CN-121999958-A - Personalized optimization method and system for physical examination scheme of artificial intelligent main sample

Abstract

The embodiment of the application provides a personalized optimization method and a personalized optimization system for a physical examination scheme of an artificial intelligent main specimen, which belong to the technical field of artificial intelligence, and are characterized in that firstly basic health information and personalized demand description of a physical examination person are acquired, analyzed and extracted to obtain a health and demand interpretation result; the method comprises the steps of developing dynamic association modeling of health requirements based on health and requirement interpretation results to generate an association weight matrix, screening adaptation items according to the association weight matrix and physical examination item attribute information to obtain a dynamic adaptation set, carrying out collaborative combination optimization on the items in the dynamic adaptation set, generating a personalized physical examination scheme first draft by combining health state change trend and resource use condition, feeding the personalized physical examination scheme first draft back to a physical examination person to collect feedback and evaluate feasibility, and integrating information to adjust the first draft to obtain a final scheme. The application can accurately formulate a personalized physical examination scheme and improve physical examination resource utilization rate and physical examination satisfaction.

Inventors

• ZHAO ZHIJIAN
• HUANG LIUHAO
• XU XIANQIANG
• KONG ZHENFENG

Assignees

• 广东康软科技股份有限公司

Dates

Publication Date

20260508

Application Date

20260128

Claims (10)

