A method and device for determining a user, an electronic device and a storage medium

Abstract

The application discloses a kind of methods for determining reception user, device, electronic equipment and storage medium, wherein, the method includes: according to the basic information and illness description information of target medical user, determine the to-be-processed medical user cluster associated with target medical user;Determine at least one historical reception user associated with the to-be-processed user cluster, and according to the reception record information of each historical reception user and at least one dimension reference attribute edited in advance, determine the to-be-used matrix corresponding to the to-be-processed medical user cluster;According to the historical medical attribute of target medical user, process to-be-used matrix, determine the target characteristic value of each historical reception user relative to target medical user;The historical reception user with target characteristic value higher than preset characteristic value threshold is used as the target reception user of target medical user. It realizes to medical user more matched reception user is recommended, improves the efficiency of effect that medical user determines reception user.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method, apparatus, electronic device and storage medium for determining a patient receiving a medical appointment. Background Technology

[0002] With the continuous improvement of the medical system, significant progress has been made in the construction of medical information technology, making it more convenient and efficient for patients to seek medical treatment.

[0003] Currently, when patients need medical care, they typically input relevant symptom information into various medical platforms. These platforms then recommend suitable doctors based on the patient's condition, such as recommending doctors with good historical patient reviews that match the patient's symptoms. However, these platforms rely on historical patient reviews of doctors to determine patient satisfaction. This evaluation method is rather general and heavily influenced by subjectivity. Therefore, when recommending doctors who have satisfied previous patients, there may be instances where the recommended doctor is not a perfect match for the individual patient.

[0004] To recommend more suitable doctors to patients, the methods for determining which doctor to see need to be improved. Summary of the Invention

[0005] This invention provides a method, apparatus, electronic device, and storage medium for determining a patient to receive treatment, thereby improving the efficiency of determining a more suitable patient to receive treatment.

[0006] In a first aspect, embodiments of the present invention provide a method for determining a patient receiving a medical appointment, comprising:

[0007] Based on the basic information and symptom description of the target patient, a cluster of patients awaiting treatment associated with the target patient is identified; wherein, the cluster of patients awaiting treatment includes multiple historical patients.

[0008] Identify at least one historical patient who is associated with the cluster of patients to be processed, and determine the matrix to be used corresponding to the cluster of patients to be processed based on the patient record information of each historical patient and at least one pre-edited dimension reference attribute; wherein, each element in the matrix to be used represents the evaluation attribute value of each historical patient to the historical patient, and the elements in the same column of the matrix to be used correspond to the same patient.

[0009] Based on the historical medical visit attributes of the target patient, the matrix to be used is processed to determine the target feature value of each historical patient relative to the target patient.

[0010] Historical patients whose target feature values ​​are higher than a preset feature value threshold are designated as target patients of the target patient.

[0011] Secondly, embodiments of the present invention also provide a device for determining a patient receiving treatment, comprising:

[0012] The pending medical user cluster determination module is used to determine the pending medical user cluster associated with the target medical user based on the target medical user's basic information and symptom description information; wherein, the pending medical user cluster includes multiple historical medical users;

[0013] The module for determining the matrix to be used is used to determine at least one historical patient who is associated with the cluster of patients to be processed, and to determine the matrix to be used corresponding to the cluster of patients to be processed based on the patient record information of each historical patient and at least one pre-edited dimension reference attribute; wherein, each element in the matrix to be used represents the evaluation attribute value of each historical patient to the historical patient, and the elements in the same column of the matrix to be used correspond to the same patient.

[0014] The target feature value determination module is used to process the matrix to be used based on the historical medical visit attributes of the target medical user, and determine the target feature value of each historical medical user relative to the target medical user.

[0015] The target patient identification module is used to identify historical patients whose target feature values ​​are higher than a preset feature value threshold as target patients of the target patient.

[0016] Thirdly, embodiments of the present invention also provide an electronic device, the electronic device comprising:

[0017] One or more processors;

[0018] Storage device for storing one or more programs.

[0019] When the one or more programs are executed by the one or more processors, the one or more processors implement the method for determining the receiving user as described in any embodiment of the present invention.

[0020] Fourthly, embodiments of the present invention also provide a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform the method for determining a patient as described in any of the embodiments of the present invention.

[0021] The technical solution of this embodiment determines a cluster of patients to be processed associated with the target patient based on the target patient's basic information and symptom description information; determines at least one historical patient associated with the cluster of patients to be processed, and determines a matrix to be used corresponding to the cluster of patients to be processed based on the patient records of each historical patient and at least one pre-edited dimensional reference attribute; processes the matrix to be used based on the historical patient attributes of the target patient to determine the target feature value of each historical patient relative to the target patient; and designates historical patients whose target feature values ​​are higher than a preset feature value threshold as target patients of the target patient. This solves the problem in the prior art where the assessment of patients based on active evaluation is inaccurate, leading to the possibility that the recommended patients may not match the patient, and achieves the recommendation of more suitable patients, improving the effectiveness of patients in determining patients' patients. Attached Figure Description

[0022] To more clearly illustrate the technical solutions of exemplary embodiments of the present invention, the accompanying drawings used in describing the embodiments are briefly introduced below. Obviously, the accompanying drawings described are only a portion of the drawings of the embodiments to be described in this invention, and not all of the drawings. For those skilled in the art, other drawings can be obtained from these drawings without any creative effort.

