Physical examination evaluation method, system and storage medium based on simulated digital patients

Abstract

The present invention discloses a physical examination evaluation method, system and storage medium based on simulated digital patients, which relates to the technical field of medical treatment. By combining the OSCE scoring standard with the relevant physical examination scoring data set, medical students can obtain highly targeted feedback to help them understand their performance in different assessment items. This data-based personalized feedback mechanism enables students to quickly identify and improve their weaknesses, thereby achieving targeted learning. In steps S3 and S4, through the hierarchical classification of medical student groups and the analysis of common weaknesses, educators can obtain in-depth insights into different learning groups. This insight not only helps to adjust teaching strategies, but also promotes the continuous improvement of teaching content to better meet the needs of different learners. The teaching plan prediction model constructed by convolutional neural network technology can integrate relative abundance and improvement to generate a prioritized teaching plan.

Description

Technical Field

[0001] The present invention relates to the technical field of medical treatment, and in particular to a physical examination evaluation method, system and storage medium based on simulating digital patients. Background Art

[0002] In the field of medical education, with the advancement of technology, especially the widespread application of virtual reality (VR) and augmented reality (AR) technologies, simulation teaching has become a key means to improve the clinical skills of medical students. These technologies enable medical students to conduct actual combat exercises in a controllable and immersive environment, thereby reducing the risks in the real clinical environment. In this context, the concept of simulated digital patients came into being. By creating a 3D digital patient examination environment, educators can provide medical students with a rich variety of disease types, help them accumulate practical experience during physical examination training, and evaluate their physical examination skills through objective structured clinical examination (OSCE) standards. This process not only improves the operational capabilities of medical students, but also enables them to analyze their own performance with the support of data.

[0003] Although physical examination training with simulated digital patients has shown many advantages in medical education, there are still some existing defects and shortcomings. First, traditional physical examination training often relies on limited clinical cases and actual patients, which makes it difficult for medical students to gain sufficiently diverse disease experience during training, limiting their comprehensive ability improvement. Second, the existing evaluation system often fails to make full use of the collected data, making it difficult for educators to accurately identify the common weaknesses and needs of medical students, resulting in educators not being very targeted when teaching medical students. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a physical examination evaluation method, system and storage medium based on simulating digital patients, which solve the problems in the above-mentioned background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: A physical examination evaluation method based on simulating a digital patient comprises the following steps:

[0006] S1. Use VR and AR technologies to create a 3D digital patient examination environment, and determine the disease type of the 3D digital patient based on the learning objectives;

[0007] S2. Medical students conduct physical examination training based on the disease types of 3D digital patients, and combine the OSCE scoring standards to collect statistics on the physical examination results of several medical students in each assessment item to obtain relevant physical examination scoring data sets;

[0008] S3. Based on the relevant physical examination score data set in S2, analyze the similarity of physical examination performance among several medical students to hierarchically classify the medical students, obtain several groups of medical students, and analyze and calculate the relative abundance Xfz;

[0009] S4. Analyze and identify the common weaknesses of medical students in several groups of medical students to obtain the degree of improvement between each assessment item in the corresponding medical student groups. , based on the lift The numerical value of is used to determine the teaching content that needs to be strengthened for medical students in the corresponding medical student group;

[0010] S5. Use convolutional neural network technology to build a teaching plan prediction model. By Input into the teaching plan prediction model to construct the priority Yxj of the teaching plan, and formulate the corresponding teaching plan based on the priority Yxj value.

[0011] Preferably, the specific steps of S1 include:

[0012] S11. First, according to the learning objectives, determine the disease type of the 3D digital patient, and use the 3D modeling software Maya to create and edit the 3D digital patient model;

[0013] S12. Combine Oculus Quest in VR technology and Microsoft HoloLens in AR technology to provide a virtual reality experience for the 3D digital patient model. The design of the 3D digital patient model will construct patients of corresponding age, gender and pathological characteristics according to the learning objectives, simulate the physiological characteristics of the corresponding disease type for the 3D digital patient model, and then use sound effects to simulate the sound of a stethoscope.

[0014] Preferably, the specific steps of S2 include:

[0015] S21. Medical students conduct physical examination training according to the set disease types. During the physical examination training, the corresponding physical examination operations and conclusions of each medical student are monitored and recorded in real time, and the scoring indicators of the physical examination training are formulated in combination with the OSCE scoring standards. The scoring indicators include: communication skills, correctness of physical examination methods, observation ability, clinical reasoning ability and patient safety awareness;

[0016] S22. Use spreadsheet software to create a scoring table to record the examiner's item-by-item evaluation of the medical students to obtain a relevant physical examination scoring data set, which includes the pass and fail status of each medical student in terms of communication skills, correctness of physical examination methods, observation ability, clinical reasoning ability, and patient safety awareness, and convert the data information in the relevant physical examination scoring data set into Boolean data.

