behavior meets the verification conditions for operational accuracy, the number of special examination orders corresponding to the operational behavior does not exceed the maximum number of special examination orders allowed for mid-level staff, and the maximum number of rounds of special examination orders corresponding to the operational behavior does not exceed the maximum number of rounds of special examination orders allowed for mid-level staff. In practical applications, the model, while ensuring the accuracy of the final diagnosis, also incorporates the dual requirements of the number of special examinations and the number of rounds.

[0064] For senior staff, the compliance assessment result is determined to be compliant if the operational behavior meets the verification conditions for operational accuracy, and the number of special examination orders corresponding to the operational behavior does not exceed the maximum number of special examination orders allowed for senior staff, and the maximum number of rounds of special examination orders corresponding to the operational behavior does not exceed the maximum number of rounds of special examination orders allowed for senior staff. Specifically, the maximum number of special examination orders allowed for senior staff is less than the maximum number of special examination orders allowed for mid-level staff, and the maximum number of rounds of special examination orders allowed for senior staff is less than the maximum number of rounds of special examination orders allowed for mid-level staff. In practical applications, the model further narrows the thresholds for the number of special examinations and rounds, while ensuring the accuracy of the final diagnosis.

[0065] The system automatically matches and calculates the diagnostic behavior sequence data of the examinee with the diagnostic path constraint model corresponding to their seniority level, thereby obtaining the diagnostic path compliance judgment result of the answer record. The calculation process includes the following four steps: determining whether the final diagnostic conclusion meets the correctness requirements corresponding to the seniority level; counting the number of special examination orders and determining whether they exceed the quantity threshold set in the constraint model of the seniority level; identifying the special examination rounds based on the time sequence of the special examination orders and determining whether they exceed the round threshold set in the constraint model of the seniority level; and combining the above three judgment results to generate a unique diagnostic path compliance indicator.

[0066] Step 3: If the compliance assessment result is compliant, retain the evaluation index values ​​corresponding to the pathology teaching test questions. When the system performs statistical calculations on the difficulty and discrimination of real case diagnostic questions, it only includes answer records with a "compliant" diagnostic path assessment result in the statistical sample. For answer records that do not meet the corresponding seniority level's diagnostic path constraints, the system marks them as "unstable samples" and excludes them from the statistical calculation. This screening mechanism avoids statistical bias introduced by candidates at different seniority levels using differentiated and reasonable diagnostic paths.

[0067] To further understand, this embodiment of the invention provides a specific application example, including: In the pathology diagnosis examination system, the real case diagnosis question type requires candidates to complete a multi-stage interactive operation based on the same case: the first stage is to make a preliminary judgment based on information such as HE slides and patient medical history; the second stage is to initiate a special examination order in the system, including immunohistochemistry, molecular detection or special staining; the third stage is to give a final pathological diagnosis conclusion by combining the aforementioned information and test results.

[0068] Research has revealed that in this multi-stage interactive diagnostic question type, directly analyzing the difficulty and discrimination of answer data without distinguishing between candidates' seniority levels will lead to the following problems: Significant differences exist among candidates of different seniority levels in the number of special examination orders, rounds, and overall diagnostic path length; mixing answer data with different path characteristics will cause significant fluctuations in question difficulty and discrimination indicators, resulting in unstable statistical results; and statistical indicators for the same question are difficult to reuse across different exam batches, affecting automatic question generation and continuous maintenance of the question bank. Therefore, a technical solution is urgently needed that can analyze the multi-stage diagnostic behavior of candidates of different seniority levels and ensure the stability of statistical data.

[0069] Diagnostic path constraint model description: The system presets thresholds for the number of standard special examinations and the number of standard special examination rounds for candidates of different seniority levels. The following explanation uses senior candidates as an example, but is not limited to this.

[0070] (a) Parameter settings: The full score of the test questions is 100 points. Points; the standard parameters for senior candidates are: the maximum number of special examination orders allowed. The maximum number of special examination orders allowed .

