Intervention mode construction method and system for pre-establishing medical care plan for elderly patient

Abstract

The invention discloses an intervention mode construction method and system for pre-establishing a medical care plan for an elderly patient, and relates to the technical field of medical assistance, and the method comprises the steps: obtaining ACP information of a target patient, extracting key information of the ACP information, and mapping each piece of key information into a willingness preference vector according to a value preference dimension; obtaining a candidate medical scheme of the target patient, extracting medical entities of the candidate medical scheme, and mapping each medical entity into a scheme preference vector according to the value preference dimension; generating a consistency comparison result of the willingness preference vector and the scheme preference vector; and the recommendation value of the candidate medical scheme is generated according to the consistency comparison result, so that the incidence relation between the ACP information and the medical scheme is established, the efficiency and transparency of doctor-patient joint decision making are improved, and the problem that ACP execution is difficult in elderly patient nursing is effectively solved.

Description

Technical Field

[0001] The present application relates to the field of medical communication assistance technology, and specifically to a method and system for constructing an intervention model for advance medical care planning for elderly patients. Background Art

[0002] Advance Care Planning (ACP) refers to a patient's expressed wishes regarding their medical care goals and treatment options for potential serious illness or disability, expressed in advance when they are able to make their own decisions. This written document is then recorded and archived. If a patient becomes disabled, healthcare professionals can develop a medical care plan based on the ACP and the actual clinical situation, thus reflecting the patient's independent wishes in medical decision-making.

[0003] Currently, the ACP still faces significant implementation challenges in its promotion and application. One key reason is that the ACP covers limited clinical scenarios. When clinical scenarios not covered by the ACP arise, medical staff still rely on subjective judgment to make medical decisions, reducing the effectiveness of the ACP. This is especially true for the elderly, whose overall physical function is gradually declining and the causes of disability are more complex and unpredictable. This further increases the difficulty of implementing the ACP among them.

[0004] However, the existing ACP management system mainly saves patients' ACP in the database, and medical staff retrieve and refer to it before making clinical decisions. It lacks guiding suggestions for medical staff to make medical plan decisions based on patients' ACP information in specific clinical scenarios. Summary of the Invention

[0005] The purpose of the present invention is to solve the technical problem of how to establish a corresponding relationship between ACP and medical plans that are not directly expressed. It provides a method and system for constructing an intervention model for advance medical care plans for elderly patients. By introducing the value preference dimension, the patient's key information in the ACP and the medical entities in the medical plan are quantified into corresponding preference vectors. According to the consistency comparison results of the intention preference vector and the plan preference vector, it is judged whether the medical plan can be used as the basis for expanding the ACP information, thereby establishing an association between the ACP information and the medical plan, effectively improving the problem of difficulty in implementing ACP in the care of elderly patients.

[0006] According to a first aspect of the present invention, the present invention claims protection for a method for constructing an intervention model for advance medical care planning for elderly patients, comprising: Obtain the target patient's ACP information, extract the key information of the ACP information, and map each key information into a willingness preference vector based on the value preference dimension; Obtain candidate medical plans for the target patient, extract the medical entities of the candidate medical plans, and map each medical entity into a plan preference vector based on the value preference dimension; Generate consistency comparison results between the intention preference vector and the solution preference vector; Generate recommended values ​​for candidate medical options based on the consistency comparison results.

[0007] Preferably, after obtaining the ACP information of the target patient, the method further includes: if the ACP information contains at least 2 versions, generating a willingness preference vector based on the latest version of the ACP information.

[0008] Preferably, generating the recommendation value of the candidate medical plan also includes: Obtain the fluctuation range of the target patient in each value preference dimension based on the willingness preference vector of each ACP information; Generate stable weights corresponding to value preference dimensions based on the fluctuation range; The weighted result of the similarity between the willingness preference vector and the scheme preference vector is calculated according to the stable weight to obtain the recommendation value.

[0009] Preferably, mapping key information into willingness preference vectors also includes matching each key information to the medical entity library separately. If the matching fails, matching the value preference interval according to the corresponding key information; when generating consistency comparison results, the corresponding consistency comparison results are generated based on whether the value of the scheme preference vector in the corresponding value preference dimension belongs to the corresponding value preference interval.