[0023] Figure 1 This is a flowchart illustrating a method for determining a patient to receive treatment, as provided in Embodiment 1 of the present invention.

[0024] Figure 2 This is a flowchart illustrating a method for determining a patient to receive treatment, provided in Embodiment 2 of the present invention.

[0025] Figure 3 This is a flowchart illustrating a method for determining a patient to receive treatment, provided in Embodiment 3 of the present invention.

[0026] Figure 4 This is a flowchart illustrating a method for determining a patient to receive treatment, provided in Embodiment 4 of the present invention.

[0027] Figure 5 This is a schematic diagram of the structure of a device for determining a patient receiving a medical appointment, provided in Embodiment 5 of the present invention;

[0028] Figure 6 This is a schematic diagram of the structure of an electronic device provided in Embodiment Six of the present invention. Detailed Implementation

[0029] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, the accompanying drawings show only the parts relevant to the present invention, and not all of the structures.

[0030] Before elaborating on this technical solution, its application scenarios will be introduced to facilitate a clearer understanding. This technical solution is a method for identifying patients, applicable to various medical institution platforms, medical mini-programs, or medical application software. When a patient has a medical need, they can log in to the relevant medical platform and input relevant symptom information according to their medical needs. Then, based on the method of this technical solution, doctors matching the patient's symptom information will be recommended to the patient.

[0031] Example 1

[0032] Figure 1 This is a flowchart illustrating a method for determining a patient to receive treatment, as provided in Embodiment 1 of the present invention. This embodiment is applicable to situations where a more suitable patient to receive treatment is recommended based on the different medical needs of the patient. This method can be executed by a device for determining the patient to receive treatment. This device can be implemented in the form of software and / or hardware. The hardware can be an electronic device, such as a mobile terminal or a PC.

[0033] like Figure 1 As shown, the method includes:

[0034] S110. Based on the basic information and symptom description information of the target patient, determine the cluster of patients to be processed associated with the target patient; wherein, the cluster of patients to be processed includes multiple historical patients.

[0035] The target patient can be understood as a patient with medical needs. Basic information refers to the target patient's identity information, such as name, gender, mobile phone number, and other basic personal information. Symptom description information refers to the target patient's symptoms, such as cough, dizziness, fatigue, colds, fever, diarrhea, and other symptoms. The pending patient cluster can be understood as a group of patients whose symptoms match those of the target patients. This pending patient cluster includes multiple historical patients.

[0036] In practical applications, the first step is to acquire complete medical data from admission to discharge for patients at various medical institutions within each region, including patient information, symptom data, examination data, and treatment plan data. Then, the obtained data undergoes deduplication, data completion, and data transformation. For example, when the collected data includes identical information, only one record of the duplicate data is retained. Finally, the preprocessed data is converted into feature vectors.

[0037] For example, various symptom information are pre-set. When the symptom data of a target patient matches a certain symptom data, the feature word corresponding to that symptom data is represented by a corresponding weight. If the target patient does not contain the corresponding symptom data, the weight of the feature word corresponding to that symptom data is recorded as "0". For example, if the pre-set feature vector corresponding to the symptom data is "cough; fever; headache; fatigue; dry mouth", and the symptom information corresponding to the target patient includes "cough" and "headache", then the feature vector corresponding to the symptom information of the target patient can be (0.8, 0, 0.9, 0, 0). The weights corresponding to the feature words are calculated using the TF-IDF (Term Frequency-Inverse Document Frequency) method. It should be noted that the above example is only for illustration and is not real feature vector data. When setting the feature vector corresponding to the symptom data, multiple symptom data can be set, and there is no limit to the specific number of symptom data.

[0038] Specifically, target patients can fill in relevant basic information and symptom descriptions corresponding to their symptoms on various medical platforms based on their medical needs. This information is then converted into feature vectors. The system identifies feature vectors from the symptom feature vectors collected from various medical institutions that match the target patient's symptom information. Multiple patients whose symptom information matches the target patient's are grouped into a cluster of patients awaiting processing, associated with the target patient. This cluster includes multiple historical patients.

[0039] Optionally, determining the cluster of patients to be processed associated with the target patient based on the target patient's basic information and symptom description information includes: obtaining symptom description information and basic information from the symptom editing control on the target application; determining at least one keyword to be used based on the symptom description information and basic information; and determining the cluster of patients to be processed from the at least one historical patient based on the at least one keyword to be used and the historical symptom descriptions of each historical patient.

[0040] The target application can be understood as a program that uses the technical methods of this solution to recommend matching patients to various users. It can be various medical platforms, medical mini-programs, or medical software. The symptom editing control can be understood as a control used by each patient in the target application to describe symptom information and basic information. This can be a text-based or voice-based editing control. Keywords to be used can be understood as keywords corresponding to each symptom and basic information. Historical patients can be understood as users who have visited various medical institutions. Historical patients can be users from different medical institutions, such as historical patients from all medical institutions in Shandong Province. Historical symptom descriptions can be understood as symptom information corresponding to each historical patient.

[0041] Specifically, when a target patient has a medical need, they fill in a description of their symptoms and basic personal information corresponding to their medical needs through a symptom editing control set on the target application. Then, keywords corresponding to the symptom description of the target patient are used as keywords. Through keyword comparison, at least one historical patient matching the keywords is identified from the historical symptom descriptions of various historical patients. This at least one historical patient matching the keywords is then grouped into a cluster of patients awaiting processing.