[0071] (II) Definition of Diagnostic Path Constraint Model: The system calculates the test score according to the following rules based on the candidate's answer data: Final diagnosis accuracy assessment: If the final diagnosis is incorrect, the score is 0; if it is correct, the path cost deduction calculation begins. Points deducted for the number of special inspections: Let the actual number of special inspections be... ,when At that time, the deduction value ;otherwise ; Points deducted for each special inspection round: Let the actual number of special inspection rounds be... ,when At that time, the deduction value ;otherwise ; Final score: When the diagnosis is correct, The minimum score is 0 points.

[0072] (III) Example of scoring results: Under the above rules, the scores of the test items are continuously distributed. For example, in the group of senior test takers, the same test item may produce the following typical scores: those with the optimal path and correct diagnosis get close to full marks; those with correct diagnosis but redundant special examinations get medium scores; those with incorrect diagnosis or severely excessive path get low scores or 0 points, thus forming a complete score distribution covering high, medium and low score segments.

[0073] Compared with existing technical solutions that score solely based on the correctness of the final diagnostic answer, the embodiments of the present invention have at least the following technical effects: transforming the behavioral processes in real clinical diagnostic pathways into quantifiable and calculable scoring indicators; introducing the ability to differentiate the rationality and cost-effectiveness of auxiliary examinations while ensuring the fairness of the examination; enabling the test item scoring results to present a natural and continuous distribution, providing a stable and reliable data foundation for the statistical calculation of question difficulty and discrimination; and being particularly suitable for real-case diagnostic examination scenarios in which candidates of different seniority levels participate together.

[0074] (2.2) Candidate feature matching degree screening, including: extracting the candidate features corresponding to each candidate in the historical exam; determining the dominant features corresponding to the historical exam based on the candidate features corresponding to each candidate; grouping the dominant features as a candidate feature group, and associating the pathology teaching test questions included in the historical exam to the candidate feature group.

[0075] In practical implementation, the key concept is to construct candidate characteristic groups, including: establishing a multi-dimensional characteristic description of the candidate group for each exam; and establishing independent statistical records for each test question grouped according to candidate characteristics.

[0076] The specific operating procedure is as follows: Feature extraction: After each exam, the system extracts the following four features from the candidate group: seniority distribution, divided into junior (0-1 years), intermediate (1-3 years), and senior (3 years or more); education distribution, divided into bachelor's degree, master's degree and above; job / professional distribution, such as doctor, technician, etc.; historical performance level, divided into high, medium and low levels based on the average score of candidates in previous exams.

[0077] Dominant Feature Determination: If any subcategory within a single feature category accounts for more than 60%, then that subcategory is marked as the "dominant feature." For example, in exam E001, if 70% of the candidates have "junior level experience," then it is marked as "junior level dominant exam." If no single subcategory accounts for more than 60%, but two subcategories combined account for more than 80%, then it is marked as a "composite feature." For example, "junior level experience" 55% + "intermediate level experience" 30% = 85%, marked as "junior-intermediate mixed." If all subcategories are evenly distributed and there is no obvious concentration trend, then it is marked as a "mixed group."

[0078] Grouped storage includes: data isolation storage, which means that only answer records that meet specific feature combinations are stored in the corresponding group, and data between different feature groups are not mixed; dynamic group creation, which means that when an undefined feature combination (such as "junior seniority + undergraduate") appears, a new statistical group is automatically created; weighted statistics, which means that within the same group, the weighted average of the difficulty and discrimination of the questions is calculated based on the number of candidates in each exam.

[0079] Based on this, when generating intelligent test papers for new exams: First, predict the characteristic distribution of candidates for this exam; then, prioritize matching statistical groups with completely identical characteristics in the question bank; if there are no completely matching groups, calculate the similarity according to the feature overlap with preset weights, and make weighted references to the most similar groups; among them, the matching weights for the three characteristics of seniority, education, and job position are 40%, 30%, and 30%, respectively.