[0010] Preferably, matching the value preference interval according to the corresponding key information also includes: Determine whether there is key information that successfully matches the medical entity database. If so, use the value of each value preference dimension corresponding to the key information that successfully matches the medical entity database as the reference value, and obtain the value preference interval based on the interval whose distance from the reference value does not exceed the value distance threshold.

[0011] Preferably, if there are at least two key information that are successfully matched with the medical entity library, after obtaining all reference values, it also includes: taking the interval whose distance from each reference value does not exceed the value distance threshold as the reference interval, and taking the intersection of all reference intervals as the value preference interval.

[0012] Preferably, the method further includes: comparing the scheme preference vector corresponding to each medical entity with all willingness preference vectors respectively, taking the comparison result with the highest consistency as the consistency comparison result of the corresponding medical entity, and generating a recommendation value for the candidate medical scheme based on the consistency comparison results of all medical entities.

[0013] According to a second aspect of the present invention, the present invention claims protection for a system for constructing an intervention model for advance medical care planning for elderly patients, comprising: An acquisition module is used to obtain the ACP information and candidate medical plans of the target patient; An extraction module, used to extract key information of ACP information and medical entities of candidate medical solutions; A mapping module, configured to map each key information into a willingness preference vector and each medical entity into a scheme preference vector according to the value preference dimension; A comparison module is used to generate a consistency comparison result between the intention preference vector and the solution preference vector; The generation module is used to generate recommendation values ​​of candidate medical solutions based on the consistency comparison results.

[0014] Preferably, the acquisition module further comprises: if the ACP information includes at least two versions, generating a willingness preference vector based on the latest version of the ACP information.

[0015] Preferably, the generating module further includes: Obtain the fluctuation range of the target patient in each value preference dimension based on the willingness preference vector of each ACP information; Generate stable weights corresponding to value preference dimensions based on the fluctuation range; The weighted result of the similarity between the willingness preference vector and the scheme preference vector is calculated according to the stable weight to obtain the recommendation value.

[0016] Preferably, the mapping module also includes matching each key information with the medical entity library respectively. If the matching fails, matching the value preference interval according to the corresponding key information; when generating the consistency comparison result, the corresponding consistency comparison result is generated according to whether the value of the scheme preference vector in the corresponding value preference dimension belongs to the corresponding value preference interval.

[0017] Preferably, the mapping module also includes: determining whether there is key information that successfully matches the medical entity library; if so, taking the value of each value preference dimension corresponding to the key information that successfully matches the medical entity library as a reference value, and obtaining the value preference interval based on the interval whose distance from the reference value does not exceed the value distance threshold.

[0018] Preferably, if there are at least two key information that are successfully matched with the medical entity library, the mapping module further includes: taking an interval whose distance from each reference value does not exceed a value distance threshold as a reference interval, and taking the intersection of all reference intervals as a value preference interval.

[0019] Preferably, the comparison module also includes comparing the scheme preference vector corresponding to each medical entity with all the willingness preference vectors respectively, and taking the comparison result with the highest consistency as the consistency comparison result of the corresponding medical entity; the generation module also includes generating a recommendation value for the candidate medical scheme based on the consistency comparison results of all medical entities.

[0020] According to the third aspect of the present invention, the present invention seeks to protect a device for constructing an intervention model for advance medical care planning for elderly patients, comprising a processor and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the steps in the method described in the first aspect above are executed.

[0021] This application has the following beneficial effects: 1. By introducing the value preference dimension, the patient's key information in the ACP and the medical entities in the medical plan are quantified into corresponding preference vectors. The consistency comparison between the intention preference vector and the plan preference vector determines whether the medical plan can be used as the basis for expanding the ACP information. Higher consistency indicates that the medical plan aligns with the patient's value preferences expressed in the ACP information. Even if the medical plan contains components not covered by the ACP information, it can still be recommended as the medical plan for that patient. Low consistency indicates that the medical plan does not align with the patient's value preferences expressed in the ACP information, necessitating corresponding adjustments to the medical plan or focusing on the dimensions with significant discrepancies with the patient and / or their family. This method establishes a correlation between ACP information and medical plans. Even when the medical plan contains content that does not align with the patient's direct expression in the ACP information, the patient's true intentions can be captured by quantifying the patient's value preferences. In clinical applications, this method enables medical staff to quickly focus on the treatment path that best aligns with the patient's value preferences, improving the efficiency and transparency of joint decision-making between doctors and patients, enhancing the operability of ACP in real-world clinical scenarios, and effectively alleviating the difficulties in implementing ACP in the care of elderly patients.