[0042] S120. Determine at least one historical patient who is associated with the cluster of patients to be processed, and determine the matrix to be used corresponding to the cluster of patients to be processed based on the patient record information of each historical patient and at least one pre-edited dimension reference attribute.

[0043] In this context, "historical patients" refers to the doctors who have treated each patient in the past. Each historical patient may have one or more doctors; for example, a single patient may have received treatment at multiple medical institutions, or at the same medical institution, different doctors may have provided services to that patient. "Patient records" refers to the information documented by each doctor in providing services to each patient, including data such as treatment duration, cost, and outcome. "Dimensional reference attributes" refers to different treatment attributes set based on the specific needs of each patient, including treatment duration, cost, and outcome. After receiving treatment from a doctor, each patient can rate the doctor, and a corresponding matrix of available resources is generated based on these ratings. It should be noted that the matrix to be used can correspond to the reference attributes of each dimension. That is to say, the matrix to be used can include the matrix to be used corresponding to the treatment duration, the matrix to be used corresponding to the treatment cost, the matrix to be used corresponding to the treatment effect, and the matrix to be used corresponding to any two reference attributes.

[0044] For example, taking treatment duration as an example, each element in the matrix to be used represents the evaluation attribute value of each historical patient to each historical patient, that is, the rating of each historical patient to each historical patient under the dimension reference attribute of treatment duration. Each row of data in the matrix to be used represents the rating of the same historical patient to different patients, and the elements in the same column correspond to the same patient.

[0045] Specifically, based on the symptom description information filled in by the target patient in the editing control, a cluster of users to be processed corresponding to the target patient is determined. This cluster includes multiple historical patients, each of whom can correspond to one or more historical patients. By using the cluster of users to be processed associated with the target patient and the consultation records of each historical patient, at least one historical patient associated with the target patient can be identified. After each historical patient completes their treatment of their respective patient, they can rate the corresponding patient from different dimensions, including at least one of treatment duration, treatment cost, and treatment effectiveness. The ratings from each historical patient based on these dimensions are then entered into the corresponding positions in a matrix, generating a matrix to be used corresponding to the reference attributes of each dimension.

[0046] Optionally, determining the matrix to be used corresponding to the cluster of patients to be processed based on the patient reception record information of each historical patient and at least one pre-edited dimension reference attribute includes: if the at least one dimension reference attribute is consistent with any one of the at least three dimensions of the patient reception record information, then obtaining the record data to be used that is consistent with the dimension reference attribute from the at least three dimensions; for each historical patient, determining the actual patient data of the current historical patient for each historical patient, and determining the vector value of the current historical patient corresponding to each historical patient based on the maximum and minimum values ​​in the actual patient data and the actual patient data of each historical patient; determining the matrix to be used based on the vector values ​​of each historical patient; wherein each column element in the matrix to be used corresponds to the same historical patient.

[0047] The data to be used can be understood as the data information of each historical patient corresponding to the dimension reference attribute selected by the target patient. For example, if the dimension reference attribute selected by the target patient is treatment cost, then the data to be used can be the treatment cost data information corresponding to each historical patient in the consultation record information. The current historical patient can be understood as the historical user from whom actual consultation data is obtained. Each historical patient can be used as the current historical patient; this distinction is only made for the historical patients from whom actual consultation data needs to be obtained. Actual consultation data can be understood as the actual consultation data of the current historical patient during the consultation process. This can include different dimension reference attributes, namely, at least one of the following: consultation cost information, consultation duration information, and consultation effect evaluation information corresponding to each historical patient. Each actual consultation data includes a maximum and minimum value. Taking consultation cost information as an example, the maximum and minimum values ​​in the actual consultation data correspond to the highest and lowest costs for each historical patient, respectively. Vector values ​​can be understood as the rating data values ​​corresponding to the current historical patient. Among them, the medical record information must be consistent in at least three dimensions, including medical fee information, treatment duration information, and treatment effect evaluation information, and at least one dimension of the reference attribute must correspond to at least three dimensions of the medical record information.

[0048] Specifically, the consultation records for each historical patient include various information such as consultation fees, treatment duration, and treatment effectiveness evaluation, with corresponding data points. A target patient can select a suitable patient based on at least one reference attribute according to their medical needs. If the reference data for at least one dimension of the target patient matches any dimension in the consultation record, the corresponding historical patient data can be retrieved from the consultation record. For example, if the target user selects treatment duration as the reference attribute, the treatment data of each historical patient corresponding to that treatment duration can be retrieved from the consultation record. Based on the reference attribute selected by the target user, the historical patients corresponding to that reference attribute are determined, and the actual treatment data corresponding to the current historical patient is also determined. Furthermore, in order to determine the rating data corresponding to the current historical patient, the rating of the historical patient corresponding to the current patient can be determined based on the maximum and minimum values ​​in the actual patient data and the actual patients of each historical patient. The obtained rating data is then used as the vector value corresponding to that patient.

[0049] In practical applications, based on the rating data of each historical patient to each historical patient, the vector values ​​in the matrix to be used can be determined, and the calculation formulas for the rating data corresponding to different dimension reference attributes can also determine the vector values ​​in the matrix to be used.

