An intelligent physical examination recommendation system and method based on health questionnaire

Abstract

The present invention discloses an intelligent physical examination recommendation system and method based on a health questionnaire, and relates to the field of medical informatics; the present invention first collects the first answer information made by the first user for the health questionnaire, and judges whether the health questionnaire is a credible answer sheet according to the answer time duration set of each question corresponding to the questionnaire question set; then, in response to the health questionnaire being a credible answer sheet, the normalized first answer time sequence is compared with the standard answer time sequence, and the credible weights of the answer options of each question corresponding to the first answer time sequence are obtained; finally, according to the credible weights of each question of the health questionnaire, a first credible answer sheet is generated, and the first credible answer sheet is input into a physical examination recommendation model, and a customized physical examination recommendation list is generated. In the present invention, considering that there may be a situation where the questionnaire survey is filled in arbitrarily, by identifying the credible questionnaire and then recommending the physical examination, the accuracy of the physical examination recommendation is effectively improved, and the risk of missing physical examination items is reduced.

Description

Technical Field

[0001] The present invention relates to the field of medical informatics, and in particular to an intelligent physical examination recommendation system and method based on a health questionnaire. Background Art

[0002] As people's quality of life gradually improves, the public's demand for health examinations is increasing, and health examinations are gradually becoming the mainstream business of medical examination centers in addition to outpatient examinations. There are many types of physical examination items. In order to generate revenue, physical examination companies are likely to arrange many physical examination items; and they will not take into account the differences among individuals undergoing physical examinations. Physical examinations cannot be conducted based on the historical health data of the individual being examined, so physical examination items are often unscientific and cannot be conducted based on the individual's actual situation.

[0003] In the prior art, on the one hand, physical examination items cannot be compiled according to the individual and the location, and on the other hand, for physical examinations based on questionnaires, some people tend to underestimate the questionnaire and answer hastily, making the questionnaire itself lose the meaning of recommending physical examination items. Summary of the invention

[0004] In view of the shortcomings of the prior art, the technical problem to be solved by the present invention is to provide an intelligent physical examination recommendation system and method based on a health questionnaire, which aims to identify credible questionnaires through the length of time for answering the questionnaire and then make physical examination recommendations, thereby effectively improving the accuracy of physical examination recommendations and reducing the risk of missing physical examination items; and, through the length of time for answering each question in the questionnaire, determining the credible weight, which is then input into the model to generate physical examination recommendations, thereby effectively avoiding the problem of poor questionnaire quality caused by users randomly skipping and filling out certain questions rather than subjectively, and effectively improving the accuracy of physical examination item recommendations.

[0005] In a first aspect of the present invention, a method for intelligent physical examination recommendation based on a health questionnaire is provided, the method comprising:

[0006] Step S1: collecting the first answer information of the first user for the health questionnaire; the health questionnaire includes a set of questionnaire questions The first answer information includes: Answer key to each question , and the time taken to answer each question ; , is the total number of questions in the health questionnaire; the health questionnaire is a multiple-choice question or a true-or-false question; the questionnaire question set 、The answer set 、The time taken to answer the questions is One to one correspondence;

[0007] Step S2: according to the questionnaire question set The time taken to answer each question is a set of , judging whether the health questionnaire is a credible answer sheet; in response to the health questionnaire not being a credible answer sheet, outputting a routine physical examination recommendation list; in response to the health questionnaire being a credible answer sheet, executing step S3;

[0008] Step S3: Normalize the time set of answering the health questionnaire , get the normalized first answer time sequence of each question , the first answer time sequence Standard answer time sequence Compare and obtain the time sequence of the first answer The credibility weight of the answer options for each question ; Among them, the standard answer time sequence It is obtained by normalizing the answering time of each question based on historical data;

[0009] Step S4: according to the trust weight of each question in the health questionnaire , generate the first credible answer sheet ; Input the first credible answer sheet into the physical examination recommendation model and generate a customized physical examination recommendation list.

[0010] In this technical solution, considering that questionnaires may be filled out randomly, by identifying credible questionnaires and then recommending physical examinations, the accuracy of physical examination recommendations is effectively improved and the risk of missing physical examination items is reduced. In addition, in this technical solution, the credible weight is determined by the time it takes to answer each question in the questionnaire, and then input into the model to generate physical examination recommendations, effectively avoiding the problem of poor questionnaire quality caused by users randomly skipping some questions rather than subjectively filling them out, and effectively improving the accuracy of physical examination item recommendations.