[0017] Preferably, the specific steps of S3 include:

[0018] S31, setting the number of cluster centers Zs, wherein the number of cluster centers Zs is set as follows: Among them, m represents the number of scoring indicator items;

[0019] S32, based on the relevant physical examination score data set after data conversion in S22, randomly assign each medical student in it to a number of cluster centers, and analyze and calculate the number of physical examination results of each medical student for each assessment item that are the same as the physical examination results of each assessment item in each cluster center, and record it as a matching value Ppz; wherein each cluster center is a combination of the physical examination results of each assessment item;

[0020] S33, presetting a threshold, and comparing the matching value Ppz with the threshold, if the matching value Ppz exceeds the threshold, the corresponding medical student is included in the corresponding cluster center; if the matching value Ppz does not exceed the threshold, the corresponding medical student is not included in the corresponding cluster center;

[0021] S34. Based on the comparison result in S33, the medical students included in each cluster center are finally determined to construct several groups of medical students, wherein the number of the several groups of medical students is the same as the number of cluster centers Zs.

[0022] Preferably, S3 also includes:

[0023] S35. According to several groups of medical students, the relative abundance Xfz of the corresponding medical student groups is analyzed and calculated, and is obtained specifically in the following manner:

[0024] ;

[0025] In the formula, It is expressed as the relative abundance of medical students in the i-th group of medical students, It is expressed as the number of medical students in the i-th group of medical students, Expressed as the number of medical students.

[0026] Preferably, the specific steps of S4 include:

[0027] S41. Calculate the physical examination results of the corresponding assessment items in each medical student group The single support , which can be obtained by:

[0028] ;

[0029] In the formula, It is expressed as the number of different physical examination results of each assessment item in the corresponding medical student group. Indicates the results of the physical examination for the corresponding assessment items The number of occurrences;

[0030] S42, pre-set the frequency threshold P, and calculate the physical examination results of the corresponding assessment items in each medical student group. The single support Compare with the frequency threshold P to screen out the relatively frequent physical examination results of the assessment items in each medical student group. The specific contents are:

[0031] If the physical examination results of the corresponding assessment items in each medical student group The single support When the frequency threshold P is exceeded, the physical examination results of the relatively frequent assessment items in the corresponding medical student group are screened out; if the physical examination results of the corresponding assessment items in each medical student group The single support When the frequency threshold value P is not exceeded, it is judged that the physical examination results of the corresponding assessment items do not have a relatively frequent situation;

[0032] S43, the physical examination results of the relatively frequent assessment items in the corresponding medical student groups screened out in S42 are counted and combined into two items to generate a frequent binomial set; the physical examination results of the corresponding assessment items in each medical student group are The single support Calculation of the transition to the two-term support , where the two-term support Calculation and single item support The same, specifically:

[0033] ;

[0034] In the formula, It is expressed as the number of occurrences of the physical examination results of the corresponding assessment items in the two groups; It is expressed as the number of different physical examination results for each assessment item within the two corresponding groups of medical students;

[0035] S44, repeat the steps in S42 and S43 until there is no combination item set whose corresponding support exceeds the frequency threshold P.

[0036] Preferably, S4 also includes:

[0037] S45. Calculate the confidence level based on the contents in S1 to S4 , which is obtained by the following formula:

[0038] ;

[0039] In the formula, It is expressed as the support degree of the physical examination results of two groups of corresponding assessment items appearing simultaneously;

[0040] S46. Analyze the common weaknesses of medical students in several groups to obtain the degree of improvement between the assessment items in the corresponding medical student groups. , which is obtained by the following formula:

[0041] ;

[0042] In the formula, Represents the physical examination results of the corresponding assessment items in each medical student group The single item support.

[0043] Preferably, the specific steps of S5 include:

[0044] S51, by the relative abundance Xfz and the degree of improvement Input into the teaching plan prediction model to construct the teaching plan priority Yxj, which is obtained by the following formula:

[0045] ;

[0046] In the formula, and are weight values, Expressed as a correction constant;

[0047] S52. Based on the numerical value of the priority level Yxj, the medical students are given reinforced homework on the corresponding teaching contents.

[0048] A physical examination and evaluation system based on simulated digital patients, including a creation module, a statistics module, a classification module, an impact module and a priority compensation module;

[0049] Create modules, use VR and AR technologies to create a 3D digital patient examination environment, and determine the type of symptoms of the 3D digital patient based on the learning objectives;

[0050] In the statistical module, medical students are trained to conduct physical examinations based on the types of symptoms of 3D digital patients, and combined with the OSCE scoring standards, the physical examination results of several medical students in each assessment item are statistically analyzed to obtain relevant physical examination scoring data sets;

[0051] The classification module analyzes the similarity of physical examination performance among several medical students based on the relevant physical examination score data set, so as to classify the medical students hierarchically, obtain several groups of medical students, and analyze and calculate the relative abundance Xfz;

[0052] Impact module, analyze and determine the common weaknesses of medical students in several groups of medical students to obtain the improvement degree between each assessment item in the corresponding medical student group , based on the lift The numerical value of is used to determine the teaching content that needs to be strengthened for medical students in the corresponding medical student group;

[0053] Priority compensation module and convolutional neural network technology are used to build a teaching plan prediction model. Input into the teaching plan prediction model to construct the priority Yxj of the teaching plan, and formulate the corresponding teaching plan based on the priority Yxj value.

[0054] The physical examination and evaluation storage medium based on the simulated digital patient stores a computer program, and the computer program stores computer instructions. When the instructions are executed by the processor, a physical examination and evaluation system based on the simulated digital patient is implemented.