[0050] For example, taking the consultation time information as an example, the scoring data corresponding to the consultation time dimension reference attribute is s_time. x :

[0051]

[0052] Among them, s_time x This indicates the treatment duration information for each historical patient. This represents the score data after normalizing the information on each treatment duration; T max T represents the maximum value of the treatment duration information. min This represents the minimum value for the consultation duration information; 5 indicates that the scoring system is a five-point scale.

[0053] Based on the above formula, we can obtain the rating data of each historical patient to each historical patient, under the parameter attribute of treatment duration information. For example, among the historical patients, the longest treatment duration is 20 hours and the shortest treatment duration is 12 hours. In this case, the difference between the longest and shortest treatment durations is not equal to 0, so the normalized rating data of the treatment duration information corresponding to the current historical patient can be obtained by substituting it into the above formula. If the difference between the longest and shortest treatment durations is equal to 0, then the normalized rating data of the treatment duration information corresponding to the current historical patient is 1. Then, multiplying the obtained normalized rating data by the pre-set rating system, we can obtain the treatment duration information corresponding to the current historical patient. Then, the rating data of each historical patient to each historical patient is used as the vector value in the matrix to be used for the treatment duration information.

[0054] It should be noted that the scoring data obtained by this technical solution is determined using a five-point system. Other scoring systems, such as a ten-point system or a hundred-point system, can also be used. When using a ten-point system, the 5 in the formula should be changed to 10. When using a hundred-point system, the 5 in the formula should be changed to 100. In this technical solution, only the five-point system is used as an example for explanation.

[0055] Similar to the aforementioned consultation duration information dimension reference attribute, another dimension reference attribute can be the consultation fee information dimension reference attribute, where the scoring data corresponding to the consultation fee information dimension reference attribute is s_cost. x :

[0056]

[0057] Wherein, s_cost x This indicates the medical expenses information for each historical patient. This represents the scoring data after normalizing the information on each patient's treatment fee; C max C represents the maximum value of the consultation fee information. min The minimum value for the consultation fee information is 5, which indicates that the scoring system is a five-point scale.

[0058] Similar to the calculation method for treatment duration information, based on the above formula, under the dimension reference attribute of consultation fee information, the maximum and minimum values ​​of consultation fees for each historical patient can be substituted into the above formula to calculate the consultation fee information corresponding to the current historical patient under the dimension reference attribute of consultation fee information, and determine the vector values ​​in the matrix to be used under this dimension.

[0059] Similar to the aforementioned reference attributes for treatment duration information, another type of reference attribute can be taken as treatment effectiveness evaluation information. The scoring data corresponding to the reference attribute for treatment effectiveness evaluation information is s_effect. x :

[0060]

[0061] Among them, s_effect x This indicates the treatment outcome evaluation information for each historical patient. E represents the score data after normalization of various treatment effectiveness evaluation information; max E represents the maximum value of information for evaluating treatment effectiveness. min The minimum value for evaluating treatment effectiveness is 5, which indicates that the scoring system is a five-point scale.

[0062] The calculation method for the treatment effect evaluation information E is as follows: Among them, DisH f The DisH is the pre-treatment health distance index value. l The post-treatment health distance index (DisH) is calculated as follows: , where t i For the patient's various test values, s i The standard deviation of each test component is given. To ensure that the corresponding indicators reach their maximum healthy level, This represents the minimum value required for the corresponding indicator to reach a healthy level.

[0063] Similar to the calculation method for treatment duration information, based on the above formula, under the dimensional reference attribute of treatment effect evaluation information, the maximum and minimum values ​​of treatment effect evaluation information for each historical patient can be substituted into the above formula to calculate the treatment effect evaluation information corresponding to the current historical patient under the dimensional reference attribute of treatment effect evaluation information, and determine the vector values ​​in the matrix to be used under this dimensional reference attribute.

[0064] S130. Based on the historical medical visit attributes of the target patient, process the matrix to be used to determine the target feature value of each historical patient relative to the target patient.

[0065] Among them, historical medical visit attributes can be understood as whether the target patient has participated in the corresponding medical visit. Target feature values ​​can be understood as the matching value of each historical patient for the target patient. The magnitude of the target feature value determines the matching degree between each historical patient and the target patient. The larger the target feature value, the higher the matching degree between the historical patient and the target patient, or the better the treatment effect of the historical patient on the target patient's corresponding ailment. It should be noted that the treatment effect here can be at least one of the following: consultation fee, consultation duration, or treatment effect.

[0066] Specifically, the historical medical visit attributes of the target patient are determined based on whether the target patient has participated in a medical visit. Then, the matrix to be used is processed based on the historical medical visit attributes to determine the target feature value of each historical patient.

[0067] Optionally, the step of processing the matrix to be used based on the historical medical visit attributes of the target patient to determine the target feature value of each historical patient relative to the target patient includes: if the historical medical visit attributes do not include participation in medical visits, determining the mean of the elements in the same column of the matrix to be used, and determining the target feature value of each historical patient.

[0068] Specifically, if the historical visit attribute corresponding to the target patient is "not including participation in visits," it means that the collected information on historical patients does not contain data corresponding to the target patient. In this case, to determine the patient who best matches the target patient, the mean of the elements in the same column of the matrix to be used can be used as the target feature value corresponding to each historical patient. That is, in the matrix to be used, each column represents the rating data of each historical patient for the same historical patient, and the average of the rating data is used as the target feature value corresponding to that historical patient.

[0069] S140. The historical patients whose target feature values ​​are higher than the preset feature value threshold are taken as the target patients of the target patients.