1. 1. A personalized optimization method for a physical examination scheme of an artificial intelligent master sample, which is characterized by comprising the following steps: Basic health information of a physical examination person and personalized demand description proposed by the physical examination person are obtained, the basic health information and the personalized demand description are input into an artificial intelligent main examination system, semantic analysis and feature extraction are carried out on the basic health information and the personalized demand description, and a health information interpretation result and a demand interpretation result are obtained; Developing dynamic association modeling of health requirements based on the health information interpretation results and the requirement interpretation results, mining potential association relations between the health information and personalized requirements, and generating an association weight matrix; Performing scene adaptation screening on physical examination items in a preset physical examination item library according to the association weight matrix and physical examination item attribute information in the preset physical examination item library, and determining adapted physical examination items by combining the health state of a physical examination person and real-time resource conditions to obtain a dynamic adaptation physical examination item set; Performing collaborative combination optimization on physical examination items in the dynamic adaptation physical examination item set, and planning physical examination item execution sequences and resource allocation modes by combining the current health state time sequence change trend and real-time physical examination resource use condition of physical examination persons to generate personalized physical examination scheme first draft comprising physical examination item execution sequences, resource allocation schemes and expected execution flows; And feeding back the personalized physical examination proposal first draft to the physical examination person, collecting real-time feedback information of the physical examination person, simultaneously carrying out feasibility assessment on the personalized physical examination proposal first draft, and integrating the real-time feedback information and the feasibility assessment result to adjust the personalized physical examination proposal first draft so as to obtain a final personalized physical examination proposal.
2. 2. The personalized optimization method for physical examination solutions of an artificial intelligent master sample according to claim 1, wherein the developing dynamic association modeling of health requirements based on the health information interpretation results and the requirement interpretation results, mining potential association relations between health information and personalized requirements, and generating an association weight matrix comprises: Extracting health characteristic items in the health information interpretation result and demand characteristic items in the demand interpretation result, wherein the health characteristic items comprise physical index characteristics, medical history characteristics and life habit characteristics of the physical examination person, and the demand characteristic items comprise description characteristics of physical examination objects of the physical examination person, physical examination period characteristics of the physical examination person and preference characteristics of physical examination items of the physical examination person, so as to obtain a health characteristic item set and a demand characteristic item set; Invoking a characteristic association rule library stored in an artificial intelligent main inspection system, and screening association rules related to the healthy characteristic item set and the required characteristic item set from the characteristic association rule library to form a characteristic association rule set; Constructing a feature association network between the health feature and the demand feature based on the feature association rule set, taking the health feature items in the health feature item set and the demand feature items in the demand feature item set as network nodes, and establishing association relations between the related nodes as network edges according to the feature association rule to form a feature association network structure; Carrying out association strength calibration on each association relation side in the feature association network structure, extracting the support degree of feature association rules corresponding to each association relation side, the co-occurrence proportion of health feature items and demand feature items in historical physical examination scheme data, and the influence coefficient of the association relation on physical examination item selection, wherein the co-occurrence proportion is obtained by calculating the proportion of the number of the historical physical examination schemes in which the health feature items and the demand feature items co-occur relative to the total number of the related historical physical examination schemes, and carrying out integrated calculation on the support degree, the co-occurrence proportion and the influence coefficient to obtain the association strength value corresponding to each association relation side; according to the feature association network structure and the association strength value of each association relation edge, determining association distribution of health feature items and demand feature items, arranging the health feature items in a health feature item set according to a fixed sequence to be used as matrix rows, arranging the demand feature items in a demand feature item set according to the fixed sequence to be used as matrix columns, filling the association strength values between the corresponding health feature items and the demand feature items into corresponding positions of the matrix, and generating an association weight matrix; And calling historical physical examination scheme data to verify the generated association weight matrix, comparing the association relation reflected by the association weight matrix with the actual condition of health requirement association in the historical physical examination scheme, calculating the difference value of the association strength reflected by the association weight matrix and the actual association strength, readjusting the association strength calibration parameter based on the difference value if the difference value exceeds a preset difference threshold value, and repeating the association strength calculation and association weight matrix generation steps until the difference value is in the preset difference threshold value range, wherein the association weight matrix accords with the association rule reflected by the historical data.