[0022] 2. Conduct dimension-level statistical analysis based on the historical version of the willingness preference vector, and set a stable weight for each value preference dimension according to the fluctuation amplitude. The higher the stable weight, the higher the confidence level of the corresponding dimension value; the lower the stable weight, the lower the confidence level of the corresponding dimension value. This takes into account both the historical stable value preferences and the latest value preferences of the target patients, reduces the impact of accidental factors in the latest value preferences on the value preferences of the target patients, and improves the accuracy.

[0023] 3. The interval-based mapping method of value preference dimensions not only allows tacit expression to obtain quantitative results, but also provides more flexible compatibility judgments in the consistency comparison process, making the matching results between candidate options and patient wishes more in line with clinical reality.

[0024] 4. The value corresponding to the successfully matched medical entity is used as a reference value to generate the corresponding value preference interval. This method of generating the value preference interval can adapt to the individual preference characteristics of the target patient, that is, the mapping representation of the value preference intention corpus to the value preference dimension and the mapping standard of the medical entity to the value preference dimension can adapt to the individual differences of the patient and improve clinical interpretability. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0026] Figure 1 A flowchart of a method for constructing an advance medical care plan intervention model for elderly patients according to an embodiment of the present application; Figure 2 This is a flow chart of a method for generating a recommendation value according to an embodiment of the present application; Figure 3 This is a flow chart for calculating the willingness preference vector involved in the embodiment of the present application; Figure 4 A schematic diagram of a system for constructing an advance medical care plan intervention model for elderly patients according to an embodiment of the present application; Figure 5 This is a schematic diagram of the structure of an electronic device involved in an embodiment of the present application. DETAILED DESCRIPTION

[0027] The present invention provides a method and system for constructing an advance care planning intervention model for elderly patients. To make the above-mentioned objects, features, and advantages of this application more readily apparent, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of this application. It should be understood that the described embodiments represent only a portion of the embodiments of this application, and not all of them. The components of the embodiments of this application generally described and illustrated in the drawings herein may be arranged and designed in a variety of different configurations. Therefore, reference to the terms "one embodiment," "some embodiments," "implementation," "embodiment," "illustrative embodiment," "example," "specific example," or "some examples," etc., in the following detailed description of the embodiments of this application provided in the drawings, is not intended to limit the scope of the claimed application, but merely indicates that the specific features, structures, or characteristics described in conjunction with such embodiment or example are included in at least one embodiment or example of the present invention. Furthermore, the specific features, structures, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. All other embodiments derived by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0028] It should be noted that similar reference numerals and letters represent similar items in the following figures. Therefore, once an item is defined in one figure, it does not need to be further defined or explained in subsequent figures. At the same time, in the description of this application, relational terms such as "first" and "second" 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.

[0029] According to the first aspect of the present invention, the present invention claims a method for constructing an intervention model for advance medical care planning for elderly patients, with reference to the attached Figure 1 As shown, including: S1: Obtain the ACP information of the target patient.

[0030] It should be noted that ACP information refers to the target patient's wishes and preferences regarding future medical care when they have the ability to make independent decisions. This information can be collected through questionnaires, text records, voice recordings, and other methods.

[0031] In this embodiment, the ACP information recorded in text form is taken as an example for detailed description. ACP information in other forms can be converted into text form for processing through voice recognition or other feasible processing methods.

[0032] In this embodiment, if the ACP information of the target patient contains at least two versions, the latest version is selected as the basis for generating the willingness preference vector according to the collection time of each ACP information to ensure that the willingness preference vector conforms to the latest value preference of the target patient under different disease courses.

[0033] S2: Extract key information of ACP information.

[0034] It should be noted that the key information refers to sentences or phrases extracted from the ACP information that directly or indirectly express the willingness for medical care, such as "willing to use a ventilator" extracted from text or voice recordings.