[0011] In a specific implementation, in step S2, the step of determining whether the health questionnaire is a reliable answer sheet includes:

[0012] Get the time set for answering the question The duration of each question , in response to If the confidence interval is smaller than the historical answer time of the question, the question is judged as an unreliable question; when the number of unreliable questions exceeds the preset number, the health questionnaire is judged not to be a reliable answer sheet, otherwise, the health questionnaire is a reliable answer sheet.

[0013] In this technical solution, by making a credibility judgment on each question separately, when the number of untrustworthy questions exceeds a preset number, the health questionnaire is judged not to be the credible answer sheet, thereby realizing the screening of credible answer sheets.

[0014] In a specific embodiment, the construction of the physical examination recommendation model includes:

[0015] Collect the data set of answers from trusted volunteers to the health questionnaire And the corresponding lesion identification set obtained after full body examination , forming a training data set ;

[0016] The training data set Conduct training to obtain the questionnaire / lesion relationship model ; Wherein, the physical examination recommendation model includes the questionnaire / lesion relationship model, the lesion and physical examination item mapping unit , Physical examination recommendation list generation unit ;

[0017] Build the lesion and physical examination item mapping unit ; The lesion and physical examination item mapping unit Used to map the lesions output by the questionnaire / lesion relationship model into physical examination items;

[0018] Build the physical examination recommendation list generation unit ; The physical examination recommendation list generation unit Used to generate the customized physical examination recommendation list after merging the multiple physical examination items corresponding to the multiple lesions.

[0019] In this technical solution, the physical examination recommendation model is constructed by collecting the answer data and lesion conditions of trusted volunteers, so as to output the physical examination recommendation items after the subsequent physical examination subjects input the questionnaire.

[0020] In a specific implementation, the trust weight is 0.6-1; wherein, the first answer time sequence The corresponding values ​​of each question in the standard answer time sequence The numerical ratios corresponding to each question The smaller it is, the smaller the credibility weight corresponding to the question is.

[0021] In a specific implementation manner, the first credible answer sheet Can be expanded to:

[0022] ;

[0023] The original answer set Corresponding to the answers filled in, the for The corresponding trust weight; , For the The alternative answers to the questions and their corresponding probability weights, ; For the The number of possible answers for each question, .

[0024] In this technical solution, the first credible answer sheet is expanded, and the probability weight assignment constraints of other options of the first credible answer sheet that are not user answers are restricted to optimize the first credible answer sheet.

[0025] Furthermore, in one embodiment, the probability weight The weighted probability is assigned according to the distribution of answers corresponding to the population corresponding to the answerer. In this technical solution, the probability weight assignment of other options of the first credible answer sheet is optimized.

[0026] In a second aspect of the present invention, an intelligent physical examination recommendation system based on a health questionnaire is provided, the system comprising:

[0027] The answer collection module is used to collect the first answer information made by the first user for the health questionnaire; the health questionnaire includes a questionnaire question set The first answer information includes: a set of answers associated with each question , and the time taken to answer each question ; , is the total number of questions in the health questionnaire; the health questionnaire is a multiple-choice question or a true-or-false question; the questionnaire question set 、The answer set 、The time taken to answer the questions is One to one correspondence;

[0028] A trustworthy judgment module is used to determine the The time taken to answer each question is a set of , judging whether the health questionnaire is a credible answer sheet; in response to the health questionnaire not being a credible answer sheet, outputting a routine physical examination recommendation list; in response to the health questionnaire being a credible answer sheet, sending the health questionnaire to a credible weight solving module;

[0029] The trust weight solving module is used to normalize the answering time set of the health questionnaire , get the normalized first answer time sequence of each question , the first answer time sequence Standard answer time sequence Compare and obtain the time sequence of the first answer The credibility weight of the answer options for each question ; Among them, the standard answer time sequence It is obtained by normalizing the answering time of each question based on historical data;

[0030] Customized physical examination recommendation module, used to determine the trustworthiness weight of each question in the health questionnaire , generate the first credible answer sheet ; Input the first credible answer sheet into the physical examination recommendation model and generate a customized physical examination recommendation list.