3. 3. The personalized optimization method for physical examination schemes of artificial intelligent main examination according to claim 1, wherein the step of performing scene adaptation screening on physical examination items in a preset physical examination item library according to the association weight matrix and physical examination item attribute information in the preset physical examination item library, and determining adapted physical examination items by combining physical examination person health status and real-time resource conditions to obtain a dynamic adaptation physical examination item set comprises the following steps: Invoking a preset physical examination item library, and extracting physical examination item attribute information of each physical examination item in the preset physical examination item library, wherein the physical examination item attribute information comprises a health characteristic adaptation item corresponding to the physical examination item, a demand characteristic adaptation item corresponding to the physical examination item, a resource type required by physical examination item execution, physical examination item execution duration and compatibility description of the physical examination item and other physical examination items, so as to obtain a physical examination item attribute information set; Matching the health feature items in the association weight matrix with the health feature adaptation items of each individual test item, counting the matching quantity of the health feature adaptation items of each individual test item and the health feature items in the association weight matrix, simultaneously matching the demand feature items in the association weight matrix with the demand feature adaptation items of each individual test item, and counting the matching quantity of the demand feature adaptation items of each individual test item and the demand feature items in the association weight matrix; Calculating the adaptation score of each body inspection item by combining the association strength values of the corresponding health feature items and the demand feature items in the association weight matrix based on the health feature matching quantity and the demand feature matching quantity of each body inspection item, summing the association strength values corresponding to each matched health feature item, summing the association strength values corresponding to each matched demand feature item, and summing the two summation results according to a preset integration proportion to obtain the adaptation score of each body inspection item; screening physical examination items with the fit scores reaching preset fit scoring criteria according to preset fit scoring criteria, wherein the preset fit scoring criteria are set based on scoring data of historical fit successful cases, and a preliminary screening physical examination item set is formed; Carrying out health status adaptation judgment on each physical examination item in the preliminary screening physical examination item set, extracting the requirement of each physical examination item on the physical state of a physical examination person, combining the current health status information of the physical examination person, comparing the current health status information of the physical examination person with the requirement of the physical examination item on the physical state, and retaining the current health status information of the physical examination person to meet the physical state requirement of the physical examination item; Performing resource adaptation judgment on physical examination items passing through the health state adaptation judgment, extracting the resource types required by each physical examination item to execute, acquiring the available conditions of each resource type in the current physical examination resource, comparing the available quantity of each resource type with the resource quantity required by the physical examination item to execute, and reserving physical examination items with the available quantity of each resource type meeting the resource quantity required by the physical examination item to execute; And carrying out compatibility analysis on the physical examination items which are judged to pass through the resource adaptation, extracting the compatibility description of each physical examination item, checking whether conflict records exist between the physical examination item and the compatibility description of other reserved physical examination items one by one, further comparing the execution time periods and the resource use conditions of the two physical examination items if the conflict records do not exist, checking whether the execution time periods have time range intersection, and whether the same resource is occupied at the same time or not in the resource use condition, reserving physical examination items which have no conflict records, have no time range intersection in the execution time periods and have no same resource simultaneous occupation condition in the resource use, and finally forming a dynamic adaptation physical examination item set.
4. 4. The personalized optimization method for physical examination schemes of artificial intelligent main examination according to claim 1, wherein the step of performing collaborative combination optimization on physical examination items in the dynamic adaptation physical examination item set, and planning physical examination item execution sequence and resource allocation mode by combining current health state time sequence variation trend and real-time physical examination resource use condition of a physical examination person, generating personalized physical examination scheme initial manuscripts comprising physical examination item execution sequence, resource allocation scheme and expected execution flow comprises the steps of: Acquiring time sequence change data of the current health state of the physical examination person, analyzing the change trend of the health state of the physical examination person along with time, determining the physical state of the physical examination person in different time periods to be suitable for executing physical examination item types, and forming a health state time sequence adaptation suggestion containing the physical state suitable for executing the physical examination item types in each time period; acquiring real-time physical examination resource use data, analyzing the use progress, idle time period and resource maintenance condition of each resource type, and recording the available states of each resource type in different time periods to form resource time sequence use suggestions containing the available time periods of each resource type; Extracting the execution duration, the execution difficulty and the physical consumption degree of each examination item in the dynamic adaptation examination item set, combining the examination item types suitable for execution in each time period in the health state time sequence adaptation suggestion, distributing each examination item to the corresponding time period suitable for execution, and preliminarily planning the suitable execution period of each examination item; According to the proper execution time period of each physical examination item and the type of resources required for execution, the available time period of each resource type in the resource time sequence use proposal is