[0035] In this embodiment, step S2 also includes: S21, performing text cleaning and word segmentation processing on each of the ACP information. S22, extracting key sentences and / or phrases from the ACP information to remove text content with low relevance. Natural language processing technology can be used to map each of the ACPs into a word vector and then input it into a pre-trained text extraction model to output the key information. The text extraction model can be a long short-term memory network model, a convolutional neural network model, a model built based on the Transformer architecture, etc.

[0036] It should be noted that step S22 also includes: pre-collecting texts related to and unrelated to medical care intentions as sample data. The sample data is pre-processed by segmenting, removing stop words, and labeling relevance to obtain a training data set. The training data set is divided into a training set and a test set. Each pre-processed sample data in the training set is input into a pre-built text extraction model, and the value of the loss function is obtained according to the gap between the corresponding relevance labeling and the output of the text extraction model. The text extraction model is trained by minimizing the loss function. The text extraction model is evaluated using the test set, and the model is optimized according to the evaluation results, and the hyperparameter values ​​and structure of the model are adjusted to obtain the final text extraction model.

[0037] It should be noted that the key information extracted may be one or more, and this embodiment does not limit the number of the key information.

[0038] S3: Map each key information into a corresponding preference vector according to the value preference dimension to obtain the willingness preference vector.

[0039] It should be noted that the value preference dimensions are used to describe a multi-dimensional set of aspects that patients value in their medical care decisions, such as life extension, financial burden, pain management, and acceptability of invasiveness. The value preference dimensions can be obtained by presetting. The values ​​corresponding to each value preference dimension can also be obtained by presetting.

[0040] In this embodiment, the willingness preference vector refers to the vector representation of the key information quantified on the value preference dimension, which is recorded as Where d represents the number of dimensions representing the value preference dimension. When the value of the corresponding dimension is -1, it indicates no preference; when the value of the corresponding dimension is 1, it indicates preference; when the value of the corresponding dimension is 0, it indicates neutrality or no mention.

[0041] In this embodiment, step S3 further includes the following sub-steps: S31. Determine the information type of each key information, that is, determine whether the key information corresponds to the medical entity's willingness to accept or the value preference. If the key information corresponds to the medical entity's willingness to accept, proceed to step S32; if the key information corresponds to the value preference, proceed to step S34.

[0042] It should be noted that the value mapping rules of "medical entity-value dimension" can be established in advance. According to the pre-set value preference dimension and scoring criteria, clinical experts are organized to score all medical entities of the medical institution. For example, in the "life extension" dimension, 0 points means neutral and / or no mention, and 1 point means significantly extended survival time; in the "pain management" dimension, negative values ​​mean increased pain or discomfort, and positive values ​​mean pain relief; in the "economic burden" dimension, negative values ​​mean high medical costs, and positive values ​​mean low medical costs; in the "acceptability of invasiveness" dimension, negative values ​​mean low acceptability, and positive values ​​mean high acceptability.

[0043] In this embodiment, the key information is matched with all medical entities in the pre-built value mapping rules through natural language processing and other methods. If the match is successful, the corresponding key information is regarded as the willingness to accept the medical entity. If the match fails, the corresponding key information is regarded as the value preference willingness.

[0044] It should be noted that determining the type of the critical information may also be accomplished using other feasible methods, such as pre-building a corpus that includes both the willingness to accept medical entities and the value preference corpus, pre-training a neural network model using the corpus, and using the neural network model to distinguish the types of the critical information. This application does not further limit the method for determining the type of critical information.

[0045] S32. The value of each value preference dimension corresponding to the medical entity that successfully matches the key information is used as a reference value.

[0046] S33. Extract from the key information whether the patient accepts or rejects the corresponding medical entity. If the key information indicates that the target patient accepts the corresponding medical entity, the key information may directly use the reference value as the score for the corresponding dimension. If the key information indicates that the target patient rejects the corresponding medical entity, the key information may directly use the opposite of the reference value as the score for the corresponding dimension. The willingness preference vector is obtained based on the values ​​of the key information across all value preference dimensions.

[0047] For example, if the score of the medical entity "mechanical ventilation" in "pain management" is -0.5, the key information "willing to accept mechanical ventilation" corresponds to a score of -0.5 in "pain management", and the key information "refusal of mechanical ventilation" corresponds to a score of 0.5 in "pain management".