[0031] In a specific implementation, the trust judgment module is configured as follows:

[0032] Get the time set for answering the question The duration of each question , in response to If the confidence interval is smaller than the historical answer time of the question, then it is judged that the health questionnaire is not the credible answer sheet; otherwise, the health questionnaire is a credible answer sheet.

[0033] In a specific embodiment, the construction of the physical examination recommendation model includes:

[0034] Collect the data set of answers from trusted volunteers to the health questionnaire And the corresponding lesion identification set obtained after full body examination , forming a training data set ;

[0035] The training data set Conduct training to obtain the questionnaire / lesion relationship model ; Wherein, the physical examination recommendation model includes the questionnaire / lesion relationship model, the lesion and physical examination item mapping unit , Physical examination recommendation list generation unit ;

[0036] Build the lesion and physical examination item mapping unit ; The lesion and physical examination item mapping unit Used to map the lesions output by the questionnaire / lesion relationship model into physical examination items;

[0037] Build the physical examination recommendation list generation unit ; The physical examination recommendation list generation unit Used to generate the customized physical examination recommendation list after merging the multiple physical examination items corresponding to the multiple lesions.

[0038] In a specific implementation, the trust weight is 0.6-1; wherein, the first answer time sequence The corresponding values ​​of each question in the standard answer time sequence The numerical ratios corresponding to each question The smaller it is, the smaller the credibility weight corresponding to the question is.

[0039] In a specific implementation manner, the first credible answer sheet Can be expanded to:

[0040] ;

[0041] The original answer set Corresponding to the answers filled in, the for The corresponding trust weight; , For the The alternative answers to the questions and their corresponding probability weights, ; For the The number of possible answers for each question, .

[0042] In a specific embodiment, the probability weight The weighted probability is assigned according to the distribution of answers corresponding to the population corresponding to the respondent.

[0043] Beneficial effects of the present invention: 1) Taking into account the possibility of random filling in of questionnaire surveys, the present invention identifies credible questionnaires and then recommends physical examinations, effectively improving the accuracy of physical examination recommendations and reducing the risk of missed physical examination items. 2) The present invention determines the credible weight by the length of time for answering each question in the questionnaire, and then inputs it into the model to generate physical examination recommendations, effectively avoiding the problem of poor questionnaire quality caused by users randomly skipping some questions rather than subjectively filling them in, and effectively improving the accuracy of physical examination item recommendations. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a flowchart of an intelligent physical examination recommendation method based on a health questionnaire in a specific implementation of the present invention. DETAILED DESCRIPTION

[0045] The embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0046] like Figure 1 As shown, an embodiment of the present invention provides an intelligent physical examination recommendation method based on a health questionnaire, the method comprising:

[0047] Step S1: collecting the first answer information of the first user for the health questionnaire; the health questionnaire includes a set of questionnaire questions The first answer information includes: Answer key to each question , and the time taken to answer each question ; , is the total number of questions in the health questionnaire; the health questionnaire is a multiple-choice question or a true-or-false question; the questionnaire question set 、The answer set 、The time taken to answer the questions is One to one correspondence;

[0048] Step S2: according to the questionnaire question set The time taken to answer each question is a set of , judging whether the health questionnaire is a credible answer sheet; in response to the health questionnaire not being a credible answer sheet, outputting a routine physical examination recommendation list; in response to the health questionnaire being a credible answer sheet, executing step S3;

[0049] Step S3: Normalize the time set of answering the health questionnaire , get the normalized first answer time sequence of each question , the first answer time sequence Standard answer time sequence Compare and obtain the time sequence of the first answer The credibility weight of the answer options for each question ; Among them, the standard answer time sequence It is obtained by normalizing the answering time of each question based on historical data;

[0050] Step S4: according to the trust weight of each question in the health questionnaire , generate the first credible answer sheet ; Input the first credible answer sheet into the physical examination recommendation model and generate a customized physical examination recommendation list.