[0055] The present invention provides a physical examination evaluation method, system and storage medium based on simulating digital patients, which have the following beneficial effects:

[0056] (1) The 3D digital patient examination environment created using VR and AR technology not only provides medical students with a realistic clinical practice scenario, but also stimulates their interest in learning and sense of participation. This immersive experience enables students to practice repeatedly in a simulated environment, develop their clinical skills and confidence, and thus improve their learning efficiency. By combining the OSCE scoring criteria with relevant examination scoring data sets, medical students can obtain highly targeted feedback to help them understand their performance in different assessment items. This data-based personalized feedback mechanism enables students to quickly identify and improve their weaknesses, thereby achieving targeted learning. In steps S3 and S4, through the hierarchical classification of medical student groups and the analysis of common weaknesses, educators can gain in-depth insights into different learning groups. This insight not only helps to adjust teaching strategies, but also promotes the continuous improvement of teaching content to better meet the needs of different learners. The teaching plan prediction model constructed using convolutional neural network technology can integrate relative abundance and improvement to generate a prioritized teaching plan. This method enables educators to quickly respond to the needs of different learning groups, realize dynamic adjustment of teaching content, and ensure that each student can receive intensive training in the areas where they need it most. At the same time, through the accumulation of large amounts of data, educators can conduct further research to explore more effective teaching methods and evaluation criteria, and promote the continued development of the field of medical education.

[0057] (2) After determining the matching value Ppz between each medical student and the cluster center, personalized feedback and guidance can be provided to the medical students. When the matching value of a medical student exceeds the preset threshold, it means that he or she has performed outstandingly in a specific assessment project, and a higher level of learning resources and challenges can be provided to him or her. Students whose matching value does not meet the standard can receive targeted support, thereby achieving teaching in accordance with the students' aptitude. By analyzing the relative abundance Xfz of the medical student group, it can help educational administrators better understand the overall performance of different medical student groups. This data-driven analysis method can provide a strong basis for curriculum design, teaching methods, and resource allocation, thereby optimizing educational management and decision-making. This method provides a quantitative evaluation standard by statistically analyzing the physical examination results of each medical student in each assessment project. When medical students are assigned to a specific cluster center, the group set formed can promote teamwork. Medical students can discuss and share each other's experience and knowledge in the group, improve the team's collaborative ability, and enhance learning effects. By systematically analyzing the physical examination results of each medical student, it is possible to better identify the strengths and weaknesses of medical students in clinical skills, so as to improve their abilities in a targeted manner.

[0058] (3) By analyzing the common weaknesses of medical students, teachers can develop targeted teaching plans to meet the needs of different student groups, improve learning outcomes, identify frequently occurring physical examination results of assessment items, and make teaching content more focused on students' weak links in practical skills or knowledge mastery, thereby improving learning efficiency and reducing time wasted in the learning process. Based on the value of the improvement degree Tsd, educational institutions can adjust the curriculum to strengthen medical students' learning in certain assessment items to ensure that they have the necessary skills and knowledge before entering clinical practice. By quantitatively analyzing the support, educational managers can make more scientific decisions based on data, thereby improving course quality and teaching outcomes.

[0059] (4) Confidence calculation is used to quantify the relationship between assessment items. By calculating the support of co-occurrence, it is possible to understand whether these items are related to each other. By calculating confidence, educators can understand which assessment items have an impact on each other, which helps to build a knowledge framework for medical students and enable them to better understand the connections and differences between knowledge. If it is found that the confidence between certain assessment items is high, teachers can consider integrating these items into the same teaching module to improve students' overall understanding and application ability. Based on the performance of different students in the assessment items, teachers can provide students with personalized feedback and learning suggestions to help them focus on improving their weak links.

[0060] (5) Through the calculation of the priority Yxj, the learning needs of medical students can be further clearly identified, the importance of each assessment item can be quantitatively evaluated, and high-priority items can be ensured to receive sufficient attention under the condition of limited teaching resources, thereby improving the overall teaching effect. It helps medical students to improve their clinical skills in a targeted manner, especially their performance in physical examination training. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 It is a flow chart of the physical examination and evaluation method based on simulating a digital patient according to the present invention;

[0062] Figure 2 This is a block diagram of the physical examination and evaluation system based on simulating digital patients according to the present invention. DETAILED DESCRIPTION

[0063] 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.

[0064] Example 1

[0065] See also Figure 1 The present invention provides a physical examination evaluation method based on simulating a digital patient, comprising the following steps:

[0066] S1. Use VR and AR technologies to create a 3D digital patient examination environment, and determine the disease type of the 3D digital patient based on the learning objectives;

[0067] S2. Medical students conduct physical examination training based on the disease types of 3D digital patients, and combine the OSCE scoring standards to collect statistics on the physical examination results of several medical students in each assessment item to obtain relevant physical examination scoring data sets;

[0068] S3. Based on the relevant physical examination score data set in S2, analyze the similarity of physical examination performance among several medical students to hierarchically classify the medical students, obtain several groups of medical students, and analyze and calculate the relative abundance Xfz;

[0069] S4. Analyze and identify the common weaknesses of medical students in several groups of medical students to obtain the degree of improvement between each assessment item in the corresponding medical student groups. , based on the lift The numerical value of is used to determine the teaching content that needs to be strengthened for medical students in the corresponding medical student group;

[0070] S5. Use convolutional neural network technology to build a teaching plan prediction model. By Input into the teaching plan prediction model to construct the priority Yxj of the teaching plan, and formulate the corresponding teaching plan based on the priority Yxj value.