combined, resources in the corresponding available time period are allocated for each physical examination item, the physical examination items with the execution time period exceeding a preset time period threshold and the resource dependence meeting high-dependence judgment conditions are preferentially allocated for the physical examination items, the preset time period threshold is statistically determined based on the historical physical examination item execution time period, and the high-dependence judgment conditions are set based on the use frequency and the use time period of the physical examination items; Performing execution sequence optimization on physical examination items after resource allocation, analyzing the sequence execution relationship among the physical examination items, determining physical examination items with the sequence execution relationship by comparing whether the execution requirement of each physical examination item contains expressions taking the results of other physical examination items as preconditions, arranging the physical examination items with the sequence execution relationship according to the precondition relationship, arranging physical examination items without the sequence execution relationship in parallel in the same time period by combining the resource use condition and the physical examination item types suitable for execution in each time period in the health state time sequence adaptation suggestion; formulating execution details of each physical examination item, including preparation requirements before execution, monitoring points in execution and result record requirements after execution, integrating physical examination item execution sequence, resource allocation condition and execution details, and forming a physical examination item execution plan; Calculating overall expected execution time according to the execution plans of all physical examination items, determining the starting time and the ending time of a personalized physical examination scheme first draft, counting the use period of each resource, recording the allocation condition of each resource and generating a resource allocation list; Integrating physical examination project execution sequences, resource allocation lists, expected execution time length and execution detail requirements to form a personalized physical examination proposal first draft containing integral execution flow description; And calculating the difference value between the integral execution time length of the personalized physical examination proposal manuscript and the basic execution sequence time length, and completing the generation of the personalized physical examination proposal manuscript when the difference value is a positive value.
5. 5. The personalized optimization method for physical examination schemes of artificial intelligent main examination samples according to claim 1, wherein the steps of feeding back the personalized physical examination scheme manuscript to a physical examination person and collecting real-time feedback information of the physical examination person, simultaneously performing feasibility assessment on the personalized physical examination scheme manuscript, and integrating the real-time feedback information and feasibility assessment results to adjust the personalized physical examination scheme manuscript to obtain a final personalized physical examination scheme comprise the following steps: the personalized physical examination proposal initial draft is presented to the physical examination person in a visual form, a feedback input interface is provided, the physical examination person is guided to provide comments on physical examination item selection, execution sequence, execution time and resource allocation in the personalized physical examination proposal initial draft, comments input by the physical examination person are received through the feedback input interface, and real-time feedback information of the physical examination person is formed; Performing time feasibility assessment on the personalized physical examination scheme first draft, analyzing the relation between the estimated execution time length of the personalized physical examination scheme first draft and the acceptable time range confirmed in advance by a physical examination person, checking whether the starting time and the ending time of the execution time of each physical examination item in the personalized physical examination scheme first draft are overlapped, and marking the time feasibility problem that the estimated execution time length exceeds the acceptable time range or the execution time is overlapped; Carrying out resource feasibility assessment, verifying the current available state of the resources distributed in the personalized physical examination scheme first draft, comparing the current available quantity with the quantity required by physical examination item execution, checking whether the starting time and the ending time of a resource use period are overlapped or not, and marking the available state of the resources as unavailable, insufficient available quantity or the problem of resource feasibility with overlapped use periods; Carrying out health adaptation feasibility assessment, re-checking the corresponding relation between the execution time periods of each physical examination item in the personalized physical examination scheme first draft and the proper execution time period in the health state time sequence adaptation suggestion, checking whether the execution sequence of the physical examination item accords with the rule of no precondition conflict among physical examination items, and marking the health adaptation feasibility problem that the execution time period does not correspond to the proper execution time period or the execution sequence has precondition conflict; collecting the problems marked in the time feasibility, the resource feasibility and the health adaptation feasibility evaluation, and sorting and forming a feasibility evaluation report according to the problem types; Integrating real-time feedback information and feasibility assessment reports of the physical examination person, preferentially processing the content which explicitly presents adjustment requirements in the real-time feedback information of the physical examination person, removing the physical examination item from the personalized physical examination scheme first draft if the physical examination person does not accept the physical examination item, and adding the physical examination item which is inferior to the physical examination item and has no conflict in the screening of the dynamic physical examination item set as a substitute physical examination item into the personalized physical examination scheme first draft; Aiming at the problems in the feasibility evaluation report, if time conflict exists, re-planning the execution time of related physical examination items in the personalized physical examination scheme first draft based on the duration of the conflict time period, or adjusting the execution sequence based on a conflict-free execution sequence rule, wherein the conflict-free execution sequence rule is summarized