[0051] Typically, the health questionnaire includes age, gender, occupation, past medical history, family medical history, stress, lifestyle surveys such as smoking, drinking, exercise, eating habits, etc., as well as height, weight, waist circumference, blood pressure measurements, and understanding the budget of the examinee, and forming a physical examination package for the examinee based on the budget. Typically, lesion identification includes various common diseases and lesions.

[0052] Furthermore, in this embodiment, in step S2, the step of determining whether the health questionnaire is a reliable answer sheet includes:

[0053] Get the time set for answering the question The duration of each question , in response to If the confidence interval is smaller than the historical answer time of the question, the question is judged as an unreliable question; when the number of unreliable questions exceeds the preset number, the health questionnaire is judged not to be a reliable answer sheet, otherwise, the health questionnaire is a reliable answer sheet.

[0054] In this embodiment, the construction of the physical examination recommendation model includes:

[0055] Collect the data set of answers from trusted volunteers to the health questionnaire And the corresponding lesion identification set obtained after full body examination , forming a training data set ;

[0056] The training data set Conduct training to obtain the questionnaire / lesion relationship model ; Wherein, the physical examination recommendation model includes the questionnaire / lesion relationship model, the lesion and physical examination item mapping unit , Physical examination recommendation list generation unit ;

[0057] Build the lesion and physical examination item mapping unit ; The lesion and physical examination item mapping unit Used to map the lesions output by the questionnaire / lesion relationship model into physical examination items;

[0058] Build the physical examination recommendation list generation unit ; The physical examination recommendation list generation unit Used to generate the customized physical examination recommendation list after merging the multiple physical examination items corresponding to the multiple lesions.

[0059] In this embodiment, the trust weight is 0.6-1; wherein, the first answer time sequence The corresponding values ​​of each question in the standard answer time sequence The numerical ratios corresponding to each question The smaller it is, the smaller the credibility weight corresponding to the question is.

[0060] Here, the trust weight The solution is given as an example. This example is only for illustration and can be modified in actual use. It only needs to meet the following requirements: The corresponding values ​​of each question in the standard answer time sequence The numerical ratios corresponding to each question The smaller it is, the smaller the credibility weight corresponding to the question is.

[0061] Typical:

[0062]

[0063] For the recommendation model of this embodiment, the first credible answer sheet can be input without supplementing other options. , generate recommended items for physical examination. In addition, you can further supplement the credible answer sheet ;

[0064] That is, in one usage scenario, the first credible answer sheet Can be expanded to:

[0065] ;

[0066] The original answer set Corresponding to the answers filled in, the for The corresponding trust weight; , For the The alternative answers to the questions and their corresponding probability weights, ; For the The number of possible answers for each question, .

[0067] The above probability weights can be evenly distributed by other options; in addition, in an optional scenario, the first credible answer sheet is expanded to When the probability weight The weighted probability is assigned according to the distribution of answers corresponding to the population corresponding to the respondent.

[0068] In this embodiment, the input of the physical examination recommendation model can be directly loaded on the input answer options according to the above-mentioned weight probability and credible weight; in another form, the first credible answer sheet can be installed with the weight probability and credible weight of each answer option of each question to expand the number of first credible answer sheets, that is, in order to meet the weight probability / credible weight distribution, a first credible answer sheet is converted into several first credible answer sheets and input into the physical examination recommendation model.

[0069] In a second embodiment of the present invention, an intelligent physical examination recommendation system based on a health questionnaire is provided, the system comprising:

[0070] The answer collection module is used to collect the first answer information made by the first user for the health questionnaire; the health questionnaire includes a questionnaire question set The first answer information includes: a set of answers associated with each question , and the time taken to answer each question ; , is the total number of questions in the health questionnaire; the health questionnaire is a multiple-choice question or a true-or-false question; the questionnaire question set 、The answer set 、The time taken to answer the questions is One to one correspondence;

[0071] A trustworthy judgment module is used to determine the The time taken to answer each question is a set of , judging whether the health questionnaire is a credible answer sheet; in response to the health questionnaire not being a credible answer sheet, outputting a routine physical examination recommendation list; in response to the health questionnaire being a credible answer sheet, sending the health questionnaire to a credible weight solving module;

[0072] The trust weight solving module is used to normalize the answering time set of the health questionnaire , get the normalized first answer time sequence of each question , the first answer time sequence Standard answer time sequence Compare and obtain the time sequence of the first answer The credibility weight of the answer options for each question ; Among them, the standard answer time sequence It is obtained by normalizing the answering time of each question based on historical data;

[0073] Customized physical examination recommendation module, used to determine the trustworthiness weight of each question in the health questionnaire , generate the first credible answer sheet ; Input the first credible answer sheet into the physical examination recommendation model and generate a customized physical examination recommendation list.

