A nursing decision support system based on intelligent terminals

Abstract

The present application discloses a nursing decision support system based on an intelligent terminal, which relates to the technical field of medical nursing information management, and includes a plan creation module, an execution tracking module, and a plan evaluation module; the diagnosis acquisition sub-module of the plan creation module obtains the target nursing diagnosis of the target patient according to the selected nursing diagnosis label by the user; the expected acquisition sub-module of the plan creation module is used to obtain the corresponding target nursing expectation according to the selected expected nursing result by the user under the corresponding target nursing diagnosis; the measure acquisition sub-module of the plan creation module is used to obtain the corresponding target nursing measure according to the selected nursing measure by the user under the corresponding target nursing expectation; the plan generation sub-module of the plan creation module is used to generate the target nursing plan of the target patient; the execution tracking module is used to supervise the execution process of the target nursing plan; the plan evaluation module is used to obtain the actual nursing result of the target nursing expectation, so as to flexibly create a personalized nursing plan.

Description

Technical Field

[0001] This application relates to the technical field of medical care information management, and particularly relates to a nursing decision support system based on intelligent terminals. Background Art

[0002] With the increasing growth of medical care needs, the nursing tasks and nursing decisions faced by nursing staff tend to become more complex. In the prior art, computer systems have been used to assist nursing staff in formulating nursing plans, tracking and managing nursing tasks, and collecting and collating nursing feedback. For example, a method, storage medium, and device for formulating a nursing plan with the patent application number CN202410605149.6 match nursing templates in a preset nursing template library according to the symptom information of patients.

[0003] Although this method effectively reduces the manual operations of nursing staff and improves the efficiency of formulating nursing plans, since the nursing templates in the preset nursing template library are usually designed for general symptoms and typical changes, and the nursing needs of patients are complex and variable, it is difficult for the nursing templates in the preset nursing template library to adapt to all patients. The nursing plans generated according to the nursing templates have a low degree of adaptation to patients, thus increasing the workload of modification for nursing staff when formulating nursing plans. For example, when the nursing needs of a patient do not completely match or even completely mismatch the nursing templates in the preset nursing template library, the nursing staff still need to spend time evaluating the degree of adaptation between the nursing templates and the patients and manually modifying the template content. Summary of the Invention

[0004] The purpose of the present invention is to solve the technical problem of how to improve the degree of adaptation between the generated nursing plan and patients, and provides a nursing decision support system based on intelligent terminals to reduce the difficulty of template modification for nursing staff.

[0005] According to the first aspect of the present invention, the present invention claims a nursing decision support system based on intelligent terminals, including a plan creation module, an execution tracking module, and a plan evaluation module;

[0006] Among them, the plan creation module includes a diagnosis acquisition sub-module, an expectation acquisition sub-module, a measure acquisition sub-module, and a plan generation sub-module; the diagnosis acquisition sub-module is used to obtain the target nursing diagnosis of the target patient according to the nursing diagnosis label selected by the user terminal; the expectation acquisition sub-module is used to obtain the target nursing expectation of the target patient according to the expected nursing result selected by the user terminal under the corresponding target nursing diagnosis; the measure acquisition sub-module is used to obtain the target nursing measure of the target patient according to the nursing measure selected by the user terminal under the corresponding target nursing expectation; the plan generation sub-module is used to generate the target nursing plan of the target patient according to all the target nursing measures;

[0007] Among them, the execution tracking module is used to supervise the execution process of the target care plan;

[0008] Among them, the plan evaluation module is used to obtain the actual care results corresponding to each target care expectation.

[0009] In an embodiment of the present application, the care measures include measure types and care contents, and the diagnosis acquisition sub-module further includes a type acquisition grandchild module, a content retrieval grandchild module, and a content planning grandchild module; the type acquisition grandchild module is used to obtain the target measure type according to the measure type selected by the user terminal under the corresponding target care expectation; the content retrieval grandchild module is used to match historical care measures according to the target care diagnosis, the target care expectation, and the target measure type to obtain candidate care measures; the content planning grandchild module is used to obtain the candidate care measures selected by the user terminal to obtain the target care measures.

[0010] In an embodiment of the present application, the execution tracking module includes a measure verification sub-module, and the measure verification sub-module further includes a content acquisition grandchild module, a standard retrieval grandchild module, and a content verification grandchild module; the content acquisition grandchild module is used to obtain the actual care video of the target person from the measure start time to the measure end time, identify the sequence of skeletal key points of each target person in the actual care video, and obtain the actual action sequence; the standard retrieval grandchild module is used to match the standard action sequence according to the target care content of the target care measure; the content verification grandchild module is used to obtain the care verification result according to the comparison result between the standard action sequence and the actual action sequence.

[0011] In an embodiment of the present application, the target persons include target care personnel and target patients, and the content verification grandchild module further includes extracting the action features of the actual action sequence to obtain actual action features; wherein, the action features include all pairwise distance features between skeletal key points; extracting the action features of the standard action sequence to obtain standard action features; obtaining the care verification result according to the action similarity between the actual action features and the standard action features.

[0012] In an embodiment of the present application, the nursing verification result includes a set of differential actions and a nursing qualification rate; the execution tracking module includes an execution reminder sub-module, and the execution reminder sub-module includes a content reminder grandchild module. The content reminder grandchild module is configured to query the historical nursing records of the target caregiver according to the target nursing content to be executed by the target patient, so as to obtain reference nursing content; the reference nursing content includes the target nursing content in the historical nursing records of the target caregiver that is the same as the target nursing content of the target patient and the target nursing content corresponding to the nursing qualification rate not greater than the qualification rate threshold; before the target execution time of the target nursing content, the content reminder grandchild module sends the standard action sequence corresponding to the reference nursing content and the corresponding set of differential actions to the target caregiver.

[0013] In an embodiment of the present application, the execution reminder sub-module further includes an expected prediction grandchild module; when the target caregiver completes the target nursing measure for the target patient, the content verification grandchild module is further configured to compare the corresponding nursing qualification rate with the qualification rate threshold. If the corresponding nursing qualification rate is not greater than the qualification rate threshold, the content verification grandchild module sends the corresponding actual action sequence to the expected prediction grandchild module. The expected prediction grandchild module obtains the first-stage nursing content according to all the actual action sequences that have been completed and all the standard action sequences that have not been executed in the corresponding target nursing expectation of the target patient, and determines whether the target patient can meet the corresponding target nursing expectation according to the first-stage nursing content and the initial characteristics of the target patient; if the target patient cannot meet the corresponding target nursing expectation, a first feedback message is sent to the user, and the first feedback message is used to remind the user to add remedial measures.

[0014] In an embodiment of the present application, the plan generation sub-module is further configured to retrieve historical nursing records according to each nursing content in the candidate nursing measures selected by the user, and generate a set of remedial measures according to the successfully matched historical nursing records; when the user adds a remedial measure for the target patient, the content modification grandchild module screens the set of remedial measures according to the corresponding target nursing content to obtain candidate remedial measures, generates second-stage nursing content according to each candidate remedial measure and the first-stage nursing content respectively, and determines whether the target patient can meet the corresponding target nursing expectation according to the second-stage nursing content and the initial characteristics of the target patient. If the target patient can meet the corresponding target nursing expectation, the corresponding candidate remedial measure is used as the recommended remedial measure.

[0015] In an embodiment of the present application, the calculation method of the nursing qualification rate includes:

[0016] ;

[0017] ;

[0018] Wherein, C represents the nursing qualification rate, I represents the indicative function, X represents the input condition of the indicative function, s represents the action similarity, ω 0 represents the weight of the action similarity, S 1 represents the similarity threshold, S 1 has a value greater than 0, x i represents the value of the actual nursing content for the i-th preset index, q i represents the passing standard for the i-th preset index, and n represents the number of preset indexes.