[0070] The preset feature value threshold can be understood as a pre-set feature value used to filter each target feature value.

[0071] Specifically, when recommending matching patients to a target patient, there may be many matching patients. The larger the target feature value, the higher the matching degree between the historical patients corresponding to that target feature value and the target patient. Based on a preset feature value threshold, the target feature values ​​corresponding to each historical patient can be filtered. Then, historical patients with target feature values ​​higher than the preset feature value threshold are selected as the target patients corresponding to the target patient.

[0072] For example, if the full score is 100 points, the preset feature value threshold can be set to 70. Based on the preset feature value threshold, the target feature values ​​corresponding to each historical patient are filtered, and historical patients with target feature values ​​higher than 70 are selected as the target patients corresponding to the target patients.

[0073] The technical solution of this embodiment determines a cluster of patients to be processed associated with the target patient based on the basic information and symptom description information of the target patient. It converts the symptom information corresponding to each historical patient into a corresponding feature vector and clusters the obtained feature vectors to obtain at least one cluster of patients to be processed. Based on the similarity between the feature vector corresponding to the target patient and the feature vectors in each cluster of patients to be processed, it determines the cluster of patients to be processed associated with the target patient. At least one historical patient is identified as associated with the cluster of patients to be processed. Based on the patient records of each historical patient and at least one pre-edited dimensional reference attribute, a matrix to be used corresponding to the cluster of patients to be processed is determined. After each patient's visit, each historical patient can be evaluated based on different dimensional reference attributes to obtain vector values ​​corresponding to each evaluation value. Then, the matrix to be used is determined based on each vector value, and the target feature value corresponding to the target patient is determined based on the matrix to be used. This solves the problem in the prior art that the evaluation of patients lacks corresponding evaluation standards, has a large subjective factor, and leads to inaccurate evaluation of each historical patient. Based on the historical patient attributes of the target patient, the matrix to be used is processed to determine the target feature value of each historical patient relative to the target patient. Based on whether the target patient includes participation in the patient's visit attribute, the target feature value of the target patient relative to each historical patient is determined. Historical patients whose target feature values ​​exceed a preset feature value threshold are designated as target patients for the target patient. A higher target feature value indicates a higher match between the historical patients corresponding to that feature value and the target patient. By using historical patients with values ​​exceeding the preset feature threshold as target patients, the target patient can select the most suitable patient from among these target patients. This solves the problem in existing technologies where subjective factors lead to mismatches in recommended target patients. It achieves the goal of recommending more suitable patients to patients, improving the effectiveness of patients in identifying suitable patients.

[0074] Example 2

[0075] As an optional embodiment of the above embodiments, Figure 2 This is a flowchart illustrating a method for determining patients according to Embodiment 2 of the present invention. Optionally, the process of determining the matrix to be used corresponding to the cluster of patients to be processed based on the patient reception record information of each historical patient and at least one pre-edited dimension reference attribute is further refined.

[0076] like Figure 2 As shown, the method includes:

[0077] S210. Based on the basic information and symptom description information of the target patient, determine the cluster of patients to be processed associated with the target patient.

[0078] S220. If the at least one dimension reference attribute is consistent with at least two dimensions of the at least three dimensions of the patient reception record information, then obtain the record data to be used that is consistent with the dimension reference attribute from the at least three dimensions.

[0079] Specifically, when a target patient selects a corresponding patient based on their medical needs, they can choose from different reference attributes, including one, two, or three. If the reference attribute selected by the target patient matches at least two of the three dimensions in the patient record information, then the corresponding record data is retrieved from the corresponding matrix.

[0080] For example, if the dimension reference attributes selected by the target patient include two dimensions, namely consultation fee information and treatment effect evaluation information, then the data to be used corresponding to the consultation fee information is obtained from the matrix to be used corresponding to the consultation fee information, and at the same time, the data to be used corresponding to the treatment effect evaluation information is obtained from the matrix to be used corresponding to the treatment effect evaluation information.

[0081] S230. For each historical patient, determine the actual medical data of the current historical patient to each historical patient, and determine the dimension vector of the current historical patient corresponding to each historical patient based on the maximum and minimum values ​​in the actual medical data and the actual medical data of each historical patient.

[0082] Here, the dimension vector can be understood as the vector value corresponding to the current historical patient when at least two dimensions are consistent, that is, the score data value corresponding to the current historical patient.

[0083] Specifically, if the dimension vector corresponding to the target patient matches at least two dimensions in the medical record information, then the actual medical data of each historical patient corresponding to the current historical patient is determined from the historical patients in the medical record. This includes the actual medical expenses, treatment duration, and treatment effectiveness evaluation information of each historical patient during their actual treatment. Similar to the case where the dimension vector corresponding to the target patient has only one dimension, the maximum and minimum values ​​in the actual medical data, along with the actual medical data of each historical patient, are used to determine the current historical patient's rating data for each historical patient under different dimensions.

[0084] S240. For each historical patient, based on the weight values ​​corresponding to the pre-edited dimensional reference attributes, process the dimension vectors of at least two dimensional reference attributes corresponding to the current historical patient to determine the feature vector to be concatenated for the current historical patient relative to each historical patient.

[0085] Among them, the rating data of each historical patient to each historical patient can be used as a dimension vector corresponding to each historical patient. Then, the dimension vectors corresponding to each historical patient are concatenated into a large dimension vector, which is used as the feature vector to be concatenated.