[0048] S34. Manual labeling can be used to set standard values ​​for each value preference dimension for each sample in the pre-constructed corpus of value preferences and intentions. For example, the sample expectations "hope for less pain," "not too much torture," and "want comfort" are marked with a value of 1 in the "pain management" dimension, and the values ​​of the remaining dimensions are marked as 0. A neural network model is pre-trained using the corpus of value preferences and intentions, and the neural network model maps the key information into the intention preference vector.

[0049] S4: Obtain candidate medical plans for the target patient, where the candidate medical plans refer to the medical plans to be selected by medical staff.

[0050] S5: Extracting medical entities of candidate medical plans, wherein the medical entities include medical plans and / or operation units that can be adopted by the patient.

[0051] It should be noted that the medical entity can be obtained in a pre-set manner, such as optional treatment options such as "endotracheal intubation", "short-term mechanical ventilation", "long-term mechanical ventilation" and "renal dialysis treatment", and available medical equipment such as "ventilator" can be used as the medical entity.

[0052] S6: Generate the preference vector corresponding to each medical entity based on the matching result of the value mapping rule of "medical entity-value dimension" to obtain the solution preference vector, which is denoted as V PALN .

[0053] S7: Generate consistency comparison results between the willingness preference vector and the solution preference vector.

[0054] It should be noted that the vector similarity between the intention preference vector and the solution preference vector may be calculated based on methods such as cosine similarity and Euclidean distance similarity to obtain the consistency comparison result.

[0055] It should be noted that the consistency comparison results can be fed back to medical staff and used as a basis for adjusting the medical plan and as a key point for communication with the target patient's family. For example, if the target patient's value preferences for the ACP information conflict with the value preferences for the medical plan, medical staff can adjust the medical plan accordingly or focus on the conflicting value preferences in communication with the patient's family.

[0056] In this embodiment, if the number of the extracted key information and the number of the medical entities are both 1, the recommendation value is directly calculated based on the consistency comparison result between the intention preference vector and the solution preference vector.

[0057] In this embodiment, if the number of key information extracted is 1, but the number of medical entities extracted is greater than 1, the scheme preference vector corresponding to each medical entity is compared with the willingness preference vector respectively, and the recommendation value of the candidate medical scheme is generated based on the consistency comparison results of all medical entities. For example, the recommendation value of the candidate medical scheme can be obtained based on the minimum value, maximum value or average value of the similarities corresponding to all medical entities.

[0058] In this embodiment, if the number of extracted key information is greater than 1 and the number of extracted medical entities is greater than or equal to 1, the plan preference vector corresponding to each medical entity is compared with all the intention preference vectors. The comparison result with the highest consistency is taken as the consistency comparison result for the corresponding medical entity. Based on the consistency comparison results of all medical entities, a recommendation value for the candidate medical plan is generated. For example, if the extracted key information includes A1 and A2, the extracted medical entity is B, the similarity between A1 and B is a1, and the similarity between A2 and B is a2, then the consistency comparison result for medical entity B is recorded as max(a1, a2), thereby reducing the risk of bias amplification due to local inconsistency.

[0059] S8: Generate a recommendation value for the candidate medical solution based on the consistency comparison result. The higher the consistency between the intention preference vector and the solution preference vector, the greater the recommendation value for the corresponding candidate medical solution; the lower the consistency between the intention preference vector and the solution preference vector, the smaller the recommendation value for the corresponding candidate medical solution.

[0060] It should be noted that the recommended values ​​can be directly displayed to medical staff and / or patients' families to provide decision-making assistance and improve communication efficiency. Medical plans can also be screened based on the ranking results of the recommended values ​​to eliminate medical communication plans that are not sufficiently consistent with the target patient's own preferences, thereby improving patient autonomy.

[0061] In a feasible embodiment, refer to the attached Figure 2 As shown, step S8 also includes the following sub-steps: S81. Obtain the fluctuation range of the target patient in each value preference dimension based on the willingness preference vector of each ACP information.

[0062] It should be noted that the fluctuation range of the value preference dimension is used to describe the degree of dispersion of the target patient's preference value for the value preference dimension in different versions of the ACP information relative to its mean value. It can be represented by statistics such as variance and standard deviation. The larger the fluctuation range, the more unstable the target patient's preference value for the value preference dimension and the easier it is to change; the smaller the fluctuation range, the more stable the target patient's preference value for the value preference dimension and the harder it is to change.