[0080] (2.3) Quality assessment: For pathology teaching questions included in the same candidate characteristic group, the following operations shall be performed: Sample stability score: If the number of times a pathology teaching test item is used is greater than or equal to a preset threshold, the corresponding evaluation index value is determined to have passed the quality assessment. In practical application, within the same feature group, a test item can only be marked as "stable and usable" if it is used three times or more; if it is used only once or twice, it is marked as "preliminary data" and is only for reference, not as the basis for test paper compilation or question bank evaluation.

[0081] Indicator Consistency Scoring: If the fluctuation of the examination results for pathology teaching test items is less than or equal to the preset fluctuation threshold, the assessment indicator value corresponding to the pathology teaching test item is determined to pass the quality assessment. In practical application, the standard deviation of the difficulty value obtained by the test item in multiple examinations within the same characteristic group is calculated; when the standard deviation is greater than 0.2, it is judged as "indicator unstable", the corresponding data is not used, and it is suggested that the group may need to be further subdivided.

[0082] For the same pathology teaching test item within the same candidate characteristic group, compared to the evaluation index values ​​of the same indicator dimension, the following operation is performed: A time decay weight is assigned to the multiple evaluation index values ​​of the pathology teaching test item relative to that indicator dimension. The multiple evaluation index values ​​of that indicator dimension are then weighted according to the time decay weight to obtain the evaluation result of the pathology teaching test item relative to that indicator dimension. For example, the data timeliness weighting rules are as follows: exam data from the most recent year has a weight of 1.0; data from 1 to 2 years ago has a weight of 0.7; and data from 3 years ago or more has a weight of 0.3, to reflect the impact of updated medical knowledge and evolving clinical skills on test item performance.

[0083] Furthermore, when conducting statistical analysis, it is also necessary to consider the consistency of exam types. For example, data from daily practice, midterm exams, and final certification exams should be collected separately and not mixed. Because students' motivation, stress levels, and behavioral patterns differ between practice and formal exams, their diagnostic paths and scores are not directly comparable.

[0084] To facilitate understanding of the foregoing embodiments, this invention provides a practical application scenario example: Scenario 1: Statistical Management of Question Q1001 The historical usage record of question Q1001 is as follows: Exam A: 70% of the candidates have junior experience, 20% have intermediate experience, and 10% have senior experience. This meets the criteria for "dominant characteristics". Their answer data is stored in the "junior dominant" group. Exam B: 70% of candidates have senior experience, 25% have intermediate experience, and 5% have junior experience; these candidates are placed in the "senior-dominated" group. Exam C: The candidates' seniority distribution is 30% junior, 40% intermediate, and 30% senior, with no dominant characteristic. They are stored in the "mixed group" group, which has a low data quality score.

[0085] Usage recommendations: When compiling test papers for the "Beginner Training Course", prioritize the "Beginner-led" group data; when compiling test papers for the "Advanced Expert Certification", prioritize the "Advanced-led" group data; the "Mixed Group" group data should only be used as supplementary reference and should be given a lower weight.

[0086] Scenario 2: Application of outlier exclusion rules: The usage record of question Q2005 is as follows: Exam D: Intermediate level exam, 100 participants, difficulty level 0.65, discrimination index 0.45, meets the requirements of stability and consistency, and is therefore retained; Exam E: Elementary exam, 25 participants, difficulty 0.90, discrimination index 0.15. Excluded because the sample size is less than 30 participants and the discrimination index deviates significantly from the historical trend of the same group. Exam F: Advanced exam, 80 participants, difficulty level 0.30, discrimination index The score of 0.10 indicates a negative discrimination index, suggesting a potential functional defect in the question, and it is therefore excluded.

[0087] Furthermore, after elimination and filtering, the system displays the question analysis results in the form of statistical cards grouped by candidate characteristics, for example: Question Q1001: Basic Python Syntax For beginner test takers (based on 5 exams, totaling 350 test takers): average difficulty 0.75 (moderate); average discrimination 0.42 (good); recommended use case: new employee training. Advanced test takers (based on 3 exams, totaling 180 test takers): Average difficulty 0.35 (difficult); Average discrimination 0.55 (excellent); Recommended use scenario expert certification.