[0071] In this embodiment, the physical examination results of medical students are statistically analyzed in combination with the OSCE scoring standard to obtain relevant physical examination scoring data sets, so that the teaching process can be customized based on actual performance, and this process ensures that the teaching content is closely integrated with the real needs of medical students. In steps S3 and S4, by analyzing the similarity of physical examination performance and identifying common weaknesses, medical students can be effectively grouped, and then the relative abundance and improvement of each group can be analyzed. This process enables teachers to quantify the learning progress of medical students, and adjust teaching strategies accordingly to optimize resource allocation. With the help of the teaching plan prediction model constructed by convolutional neural network technology, a teaching plan based on priority can be automatically generated. By integrating relative abundance and improvement into the model, not only the scientificity and effectiveness of the teaching plan are improved, but also the key content can be strengthened, thereby improving the learning effect of medical students. This method systematically evaluates and analyzes the learning performance of medical students, can timely discover learning bottlenecks and conduct targeted interventions, and significantly improve learning efficiency. Through a quantitative feedback mechanism, medical students can clearly understand their weaknesses and quickly adjust their learning strategies to achieve more efficient skill mastery. In summary, the physical examination and evaluation method based on simulated digital patients not only improves the quality and effectiveness of medical education, but also promotes personalized, data-driven teaching reforms, laying a solid foundation for the professional development of medical students.

[0072] Example 2

[0073] Please refer to Figure 1 , specifically: S1 specific steps include:

[0074] S11. First, according to the learning objectives, determine the type of disease of the 3D digital patient (such as heart disease, diabetes, cold, etc.), and use the 3D modeling software Maya to create and edit the 3D digital patient model;

[0075] S12. Combine Oculus Quest in VR technology and Microsoft HoloLens in AR technology to provide a virtual reality experience for the 3D digital patient model. The design of the 3D digital patient model will construct patients of corresponding age, gender and pathological characteristics according to the learning objectives, ensure that the model has enough details to display different physiological states, and simulate the physiological characteristics of the corresponding disease type for the 3D digital patient model, including but not limited to breathing and heartbeat; then use sound effects to simulate the sound of a stethoscope.

[0076] The specific steps of S2 include:

[0077] S21. Medical students conduct physical examination training according to the set disease types. During the physical examination training, the corresponding physical examination operations and conclusions of each medical student are monitored and recorded in real time, and the scoring indicators of the physical examination training are formulated in combination with the OSCE scoring standards. The scoring indicators include: communication skills, correctness of physical examination methods, observation ability, clinical reasoning ability and patient safety awareness;

[0078] S22. Use spreadsheet software (such as Excel or Google Sheets) to create a scoring table to record the examiner's item-by-item evaluation of the medical students to obtain a relevant physical examination scoring data set, which includes the pass and fail status of each medical student in communication skills, correctness of physical examination methods, observation ability, clinical reasoning ability and patient safety awareness, and convert the data information in the relevant physical examination scoring data set into Boolean data. The main purpose is to convert the original data into binary data (i.e., there are only two possible values, usually "0" and "1").

[0079] In this embodiment, by using the 3D modeling software Maya to create a detailed 3D digital patient model, medical students can experience a highly realistic clinical scenario. The 3D digital patient model can not only display different types of symptoms (such as heart disease, diabetes, and colds), but also design patients with specific ages, genders, and pathological characteristics according to learning objectives, further enhancing the sense of reality and engagement in learning, so that students can better understand the physiological characteristics of the disease during the physical examination. Combining Oculus Quest in VR technology and Microsoft HoloLens in AR technology, medical students can not only see the 3D model, but also obtain a multi-sensory learning experience through sound simulation. This multi-dimensional sensory stimulation can enhance students' memory and understanding ability, and help them react and respond more quickly in actual operations. During the physical examination training process, real-time monitoring and recording of medical students' operations and conclusions can provide students with immediate feedback. This timely evaluation mechanism allows students to quickly identify their shortcomings, so that they can make timely adjustments and improvements in practice, thereby improving learning efficiency. The physical examination training scoring indicators (including communication skills, correctness of physical examination methods, observation ability, clinical reasoning ability and patient safety awareness) developed in combination with the OSCE scoring standards provide a comprehensive and standardized evaluation framework for medical students. Such an evaluation system helps teachers to fully understand students' performance and provide data support for subsequent teaching. By using spreadsheet software to record the examiner's item-by-item evaluation of medical students, the relevant physical examination scoring data set formed not only provides a clear view of each student's performance, but also converts the data into Boolean data (0 and 1) for easy statistical analysis. This data processing method provides a scientific basis for teaching improvement. Teachers can adjust course content and teaching methods based on the results of data analysis to better meet students' learning needs.