and set through historical conflict-free execution cases; Generating an adjusted personalized physical examination scheme after the adjustment is completed, feeding back the adjusted personalized physical examination scheme to the physical examination person again, collecting new comments of the physical examination person through a feedback input interface, and repeatedly executing a feasibility evaluation step on the adjusted personalized physical examination scheme, wherein if the new comments of the physical examination person have no adjustment requirement and the feasibility evaluation has no problem, the adjusted personalized physical examination scheme is determined to be a final personalized physical examination scheme; If the adjustment requirement or feasibility assessment still has problems in the new opinion of the physical examination person, repeating the real-time feedback information collection of the physical examination person, the feasibility assessment of the personalized physical examination scheme and the adjustment step of the personalized physical examination scheme until the personalized physical examination scheme meets the requirement of the physical examination person and the implementation is feasible.
6. 6. The personalized optimization method for physical examination schemes of artificial intelligent main examination according to claim 2, wherein the performing association strength calibration on each association relation side in the feature association network structure, extracting the feature association rule support degree corresponding to each association relation side, the co-occurrence frequency of the health feature item and the demand feature item in the historical physical examination scheme data, and the influence coefficient of the association relation on the physical examination item selection, and performing the integration calculation on the support degree, the co-occurrence frequency and the influence coefficient to obtain the association strength value corresponding to each association relation side, comprises: Invoking a historical physical examination scheme database in the artificial intelligent main examination system, and extracting historical physical examination scheme data containing health feature items and demand feature items in the feature association network structure from the historical physical examination scheme database; Determining health feature items and demand feature items connected by the association relation edges aiming at each association relation edge in the feature association network structure, and counting the number of historical physical examination schemes simultaneously containing the health feature items and the demand feature items from historical physical examination scheme data; Counting the total number of the history physical examination schemes containing the health feature items and the total number of the history physical examination schemes containing the demand feature items in the history physical examination scheme data, calculating the proportion of the total number of the history physical examination schemes containing the health feature items and the demand feature items to the total number of the history physical examination schemes containing the health feature items and the proportion to the total number of the history physical examination schemes containing the demand feature items, adding the two proportions, and dividing the sum by two to obtain the co-occurrence frequency; Acquiring a characteristic association rule corresponding to the association relation edge from a characteristic association rule library, counting the number of the history physical examination schemes which accord with the characteristic association rule in the history physical examination scheme data, and calculating the proportion of the number of the history physical examination schemes which accord with the characteristic association rule to the total number of the history physical examination schemes as the support degree of the characteristic association rule; Analyzing the influence of the association relation on the selection of physical examination items in the historical physical examination schemes, counting the number of the historical physical examination schemes for selecting specific physical examination items due to the existence of the association relation, and calculating the proportion of the number of the historical physical examination schemes to the total number of the historical physical examination schemes containing the association relation as an influence coefficient; Setting the integration proportion of the support degree, the co-occurrence frequency and the influence coefficient, wherein the integration proportion is confirmed through the verification of the historical integration effect data, the historical integration effect data is obtained by comparing the matching degree of the association strength value and the actual association condition under different integration proportions, and the support degree, the co-occurrence frequency and the influence coefficient are multiplied by the corresponding proportion coefficient respectively according to the integration proportion and then added to obtain the association strength value corresponding to the association relation side; Repeating the steps for all the association relation sides, calculating the association strength value of each association relation side, and simultaneously carrying out unified processing on all the association strength values to ensure that all the association strength values are in a preset standard numerical value interval, wherein the preset standard numerical value interval is statistically determined through historical association strength data, so that the generation and the use of a subsequent association weight matrix are facilitated.
7. 7. The personalized optimization method for physical examination schemes of artificial intelligent main examination items according to claim 3, wherein the method is characterized in that compatibility analysis is carried out on physical examination items which are judged to pass through resource adaptation, compatibility description of each physical examination item is extracted, whether conflict records exist between the physical examination item and the compatibility description of other reserved physical examination items is checked one by one, if no conflict records exist, execution time periods and resource use conditions of the two physical examination items are further compared, whether time ranges of the execution time periods are crossed, whether the same resource is occupied at the same time or not in the resource use conditions are checked, physical examination items which have no conflict records and no time ranges of the execution time periods are crossed and the same resource is occupied at the same time are reserved, and finally a dynamic adaptation physical examination item set is formed, and the method comprises the following steps: extracting compatibility description of each physical examination item which passes the resource adaptation judgment, wherein the compatibility description comprises conflict condition records of the