[0088] (3) Adaptive Test Paper Generation: Upon receiving a user's uploaded test paper generation request, the system generates a pathology teaching test paper corresponding to the request, based on the pathology teaching test questions and their assessment results. For example, when the target candidate group is "beginner," the system automatically selects questions with a difficulty value between 0.6 and 0.8 and a discrimination index greater than 0.3; when the target candidate group is "mixed," the system prompts: "This group has dispersed characteristics. It is recommended to clarify the examination positioning and ability assessment objectives before generating the test paper."

[0089] Furthermore, the system provides a visual configuration interface, allowing administrators to flexibly adjust the following parameters: Minimum sample size threshold: default value is 30, can be set to 20, 50 or 100; Dominant feature determination threshold: default value is 60%, can be set to 50% or 70%; Data timeliness weight: adjust the participation weight of data from different years by setting the time decay coefficient; Forced exclusion rules: include whether to automatically exclude data with a discrimination index less than 0, and whether to automatically exclude data with an accuracy rate of 100% or 0%; Grouping and merging rules: include merging conditions for groups with similar features (such as merging "Junior - Undergraduate" and "Junior - Associate" into "Junior"), as well as conditions for creating new groups.

[0090] In summary, the embodiments of the present invention have at least the following characteristics: The difficulty of the questions is group-dependent; there is no "absolute difficulty" detached from the characteristics of test takers, only "difficulty targeted at a specific group of test takers." Statistical results must be strictly bound to specific test taker characteristics and examination scenarios; indicators out of context are meaningless. Data quality takes precedence over quantity, favoring a small number of homogeneous, stable, and compliant samples rather than a rough mean after a large number of mixed features. The statistical profile of the questions is dynamically optimized as the frequency of use increases, showing a gradual trend towards precision. All statistical work is aimed at practicality, ultimately serving accurate test paper compilation and effective teaching feedback.

[0091] This logical system ensures that the statistical indicators for the questions are both accurate and practical, and can effectively avoid biases caused by the mixed use of data by different groups of test takers, such as "basic exam data lowering the difficulty of expert questions" or "advanced exam data raising the difficulty of basic questions".

[0092] Based on the foregoing embodiments, this invention provides an adaptive test paper generation device for pathology teaching, see [link to related document]. Figure 3 The diagram shows a structural schematic of an adaptive test paper generation device for pathology teaching. The device mainly includes the following parts: The indicator determination module 302 is used to determine the evaluation indicator value of the pathology teaching test questions contained in the pathology teaching test question bank in at least one historical examination. The pathology teaching test questions are constructed based on case data. The same case data corresponds to one or more pathology teaching test questions. The evaluation indicator value is determined based on the examinee's examination results for the pathology teaching test questions in the historical examination. The integrity verification module 304 is used to perform data integrity verification on the evaluation index values ​​corresponding to pathology teaching test questions. Feature grouping module 306 is used to establish candidate feature groups after passing the data integrity check, and to associate pathology teaching test questions with their matching candidate feature groups; The quality assessment module 308 is used to assess the quality of the assessment index values ​​corresponding to the pathology teaching test items associated with the candidate characteristic groups, and to determine the assessment results of the pathology teaching test items in their associated candidate characteristic groups based on the assessment index values ​​that have passed the quality assessment. The test paper generation module 310 is used to generate a pathology teaching test paper corresponding to the test paper generation request based on the pathology teaching test questions and their evaluation results when a user uploads a test paper generation request.

[0093] The adaptive test paper generation device for pathology teaching provided in this invention realizes personalized, hierarchical, and data-driven adaptive test paper generation for pathology teaching scenarios by integrating evaluation index analysis of historical examination data, data integrity verification, student characteristic grouping modeling, and dynamic evaluation of test paper quality. This significantly improves the scientific nature of case test paper selection, the accuracy of hierarchical matching, and the automation and teaching adaptability of the test paper generation process.