[0086] Specifically, when each historical patient selects dimensions that match at least two of the three dimensions in their medical records, the weight values ​​corresponding to the reference attributes of each dimension can be determined based on their own medical needs. Then, the dimension vectors corresponding to each dimension are multiplied by their respective weight values, and the result is used as the dimension vector corresponding to each historical patient. Finally, the dimension vectors of the current historical patient and each historical patient are used as the feature vector values ​​to be concatenated, so that the dimension vectors corresponding to each historical patient can be concatenated to obtain a larger dimension vector.

[0087] For example, the reference dimensions for each historical patient include two dimensions, such as treatment duration information and treatment effect evaluation information. Each historical patient determines the weight value corresponding to each dimension's reference attribute based on their medical needs. For instance, the weight values ​​for each dimension's reference attribute could be 0.5 and 0.5, or 0.3 and 0.7, with the sum of the weight values ​​being 1. The specific weight values ​​are determined based on the actual situation. Based on the dimension vectors of the current historical patient relative to the current historical patient under different dimensions, each dimension vector is multiplied by the weight value of its corresponding dimension reference attribute, and the results are summed to obtain the feature vector to be concatenated relative to each historical patient.

[0088] For example, if the reference attributes corresponding to the target patient include at least two dimensions, the comprehensive score data can be determined based on the weight values ​​corresponding to each reference attribute. If the reference attributes corresponding to the target patient include three dimensions, then the score data s_all x It can be calculated using the following formula:

[0089]

[0090] Among them, s_all x This represents the comprehensive score data across three dimensions; w timeThe weight value of the dimension reference attribute representing the consultation duration information; w cost The weight value of the dimension reference attribute representing the consultation fee information, w effect This indicates the weight value of the reference attribute for the treatment effect dimension. .

[0091] S250. Determine the matrix to be used based on the feature vectors to be spliced ​​for each historical patient.

[0092] Specifically, under different dimensional reference attributes, the final dimensional vector corresponding to the dimensional vector of each historical patient in each dimensional reference attribute will be used to obtain the feature vector to be spliced, and each feature vector to be spliced ​​will be filled into the matrix to be used to determine the matrix to be used.

[0093] It should be noted that in the matrix to be used, each row of data represents the feature vector to be concatenated for each historical patient and each historical patient.

[0094] For example, consider the matrix to be used corresponding to the treatment duration dimension reference attribute:

[0095]

[0096] Where m represents the total number of patients, n represents the total number of doctors, and r ij This represents the rating data of the i-th patient for the j-th doctor.

[0097] If the same historical patient has multiple rating records for the same historical patient, the latest rating record will be used as the rating record for that historical patient for that historical patient.

[0098] Similarly, the reference attribute for the consultation fee information dimension corresponds to the matrix R to be used. cost The matrix R to be used corresponds to the reference attributes of the treatment effect evaluation information dimensions. effect and the matrix R to be used corresponding to at least two reference attributes. all The method for determining the matrix to be used is the same as the method for determining the reference attribute of the treatment duration dimension, and will not be repeated here.

[0099] S260. Based on the historical medical visit attributes of the target patient, process the matrix to be used to determine the target feature value of each historical patient relative to the target patient.

[0100] S270. The historical patients whose target feature values ​​are higher than the preset feature value threshold are taken as the target patients of the target patients.

[0101] In this embodiment, if the at least one dimension reference attribute matches at least two of the at least three dimensions of the patient record information, then the record data to be used that matches the dimension reference attribute is obtained from the at least three dimensions. Based on the dimension parameter attributes corresponding to the target patient, the record data to be used corresponding to each dimension parameter attribute is obtained. For each historical patient, the actual patient data of the current historical patient for each historical patient is determined. Based on the maximum and minimum values ​​in the actual patient data, and the actual patient data of each historical patient, the dimension vector of the current historical patient corresponding to each historical patient is determined. Then, based on the weight values ​​corresponding to each dimension parameter attribute and the actual patient data of each historical patient, the vector values ​​corresponding to each dimension parameter attribute are determined. For each historical patient, based on the weight values ​​corresponding to pre-edited dimensional reference attributes, the dimensional vectors of at least two dimensional reference attributes corresponding to the current historical patient are processed to determine the feature vector to be concatenated relative to each historical patient. Based on the preset weight values ​​of each dimensional reference attribute and the dimensional vector values ​​of each historical patient, a vector matrix to be concatenated is determined. Based on the feature vectors to be concatenated for each historical patient, a matrix to be used is determined, and the target patient is determined based on the target feature values ​​of the matrix to be used. This solves the problem that when recommending patients to patients from a single dimension, the patient and the patient may not be well-matched, achieving the effect of recommending more suitable patients.

[0102] Example 3

[0103] As an optional embodiment of the above embodiments, Figure 3 This is a flowchart illustrating a method for determining patients according to Embodiment 3 of the present invention. Optionally, the matrix to be used is processed based on the historical medical attributes of the target patient to refine the target feature values ​​of each historical patient relative to the target patient.

[0104] S310. Based on the basic information and symptom description information of the target patient, determine the cluster of patients to be processed associated with the target patient.

[0105] S320. Determine at least one historical patient who is associated with the cluster of patients to be processed, and determine the matrix to be used corresponding to the cluster of patients to be processed based on the patient record information of each historical patient and at least one pre-edited dimension reference attribute.