[0080] Example 3

[0081] Please refer to Figure 1 , specifically: S3 specific steps include:

[0082] S31, setting the number of cluster centers Zs, wherein the number of cluster centers Zs is set as follows: Among them, m represents the number of scoring index items. There are five scoring indexes here, so m is 5;

[0083] S32, based on the relevant physical examination score data set after data conversion in S22, randomly assign each medical student in it to a number of cluster centers, and analyze and calculate the number of physical examination results of each medical student for each assessment item that are the same as the physical examination results of each assessment item in each cluster center, and record it as a matching value Ppz; wherein each cluster center is a combination of the physical examination results of each assessment item;

[0084] Among them, each assessment item corresponds to each scoring indicator;

[0085] S33, presetting a threshold, and comparing the matching value Ppz with the threshold, if the matching value Ppz exceeds the threshold, the corresponding medical student is included in the corresponding cluster center; if the matching value Ppz does not exceed the threshold, the corresponding medical student is not included in the corresponding cluster center;

[0086] It should be noted that the threshold here can be obtained in the following way: analyze and calculate the number of physical examination results of each medical student in each assessment item that are the same as the physical examination results of each assessment item in each cluster center, and record them as matching value Ppz, and use a statistical algorithm to obtain the average matching value and the standard deviation of the matching value for several groups of matching values ​​Ppz, and set the threshold based on the average matching value and the standard deviation of the matching value: threshold = average matching value + k* standard deviation of the matching value; wherein k is represented as a constant, usually taking a value of 1-3, corresponding to different confidence levels, and the specific value is set by the user (according to actual conditions); similarly, the frequency threshold P in step S4 is obtained in the same way.

[0087] S34. Based on the comparison result in S33, the medical students included in each cluster center are finally determined to construct several groups of medical students, wherein the number of the several groups of medical students is the same as the number of cluster centers Zs.

[0088] S3 also includes:

[0089] S35. According to several groups of medical students, the relative abundance Xfz of the corresponding medical student groups is analyzed and calculated, and is obtained specifically in the following manner:

[0090] ;

[0091] In the formula, It is expressed as the relative abundance of medical students in the i-th group of medical students, It is expressed as the number of medical students in the i-th group of medical students, Expressed as the number of medical students.

[0092] In this embodiment, by setting the number of cluster centers Zs, the method can determine the basic framework of the model analysis, and the specific number is based on the number of scoring index items. This method ensures that the selection of cluster centers is data-driven and reflects the dimensions of the actual evaluation. Medical students are randomly assigned to cluster centers, and the matching value Ppz of each assessment item is calculated, so that the similarity between the performance of medical students on each assessment item and the cluster center can be quantified. This matching process can provide in-depth insights into the learning outcomes of medical students. By setting a threshold and comparing the matching values, outstanding medical students can be classified into the corresponding cluster centers, thereby realizing dynamic grouping based on data. This mechanism can promote the formulation of personalized learning paths. Finally, the medical students included in each cluster center are determined, so as to construct several groups of medical students. This process helps to form a diverse learning group, which is convenient for targeted training and communication. Analyzing the relative abundance Xfz of each medical student group can help educators understand the performance differences of different groups and provide data support for further education reform and resource allocation. By grouping medical students, educators can formulate targeted courses and learning resources for each group to improve learning effects. During the teaching process, by continuously monitoring the matching values ​​and relative abundances, teachers can obtain timely feedback information and adjust their teaching strategies, thus forming a closed-loop feedback mechanism that helps to continuously optimize teaching results.

[0093] Example 4

[0094] Please refer to Figure 1 , specifically: S4 specific steps include:

[0095] S41. Calculate the physical examination results of the corresponding assessment items in each medical student group The single support , that is, the frequency of occurrence, is obtained in the following way:

[0096] ;

[0097] In the formula, It is expressed as the number of different physical examination results of each assessment item in the corresponding medical student group. Indicates the results of the physical examination for the corresponding assessment items The number of occurrences;

[0098] S42, pre-set the frequency threshold P, and calculate the physical examination results of the corresponding assessment items in each medical student group. The single support Compare with the frequency threshold P to screen out the relatively frequent physical examination results of the assessment items in each medical student group. The specific contents are:

[0099] If the physical examination results of the corresponding assessment items in each medical student group The single support When the frequency threshold P is exceeded, the physical examination results of the relatively frequent assessment items in the corresponding medical student group are screened out; if the physical examination results of the corresponding assessment items in each medical student group The single support When the frequency threshold value P is not exceeded, it is judged that the physical examination results of the corresponding assessment items do not have a relatively frequent situation;

[0100] S43, the physical examination results of the relatively frequent assessment items in the corresponding medical student groups screened out in S42 are counted and combined into two items to generate a frequent binomial set; the physical examination results of the corresponding assessment items in each medical student group are The single support Calculation of the transition to the two-term support , where the two-term support Calculation and single item support The same, specifically:

[0101] ;

[0102] In the formula, It is expressed as the number of occurrences of the physical examination results of the corresponding assessment items in the two groups; It is expressed as the number of different physical examination results for each assessment item within the two corresponding groups of medical students;

[0103] S44, repeat the steps in S42 and S43 until there is no combination item set whose corresponding support exceeds the frequency threshold P.

[0104] Specific example: Suppose we have a medical dataset that records the physical examination results of multiple medical students on simulated patients. Each medical student's performance on several examination items (such as normal or abnormal) is recorded, and our goal is to identify the correlation between these performances.