[0094] In one implementation, a test item screening module is also included, for: When pathology teaching exam questions fall under the multi-stage diagnostic pathway analysis type, the following steps should be taken: To obtain the candidates' operational behavior regarding pathology teaching test questions; The diagnostic path constraint model corresponding to the candidate's seniority level is invoked, and the operational behavior is matched with the diagnostic path constraint model to obtain the compliance judgment result corresponding to the operational behavior. If the compliance assessment result is compliant, the evaluation index values ​​corresponding to the pathology teaching test questions will be retained.

[0095] In one implementation, the seniority hierarchy includes at least three levels: low seniority, middle seniority, and high seniority. The diagnostic path constraint model includes at least one or a combination of the following parameters: the maximum number of allowed special examination orders, the maximum number of allowed rounds of special examination orders, and verification conditions for operational accuracy. The test item screening module is specifically used for: For junior staff, if the operation meets the verification conditions for operational accuracy, the corresponding compliance judgment result of the operation is determined to be compliant. For mid-level staff, if the operation meets the verification conditions for operational accuracy, and the number of special medical orders corresponding to the operation does not exceed the maximum number of special medical orders allowed for mid-level staff, and the maximum number of rounds of special medical orders corresponding to the operation does not exceed the maximum number of rounds of special medical orders allowed for mid-level staff, the compliance judgment result corresponding to the operation is determined to be compliant. For senior staff, if the operation meets the verification conditions for operational accuracy, and the number of special examination orders corresponding to the operation does not exceed the maximum number of special examination orders allowed for senior staff, and the maximum number of rounds of special examination orders corresponding to the operation does not exceed the maximum number of rounds of special examination orders allowed for senior staff, the compliance judgment result corresponding to the operation is determined to be compliant.

[0096] In one implementation, the feature grouping module 306 is specifically used for: Extract the candidate characteristics corresponding to each candidate in historical exams; Based on the candidate characteristics corresponding to each candidate, the dominant characteristics corresponding to the historical examination are determined. The dominant feature is grouped as a candidate feature group, and the pathology teaching questions included in the historical exam are associated with the candidate feature group.

[0097] In one implementation, the quality assessment module 308 is specifically used for: For pathology teaching questions included in the same candidate's characteristic group, the following operations shall be performed: If the number of times a pathology teaching test item is used is greater than or equal to a preset threshold, the corresponding evaluation index value of the pathology teaching test item is determined to pass the quality assessment. And / or, if the fluctuation of the examination results of the pathology teaching test items is less than or equal to the preset fluctuation threshold, the evaluation index value corresponding to the pathology teaching test items is determined to pass the quality assessment.

[0098] In one implementation, the quality assessment module 308 is specifically used for: For the same candidate characteristic group, when comparing the evaluation index value of the same pathology teaching test item with the same indicator dimension, the following operations are performed: A time decay weight is assigned to the pathology teaching test item relative to the multiple evaluation index values ​​of this indicator dimension. The multiple evaluation index values ​​of this indicator dimension are then weighted according to the time decay weight to obtain the evaluation result of the pathology teaching test item relative to this indicator dimension.

[0099] In one implementation, the integrity verification module 304 is specifically used for: In cases where the number of candidates in the historical exam is lower than the preset threshold, or the results of the historical exam are abnormal, or there is a validity failure event in the historical exam, the determination of the evaluation index value corresponding to the pathology teaching test questions included in the historical exam fails the data integrity check.

[0100] The device provided in this embodiment of the invention has the same implementation principle and technical effect as the aforementioned method embodiment. For the sake of brevity, any parts not mentioned in the device embodiment can be referred to the corresponding content in the aforementioned method embodiment.

[0101] This invention provides an electronic device, specifically, the electronic device includes a processor and a memory; the memory stores a computer program, which, when run by the processor, executes the method described in any of the above embodiments.