physical examination item and other physical examination items in the requirements of execution time, execution resources and physical state; Combining all physical examination items which are subjected to resource adaptation judgment in pairs to form physical examination item combination pairs; For each physical examination item combination pair, extracting a compatibility description of a first individual examination item in the physical examination item combination pair, checking whether a conflict record with a second individual examination item in the physical examination item combination pair is contained therein, and simultaneously extracting a compatibility description of a second individual examination item in the physical examination item combination pair, checking whether a conflict record with the first individual examination item in the physical examination item combination pair is contained therein; If the compatibility descriptions of the two physical examination items in the physical examination item combination pair have no conflict record, extracting the execution time periods of the two physical examination items, comparing the starting time and the ending time of the execution time period of the first physical examination item with the starting time and the ending time of the execution time period of the second physical examination item, judging whether the condition that the starting time of one time period is between the starting time and the ending time of the other time period exists or not, and if the condition exists, judging that the execution time periods are crossed; Simultaneously extracting the resource types required by the execution of the two body inspection items, checking whether the same resource type exists, if so, further comparing the use time periods of the resource type, judging whether the use time period overlap exists, and determining the use time period overlap by comparing the start time and the end time of the two use time periods; if the conflict record exists between two physical examination items in the physical examination item combination pair, the time range of the execution period is crossed, or the condition that the same resource is occupied at the same time exists in the use of the resource, the physical examination item combination pair is marked to have the execution conflict, and if the condition does not exist, the physical examination item combination pair is marked to have no conflict; After the conflict analysis is completed on all the physical examination item combination pairs, counting the number of physical examination item combination pairs in which each physical examination item participates and conflict is executed; If the number of the physical examination item combination pairs with the execution conflict, which are participated in by a certain physical examination item, exceeds a preset conflict number standard, wherein the preset conflict number standard is set based on the conflict number statistics of a historical conflict-free scheme, removing the physical examination item from a physical examination item list, and if the number of the physical examination item combination pairs with the execution conflict does not exceed the preset conflict number standard, reserving the physical examination item; And carrying out combination of the reserved physical examination items again, repeatedly executing a conflict analysis step, confirming that all physical examination item combination pairs formed by combination are marked as conflict-free, and finally forming a dynamic aptamer physical examination item set.
8. 8. The personalized optimization method for physical examination schemes of artificial intelligent main examination according to claim 4, wherein the performing sequence optimization is performed on physical examination items after allocation of resources, the sequential execution relationship among the physical examination items is analyzed, whether the physical examination items with the sequential execution relationship are contained in the execution requirement of each physical examination item or not is determined by comparing whether the physical examination items with other physical examination item results as preconditions, the physical examination items with the sequential execution relationship are arranged according to the precondition relationship sequence, and the physical examination items without the sequential execution relationship are combined with the physical examination item types suitable for execution in each time period in the time sequence adaptation suggestion of the resource use condition and the health state, and the physical examination items which use different resources and are in the same suitable execution time period are arranged to be executed in parallel in the same time period, and the personalized optimization method comprises the following steps: Extracting the execution requirement of each physical examination item after resource allocation, checking whether the execution requirement of each physical examination item contains expressions on the premise of other physical examination items, and if so, determining that the physical examination item has a sequential execution relationship with the physical examination item mentioned in the expressions; Constructing a physical examination item execution sequence directed graph according to the identified sequential execution relationship, taking physical examination items as nodes, directing the physical examination items mentioned in the expression to the direction of the directed edges of the current physical examination items, and establishing the directed edges to form an initial execution sequence frame; Performing topological ordering on nodes in the initial execution sequence frame, and sequentially arranging the nodes according to the direction of the directed edge so that all physical examination items with sequential execution relations are arranged according to the order of the preconditions to form a basic execution sequence; Screening physical examination items without directed edge connection from the basic execution sequence to form physical examination item groups without sequential execution relationship, wherein the physical examination items in each physical examination item group are connected without any directed edge connection; Extracting the type of resources required by the execution of each physical examination item in each physical examination item group and the proper execution time period corresponding to the physical examination item in the health state time sequence adaptation suggestion for each physical examination item group without a sequential execution relationship; Counting the number of different resource types in a group, classifying physical examination items which use different resource types and have the same suitable execution time period, checking whether the resources corresponding to the classified physical examination items are in an available state in the suitable execution time period, and if the resources are in the available state, arranging the physical examination items to be executed in parallel in the suitable execution time period; If the types of resources