[0019] In an embodiment of the present application, the execution tracking module further includes a personnel allocation sub-module, and the personnel allocation sub-module further includes a patience acquisition sub-module, a skill threshold matching sub-module, a skill level acquisition sub-module, and a personnel matching sub-module. The patience acquisition sub-module is used to obtain the patience level of the target patient according to the satisfaction evaluation results of the target patient for each candidate caregiver in all completed target nursing measures. The skill threshold matching sub-module is used to match the corresponding skill level threshold according to the patience level of the target patient. The skill level acquisition sub-module is used to obtain the skill level of each candidate caregiver for the corresponding target nursing content. The personnel matching sub-module is used to compare the skill level of each candidate caregiver with the skill level threshold respectively, and use the candidate caregivers whose skill level is not less than the corresponding skill level threshold as the recommended caregivers.

[0020] The present application has the following beneficial effects:

[0021] 1. The retrieval and generation method with a hierarchical structure allows users to select step by step according to nursing diagnoses, expected nursing goals, and nursing measures, enabling users to gradually screen and flexibly combine nursing diagnoses, expected nursing effects, and nursing measures. While providing personalized nursing plans, it reduces the generation of nursing plans with low fitness to the target patient, and reduces the workload of modifying nursing templates at the user end during the nursing plan creation stage.

[0022] 2. On the basis of ensuring that the nursing plan can flexibly combine nursing diagnoses, expected nursing outcomes, and nursing measures, this system also recommends nursing measures according to the retrieval path of nursing diagnosis - expected nursing outcome - measure type - nursing content. By analyzing historical cases, it obtains specific nursing plan recommendations for nursing measures based on the conditions of similar patients, reducing the probability of unnecessary repetition or omission in the process of formulating the nursing plan and minimizing the impact of problems such as empirical bias or insufficient information on the nursing quality of patients. At the same time, retrieving historical nursing measures during the formulation of the nursing plan makes the formulated nursing plan more in line with the characteristics of the patient population and medical resources of the medical structure itself. The nursing plan evidenced by historical nursing records is more scientific, improving the quality of nursing and the accuracy of decision-making.

[0023] 3. The measure verification sub-module compares the actions of the target nurse in the actual nursing process with the actions in the standard nursing content. The nursing verification result can be used to determine whether there are any non-standard behaviors of the nurse during the nursing operation, checking for deviations in the nursing operation, which helps reduce nursing non-standard behaviors caused by insufficient understanding of the patient, lack of experience, negligence, cognitive bias, etc.

[0024] 4. When extracting the actual action features and standard action features, not only the distance features between every two of all the skeletal key points of the nurse and the distance features between every two of all the skeletal key points of the patient are extracted, but also the distance features between each skeletal key point of the nurse and each skeletal key point of the patient are extracted. The actual action features and the standard action features not only describe the actions of the nurse and the patient, but also describe the temporal relationship between the actions of the nurse and the patient. For example, whether the actions of the nurse and the patient occur successively or simultaneously. This method improves the accuracy of judging the standard degree of the nurse's nursing operation, helping to improve the comfort and safety of the patient during the nursing process.

[0025] 5. The content reminder sub-module personalizedly pushes the specification requirements of the target nursing content for the target patient and the actions of the target nurse that are inconsistent with the nursing specification requirements in the historical nursing records before the target nurse executes the target nursing content for the target patient, thereby assisting the nurse to improve the standard degree of nursing operation and reducing the risk of poor nursing effects for the target patient caused by non-standard nursing operations of the nurse.

[0026] 6. When there are non-standard nursing operations of the target nurse, the expected prediction sub-module judges whether the non-standard operation will affect the nursing effect of the patient according to the physiological characteristics of the patient and the nursing content in the first stage, and reminds the user to add remedial measures to timely remedy the adverse effects of non-standard nursing operations.

[0027] 7. The identification acquisition sub-module sets a remedy identification or a non-remedy identification for the nursing measures newly added in the target nursing plan process, so as to assist the nursing staff in judging the necessity of the nursing measures during the process of creating the nursing plan and reduce the probability of over-nursing.

[0028] 8. The actions in the actual nursing video are screened according to the comparison result between the action similarity and the first preset threshold, so as to judge whether the actual nursing content in the actual nursing video is the same as the target nursing content. If the action similarity is not greater than the first preset threshold, it is considered that the actual nursing content is different from the target nursing content, and there is no need to calculate whether the values of other preset indicators meet the specifications, thus reducing the computing amount of the computer system.

[0029] 9. Recommending a caregiver for the target patient based on the patience level of the target patient and the skill level of the candidate caregiver can not only ensure the cooperation degree of the target patient during the nursing process, but also facilitate arranging a caregiver with a lower skill level to perform nursing operations when the patient has a higher patience level, so as to specifically improve the nursing skill level of the caregiver. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The drawings herein are incorporated into the specification and form a part of this specification, showing embodiments consistent with the present disclosure, and are used together with the specification to explain the principles of the present disclosure.

[0031] Figure 1 It is a schematic structural diagram of an implementation manner of a nursing decision support system based on an intelligent terminal according to an embodiment of the present application;

[0032] Figure 2 It is a schematic structural diagram of another implementation manner of a nursing decision support system based on an intelligent terminal according to an embodiment of the present application;

[0033] Figure 3 It is a schematic structural diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0034] The present invention provides a nursing decision support system based on an intelligent terminal. To make the above objects, features, and advantages of the present application more apparent and understandable, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all embodiments. Usually, the components of the embodiments of the present application described and illustrated herein can be arranged and designed in various different configurations. Therefore, with reference to terms such as "one embodiment", "some embodiments", "implementation manners", "embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples", etc., the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the present application to be protected, but only represents that the specific features, structures, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. Moreover, the specific features, structures, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present application.

[0035] It should be noted that: similar reference numerals and letters denote similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of the present application, relational terms such as "first", "second", etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations.

[0036] According to a first aspect of the present invention, the present invention claims protection for a nursing decision support system based on an intelligent terminal. Referring to the attached Figure 1 as shown, it includes a plan creation module, an execution tracking module, a plan evaluation module, and a data storage module. Among them, the plan creation module is used to create a target nursing plan for a target patient, the execution tracking module is used to supervise the execution process of the target nursing plan, and the plan evaluation module is used to obtain the actual nursing results corresponding to each target nursing expectation.

[0037] In this embodiment, the target care plan includes several target care measures, and of course, it may not be limited thereto. The plan creation module includes a label retrieval sub-module, a diagnosis acquisition sub-module, an expectation acquisition sub-module, a measure acquisition sub-module, and a plan generation sub-module, and of course, it may not be limited thereto. The label retrieval sub-module is used to retrieve corresponding secondary labels according to the labels selected by the user terminal. The diagnosis acquisition sub-module is used to obtain the target care diagnosis of the target patient according to the care diagnosis label selected by the user terminal. The expectation acquisition sub-module is used to obtain the target care expectation of the target patient according to the expected care result selected by the user terminal under the corresponding target care diagnosis. The target care measures of the target patient are obtained according to the care measures selected by the user terminal under the corresponding target care expectation. The plan generation sub-module responds to the plan generation task request initiated by the user terminal and obtains the target care plan of the target patient according to all the target care measures.

[0038] It should be noted that the form of the user terminal may be a client and / or a web terminal, etc. The present application does not further limit the specific form of the user terminal. The identity identifiers of the users include the responsible nurse, the assisting nurse, and other nursing staff, and of course, it may not be limited thereto. The other nursing staff may be a nursing worker or a family member of the target patient. Different operation permissions can be set for users with different identity identifiers. For example, it can be set that only the responsible nurse of the target patient has the permission to create a target care plan, etc.