[0102] Figure 4 The present invention provides a schematic diagram of the structure of an electronic device 100, which includes a processor 40, a memory 41, a bus 42 and a communication interface 43. The processor 40, the communication interface 43 and the memory 41 are connected through the bus 42. The processor 40 is used to execute executable modules, such as computer programs, stored in the memory 41.

[0103] The memory 41 may include high-speed random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 43 (which can be wired or wireless), such as the Internet, wide area network, local area network, metropolitan area network, etc.

[0104] Bus 42 can be an ISA bus, PCI bus, or EISA bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 4 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0105] The memory 41 is used to store programs. After receiving an execution instruction, the processor 40 executes the program. The method executed by the device for defining the flow process disclosed in any of the foregoing embodiments of the present invention can be applied to the processor 40 or implemented by the processor 40.

[0106] Processor 40 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of processor 40 or by instructions in software form. Processor 40 can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this invention can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in memory 41. The processor 40 reads the information in memory 41 and, in conjunction with its hardware, completes the steps of the above method.

[0107] The computer program product of the readable storage medium provided in the embodiments of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods described in the foregoing method embodiments. For specific implementation, please refer to the foregoing method embodiments, which will not be repeated here.

[0108] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0109] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. An adaptive test paper generation method for pathology teaching, characterized in that, include: The evaluation index value corresponding to the pathology teaching test questions included in the pathology teaching test question bank in at least one historical examination is determined. The pathology teaching test questions are constructed based on case data. The same case data corresponds to one or more of the pathology teaching test questions. The evaluation index value is determined based on the examinee's examination results for the pathology teaching test questions in the historical examination. The data integrity of the assessment index values ​​corresponding to the pathology teaching test questions was checked. If the data integrity check is passed, candidate feature groups are established, and the pathology teaching test questions are associated with the matching candidate feature groups. The evaluation index values ​​corresponding to the pathology teaching test items associated with the candidate feature groups are evaluated for quality, and the evaluation result of the pathology teaching test items in their associated candidate feature groups is determined based on the evaluation index values ​​that pass the quality evaluation. Upon receiving a user's request to create a test paper, a pathology teaching test paper corresponding to the request is generated based on the pathology teaching test questions and the evaluation results.

2. The adaptive test paper generation method for pathology teaching according to claim 1, characterized in that, Before establishing candidate characteristic groups and associating the pathology teaching test questions with their matching candidate characteristic groups, the method further includes: If the pathology teaching test questions belong to the multi-stage diagnostic pathway analysis type, the following operations shall be performed: Acquire the candidates' operational behavior regarding the aforementioned pathology teaching test questions; The diagnostic path constraint model corresponding to the candidate's seniority level is invoked, and the operation behavior is matched with the diagnostic path constraint model to obtain the compliance judgment result corresponding to the operation behavior; If the compliance assessment result is compliant, the evaluation index value corresponding to the pathology teaching test question shall be retained.

3. The adaptive test paper generation method for pathology teaching according to claim 2, characterized in that, The seniority hierarchy includes at least a low seniority hierarchy, a middle seniority hierarchy, and a high seniority hierarchy. The diagnostic path constraint model includes at least one of the following parameters or a combination thereof: the maximum number of allowed special examination orders, the maximum number of allowed special examination orders, and the verification conditions for operational accuracy. The operation behavior is matched with the diagnostic path constraint model to obtain the compliance judgment result corresponding to the operation behavior, including: For the lower seniority level, if the operation meets the verification conditions for the accuracy of the operation, the compliance judgment result corresponding to the operation is determined to be compliant; For the middle-aged and senior level, if the operation meets the verification conditions for the accuracy of the operation, and the number of special medical orders corresponding to the operation does not exceed the maximum number of special medical orders allowed for the middle-aged and senior level, and the maximum number of rounds of special medical orders corresponding to the operation does not exceed the maximum number of rounds of special medical orders allowed for the middle-aged and senior level, then the compliance judgment result corresponding to the operation is determined to be compliant. For the senior level, if the operation meets the verification conditions for the accuracy of the operation, and the number of special examination orders corresponding to the operation does not exceed the maximum number of special examination orders allowed for the senior level, and the maximum number of rounds of special examination orders corresponding to the operation does not exceed the maximum number of rounds of special examination orders allowed for the senior level, then the compliance judgment result corresponding to the operation is determined to be compliant.