[0105] Step 1: Data preprocessing: Convert the data set into Boolean type, 1 means "normal" and 0 means "abnormal": as shown in Table 1:

[0106] Table 1

[0107] Medical Student ID Communication skills Observational skills Clinical reasoning skills 1 Normal (1) Normal (1) Exception (0) 2 Exception (0) Normal (1) Normal (1) 3 Exception (0) Exception (0) Normal (1) 4 Normal (1) Exception (0) Exception (0) 5 Normal (1) Normal (1) Normal (1) 6 Normal (1) Normal (1) Exception (0) 7 Exception (0) Exception (0) Exception (0)

[0108] Step 2: 1. Calculate the support of each item. Calculate the support of each inspection item (i.e. the frequency of occurrence):

[0109] Normal communication skills: 4 / 7 = 0.57; Abnormal communication skills: 3 / 7 = 0.43; Normal observation skills: 4 / 7 = 0.57; Abnormal observation skills: 3 / 7 = 0.43; Normal clinical reasoning skills: 3 / 7 = 0.43; Abnormal clinical reasoning skills: 4 / 7 = 0.57;

[0110] Step 3: 1. Generate frequent item sets:

[0111] First, a single frequent item set is a single frequent item set. Corresponding situation: We set the threshold to 0.4. According to the above calculation, all item sets are frequent. The specific single frequent item sets are: {Communication skills = normal} (support = 0.57); {Communication skills = abnormal} (support = 0.43); {Observation ability = normal} (support = 0.57); {Observation ability = abnormal} (support = 0.43); {Clinical reasoning ability = normal} (support = 0.43); {Clinical reasoning ability = abnormal} (support = 0.57);

[0112] 2. Generate frequent binomial sets:

[0113] Generate all binary combinations (such as {normal communication skills, normal observation ability}, {normal observation ability, abnormal clinical reasoning ability}, etc.), calculate their support, and then compare them with the threshold value of 0.4 to screen out binary combinations composed of relatively frequent physical examination results of assessment items;

[0114] 3. Generate three candidate item sets. Here, they will be screened in the same way as single frequent item sets and frequent binary item sets until there is no item set that meets the support threshold greater than 0.4.

[0115] This process is not only applicable to medical data, but can also be applied to data analysis in other fields to help identify patterns and relationships hidden behind the data.

[0116] In this embodiment, by analyzing the physical examination results of different assessment items in the medical student group, the weaknesses and strengths of each student can be identified, which makes teaching more personalized, and teachers can adjust the course content according to the needs of each student to improve learning efficiency. By analyzing the common weaknesses of the medical student group, educational institutions can allocate teaching resources more effectively, identify which assessment items need to be strengthened, and concentrate resources to improve them, thereby improving the overall teaching quality. By calculating the support, educators can clearly understand which assessment items are more frequent, so that they can design the assessment content more targeted, making the assessment more effective and reliable. By improving the ability of medical students in key assessment items, the competitiveness of graduates in the job market will be significantly enhanced, which not only improves students' self-confidence, but also increases their chances of success in the early stages of their careers. The analysis process can be used as a feedback mechanism to regularly evaluate the learning progress of medical students and adjust teaching methods in a timely manner to ensure the continuous improvement of teaching quality. Education managers can adjust curriculum settings and teaching strategies based on the analysis results to ensure that these decisions are more scientific and reasonable. Through the analysis of the medical student group, it is possible to encourage collaboration and communication among students and jointly overcome difficulties in learning. This interaction not only promotes the depth of learning, but also cultivates students' teamwork ability. In general, the analysis based on the improvement degree Tsd can improve the quality of medical education, help cultivate higher-quality medical workers, and then improve the level of medical services, which has a positive impact on society. Through continuous iteration and optimization of teaching content and methods, an adaptable teaching model can be formed, which can quickly adjust strategies according to changes in the medical student group to adapt to new educational needs and challenges. By implementing the above steps and methods, it can not only improve the learning effect and teaching quality of medical students, but also lay a solid foundation for the future career development of medical students.

[0117] Example 5

[0118] Please refer to Figure 1 Specifically, S4 also includes:

[0119] S45. Calculate the confidence level based on the contents in S1 to S4 , which is obtained by the following formula:

[0120] ;

[0121] In the formula, It is expressed as the support degree of the physical examination results of two groups of corresponding assessment items appearing simultaneously;

[0122] S46. Analyze the common weaknesses of medical students in several groups to obtain the degree of improvement between the assessment items in the corresponding medical student groups. , which is obtained by the following formula:

[0123] ;

[0124] In the formula, Represents the physical examination results of the corresponding assessment items in each medical student group The single item support.

[0125] in, It refers to the degree of improvement from the appearance of physical examination result A of the assessment item to the appearance of physical examination result B of the assessment item.

[0126] In this embodiment, by calculating the confidence, the teaching team can better identify which assessment items have a strong correlation, so that the teaching of relevant knowledge can be strengthened in a targeted manner in future teaching; according to the analysis of the confidence, personalized learning plans are formulated for different student groups to help students make breakthroughs in their weaknesses; by analyzing the common weaknesses of each medical student group, the teaching team can identify common problems and concentrate resources for targeted teaching. By improving and strengthening weaknesses in a targeted manner, the assessment performance of medical students is improved as a whole, and the clinical ability and comprehensive quality of graduates are improved. Showing the data analysis results to medical students makes them aware of their weaknesses and progress, and encourages them to be more proactive in learning. Through the above analysis framework and creative effects, the learning process of medical students will be more systematic and personalized, which can not only improve their clinical skills, but also enhance their learning motivation, laying a solid foundation for their future career.