[0039] It should be noted that the data storage module includes a label storage sub-module, and the label storage sub-module is used to store the hierarchical structure of the care plan. The label storage sub-module can be constructed according to the foreign key in the relational database, or according to the tree storage structure, or constructed according to the graph database, or other feasible methods. The hierarchical structure can be obtained through a preset method or other feasible methods.

[0040] In this embodiment, the hierarchical structure is preset specialty classification - nursing label - expected care result - care measure, specifically referring to the appendix Figure 2As shown. The initial value of the hierarchical structure obtains the initial content of the system according to the preset specialty classification and the nursing terminology classification system published by the North American Nursing Diagnosis Association (NANDA-I) to ensure the accuracy of nursing diagnoses and the continuity of nursing plans. The preset specialty classification includes general categories, respiratory categories, urinary categories, endocrine categories, neurological categories, digestive categories, circulatory categories, and specialized disease diagnosis categories, and of course it can be not limited to this. According to the preset specialty classification, the nursing diagnoses in the nursing terminology classification system published by the North American Nursing Diagnosis Association (NANDA-I) are divided into 8 nursing diagnosis categories, and the nursing diagnosis categories correspond to the preset specialties one by one to improve the retrieval efficiency of the client when selecting diagnosis labels. According to the nursing terminology classification system published by the North American Nursing Diagnosis Association (NANDA-I), nursing diagnosis labels, expected nursing outcomes, nursing measures, hierarchical relationships between different nursing diagnosis labels, hierarchical relationships between nursing diagnosis labels and expected nursing outcomes, hierarchical relationships between different expected nursing outcomes, hierarchical relationships between expected nursing outcomes and nursing measures, and hierarchical relationships between different nursing measures are obtained, and of course it can be not limited to this.

[0041] It should be noted that the expected nursing outcomes include the achievement time limit and evaluation criteria, and of course it can be not limited to this. For example, within one week after the operation, the target patient can walk independently for a distance of not less than 50 meters. The nursing measures include nursing content and execution time, and of course it can be not limited to this.

[0042] In this embodiment, the label retrieval sub-module pushes all the preset specialty classifications to the client. The label retrieval sub-module queries the hierarchical structure in the label storage sub-module according to the preset specialty classification selected by the client, obtains the corresponding candidate nursing diagnosis labels, and pushes all the candidate nursing diagnosis labels to the client. The diagnosis acquisition sub-module acquires the nursing diagnosis labels selected by the client. The label retrieval sub-module queries the hierarchical structure in the label storage sub-module according to the nursing diagnosis labels selected by the client, and obtains the corresponding optional secondary diagnosis labels and / or expected nursing outcomes. When the query result includes the corresponding secondary diagnosis labels, the user can continue to select the nursing diagnosis labels corresponding to the target patient. When the query result includes the corresponding expected nursing outcomes, the user can select the expected nursing outcomes that the target patient will achieve. The label retrieval sub-module queries the hierarchical structure in the label storage sub-module according to the expected nursing outcomes selected by the client, and obtains the corresponding optional secondary expected nursing outcomes and / or nursing measures. When the query result includes the corresponding secondary expected nursing outcomes, the user can continue to select the expected nursing outcomes corresponding to the target patient. When the query result includes the corresponding nursing measures, the user can continue to select the nursing measures that the target patient will adopt. The plan generation sub-module responds to the plan generation task request initiated by the client, generates the target nursing plan according to all the target nursing measures selected by the client, returns the target nursing plan to the plan creation module, and the plan creation module sends the target nursing plan to the execution tracking module.

[0043] It should be noted that after the client selects the nursing measures, the selection interface of the nursing diagnosis labels and / or expected nursing outcomes and / or nursing measures can be returned to continue adding the target diagnosis labels, target nursing expectations, and target nursing measures corresponding to the target patient. If the corresponding diagnosis labels and / or expected nursing outcomes and / or nursing measures cannot be found by the user, the nursing staff can report to the management personnel such as the nursing department, and update the hierarchical structure of the label storage sub-module through the label update sub-module.

[0044] It should be noted that the data storage module further includes a history storage sub-module. The history storage sub-module is used to store the historical nursing diagnoses, historical nursing expectations, historical nursing plans, historical nursing effects, and historical plan statuses corresponding to historical patients, and of course, it is not limited thereto. Among them, the historical plan status is used to describe the execution status of the corresponding historical nursing plan.

[0045] In this embodiment, the diagnosis acquisition sub-module also responds to a plan generation task request initiated by the client, generates the target nursing diagnosis according to all the nursing diagnosis labels selected by the client, and sends the target nursing diagnosis to the historical storage sub-module to update the historical nursing diagnosis. The expected acquisition sub-module also responds to a plan generation task request initiated by the client, obtains the target nursing expectation according to all the expected nursing outcomes selected by the user, and sends the target nursing expectation to the historical storage sub-module and the plan evaluation module respectively.

[0046] It should be noted that the execution tracking module includes a plan modification sub-module and a plan viewing sub-module, and of course it may not be limited to this. The plan modification sub-module is used to modify the detailed information of the target nursing plan. The detailed information of the nursing plan includes the plan execution status, the content of each nursing measure corresponding to the plan, the corresponding measure execution status, and the corresponding measure execution time, and of course it may not be limited to this. The plan execution status includes in execution, completed, and terminated, and of course it may not be limited to this. The measure execution status includes not executed, in execution, and executed, and of course it may not be limited to this.

[0047] In this embodiment, the plan modification sub-module includes a status modification sub-sub-module. When the execution tracking module obtains the target nursing plan sent by the plan creation module, the status modification sub-sub-module sets the execution status of the target nursing plan to in execution.

[0048] In this embodiment, when the number of nursing measures completed in the target nursing plan is equal to the total number of nursing measures in the target nursing plan, the status modification sub-sub-module sends feedback information to the corresponding client to remind the corresponding user whether to modify the plan execution status of the target nursing plan to completed. The status modification sub-sub-module responds to the user's selection result. If the user selects "yes", the status modification sub-sub-module modifies the plan execution status of the target nursing plan to completed and sends feedback information for obtaining the prognosis information of the target patient to the corresponding client. If the user selects "no", the status modification sub-sub-module sends feedback information to the client to remind the user to add the target nursing measure in the target nursing plan.

[0049] In this embodiment, the status modification sub-sub-module responds to the user's plan termination task request. For example, when the target patient has an early discharge or a transfer, the nursing staff modifies the plan execution status of the corresponding target nursing plan to terminated through the plan status modification sub-sub-module and sends feedback information for obtaining the prognosis information of the target patient and the reason for plan termination to the user.

[0050] In this embodiment, when the execution tracking module receives the target care plan sent by the plan creation module, the status modification sub-module sets the execution status of all target care measures corresponding to the target care plan to not executed. The status modification sub-module responds to the measure execution task request initiated by the user terminal. When the user clicks to start executing the corresponding target care measure, the status modification sub-module modifies the execution status of the corresponding target care measure to in execution. The status modification sub-module responds to the measure end task request initiated by the user terminal. When the user clicks to end the execution of the corresponding target care measure, the status modification sub-module modifies the execution status of the corresponding target care measure to executed.

[0051] It should be noted that the plan status modification sub-module can also respond to the status modification task request of the user, modify the plan execution status of the target care plan to the plan execution status specified by the user, and / or modify the execution status of the target care measure to the measure execution status specified by the user, so as to reduce the impact of the user's misoperation on the execution process of the target care plan.

[0052] It should be noted that the plan modification sub-module further includes a content modification sub-module, and the content modification sub-module is used to add target care measures to the target care plan, modify the content, execution time and execution status of the existing target care measures, and of course it can be more than this.