required by the execution of the partial physical examination items in the group are the same, extracting the execution time length of the physical examination items, arranging the physical examination items in order from small to large according to the execution time length, comparing the execution time length, determining by comparing the numerical values of the execution time lengths of the physical examination items, inquiring the idle time periods of the resources corresponding to the physical examination items, and sequentially distributing the arranged physical examination items to the continuous idle time periods; combining with a time interval range in which physical states of the physical detectors in the health state time sequence adaptation advice are suitable for execution, and adjusting the parallel execution time interval to be within the time interval range so that the physical states of the physical detectors in the parallel execution time interval meet the requirements of all the parallel execution physical examination projects; integrating the optimized execution sequence with parallel arrangement to form a final physical examination item execution sequence; and calculating a difference value between the overall execution time length corresponding to the final physical examination item execution sequence and the time length corresponding to the basic execution sequence, and completing the execution sequence optimization when the difference value is a positive value.
9. 9. The personalized optimization method for physical examination protocols of artificial intelligent main examination according to claim 5, wherein if there is still a problem in adjustment requirement or feasibility assessment in new opinion of the physical examination person, repeating the steps of real-time feedback information collection of the physical examination person, feasibility assessment of personalized physical examination protocol execution and feasibility adjustment of personalized physical examination protocol until the personalized physical examination protocol meets the requirement of the physical examination person and is feasible to execute, comprising: The adjusted personalized physical examination proposal is presented to the physical examination person again in a visual form, new ideas input by the physical examination person are received through a feedback input interface, new adjustment requirements of the physical examination person on physical examination item selection, execution sequence, execution time and resource allocation in the adjusted personalized physical examination proposal are recorded, and new real-time feedback information of the physical examination person is formed; Performing execution feasibility assessment again on the adjusted personalized physical examination scheme, firstly performing time feasibility assessment, extracting execution time of each physical examination item in the adjusted personalized physical examination scheme, comparing start time and end time of the execution time of each physical examination item, checking whether overlap exists, and simultaneously comparing the whole execution time with an acceptable time range confirmed in advance by a physical examination person, and marking time feasibility problems; Extracting resource allocation information of each physical examination item in the adjusted personalized physical examination scheme, inquiring the current available state and available quantity of each resource, comparing the available quantity with the quantity required by physical examination item execution, checking whether the resource use time period is overlapped or not, and marking the feasibility problem of the resource; Extracting the execution time period of each physical examination item in the adjusted personalized physical examination scheme, comparing the time periods suitable for execution in the health state time sequence adaptation suggestion, checking whether the execution sequence accords with the preconditionless conflict rule, and marking the health adaptation feasibility problem; collecting the marked problems in the new feasibility evaluation result, integrating the marked problems with new real-time feedback information of the physical examination person, and sorting the marked problems according to the principle of giving priority to the feedback requirement of the physical examination person to form a new adjustment basis; Aiming at the adjustment requirement in the new real-time feedback information of the physical examination person, if an additional physical examination item is proposed, screening the physical examination item which meets the preset standard and has no conflict with the existing physical examination item in the adjusted personalized physical examination scheme from the dynamic physical examination item set, and adding the adjusted personalized physical examination scheme; Aiming at the problems in the new feasibility evaluation result, if the problems are time conflicts, the proper execution time periods corresponding to the conflict physical examination items are queried again, and one physical examination item is adjusted to other proper execution time periods; if the physical examination item is a health adaptation problem, the physical examination item is reassigned to a period suitable for execution in the health state time sequence adaptation suggestion; after finishing the new round of adjustment, generating a readjusted personalized physical examination scheme, feeding back the readjusted personalized physical examination scheme to the physical examination person, collecting new comments of the physical examination person, and repeatedly executing the feasibility evaluation step; Recording the change content of the personalized physical examination proposal and the feasibility evaluation result after each adjustment, counting the continuous adjustment times, and stopping adjustment if the number of the continuous adjustment times reaches a preset adjustment times threshold value, the feedback opinion number of the physical examination person is lower than the preset opinion number threshold value and the feasibility problem number is zero; If the continuous adjustment times reach the preset adjustment times threshold value and still have feedback comments or feasibility problems of the physical examination person, an expert suggestion module in the artificial intelligent main examination system is called to acquire adjustment suggestions of the expert on the personalized physical examination scheme, the personalized physical examination scheme is adjusted by combining the expert suggestions, and feasibility evaluation and physical examination person comment collection are executed again until the personalized physical examination scheme meets the requirements.
10. 10. A physical examination protocol personalized optimization system for an artificial intelligent master, comprising a processor and a readable storage medium, wherein the readable storage medium stores a program which, when executed by the processor, implements the physical examination protocol personalized optimization method for an artificial intelligent master according to any one of claims 1 to 9.