[0106] S330. If the historical medical visit attributes include participation in medical visits attributes, then obtain the row data corresponding to the target medical visit user from the matrix to be used, as the feature vector to be registered.

[0107] Specifically, if the historical medical visit attributes corresponding to the target patient include participation in medical visits, meaning the medical record information includes historical data information corresponding to the target patient, then, in order to determine the medical attendants who are a good match for the target patient, the row data corresponding to the target patient can be determined from the matrix to be used based on the user identifier information corresponding to the target patient. This row data is the feature vector to be concatenated for each historical medical attendant relative to the target patient. The feature vector to be concatenated for the target patient is then used as the feature vector to be registered, and the similarity between the feature vector to be registered and the feature vectors corresponding to each historical medical attendant is determined according to a similarity algorithm.

[0108] S340. Determine the similarity between the feature vector to be registered and the row vector to be called corresponding to each row element in the matrix to be used.

[0109] Here, the row vector to be called can be understood as the vector formed by the feature vectors to be concatenated corresponding to each historical patient in the matrix to be used. Each row element in the matrix represents the feature vector to be concatenated for each historical patient. Similarity can be understood as the similarity between the feature vector to be registered and each row vector to be called, which can be obtained through a similarity algorithm, such as the cosine similarity algorithm.

[0110] Specifically, based on the similarity algorithm, the Euclidean distance between the feature vectors to be registered in the matrix to be used can be calculated. This determines the similarity between the feature vector to be registered corresponding to the target patient and each row vector to be called. Then, based on the similarity between each row vector to be called and the feature vector to be registered, the similarity between each historical patient and the target patient can be determined. The higher the similarity of the row vector to be called, the higher the similarity between the historical patient and the target patient. That is, the historical patient's symptom information is consistent with the target patient's, and the corresponding dimensional reference attributes are consistent with the target patient's, making it a better match for the target patient's medical needs. In other words, the higher the similarity of a historical patient to the target patient's dimensional vectors, the stronger the match between that historical patient and the target patient. Therefore, historical patients with higher ratings can be recommended to the target patient.

[0111] S350. Select the row vectors to be called with similarity values ​​higher than the preset similarity threshold as the target row vectors.

[0112] The preset similarity threshold can be understood as a pre-set similarity value. Based on the preset similarity threshold, each similarity can be filtered, and the row vector to be called corresponding to the similarity value higher than the similarity threshold is taken as the target row vector.

[0113] Specifically, the matrix to be used includes multiple row vectors to be invoked. Based on the similarity between each row vector to be invoked and the feature vector to be registered corresponding to the target patient, historical patients with a high similarity to the target patient can be identified. To determine the similarity between the target patient and each historical patient, row vectors to be invoked with similarity values ​​higher than a preset similarity threshold can be used as target row vectors, and the historical patients corresponding to the target row vectors can be considered as historical patients with a high matching degree to the target patient.

[0114] S360. Based on the target row vector, determine the target feature value of each historical patient relative to the target patient.

[0115] In practical applications, in order to determine the receiving user corresponding to the target receiving user, the step of determining the target feature value of each historical receiving user relative to the target receiving user based on the target row vector includes: determining the target prediction matrix based on each target row vector; and determining the target feature value of each historical receiving user relative to the target receiving user through prediction processing of the target prediction matrix.

[0116] The target prediction matrix can be understood as a matrix based on the row vectors of each target user. Within this matrix, there are feature vectors representing the characteristics of each historical patient who is highly similar to the target patient. Prediction processing can employ algorithms such as collaborative filtering.

[0117] Specifically, based on the target prediction matrix formed by each target row vector, and processed using collaborative filtering algorithms, the evaluation of each historical patient by the target patient is largely consistent with the evaluations of other historical patients by those with high similarity to the target patient. Therefore, based on the feature vectors of each historical patient corresponding to each target row vector, the most satisfactory historical patient can be identified. Then, based on the collaborative filtering algorithm and the feature vectors of each historical patient with high matching degree, the target feature values ​​of each historical patient relative to the target patient can be predicted.

[0118] S370. Historical patients whose target feature values ​​are higher than a preset feature value threshold are designated as target patients of the target patient.

[0119] In this embodiment, if the historical medical visit attributes include participation attributes, then the row data corresponding to the target medical visit user is obtained from the matrix to be used as the feature vector to be registered. If the medical record information contains historical data information corresponding to the target medical visit user, then the data information corresponding to the target medical visit user in the matrix to be used is obtained as the feature vector to be registered. The similarity between the feature vector to be registered and each feature vector is determined according to a similarity algorithm. The similarity between the feature vector to be registered and the row vector to be called corresponding to each row element in the matrix to be used is determined. Based on the similarity, each historical medical visit user with a high matching degree with the target medical visit user is determined, and the row vector to be called corresponding to each historical medical visit user is obtained. The row vector to be called with a similarity value higher than a preset similarity threshold is used as the target row vector, and the target row vector is determined to determine the corresponding target feature value. Based on the target row vector, target feature values ​​of each historical patient relative to the target patient are determined. Based on these target feature values, historical patients who are a good match for the target patient are identified, and these historical patients are used as target patients. The historical patients are then ranked based on similarity values, allowing the target patient to select the most suitable patient from among the target patients. This solves the problem in existing technologies where proactive assessment of patient selection is inaccurate, leading to potentially mismatched recommended patients. It achieves the effect of recommending more suitable patients to the patient.