[0053] In this embodiment, the content modification sub-module responds to the plan modification task request of the user, generates a new target care plan according to the nursing diagnosis label, expected nursing result and nursing measure selected by the user terminal; responds to the plan deletion task request of the user, deletes the corresponding target care plan or deletes the target care measures in the target care plan.

[0054] In this embodiment, the execution tracking module further includes an execution reminder sub-module, and the execution reminder sub-module further includes a time reminder sub-module, and the time reminder sub-module is used to remind the user to complete the corresponding nursing content according to the execution time of each target care measure. The time reminder sub-module can send a second feedback message to the user terminal according to the reminder time preset by the user, and the second feedback message is used to remind the user to complete the corresponding nursing content according to the execution time of the target care measure. For example, send the second feedback message to the user 10 minutes before the execution time of the target care measure and / or send the second feedback message to the user at the time point corresponding to the execution time. Taking sending the second feedback message to the user at the time point corresponding to the execution time as an example, the time reminder module compares the current system time with the execution time of the target care measure in real time, and when the current system time is the same as the execution time, sends the second feedback message to the user terminal.

[0055] It should be noted that the user can view the detailed information of the target care plan through the aforementioned plan viewing sub-module.

[0056] It should be noted that the data storage module further includes a log storage sub-module, which is used to store the operation records of all users for each care plan. After the user operates on the target care plan through the execution tracking module, the execution tracking module sends the corresponding operation record to the log storage sub-module and / or the historical storage sub-module. For example, when a new symptom appears in a patient and the user adds a care measure to the target care plan through the content modification sub-module, the execution tracking module sends the new operation record to the log storage sub-module to store the modification record of the target care plan, and at the same time sends the new target care plan obtained after the new operation is completed to the historical storage sub-module to update the historical care plan.

[0057] In this embodiment, the plan evaluation module includes an evaluation acquisition sub-module. The evaluation acquisition sub-module is used to obtain the actual care results corresponding to each of the target care expectations. Each of the actual care results includes achieving the corresponding expected care result or not achieving the corresponding expected care result, and of course it can be more than this.

[0058] In this embodiment, the plan evaluation module further includes an evaluation reminder sub-module, which is used to remind the user to record whether the target patient has achieved the corresponding evaluation criteria according to the achievement time limit of each of the target care expectations. For example, for a patient with walking disabilities, the target care expectation may include that the patient can walk independently for 50 meters within one week after the start of the care measures related to walking recovery. When the patient can walk independently for 50 meters within this achievement time limit, the caregiver can actively record the actual care result corresponding to the target patient through the evaluation acquisition sub-module, and the evaluation reminder sub-module can also send a message to remind the user to record the actual care result corresponding to the target care expectation at the achievement time limit.

[0059] It should be noted that the plan evaluation module also responds to operations such as adding, modifying, and deleting the expected care results of the target care plan by the user in the execution tracking module. For example, due to special circumstances such as the deterioration of the patient's condition, it is necessary to postpone the start execution time of the care measures related to the patient's walking recovery. The plan evaluation module responds to the modified content of the target care measure by the user and modifies the achievement time limit of the corresponding target care expectation according to the modified execution time.

[0060] It should be noted that the retrieval and generation method of the hierarchical structure allows users to select step by step according to the preset specialty classification, nursing diagnosis, expected nursing goals, and nursing measures. It allows the user side to gradually screen and combine nursing diagnoses, expected nursing effects, and nursing measures according to the characteristic information of the target patient. While providing a personalized nursing plan, it reduces the generation of nursing plans with low compatibility with the target patient, and reduces the workload of modifying the nursing template at the stage of creating a nursing plan by the user side.

[0061] In this embodiment, each nursing measure further includes a measure type, which of course may not be limited thereto. The measure type is obtained according to the nursing terminology classification system of the North American Nursing Diagnosis Association (NANDA-I). For example, monitoring body temperature, keeping the surgical site clean and dry, clearing respiratory secretions, using diuretics, etc. The nursing content is the specific content of the nursing measure. For example, when the measure type is to clear respiratory secretions, the corresponding nursing content can be chest percussion, or it can be postural drainage, or it can be the use of a nebulizer, or it can be other operation methods for clearing respiratory secretions, or a combination of multiple operation methods for clearing respiratory secretions. The nursing content can be obtained according to historical nursing measures.

[0062] In this embodiment, the diagnosis acquisition sub-module further includes a type acquisition sub-sub-module, a content retrieval sub-sub-module, and a content planning sub-sub-module, which of course may not be limited thereto. The type acquisition sub-sub-module is used to obtain the target measure type according to the measure type selected by the user side under the corresponding target nursing expectation. The content retrieval sub-sub-module is used to match the historical nursing measures of historical patients according to the target nursing diagnosis, the target nursing expectation, and the target measure type to obtain candidate nursing measures. The nursing diagnosis corresponding to the candidate nursing measures is the same as the target nursing diagnosis, the expected nursing result of the candidate nursing measures is the same as the target nursing expectation, and the measure type of the candidate nursing measures is the same as the target measure type. The content planning sub-sub-module is used to obtain the target nursing measure by obtaining the candidate nursing measures selected by the user side.

[0063] In this embodiment, the label retrieval sub-module queries the hierarchical structure in the label storage sub-module according to the expected nursing result selected by the user side to obtain the corresponding optional secondary expected nursing results and / or measure types. When the query result includes the corresponding measure type, the user can continue to select the measure type to be adopted by the target patient. The type acquisition module obtains the target measure type selected by the user. The content retrieval sub-sub-module queries the historical storage sub-module according to the target nursing diagnosis, the target nursing expectation, and the target measure type to obtain the candidate nursing measures. The content planning sub-sub-module obtains the candidate nursing measures selected by the user to obtain the target nursing measure.

[0064] It should be noted that on the basis of ensuring that the nursing plan can flexibly combine nursing diagnoses, expected nursing outcomes, and nursing measures, this system also recommends nursing measures according to the retrieval path of nursing diagnosis - expected nursing outcome - measure type - nursing content. By analyzing historical cases, specific nursing plan recommendations for nursing measures based on the situations of similar patients are obtained, reducing the probability of unnecessary repetition or omission in the process of formulating nursing plans and minimizing the impact of problems such as empirical deviation or insufficient information on the nursing quality of patients. At the same time, retrieving historical nursing measures during the formulation of the nursing plan makes the formulated nursing plan more in line with the characteristics of the patient population and medical resources of the medical structure itself. The nursing plan evidence-based by historical nursing records is more scientific, improving the nursing quality and the accuracy of decision-making.

[0065] In a feasible implementation manner, referring to the attached Figure 2 As shown, the execution tracking module includes a measure verification sub-module, and the measure verification sub-module further includes a content acquisition grandchild module, a standard retrieval grandchild module, and a content verification grandchild module; the content acquisition grandchild module is used to obtain the actual nursing video of the target person from the start time to the end time of the measure, identify the sequence of skeletal key points of the target person in the actual nursing video, and obtain the actual action sequence; the standard retrieval grandchild module is used to match the standard action sequence according to the target nursing content; the content verification grandchild module is used to obtain the nursing verification result according to the comparison result between the standard action sequence and the actual action sequence.

[0066] It should be noted that the target person includes the target nursing staff and the target patient, and of course it can be not limited to this. The data storage module further includes a standard measure storage sub-module, and the standard measure storage sub-module is used to store the standard sequence of skeletal key points corresponding to all nursing contents. The sequence of skeletal key points for each nursing content can be obtained by acquiring the standard demonstration video corresponding to the nursing content, performing frame-by-frame operation on the standard demonstration video to obtain standard action images, and identifying the skeletal key points of the nursing staff and the patient in each standard action image through a neural network model. The recognition results of all standard action images are combined to obtain the standard sequence of skeletal key points corresponding to the nursing content. The standard retrieval grandchild module queries the standard measure storage sub-module according to each target nursing content to obtain the standard action sequence. The nursing content of the standard action sequence is the same as the corresponding target nursing content. The skeletal key points that can be recognized by the neural network model can be preset.