4. The adaptive test paper generation method for pathology teaching according to claim 1, characterized in that, Establish candidate characteristic groups and associate the pathology teaching test questions with their matching candidate characteristic groups, including: Extract the candidate characteristics corresponding to each candidate in the historical exams; Based on the candidate characteristics corresponding to each candidate, the dominant characteristics corresponding to the historical examination are determined; The dominant feature is grouped as a candidate feature group, and the pathology teaching questions included in the historical exam are associated with the exam feature group.

5. The adaptive test paper generation method for pathology teaching according to claim 1, characterized in that, The quality assessment of the evaluation index values ​​corresponding to the pathology teaching test items associated with the candidate characteristic group includes: For the pathology teaching test questions included in the same candidate characteristic group, the following operations are performed: If the number of times the pathology teaching test question is used is greater than or equal to a preset threshold, the evaluation index value corresponding to the pathology teaching test question is determined to have passed the quality assessment. And / or, if the fluctuation of the examination results of the pathology teaching test item is less than or equal to a preset fluctuation threshold, the evaluation index value corresponding to the pathology teaching test item is determined to have passed the quality assessment.

6. The adaptive test paper generation method for pathology teaching according to claim 1, characterized in that, Based on the evaluation index values ​​obtained through quality assessment, the evaluation results of the pathology teaching test items in their associated candidate characteristic groups are determined, including: For the same candidate characteristic group, when comparing the same pathology teaching test item with the evaluation index value of the same indicator dimension, the following operations are performed: A time decay weight is assigned to the pathology teaching test item relative to the multiple evaluation index values ​​of the indicator dimension, and the multiple evaluation index values ​​of the indicator dimension are weighted according to the time decay weight to obtain the evaluation result of the pathology teaching test item relative to the indicator dimension.

7. The adaptive test paper generation method for pathology teaching according to claim 1, characterized in that, The data integrity of the assessment index values ​​corresponding to the pathology teaching test questions was checked, including: If the number of candidates in the historical exam is lower than a preset threshold, or if the results of the historical exam are abnormal, or if there is a validity failure event in the historical exam, the evaluation index value corresponding to the pathology teaching test questions included in the historical exam will be determined to have failed the data integrity check.

8. An adaptive test paper generation device for pathology teaching, characterized in that, include: The indicator determination module is used to determine the evaluation indicator value corresponding to the pathology teaching test questions contained in the pathology teaching test question bank in at least one historical examination. The pathology teaching test questions are constructed based on case data. The same case data corresponds to one or more of the pathology teaching test questions. The evaluation indicator value is determined based on the examinee's examination results for the pathology teaching test questions in the historical examination. The integrity verification module is used to perform data integrity verification on the evaluation index values ​​corresponding to the pathology teaching test questions. The feature grouping module is used to establish candidate feature groups after passing the data integrity check, and to associate the pathology teaching test questions with the matching candidate feature groups. The quality assessment module is used to assess the quality of the assessment index values ​​corresponding to the pathology teaching test questions associated with the candidate feature groups, and to determine the assessment result of the pathology teaching test questions in their associated candidate feature groups based on the assessment index values ​​that have passed the quality assessment. The test paper generation module is used to generate a pathology teaching test paper corresponding to the test paper generation request based on the pathology teaching test questions and the evaluation results, upon receiving a test paper generation request uploaded by a user.

9. An electronic device, characterized in that, The method includes a processor and a memory, the memory storing computer-executable instructions executable by the processor, the processor executing the computer-executable instructions to implement the method of any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions that, when invoked and executed by a processor, cause the processor to perform the method according to any one of claims 1 to 7.

Images