[0063] In this embodiment, all the ACP information of the target patient is recorded as ACP in chronological order. t , where t=1,…,i,…,T. ACP i Indicates the ACP information of the i-th version, ACP T The value preference dimension is denoted as K, where K = 1, ..., k, ..., d. i The value in the kth dimension is denoted as v i，k The values ​​of all the ACPs in the kth dimension form a time series value V k , where Vj={v 1，k ,…,v i，k ,…,v T，k}.

[0064] In this embodiment, the standard deviation is used as an example to explain in detail the calculation standard of the fluctuation range. The calculation method is: ; ; in, represents the mean value of the target patient's ACP in the kth dimension.

[0065] S82. Generate the stable weight of the corresponding value preference dimension according to the fluctuation range. The stable weight is recorded as S, S={s1,…,s k ,…,s d}. Among them, s k represents the stable weight of the k-th dimension.

[0066] It should be noted that the value of the stabilization weight is the result of mapping the fluctuation amplitude of the corresponding dimension into a preset weight value range. The value of the stabilization weight is inversely proportional to the value of the fluctuation amplitude. The smaller the fluctuation amplitude, the larger the value of the stabilization weight; the larger the fluctuation amplitude, the larger the value of the stabilization weight. The fluctuation amplitude can be converted into the stabilization weight by serving as the input of a mapping method such as linear mapping, parametric mapping, or Sigmoid mapping.

[0067] In this embodiment, the range of the stable weight is [0, 1], and the sum of all the stable weights is 1.

[0068] S83. Calculate the weighted result of the similarity between the willingness preference vector and the solution preference vector based on the stable weight to obtain a recommendation value.

[0069] In this embodiment, the method for calculating the recommended value Rec includes: ; Wherein, Dk represents the difference between the value of the willingness preference vector in the kth dimension and the value of the solution preference vector in the kth dimension.

[0070] In this embodiment, when the target patient has only one version of the ACP information, s k The default value can be set to .

[0071] In a feasible embodiment, refer to the attached Figure 3 As shown, step S31 matches each key information with the medical entity library respectively. If the match fails, continue to step S34. Step S34 matches the value preference interval according to the corresponding key information; in step S7, the corresponding consistency comparison result is generated according to whether the value of the solution preference vector in the corresponding value preference dimension belongs to the corresponding value preference interval. If the value of the solution preference vector in the corresponding value preference dimension does not belong to the value preference interval, the comparison result of the corresponding value preference dimension is marked as dissimilar, that is, the corresponding similarity value is 0; if the value of the solution preference vector in the corresponding value preference dimension belongs to the value preference interval, the comparison result of the corresponding value preference dimension is marked as similar, that is, the corresponding similarity value is 1.

[0072] It should be noted that the medical entity library is obtained based on the preset of all the medical entities of the medical institution, and can also be obtained by integrating all the medical entities in the value mapping rule of "medical entity-value dimension".

[0073] It should be noted that a corresponding value range can be pre-set for each dimension of each sample corpus in the value preference intention corpus. The feature vector of the key information and the feature vector of each sample corpus can be extracted using a neural network model. The cosine similarity between the feature vector of the key information and the feature vector of each sample corpus can be calculated. The value range corresponding to the sample corpus with the largest cosine similarity can be used as the value preference range of the key information. Of course, other feasible implementation methods are also possible.

[0074] In this embodiment, step S34 also includes: S341, judging whether there is key information that successfully matches the medical entity database. If so, continue to step S342; if not, continue to step S344. S342, taking the value of each value preference dimension corresponding to the key information that successfully matches the medical entity database as a reference value, recorded as R k S343: Obtain a value preference interval based on the interval whose distance from the reference value does not exceed the value distance threshold, that is, the value preference interval can be identified as [R k -δ, R k+δ]. Wherein, δ represents the value distance threshold, and its specific value can be obtained by presetting. S344. Preset a corresponding value interval for each sample corpus, extract the feature vector of the key information and the feature vector of each sample corpus in the pre-collected value preference intention corpus through a neural network model, calculate the cosine similarity between the feature vector of the key information and the feature vector of each sample corpus, and take the value interval corresponding to the sample corpus with the largest cosine similarity as the value preference interval of the key information.