[0067] In this embodiment, the skeletal key points include the head, neck, shoulders, elbows, wrists, fingertips, hips, knees, ankles, toes, chest, waist, spine, and heels, and of course, it is not limited thereto. The content verification sub-module extracts the action features of the standard action sequence through a neural network model to obtain standard action features. The action features of the actual action sequence are extracted to obtain actual action features. The action features include all pairwise distance features between the skeletal key points. For example, if the identified skeletal key points are A, B, and C, the action features include the distance features between A and B, B and C, and A and C at all times.

[0068] It should be noted that the nursing verification result can be obtained based on the action similarity. The action similarity includes the similarity between the actual action features and the standard action features. The specific calculation method of the action similarity can be calculated through the dynamic time warping algorithm, or can be calculated through methods such as cosine similarity and Euclidean distance, or other feasible calculation methods.

[0069] It should be noted that by comparing the actions of the target caregiver in the actual care process with the actions in the standard care content through the measure verification sub-module, the nursing verification result can be used to determine whether there are non-standard behaviors in the caregiver's care operation process, check for deviations in the care operation, and is conducive to reducing care non-standard behaviors caused by insufficient understanding of the patient, lack of experience, negligence, cognitive bias, etc. Especially when the care plan for the same patient needs to be completed by multiple caregivers, such as the primary nurse, assistant nurse, nursing assistant, and / or patient's family members, due to insufficient understanding of the patient's actual situation and cognitive bias, other caregivers may have misunderstandings or omissions of key information in the care measures formulated by the primary nurse. The measure verification sub-module can assist the primary nurse in capturing the differences between the actual care content and the target care content.

[0070] It should be noted that when extracting the actual action features and the standard action features, not only the distance features between each pair of all the skeletal key points of the caregiver and the distance features between each pair of all the skeletal key points of the patient are extracted, but also the distance features between each skeletal key point of the caregiver and each skeletal key point of the patient are extracted. The actual action features and the standard action features not only describe the actions of the caregiver and the patient, but also describe the temporal relationship between the actions of the caregiver and the patient. For example, whether the actions of the caregiver and the patient occur successively or simultaneously. This method improves the accuracy of judging the standard degree of the caregiver's care operation and helps to improve the comfort and safety of the patient during the care process.

[0071] In a feasible implementation, the nursing verification result includes a set of differential actions and a nursing qualification rate, which of course is not limited thereto. The set of differential actions includes the parts where the actual action sequence differs from the standard action sequence. The nursing qualification rate is obtained based on the action similarity between the actual action features and the standard action features and the achievement rate of preset indicators. The preset indicators are set in advance, such as the frequency, duration, number of specified actions, etc., which of course is not limited thereto. Specified actions include, for example, chest percussion, local massage, etc., which of course is not limited thereto.

[0072] In this implementation, taking chest percussion as an example for detailed description, the preset indicators of chest percussion include the chest percussion frequency and the chest percussion duration. The actual action sequence and the standard action sequence are aligned through the dynamic time warping algorithm. According to the alignment result, the distance between each feature point in the actual action sequence and the corresponding feature point in the standard action sequence is calculated respectively to obtain the action distance. The action distance is compared with the action threshold. If the action distance is not greater than the action threshold, then this feature point belongs to the part where the actual action sequence and the standard action sequence do not differ, that is, this feature point does not belong to the elements of the set of differential actions; if the action distance is greater than the action threshold, then this feature point belongs to the part where the actual action sequence and the standard action sequence differ, that is, this feature point belongs to the elements of the set of differential actions. The action similarity is obtained based on the action distances corresponding to all feature points in the actual action sequence.

[0073] It should be noted that the action threshold can be obtained by setting in advance.

[0074] In this implementation, the specific calculation method of the nursing verification result c of chest percussion is as follows:

[0075] ;

[0076] ;

[0077] Among them, ω 0 represents the weight of the action similarity, s represents the action similarity, S1 represents the first preset threshold, ω 1 represents the weight of the achievement rate of the chest percussion frequency, f represents the chest percussion frequency of the nursing staff in the actual nursing video, F 1 represents the minimum value of the chest percussion frequency in the nursing standard, F 2 represents the maximum value of the chest percussion frequency in the nursing standard, ω 2 represents the chest percussion duration of the nursing staff in the actual nursing video, D 1 represents the minimum value of the chest percussion duration in the nursing standard, D 2Represents the maximum value of the chest percussion duration in the nursing standard, and X represents the input condition of the indicative function I.

[0078] In this embodiment, F 1 , F 2 , D 1 and D 2 can be obtained by means of presetting. For example, the nursing standard requires that the chest percussion frequency is 120 - 180 times per minute, and the chest percussion duration is 1 - 3 minutes. If the chest percussion frequency of the nursing staff in the actual nursing video belongs to [120, 180], the calculation result of the chest percussion frequency in the indicative function is 1; if the chest percussion frequency of the nursing staff in the actual nursing video does not belong to [120, 180], the calculation result of the chest percussion frequency in the indicative function is 0. The calculation results of other preset indicators in the indicative function are similar, and will not be described in detail here.

[0079] In this embodiment, when the action similarity of the target person is less than the first preset threshold, the corresponding value of the nursing verification result is 0, that is, it is considered that the actual nursing content of the target person is different from the target nursing content, and there is no need to judge whether other preset indicators meet the standards. For example, the target nursing content is chest percussion, but the actual nursing content of the nursing staff is using a nebulizer, then there is no need to extract the chest percussion frequency and chest percussion duration in the actual operation. When the action similarity of the target person is not less than the first preset threshold, that is, it is considered that the actual nursing content of the target person is the same as the target nursing content. The closer the value of the action similarity is to 1, the higher the standard degree of the nursing actions actually performed by the target nursing staff. It is also necessary to calculate the nursing verification result according to the preset indicators. Through the comparison result of the action similarity and the first preset threshold, the actions in the actual nursing video are screened to judge whether the actual nursing content in the actual nursing video is the same as the target nursing content. If the action similarity is not greater than the first preset threshold, it is considered that the actual nursing content is different from the target nursing content, and there is no need to calculate whether the values of other preset indicators meet the specifications, so as to reduce the calculation amount of the computer system.

[0080] It should be noted that the specific value of the first preset threshold can be obtained by means of presetting. The weights of the action similarity, the chest percussion frequency, and the chest percussion duration can be obtained by presetting according to the importance and the specific value range of the nursing verification result. For example, the value range of the nursing verification result is [0, 1], then ω 0 , ω 1 and ω 2 can sum up to 1. The nursing standard can also be obtained according to the input from the user side, or other feasible methods.

[0081] It should be noted that the actual action sequence can be input into a pre-trained first machine learning model to obtain an action label sequence. The action label sequence includes a sequence composed of action category labels corresponding to each time point. The duration of the specified action is calculated according to the proportion of the specified action in all action labels and the total duration of the actual action sequence.

[0082] It should be noted that the first machine learning model can be constructed based on neural network models such as 3D convolutional neural network models and recurrent neural network models. Video images of different nursing processes in the historical monitoring records of medical institutions are collected. Action labels are manually set for each video image. For example, the action label for the first video in the time period from t 0 to t 1 is chest percussion, and the action label for the time period from t 1 to t 2 is axial turning. According to the manually set action labels, a true label sequence corresponding to t 0 to t 2 is generated. The first machine learning model is trained according to all video images and the corresponding true label sequences. The value of the loss function is calculated according to the difference between the predicted label sequence output by the first machine learning model and the true label sequence. When the value of the loss function is minimized, the training of the first machine learning model is completed.