[0074] In this embodiment, the trust judgment module is configured as follows:

[0075] Get the time set for answering the question The duration of each question , in response to If the confidence interval is smaller than the historical answer time of the question, then it is judged that the health questionnaire is not the credible answer sheet; otherwise, the health questionnaire is a credible answer sheet.

[0076] In this embodiment, the construction of the physical examination recommendation model includes:

[0077] Collect the data set of answers from trusted volunteers to the health questionnaire And the corresponding lesion identification set obtained after full body examination , forming a training data set ;

[0078] The training data set Conduct training to obtain the questionnaire / lesion relationship model ; Wherein, the physical examination recommendation model includes the questionnaire / lesion relationship model, the lesion and physical examination item mapping unit , Physical examination recommendation list generation unit ;

[0079] Build the lesion and physical examination item mapping unit ; The lesion and physical examination item mapping unit Used to map the lesions output by the questionnaire / lesion relationship model into physical examination items;

[0080] Build the physical examination recommendation list generation unit ; The physical examination recommendation list generation unit Used to generate the customized physical examination recommendation list after merging the multiple physical examination items corresponding to the multiple lesions.

[0081] In this embodiment, the trust weight is 0.6-1; wherein, the first answer time sequence The corresponding values ​​of each question in the standard answer time sequence The numerical ratios corresponding to each question The smaller it is, the smaller the credibility weight corresponding to the question is.

[0082] In this embodiment, the first credible answer sheet Can be expanded to:

[0083] ;

[0084] The original answer set Corresponding to the answers filled in, the for The corresponding trust weight; , For the The alternative answers to the questions and their corresponding probability weights, ; For the The number of possible answers for each question, .

[0085] In this embodiment, the probability weight The weighted probability is assigned according to the distribution of answers corresponding to the population corresponding to the respondent.

[0086] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. An intelligent physical examination recommendation method based on health questionnaire, characterized in that: The method comprises: Step S1: collecting the first answer information of the first user for the health questionnaire; the health questionnaire includes a set of questionnaire questions The first answer information includes: Answer key to each question , and the time taken to answer each question ; , is the total number of questions in the health questionnaire; the health questionnaire is a multiple-choice question or a true-or-false question; the questionnaire question set 、The answer set 、The time taken to answer the questions is One to one correspondence; Step S2: according to the questionnaire question set The time taken to answer each question is a set of , judging whether the health questionnaire is a credible answer sheet; in response to the health questionnaire not being a credible answer sheet, outputting a routine physical examination recommendation list; in response to the health questionnaire being a credible answer sheet, executing step S3; Step S3: Normalize the time set of answering the health questionnaire , get the normalized first answer time sequence of each question , the first answer time sequence Standard answer time sequence Compare and obtain the time sequence of the first answer The credibility weight of the answer options for each question ; Among them, the standard answer time sequence It is obtained by normalizing the answering time of each question based on historical data; Step S4: according to the trust weight of each question in the health questionnaire , generate the first credible answer sheet ; Input the first credible answer sheet into the physical examination recommendation model, and generate a customized physical examination recommendation list; The first credible answer sheet Expands to: ； The original answer set Corresponding to the answers filled in, the for The corresponding trust weight; , For the The alternative answers to the questions and their corresponding probability weights, ; For the The number of possible answers for each question, .

2. The intelligent physical examination recommendation method based on health questionnaire as claimed in claim 1, characterized in that: In step S2, the step of determining whether the health questionnaire is a reliable answer sheet includes: Get the time set for answering the question The duration of each question , in response to If the confidence interval is smaller than the historical answer time of the question, the question is judged as an unreliable question; when the number of unreliable questions exceeds the preset number, the health questionnaire is judged not to be a reliable answer sheet, otherwise, the health questionnaire is a reliable answer sheet.