[0127] Example 6

[0128] Please refer to Figure 1 Specifically: S5 includes the following steps:

[0129] S51, by the relative abundance Xfz and the degree of improvement Input into the teaching plan prediction model to construct the teaching plan priority Yxj, which is obtained by the following formula:

[0130] ;

[0131] In the formula, and are weight values, where 0 < ≤1,0< ≤1, and =1, Expressed as a correction constant;

[0132] S52. Based on the numerical value of the priority Yxj, medical students are given reinforcement assignments on the corresponding teaching contents in turn to ensure that more time and energy are allocated to those assessment items with higher priority when teaching resources are limited, with an emphasis on strengthening their weaknesses shown in physical examination training, thereby improving their overall clinical skills and physical examination capabilities.

[0133] In this embodiment, by quantifying the relative abundance and improvement, the teaching team can objectively evaluate the priority of each assessment item and further avoid the waste of resources caused by subjective judgment. According to the priority, investing more time and resources in high-priority assessment items will help improve the overall teaching effect and ensure that medical students make significant progress in clinical skills and physical examination ability. Focusing on high-priority items will help to strengthen training for medical students' weaknesses, so that they can perform better in clinical practice. According to the priority value, intensive homework is arranged for different medical students to ensure that high-priority items receive sufficient attention and practice.

[0134] Example 7

[0135] Please refer to Figure 2 ,Specifically: A physical examination and evaluation system based on simulated digital patients includes a creation module, a statistics module, a classification module, an impact module and a priority compensation module;

[0136] Create modules, use VR and AR technologies to create a 3D digital patient examination environment, and determine the type of symptoms of the 3D digital patient based on the learning objectives;

[0137] In the statistical module, medical students are trained to conduct physical examinations based on the types of symptoms of 3D digital patients, and combined with the OSCE scoring standards, the physical examination results of several medical students in each assessment item are statistically analyzed to obtain relevant physical examination scoring data sets;

[0138] The classification module analyzes the similarity of physical examination performance among several medical students based on the relevant physical examination score data set, so as to classify the medical students hierarchically, obtain several groups of medical students, and analyze and calculate the relative abundance Xfz;

[0139] Impact module, analyze and determine the common weaknesses of medical students in several groups of medical students to obtain the improvement degree between each assessment item in the corresponding medical student group , based on the lift The numerical value of is used to determine the teaching content that needs to be strengthened for medical students in the corresponding medical student group;

[0140] Priority compensation module and convolutional neural network technology are used to build a teaching plan prediction model. Input into the teaching plan prediction model to construct the priority Yxj of the teaching plan, and formulate the corresponding teaching plan based on the priority Yxj value.

[0141] The physical examination and evaluation storage medium based on the simulated digital patient stores a computer program, and the computer program stores computer instructions. When the instructions are executed by the processor, the physical examination and evaluation system based on the simulated digital patient is implemented.

[0142] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A physical examination and evaluation method based on a simulated digital patient, characterized in that: The following steps are included: S1. Use VR and AR technologies to create a 3D digital patient examination environment, and determine the disease type of the 3D digital patient based on the learning objectives; S2. Medical students conduct physical examination training based on the disease types of 3D digital patients, and combine the OSCE scoring standards to collect statistics on the physical examination results of several medical students in each assessment item to obtain relevant physical examination scoring data sets; S3. Based on the relevant physical examination score data set in S2, analyze the similarity of physical examination performance among several medical students to hierarchically classify the medical students, obtain several groups of medical students, and analyze and calculate the relative abundance Xfz; The specific steps of S3 include: S31, setting the number of cluster centers Zs, wherein the number of cluster centers Zs is set as follows: Among them, m represents the number of scoring indicator items; S32, based on the relevant physical examination score data set after data conversion in S22, randomly assign each medical student in it to a number of cluster centers, and analyze and calculate the number of physical examination results of each medical student for each assessment item that are the same as the physical examination results of each assessment item in each cluster center, and record it as a matching value Ppz; wherein each cluster center is a combination of the physical examination results of each assessment item; S4. Analyze and identify the common weaknesses of medical students in several groups of medical students to obtain the degree of improvement between each assessment item in the corresponding medical student groups. , based on the lift The numerical value of is used to determine the teaching content that needs to be strengthened for medical students in the corresponding medical student group; S4 includes: S45. Calculate confidence , which is obtained by the following formula: ； In the formula, It is expressed as the support degree of the physical examination results of two groups of corresponding assessment items appearing simultaneously; S46. Analyze the common weaknesses of medical students in several groups to obtain the degree of improvement between the assessment items in the corresponding medical student groups. , which is obtained by the following formula: ； In the formula, Represents the physical examination results of the corresponding assessment items in each medical student group The single item support of S5. Use convolutional neural network technology to build a teaching plan prediction model. By Input into the teaching plan prediction model to construct the priority Yxj of the teaching plan, and formulate the corresponding teaching plan based on the priority Yxj value; The specific steps of S5 include: S51, by the relative abundance Xfz and the degree of improvement Input into the teaching plan prediction model to construct the teaching plan priority Yxj, which is obtained by the following formula: ； In the formula, and are weight values, Expressed as a correction constant; S52. Based on the numerical value of the priority level Yxj, the medical students are given reinforced homework on the corresponding teaching contents.

2. The physical examination and evaluation method based on simulating a digital patient according to claim 1, characterized in that: The specific steps of S1 include: S11. First, according to the learning objectives, determine the disease type of the 3D digital patient, and use the 3D modeling software Maya to create and edit the 3D digital patient model; S12. Combine Oculus Quest in VR technology and Microsoft HoloLens in AR technology to provide a virtual reality experience for the 3D digital patient model. The design of the 3D digital patient model will construct patients of corresponding age, gender and pathological characteristics according to the learning objectives, simulate the physiological characteristics of the corresponding disease type for the 3D digital patient model, and then use sound effects to simulate the sound of a stethoscope.