[0083] It should be noted that the time points corresponding to the specified action are screened out from the action label sequence. According to the time points corresponding to the specified action, the action sequence corresponding to the specified action is intercepted from the actual action sequence to obtain a first action sequence. The determination rule for the specified action is obtained. The determination rule can be obtained by pre-setting according to the specific content of the specified action. According to the preset determination rule, the actual execution times of the execution action are screened out from the first action sequence. The frequency of the specified action is calculated according to the ratio between the actual execution times and the duration corresponding to the first action sequence.

[0084] In this embodiment, the determination rule for chest percussion can be that the distance from the caregiver's hand to the patient's chest is not greater than a third preset threshold. Among them, the value of the third preset threshold can be obtained by pre-setting. The distance from the hand bone key point of the target caregiver to the chest bone key point of the target patient corresponding to each moment in the first action sequence is calculated respectively to obtain the target distance. All the minimum values in the target distance are obtained. The number of minimum values not greater than the third preset threshold among all the minimum values is counted to obtain the target percussion times. The frequency of the specified action is obtained according to the ratio between the target percussion times and the duration corresponding to the first action sequence.

[0085] In this embodiment, refer to the attached Figure 2 As shown, the execution reminder submodule also includes a content reminder grandson module. The historical storage submodule is also used to store historical nursing records corresponding to each historical patient. The historical nursing records include the nursing staff of each historical nursing measure, the actual nursing video and the nursing verification result. The content reminder grandson module is used to query the historical storage submodule according to the identity information of the target nursing staff and the target nursing content of the target patient to obtain the reference nursing content. Among them, the reference nursing content at least includes the target nursing content of the historical nursing records that simultaneously meet the following three conditions: the nursing staff of the reference nursing content is the target nursing staff, the target nursing content of the historical nursing measures is the same as the target nursing content of the target patient, and the nursing qualification rate corresponding to the reference nursing content is not greater than the qualification rate threshold. The content reminder grandson module sends the standard action sequence corresponding to the reference nursing content and the difference action set corresponding to each reference nursing content to the target nursing staff before the target execution time of the target nursing content.

[0086] It should be noted that the content reminder module can send the reference care content to the target user according to the reference content query request initiated by the user terminal of the target user, or send the reference care content to the target user according to the time point preset by the target user, or send it to the target user together with the second feedback information, or at other feasible time points. The pass rate threshold can be obtained in a preset manner.

[0087] In this embodiment, the content reminder module retrieves the nursing measures that the target nursing staff has completed in the historical nursing records that are the same as the target nursing content of the target patient, so as to obtain the degree of standardization of the target nursing staff when performing the nursing measures of the same nursing content, and personalized pushes the standard requirements of the target nursing content of the target patient and the actions of the target nursing staff that are inconsistent with the nursing standard requirements in the historical nursing records before the target nursing staff performs the target nursing content for the target patient, thereby assisting the nursing staff to improve the standardization of nursing operations and reduce the risk of poor nursing effects on the target patient due to non-compliant operations of the nursing staff.

[0088] In a feasible implementation, the execution reminder submodule also includes an expected prediction submodule. When the target nursing staff completes the target nursing measures for the target patient, the content verification submodule is also used to compare the corresponding nursing qualification rate with the qualification rate threshold. If the corresponding nursing qualification rate is not greater than the qualification rate threshold, it means that the nursing operation of the target nursing staff is not standardized, and the content verification submodule sends the corresponding actual action sequence to the expected prediction submodule.

[0089] In this embodiment, the expected prediction sub-module obtains the target nursing expectation and the target nursing diagnosis corresponding to the actual action sequence, and obtains the first-stage nursing content according to the actual action sequence of all executed nursing measures and the standard action sequences of all unexecuted nursing measures corresponding to the target patient for the target nursing expectation. It determines whether the target patient can achieve the corresponding target nursing expectation according to the first-stage nursing content and the initial characteristics of the target patient; if the target patient cannot achieve the corresponding target nursing expectation, it sends a first feedback message to the user of the target patient, and the first feedback message is used to remind the user to add remedial measures.

[0090] It should be noted that the initial characteristics of the patient include the characteristic information of the patient before performing the corresponding nursing measures to achieve the target nursing expectation. The characteristic information includes disease characteristics, physiological characteristics, psychological characteristics, living habit characteristics, and other characteristics, and of course it may not be limited to this. The physiological characteristics include vital signs and physical function status, and of course it may not be limited to this. The vital signs include heart rate, blood pressure, respiratory rate, body temperature, blood oxygen saturation, and breath sounds, and of course it may not be limited to this. The physical function status includes the patient's activity ability, whether they can take care of themselves, and whether they need auxiliary equipment, and of course it may not be limited to this.

[0091] It should be noted that the first feedback message can be sent to a designated user according to the user's permission scope. For example, when only the responsible nurse has the permission to create and modify the nursing plan, the content verification sub-module can send the first feedback message only to the client corresponding to the responsible nurse of the target patient.

[0092] In this embodiment, the expected prediction sub-module can construct a classification model for whether a patient can achieve the expected nursing result, that is, an expected risk prediction model, such as a recurrent neural network model, according to a machine learning model that can process temporal information. The input of the expected risk prediction model includes all action sequences taken to achieve the specified expected nursing result and the initial characteristics of the patient, and the output includes whether the patient achieves the expected nursing result.

[0093] In this embodiment, historical nursing records are classified according to expected nursing outcomes to obtain subsets of historical data. All nursing diagnoses and expected nursing outcomes corresponding to the historical nursing records in the same subset of historical data are the same. For example, the nursing diagnosis for all historical nursing records in the first subset of historical data is the first nursing diagnosis, and at the same time, the expected nursing outcome for all historical nursing records in this subset of historical data is the first expected nursing outcome. The corresponding expected risk prediction models are trained respectively according to each subset of historical data. Patient samples with actual nursing outcomes that meet the expected nursing outcomes are set as positive example samples, and patient samples with actual nursing outcomes that do not meet the expected nursing outcomes are set as negative example samples. The expected risk prediction models are trained according to a preset loss function.

[0094] In this embodiment, the expected prediction sub-module matches the target expected risk prediction model according to the target nursing expectation and the target nursing diagnosis corresponding to the actual action sequence. The first-stage nursing content and the initial characteristics of the target patient are input into the target expected risk prediction model to obtain a classification result. If the classification result of the target expected risk prediction model is a positive example, the first feedback information does not need to be sent to the responsible nurse; if the classification result of the target expected risk prediction model is a negative example, the first feedback information is sent to the responsible nurse. When there are irregularities in the nursing operations of the target nursing staff, the expected prediction sub-module determines whether the irregular operation will affect the nursing effect of the patient according to the physiological characteristics of the patient and the first-stage nursing content. If the irregular operation will cause the target patient to fail to achieve the target nursing expectation, the expected prediction module reminds the responsible user to add remedial measures to timely remedy the adverse effects of the irregular nursing operation.

[0095] In a feasible embodiment, the plan modification sub-module further includes an identification acquisition sub-module. The identification acquisition sub-module is used to obtain the nursing measures newly added by the responsible user and set a remedial identification or a non-remedial identification for the newly added nursing measures according to the first feedback information.

[0096] It should be noted that if the user terminal adds nursing measures according to the received first feedback information, the identification acquisition sub-module sets the remedial identification for the newly added nursing measures; if the user terminal adds nursing measures not according to the received first feedback information, the identification acquisition sub-module sets the non-remedial identification for the newly added nursing measures.