[0075] It should be noted that the value range corresponding to each sample corpus in the value preference intention corpus can be obtained by manual pre-setting.

[0076] It should be noted that the value corresponding to the successfully matched medical entity is used as a reference value to generate the corresponding value preference interval. This method of generating the value preference interval can adapt to the individual preference characteristics of the target patient, that is, the mapping representation of the corpus of value preference intention to the value preference dimension and the mapping standard of the medical entity to the value preference dimension can adapt to the individual differences of the patient and improve clinical interpretability.

[0077] In this embodiment, if there are at least two key information that are successfully matched with the medical entity library, after obtaining all reference values, it also includes: taking the interval whose distance from each reference value does not exceed the value distance threshold as the reference interval, and taking the intersection of all reference intervals as the value preference interval, so as to achieve the extraction of preference intervals that are unanimously supported by multiple parties from the mapping of multiple specific medical entities, thereby more strictly "consensus screening" to obtain the same intentions of the target patients on this dimension.

[0078] According to the second aspect of the present invention, the present invention claims a system for constructing an intervention model for an advance medical care plan for elderly patients, Figure 4 As shown, including: An acquisition module is used to obtain the ACP information and candidate medical plans of the target patient; An extraction module, used to extract key information of ACP information and medical entities of candidate medical solutions; A mapping module, configured to map each key information into a willingness preference vector and each medical entity into a scheme preference vector according to the value preference dimension; A comparison module is used to generate a consistency comparison result between the intention preference vector and the solution preference vector; The generation module is used to generate recommendation values ​​of candidate medical solutions based on the consistency comparison results.

[0079] In a feasible implementation, the acquisition module further includes: if the ACP information includes at least two versions, generating a willingness preference vector based on the latest version of the ACP information.

[0080] In a feasible implementation, the generation module further includes: Obtain the fluctuation range of the target patient in each value preference dimension based on the willingness preference vector of each ACP information; Generate stable weights corresponding to value preference dimensions based on the fluctuation range; The weighted result of the similarity between the willingness preference vector and the scheme preference vector is calculated according to the stable weight to obtain the recommendation value.

[0081] In a feasible implementation, the mapping module also includes matching each key information with the medical entity library separately. If the matching fails, matching the value preference interval according to the corresponding key information; when generating the consistency comparison result, the corresponding consistency comparison result is generated based on whether the value of the scheme preference vector in the corresponding value preference dimension belongs to the corresponding value preference interval.

[0082] In a feasible implementation, the mapping module also includes: determining whether there is key information that successfully matches the medical entity library; if so, taking the value of each value preference dimension corresponding to the key information that successfully matches the medical entity library as a reference value, and obtaining the value preference interval based on an interval whose distance from the reference value does not exceed a value distance threshold.

[0083] In a feasible implementation, if there are at least two key information that are successfully matched with the medical entity library, the mapping module further includes: taking an interval whose distance from each reference value does not exceed a value distance threshold as a reference interval, and taking the intersection of all reference intervals as a value preference interval.

[0084] In a feasible embodiment, the comparison module also includes comparing the scheme preference vector corresponding to each medical entity with all the willingness preference vectors respectively, and taking the comparison result with the highest consistency as the consistency comparison result of the corresponding medical entity; the generation module also includes generating a recommendation value of the candidate medical scheme based on the consistency comparison results of all medical entities.

[0085] Refer to the attached Figure 5 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, and when the computing device is running, the processor executes the computer program to execute the system in any optional implementation mode of the above embodiment.

[0086] The embodiment of the present application provides a storage medium, and when the computer program is executed by a processor, the system of any optional implementation of the above embodiment is executed. 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 (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0087] In the embodiments provided in this application, it should be understood that the disclosed system can be implemented in other ways. The system embodiments described above are merely schematic. For example, the division of the modules is only a logical function division, and can be implemented in another way. For 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 mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interface, indirect coupling or communication connection of the system or unit, which can be electrical, mechanical or other forms.

[0088] In addition, the units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0089] Furthermore, the functional modules in the various embodiments of the present application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0090] 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 following steps do not necessarily need to be performed in exact order. Instead, the various steps may be evaluated in reverse order or simultaneously. Furthermore, other operations may be added to these processes.