3. The physical examination and evaluation method based on simulating a digital patient according to claim 1, characterized in that: The specific steps of S2 include: S21. Medical students conduct physical examination training according to the set disease types. During the physical examination training, the corresponding physical examination operations and conclusions of each medical student are monitored and recorded in real time, and the scoring indicators of the physical examination training are formulated in combination with the OSCE scoring standards. The scoring indicators include: communication skills, correctness of physical examination methods, observation ability, clinical reasoning ability and patient safety awareness; S22. Use spreadsheet software to create a scoring table to record the examiner's item-by-item evaluation of the medical students to obtain a relevant physical examination scoring data set, which includes the pass and fail status of each medical student in terms of communication skills, correctness of physical examination methods, observation ability, clinical reasoning ability, and patient safety awareness, and convert the data information in the relevant physical examination scoring data set into Boolean data.

4. The physical examination and evaluation method based on simulating a digital patient according to claim 3, characterized in that: The specific steps of S3 also include: S33, presetting a threshold, and comparing the matching value Ppz with the threshold, if the matching value Ppz exceeds the threshold, the corresponding medical student is included in the corresponding cluster center; if the matching value Ppz does not exceed the threshold, the corresponding medical student is not included in the corresponding cluster center; S34. Based on the comparison result in S33, the medical students included in each cluster center are finally determined to construct several groups of medical students, wherein the number of the several groups of medical students is the same as the number of cluster centers Zs.

5. The physical examination and evaluation method based on simulating a digital patient according to claim 3, characterized in that: S3 also includes: S35. According to several groups of medical students, the relative abundance Xfz of the corresponding medical student groups is analyzed and calculated, and is obtained specifically in the following manner: ； In the formula, It is expressed as the relative abundance of medical students in the i-th group of medical students, It is expressed as the number of medical students in the i-th group of medical students, Expressed as the number of medical students.

6. The physical examination and evaluation method based on simulating a digital patient according to claim 5, characterized in that: The specific steps of S4 include: S41. Calculate the physical examination results of the corresponding assessment items in each medical student group The single support , which can be obtained by: ； In the formula, It is expressed as the number of different physical examination results of each assessment item in the corresponding medical student group. Indicates the results of the physical examination for the corresponding assessment items The number of times it occurs; S42, pre-set the frequency threshold P, and calculate the physical examination results of the corresponding assessment items in each medical student group. The single support Compare with the frequency threshold P to screen out the relatively frequent physical examination results of the assessment items in each medical student group. The specific contents are: If the physical examination results of the corresponding assessment items in each medical student group The single support When the frequency threshold P is exceeded, the physical examination results of the relatively frequent assessment items in the corresponding medical student group are screened out; if the physical examination results of the corresponding assessment items in each medical student group The single support When the frequency threshold value P is not exceeded, it is judged that the physical examination results of the corresponding assessment items do not have a relatively frequent situation; S43, the physical examination results of the relatively frequent assessment items in the corresponding medical student groups screened out in S42 are counted and combined into two items to generate a frequent binomial set; the physical examination results of the corresponding assessment items in each medical student group are The single support Calculation of the transition to the two-term support , where the two-term support Calculation and single item support The same, specifically: ； In the formula, It is expressed as the number of occurrences of the physical examination results of the corresponding assessment items in the two groups; It is expressed as the number of different physical examination results for each assessment item within the two corresponding groups of medical students; S44, repeat the steps in S42 and S43 until there is no combination item set whose corresponding support exceeds the frequency threshold P.

7. A physical examination and evaluation system based on a simulated digital patient, used to implement the physical examination and evaluation method based on a simulated digital patient as described in any one of claims 1 to 6, characterized in that: It includes creation module, statistics module, classification module, impact module and priority compensation module; Create modules, use VR and AR technologies to create a 3D digital patient examination environment, and determine the type of symptoms of the 3D digital patient based on the learning objectives; In the statistical module, medical students are trained to conduct physical examinations based on the types of symptoms of 3D digital patients, and combined with the OSCE scoring standards, the physical examination results of several medical students in each assessment item are statistically analyzed to obtain relevant physical examination scoring data sets; The classification module analyzes the similarity of physical examination performance among several medical students based on the relevant physical examination score data set, so as to classify the medical students hierarchically, obtain several groups of medical students, and analyze and calculate the relative abundance Xfz; Impact module, analyze and determine the common weaknesses of medical students in several groups of medical students to obtain the improvement degree between each assessment item in the corresponding medical student group , based on the lift The numerical value of is used to determine the teaching content that needs to be strengthened for medical students in the corresponding medical student group; Priority compensation module and convolutional neural network technology are used to build a teaching plan prediction model. Input into the teaching plan prediction model to construct the priority Yxj of the teaching plan, and formulate the corresponding teaching plan based on the priority Yxj value.

8. A storage medium for physical examination and evaluation based on a simulated digital patient, on which a computer program is stored, characterized in that: The computer program stores computer instructions, and when the instructions are executed by the processor, the physical examination and evaluation system based on simulating a digital patient as described in claim 7 is implemented.

Images