3. The intelligent physical examination recommendation method based on health questionnaire as claimed in claim 1, characterized in that: The construction of the physical examination recommendation model includes: Collect the data set of answers from trusted volunteers to the health questionnaire And the corresponding lesion identification set obtained after full body examination , forming a training data set ; The training data set Conduct training to obtain the questionnaire / lesion relationship model ; Wherein, the physical examination recommendation model includes the questionnaire / lesion relationship model, the lesion and physical examination item mapping unit , Physical examination recommendation list generation unit ; Build the lesion and physical examination item mapping unit ; The lesion and physical examination item mapping unit Used to map the lesions output by the questionnaire / lesion relationship model into physical examination items; Build the physical examination recommendation list generation unit ; The physical examination recommendation list generation unit Used to generate the customized physical examination recommendation list after merging the multiple physical examination items corresponding to the multiple lesions.

4. The intelligent physical examination recommendation method based on health questionnaire as claimed in claim 1, characterized in that: The trust weight is 0.6-1; wherein, the first answer time sequence The corresponding values ​​of each question in the standard answer time sequence The numerical ratios corresponding to each question The smaller it is, the smaller the credibility weight corresponding to the question is.

5. The intelligent physical examination recommendation method based on health questionnaire as claimed in claim 4, characterized in that: The probability weight The weighted probability is assigned according to the distribution of answers corresponding to the population corresponding to the respondent.

6. An intelligent physical examination recommendation system based on health questionnaire, characterized in that: The system comprises: The answer collection module is used to collect the first answer information made by the first user for the health questionnaire; the health questionnaire includes a questionnaire question set The first answer information includes: a set of answers associated with each question , and the time taken to answer each question ; , is the total number of questions in the health questionnaire; the health questionnaire is a multiple-choice question or a true-or-false question; the questionnaire question set 、The answer set 、The time taken to answer the questions is One to one correspondence; A trustworthy judgment module is used to determine the The time taken to answer each question is a set of , judging whether the health questionnaire is a credible answer sheet; in response to the health questionnaire not being a credible answer sheet, outputting a routine physical examination recommendation list; in response to the health questionnaire being a credible answer sheet, sending the health questionnaire to a credible weight solving module; The trust weight solving module is used to normalize the answering time set of the health questionnaire , get the normalized first answer time sequence of each question , the first answer time sequence Standard answer time sequence Compare and obtain the time sequence of the first answer The credibility weight of the answer options for each question ; Among them, the standard answer time sequence It is obtained by normalizing the answering time of each question based on historical data; Customized physical examination recommendation module, used to determine the trustworthiness weight of each question in the health questionnaire , generate the first credible answer sheet ; Input the first credible answer sheet into the physical examination recommendation model, and generate a customized physical examination recommendation list; The first credible answer sheet Can be expanded to: ； The original answer set Corresponding to the answers filled in, the for The corresponding trust weight; , For the The alternative answers to the questions and their corresponding probability weights, ; For the The number of possible answers for each question, .

7. The intelligent physical examination recommendation system based on health questionnaire as claimed in claim 6, characterized in that: The trust judgment module is configured as follows: Get the time set for answering the question The duration of each question , in response to If the confidence interval is smaller than the historical answer time of the question, then it is judged that the health questionnaire is not the credible answer sheet; otherwise, the health questionnaire is a credible answer sheet.

8. The intelligent physical examination recommendation system based on health questionnaire as claimed in claim 6, characterized in that: The construction of the physical examination recommendation model includes: Collect the data set of answers from trusted volunteers to the health questionnaire And the corresponding lesion identification set obtained after full body examination , forming a training data set ; The training data set Conduct training to obtain the questionnaire / lesion relationship model ; Wherein, the physical examination recommendation model includes the questionnaire / lesion relationship model, the lesion and physical examination item mapping unit , Physical examination recommendation list generation unit ; Build the lesion and physical examination item mapping unit ; The lesion and physical examination item mapping unit Used to map the lesions output by the questionnaire / lesion relationship model into physical examination items; Build the physical examination recommendation list generation unit ; The physical examination recommendation list generation unit Used to generate the customized physical examination recommendation list after merging the multiple physical examination items corresponding to the multiple lesions.

Images