Description

Personalized optimization method and system for physical examination scheme of artificial intelligent main sample Technical Field The application relates to the technical field of artificial intelligence, in particular to a personalized optimization method and a personalized optimization system for a physical examination scheme of an artificial intelligence main sample. Background In the current physical examination service field, the traditional physical examination scheme making mode has a plurality of limitations. In one aspect, physical examination institutions often employ standardized physical examination packages that are often set based on common health problems of the general population, failing to adequately account for the unique health and individualization needs of each individual. For example, for a physical examination person with a particular family or past medical history, a standard package may not be targeted to cover the items of examination critical to their health risk assessment, resulting in some potential health problems not being discovered in time. On the other hand, even if part of physical examination institutions provide personalized services to a certain extent, the physical examination scheme is formulated mainly by means of manual work. The manual formulation scheme is not only low in efficiency, but also is easily influenced by experience and subjective judgment of doctors, and is difficult to comprehensively and accurately grasp the health requirements and resource conditions of physical examination persons, so that an optimal physical examination scheme cannot be generated. In addition, in the prior art, when a physical examination scheme is formulated, dynamic analysis on the time sequence change trend of the health state of the physical examination person is lacking, and comprehensive consideration of the use condition of physical examination resources in real time is carried out, so that the execution sequence and the resource allocation of the physical examination scheme are unreasonable, and the physical examination efficiency and the experience of the physical examination person are affected. Disclosure of Invention In view of the above, the application aims to provide a personalized optimization method and a personalized optimization system for a physical examination scheme of an artificial intelligent main examination. According to a first aspect of the present application, there is provided a method for personalized optimization of a physical examination protocol of an artificial intelligent master sample, the method comprising: Basic health information of a physical examination person and personalized demand description proposed by the physical examination person are obtained, the basic health information and the personalized demand description are input into an artificial intelligent main examination system, semantic analysis and feature extraction are carried out on the basic health information and the personalized demand description, and a health information interpretation result and a demand interpretation result are obtained; Developing dynamic association modeling of health requirements based on the health information interpretation results and the requirement interpretation results, mining potential association relations between the health information and personalized requirements, and generating an association weight matrix; Performing scene adaptation screening on physical examination items in a preset physical examination item library according to the association weight matrix and physical examination item attribute information in the preset physical examination item library, and determining adapted physical examination items by combining the health state of a physical examination person and real-time resource conditions to obtain a dynamic adaptation physical examination item set; Performing collaborative combination optimization on physical examination items in the dynamic adaptation physical examination item set, and planning physical examination item execution sequences and resource allocation modes by combining the current health state time sequence change trend and real-time physical examination resource use condition of physical examination persons to generate personalized physical examination scheme first draft comprising physical examination item execution sequences, resource allocation schemes and expected execution flows; And feeding back the personalized physical examination proposal first draft to the physical examination person, collecting real-time feedback information of the physical examination person, simultaneously carrying out feasibility assessment on the personalized physical examination proposal first draft, and integrating the real-time feedback information and the feasibility assessment result to adjust the personalized physical examination proposal first draft so as to obtain a final personalized physical examination proposal.