[0091] Unless otherwise defined, all terms used herein have the same meaning as commonly understood by those skilled in the art to which the present disclosure belongs. It should also be understood that terms such as those defined in common dictionaries 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 highly formal sense unless expressly defined as such herein.

[0092] The above is a detailed introduction to the method and system for constructing an intervention model for pre-established medical care plans for elderly patients. Specific examples are used in this article to illustrate the principles and implementation methods of this application. The description of the above embodiments is only an embodiment of this application and is only used to help understand the method and system for constructing an intervention model for pre-established medical care plans for elderly patients of this application, and is not used to limit the scope of protection of this application. At the same time, for those skilled in the art, this application can be modified and changed in various ways. Any modifications and equivalent substitutions made within the spirit and principles of this application should be included in the scope of protection of this application.

Claims

1. A method for constructing an intervention model for advance medical care planning for elderly patients, characterized in that: include: Obtain the target patient's ACP information, extract the key information of the ACP information, and map each key information into a willingness preference vector based on the value preference dimension; Obtain candidate medical plans for the target patient, extract the medical entities of the candidate medical plans, and map each medical entity into a plan preference vector based on the value preference dimension; Generate consistency comparison results between the intention preference vector and the solution preference vector; Generate recommended values ​​for candidate medical options based on the consistency comparison results.

2. The method for constructing an intervention model for advance medical care planning for elderly patients according to claim 1, characterized in that: After obtaining the ACP information of the target patient, the method further includes: if the ACP information includes at least two versions, generating a willingness preference vector based on the latest version of the ACP information.

3. The method for constructing an intervention model for advance medical care planning for elderly patients according to claim 2, characterized in that: The recommended values ​​for generating candidate medical treatment options also include: Obtain the fluctuation range of the target patient in each value preference dimension based on the willingness preference vector of each ACP information; Generate stable weights corresponding to value preference dimensions based on the fluctuation range; The weighted result of the similarity between the willingness preference vector and the scheme preference vector is calculated according to the stable weight to obtain the recommendation value.

4. The method for constructing an intervention model for advance medical care planning for elderly patients according to claim 1, characterized in that: Mapping key information into willingness preference vectors also includes matching each key information to the medical entity database. If the matching fails, matching the value preference interval according to the corresponding key information; when generating consistency comparison results, the corresponding consistency comparison results are generated based on whether the value of the scheme preference vector in the corresponding value preference dimension belongs to the corresponding value preference interval.

5. The method for constructing an intervention model for advance medical care planning for elderly patients according to claim 4, characterized in that: Matching the value preference interval based on the corresponding key information also includes: Determine whether there is key information that successfully matches the medical entity database. If so, use the value of each value preference dimension corresponding to the key information that successfully matches the medical entity database as the reference value, and obtain the value preference interval based on the interval whose distance from the reference value does not exceed the value distance threshold.

6. A method for constructing an intervention model for advance medical care planning for elderly patients according to claim 5, characterized in that: If there are at least two key information that are successfully matched with the medical entity database, after obtaining all reference values, it also includes: taking the interval whose distance from each reference value does not exceed the value distance threshold as the reference interval, and taking the intersection of all reference intervals as the value preference interval.

7. A method for constructing an intervention model for advance medical care planning for elderly patients according to any one of claims 1 to 6, characterized in that: The method further includes: comparing the scheme preference vector corresponding to each medical entity with all intention preference vectors respectively, taking the comparison result with the highest consistency as the consistency comparison result of the corresponding medical entity, and generating a recommendation value of the candidate medical scheme based on the consistency comparison results of all medical entities.

8. A system for constructing an intervention model for advance medical care planning for elderly patients, characterized by: include: An acquisition module is used to obtain the ACP information and candidate medical plans of the target patient; An extraction module, used to extract key information of ACP information and medical entities of candidate medical solutions; A mapping module, configured to map each key information into a willingness preference vector and each medical entity into a scheme preference vector according to the value preference dimension; A comparison module is used to generate a consistency comparison result between the intention preference vector and the solution preference vector; The generation module is used to generate recommendation values ​​of candidate medical solutions based on the consistency comparison results.

Images