[0097] In this embodiment, the content modification sub-module includes a module for adding remedial measures. When the content verification sub-module sends the first feedback information to the user, it can also send the module interface for adding remedial measures to the user. The module for adding remedial measures is used to add remedial measures to the target care plan of the target patient. If the care measures are added by the user through the module for adding remedial measures, the identification acquisition sub-module sets the remedial identification for the newly added care measures; if the care measures are not added by the user through the module for adding remedial measures, the identification acquisition sub-module sets the non-remedial identification for the newly added care measures.

[0098] In this embodiment, the content retrieval sub-module matches historical care measures according to the target care diagnosis, the target care expectation, and the target measure type, and obtains all the care measures with the non-remedial identification in each successfully matched historical care measure to obtain the corresponding candidate care measures. For example, the successfully matched historical care measures include historical care measure 1 and historical care measure 2. Historical care measure 1 correspondingly includes target care content 1, target care content 2 (remedial measure for target care content 3), and target care content 3. Historical care measure 2 correspondingly includes target care content 2 (remedial measure for target care content 3), target care content 3, and target care content 4 (remedial measure for target care content 3). Therefore, target care content 1 and target care content 3 have non-remedial identifications, and target care content 2 and target care content 4 have remedial identifications. The content retrieval module extracts target care content 1 and target care content 3 from historical care measure 1 to obtain candidate care measure 1, and extracts target care content 3 from historical care measure 2 to obtain candidate care measure 2.

[0099] It should be noted that the care measures with remedial identifications are remedial measures added due to human factors that prevent the achievement of the expectation. For patients who have not yet started care, directly adding remedial measures to the care plan of the target patient will increase the risk of over-care and poor care effects for the patient. The identification acquisition sub-module sets remedial or non-remedial identifications for the newly added care measures during the target care plan process to assist nursing staff in judging the necessity of care measures during the creation of the care plan and reduce the probability of over-care.

[0100] In a feasible implementation, the plan generation sub-module is further configured to match historical nursing records according to each piece of nursing content in the candidate nursing measures selected by the user, and generate a set of remedial measures based on the nursing measures with the remedial identifier in all successfully matched historical nursing records. For example, if the user selects candidate nursing measure 1 as the target nursing measure for the target patient, the plan generation sub-module retrieves the historical nursing records for the target nursing content 1 in candidate nursing measure 1, and the set of remedial measures for the target nursing content 1 is an empty set. When retrieving the historical nursing records for the target nursing content 3 in candidate nursing measure 1, the set of remedial measures for the target nursing content 3 includes the set of remedial measures {target nursing content 2} and {target nursing content 2, target nursing content 4}. Therefore, the set of remedial measures generated by the plan generation sub-module includes the set of remedial measures 1 corresponding to the target nursing content 1 and the set of remedial measures 2 corresponding to the target nursing content 3, where the set of remedial measures 1 is an empty set, and the set of remedial measures 2 includes {target nursing content 2} and {target nursing content 2, target nursing content 4}.

[0101] In this implementation, when the user adds a remedial measure for the target patient, the content modification sub-module filters the set of remedial measures according to the corresponding target nursing content to obtain candidate remedial measures. For example, when the content verification sub-module determines that there are irregular operations in the execution of the target nursing content 3 by the target nurse, which may cause the target patient not to achieve the target nursing expectation, the content modification sub-module will filter out the set of remedial measures 2 from all sets of remedial measures. The candidate remedial measure 1 is {target nursing content 2}, and the candidate remedial measure 2 is {target nursing content 2, target nursing content 4}.

[0102] In this implementation, the second-stage nursing content is generated according to each candidate remedial measure and the first-stage nursing content respectively. Each piece of the second-stage nursing content and the initial characteristics of the target patient are input into the target expected risk prediction model to obtain a classification result. The candidate remedial measures with a positive classification result are used as recommended remedial measures. The content modification sub-module sends the recommended remedial measures while sending the first feedback information to the user to reduce the operation difficulty of the user adding a remedial measure.

[0103] In a feasible implementation, the execution tracking module further includes a personnel allocation sub-module. The personnel allocation sub-module further includes a patience acquisition sub-module, a skill threshold matching sub-module, a skill level acquisition sub-module, and a personnel matching sub-module.

[0104] It should be noted that the patience acquisition sub-module is used to obtain the patience level of the target patient based on the satisfaction evaluation results of the target patient for each candidate caregiver in all completed target care measures. The skill threshold matching sub-module is used to match the corresponding skill level threshold according to the patience level of the target patient. The skill level acquisition sub-module is used to obtain the skill level of each candidate caregiver in the corresponding target care content. The personnel matching sub-module is used to compare the skill level of each candidate caregiver with the corresponding skill level threshold respectively, and take the candidate caregiver whose skill level is not less than the corresponding skill level threshold as the recommended caregiver.

[0105] In this embodiment, the satisfaction evaluation results of the patient for the caregiver include satisfied, average, and dissatisfied. The patience level of the patient includes high patience level, medium patience level, and low patience level. The default initial value of the patience level of the patient is medium patience level. If the satisfaction evaluation result of the patient for the caregiver is satisfied, the patience level of the patient is set to be greater than the default initial value, that is, high patience level. If the satisfaction evaluation result of the patient for the caregiver is average, the patience level of the patient is set to the default initial value, that is, medium patience level. If the satisfaction evaluation result of the patient for the caregiver is dissatisfied, the patience level of the patient is set to be less than the default initial value, that is, low patience level.

[0106] In this embodiment, the skill level thresholds include high level, medium level, and low level. Patients with a high patience level correspond to the low level, patients with a medium patience level correspond to the medium level, and patients with a low patience level correspond to the high level. The skill levels of each candidate caregiver in completing the target care content are divided into high level, medium level, and low level. When the satisfaction evaluation result of the target patient for the candidate caregiver is dissatisfied, the patience level of the target patient is set to low patience level. Then, when allocating a caregiver to execute the target care content for the target patient, the skill level of the recommended caregiver is at least high level.

[0107] It should be noted that the candidate caregivers can be set in advance, or the caregivers participating in the duty according to the execution time of the care measures can be screened according to the caregiver's schedule. The skill level of each candidate caregiver in completing the target care content can be obtained according to the skill assessment results of the caregiver.

[0108] It should be noted that recommending a caregiver for the target patient based on the patience level of the target patient and the skill level of the candidate caregiver can not only ensure the cooperation degree of the target patient during the care process, but also facilitate arranging caregivers with a lower skill level for care operations when the patience level of the patient is relatively high, so as to specifically improve the care skill level of the caregiver.

[0109] Referring to the attached Figure 3 As shown, an embodiment of the present application provides an electronic device, including: a processor and a memory. The processor and the memory are interconnected and communicate with each other through a communication bus and / or other forms of connection mechanisms (not shown). The memory stores a computer program executable by the processor. When the computing device runs, the processor executes the computer program to execute the system in any optional implementation manner of the above embodiment.

[0110] An embodiment of the present application provides a storage medium. When the computer program is executed by the processor, it executes the system in any optional implementation manner of the above embodiment. Among them, the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (abbreviation: SRAM), electrically erasable programmable read-only memory (abbreviation: EEPROM), erasable programmable read-only memory (abbreviation: EPROM), programmable read-only memory (abbreviation: PROM), read-only memory (abbreviation: ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0111] In the embodiments provided by the present application, it should be understood that the disclosed system can be implemented in other ways. The system embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division. In actual implementation, there may be other division methods. For another example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some communication interfaces. The indirect couplings or communication connections of the systems or units can be in electrical, mechanical or other forms.

[0112] In addition, the units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0113] Furthermore, each functional module in the embodiments of the present application may be integrated together to form an independent part, or each module may exist alone, or two or more modules may be integrated to form an independent part.

[0114] Flowcharts are used herein to illustrate the steps of the methods of the embodiments of the present disclosure. It should be understood that the preceding or subsequent steps do not necessarily have to be carried out precisely in sequence. On the contrary, various steps may be evaluated in reverse order or simultaneously. At the same time, other operations may also be added to these processes.

[0115] Unless otherwise defined, all terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs. It should also be understood that terms such as those defined in a general dictionary should be interpreted as having a meaning consistent with their meaning in the context of the relevant art, and should not be interpreted in an idealized or overly formal sense unless expressly so defined herein.

[0116] The above provides a detailed introduction to a nursing decision support system based on an intelligent terminal. Specific examples are used herein to elaborate on the principles and implementation manners of the present application. The description of the above embodiments is only for the embodiments of the present application, and is only used to help understand a nursing decision support system based on an intelligent terminal of the present application, and is not used to limit the protection scope of the present application; at the same time, for those skilled in the art, various changes and modifications can be made to the present application. Any modification and equivalent replacement made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A nursing decision support system based on intelligent terminal, characterized in that: It includes plan creation module, execution tracking module and plan evaluation module; Wherein, the plan creation module includes a diagnosis acquisition submodule, an expectation acquisition submodule, a measure acquisition submodule and a plan generation submodule; the diagnosis acquisition submodule is used to obtain the target nursing diagnosis of the target patient according to the nursing diagnosis label selected by the user end; the expectation acquisition submodule is used to obtain the target nursing expectation of the target patient according to the expected nursing result selected by the user end under the corresponding target nursing diagnosis; the measure acquisition submodule is used to obtain the target nursing measure of the target patient according to the nursing measure selected by the user end under the corresponding target nursing expectation; the plan generation submodule is used to generate the target nursing plan of the target patient according to all the target nursing measures; Wherein, the execution tracking module is used to supervise the execution process of the target nursing plan; the execution tracking module includes a measure verification submodule, and the measure verification submodule also includes a content acquisition submodule, a standard retrieval submodule and a content verification submodule; the content acquisition submodule is used to obtain the actual nursing video of the target person from the start time of the measure to the end time of the measure, identify the skeleton key point sequence of each target person in the actual nursing video, and obtain the actual action sequence; the standard retrieval submodule is used to obtain the standard action sequence according to the target nursing content matching of the target nursing measure; the content verification submodule is used to obtain the nursing verification result according to the comparison result of the standard action sequence and the actual action sequence; The target personnel include target nursing staff and target patients, and the content verification module further includes extracting the action features of the actual action sequence to obtain the actual action features; wherein the action features include all paired distance features between skeleton key points; extracting the action features of the standard action sequence to obtain the standard action features; and obtaining the nursing verification result according to the action similarity between the actual action features and the standard action features; Among them, the plan evaluation module is used to obtain the actual nursing results corresponding to each of the target nursing expectations.

2. According to claim 1, a nursing decision support system based on an intelligent terminal is characterized in that: The nursing measures include measure types and nursing contents, and the diagnosis acquisition submodule also includes a type acquisition submodule, a content retrieval submodule and a content planning submodule; the type acquisition submodule is used to obtain the target measure type according to the measure type selected by the user terminal under the corresponding target nursing expectation; the content retrieval submodule is used to match historical nursing measures according to the target nursing diagnosis, the target nursing expectation and the target measure type to obtain candidate nursing measures; the content planning submodule is used to obtain the candidate nursing measures selected by the user terminal to obtain the target nursing measures.

3. According to claim 2, a nursing decision support system based on an intelligent terminal is characterized in that: The nursing verification result includes a difference action set and a nursing qualification rate; the execution tracking module includes an execution reminder submodule, the execution reminder submodule includes a content reminder submodule, and the content reminder submodule is used to query the historical nursing records of the target nursing staff according to the target nursing content to be performed on the target patient to obtain reference nursing content; The reference nursing content includes the target nursing content in the historical nursing record of the target nursing staff which is the same as the target nursing content of the target patient and corresponds to the target nursing content whose nursing qualification rate is not greater than the qualification rate threshold; Before the target execution time of the target nursing content, the content reminder grandchild module sends the standard action sequence corresponding to the reference nursing content and the corresponding difference action set to the target nursing staff.

4. According to claim 3, a nursing decision support system based on an intelligent terminal is characterized in that: The execution reminder submodule also includes an expectation prediction submodule; when the target nursing staff completes the target nursing measures for the target patient, the content verification submodule is also used to compare the corresponding nursing qualification rate with the qualification rate threshold. If the corresponding nursing qualification rate is not greater than the qualification rate threshold, the content verification submodule sends the corresponding actual action sequence to the expectation prediction submodule. The expectation prediction submodule obtains the first-stage nursing content based on all the actual action sequences executed and all the standard action sequences not executed by the target patient in the corresponding target nursing expectation, and judges whether the target patient can meet the corresponding target nursing expectation based on the first-stage nursing content and the initial characteristics of the target patient; If the target patient cannot achieve the corresponding target nursing expectation, first feedback information is sent to the user, and the first feedback information is used to remind the user to add new remedial measures.

5. According to claim 4, a nursing decision support system based on an intelligent terminal is characterized in that: The execution tracking module includes a plan modification submodule, and the plan modification submodule includes a content modification grandchild module and an identification acquisition grandchild module; The content modification submodule includes a module for adding new nursing measures to the target nursing plan for the target patient, and the identifier acquisition submodule sets a remedial identifier or a non-remedial identifier for the newly added nursing measures according to the first feedback information; the content retrieval submodule is also used to identify the candidate nursing measures corresponding to the nursing measures with the non-remedial identifier in each historical nursing measure that is successfully matched.

6. A nursing decision support system based on an intelligent terminal according to claim 5, characterized in that: The plan generation submodule is also used to retrieve historical nursing records according to each nursing content in the candidate nursing measures selected by the user, and generate a remedial measure set according to the successfully matched historical nursing records; when the user adds a remedial measure for the target patient, the content modification submodule filters the remedial measure set according to the corresponding target nursing content to obtain candidate remedial measures, and generates second-stage nursing content according to each candidate remedial measure and the first-stage nursing content, and judges whether the target patient can meet the corresponding target nursing expectations based on the second-stage nursing content and the initial characteristics of the target patient; if the target patient can meet the corresponding target nursing expectations, the corresponding candidate remedial measure will be used as a recommended remedial measure.

7. A nursing decision support system based on an intelligent terminal according to any one of claims 3 to 6, characterized in that: The calculation method of the nursing qualification rate includes: ； ； Wherein, C represents the nursing qualification rate, I represents the indicative function, X represents the input condition of the indicative function, s represents the action similarity, ω0 represents the weight of the action similarity, S1 represents the similarity threshold, and the value of S1 is greater than 0. i Indicates the value of the actual nursing content at the i-th preset indicator, q i represents the qualification standard of the i-th preset indicator, and n represents the number of preset indicators.

8. The nursing decision support system based on intelligent terminal according to claim 7 is characterized in that: The execution tracking module also includes a personnel allocation submodule, which also includes a patience acquisition submodule, a skill threshold matching submodule, a skill level acquisition submodule and a personnel matching submodule. The patience acquisition submodule is used to obtain the patience of the target patient based on the satisfaction evaluation results of the target patient on each candidate caregiver in all the executed target nursing measures. The skill threshold matching submodule is used to match the patience of the target patient to obtain the corresponding skill level threshold. The skill level acquisition submodule is used to obtain the skill level of each candidate caregiver in the corresponding target nursing content. The personnel matching submodule is used to compare the skill level of each candidate caregiver with the skill level threshold, and recommend the candidate caregivers whose skill levels are not less than the corresponding skill level threshold as recommended caregivers.

Images