[0120] Example 4

[0121] In a specific example, such as Figure 4 As shown, to recommend a suitable doctor (i.e., the receiving user) to the patient (i.e., the target patient), the system first collects symptom descriptions and related treatment data from historical patients at various medical institutions within the region. This includes symptom data, testing data, and treatment plan data from admission to discharge for each historical patient. Then, the collected data from each historical patient is preprocessed, such as deduplication, data completion, and data transformation. For example, duplicate data in the symptom data of each historical patient is deleted. Next, based on keyword information (i.e., keywords to be used), the symptom descriptions of each historical patient are converted into corresponding feature vectors. Based on the feature vectors corresponding to the symptom information of the target patient, cluster analysis is performed on each patient to determine clusters of patients who are similar to and closely matched to the target patient. For example, if the target patient's symptom information includes cough and fever, then the symptom information of each historical patient in the cluster of patients matching the target patient will also be cough and fever.

[0122] It should be noted that after each historical patient completes their consultation, they can rate the corresponding historical patient. The resulting ratings will serve as vector values ​​for each historical patient's experience with respect to their respective historical patients. Each historical patient can evaluate their historical patient using different dimensions and reference attributes, including at least one of the following: consultation fees, consultation duration, and treatment effectiveness assessment. The rating data can be calculated using the following formula:

[0123] Taking treatment duration information as an example, the scoring data corresponding to the reference attribute for the treatment duration dimension is s_time. x :

[0124]

[0125] Among them, s_time x This indicates the treatment duration information for each historical patient. This represents the score data after normalizing the information on each treatment duration; T max T represents the maximum value of the treatment duration information. min This represents the minimum value for the consultation duration information; 5 indicates that the scoring system is a five-point scale.

[0126] It should be noted that the scoring data obtained by this technical solution is determined using a five-point system. Other scoring systems, such as a ten-point system or a hundred-point system, can also be used. When using a ten-point system, the 5 in the formula should be changed to 10. When using a hundred-point system, the 5 in the formula should be changed to 100. In this technical solution, only the five-point system is used as an example for explanation.

[0127] Taking consultation fee information as an example, the scoring data corresponding to the reference attribute of consultation fee information dimension is s_cost. x :

[0128]

[0129] Wherein, s_cost x This indicates the medical expenses information for each historical patient. This represents the scoring data after normalizing the information on each patient's treatment fee; C max C represents the maximum value of the consultation fee information. min The minimum value for the consultation fee information is 5, which indicates that the scoring system is a five-point scale.

[0130] Taking treatment effectiveness evaluation information as an example, the scoring data corresponding to the reference attributes of the treatment effectiveness evaluation information dimensions is s_effect. x :

[0131]

[0132] Among them, s_effect x This indicates the treatment outcome evaluation information for each historical patient. E represents the score data after normalization of various treatment effectiveness evaluation information; max E represents the maximum value of information for evaluating treatment effectiveness. min The minimum value for evaluating treatment effectiveness is 5, which indicates that the scoring system is a five-point scale.

[0133] The calculation method for the treatment effect evaluation information E is as follows: Among them, DisH f The DisH is the pre-treatment health distance index value. l The post-treatment health distance index (DisH) is calculated as follows: , where t i For the patient's various test values, s i The standard deviation of each test component is given. To ensure that the corresponding indicators reach their maximum healthy level, This represents the minimum value required for the corresponding indicator to reach a healthy level.

[0134] If the target patient's reference attributes include at least two dimensions, the comprehensive score data can be determined based on the weight values ​​corresponding to each reference attribute. For example, if the target patient's reference attributes include three dimensions, then the score data s_all... x It can be calculated using the following formula:

[0135]

[0136] Among them, s_all x This represents the comprehensive score data across three dimensions; w time The weight value of the dimension reference attribute representing the consultation duration information; w cost The weight value of the dimension reference attribute representing the consultation fee information, w effect This indicates the weight value of the reference attribute for the treatment effect dimension. .

[0137] Based on the rating data of each historical patient and doctor from different reference attributes, a multi-dimensional patient-doctor rating data matrix (i.e., the matrix to be used) can be obtained. Taking the matrix to be used corresponding to the treatment duration dimension reference attribute as an example:

[0138]

[0139] Where m represents the total number of patients, n represents the total number of doctors, and r ijThis represents the rating data of the i-th patient for the j-th doctor.

[0140] If the same historical patient has multiple rating records for the same historical patient, the latest rating record will be used as the rating record for that historical patient for that historical patient.

[0141] Similarly, the reference attribute for the consultation fee information dimension corresponds to the matrix R to be used. cost The matrix R to be used corresponds to the reference attributes of the treatment effect evaluation information dimensions. effect and the matrix R to be used corresponding to at least two reference attributes. all The method for determining the matrix to be used is the same as the method for determining the reference attribute of the treatment duration dimension, and will not be repeated here.

[0142] The matrix to be used, generated based on the rating data of historical patients to historical patients under different reference attributes, can be used with recommendation algorithms, such as collaborative filtering, to predict the target patient's rating data for each historical patient. Then, the historical patients are sorted, with those having higher predicted ratings placed at the top. A corresponding recommendation list of patients is generated from these top-ranked historical patients. The target patient selects a suitable recommendation list based on their medical needs and uses the selected historical patients as their target patients. After the target patient's visit, they can further rate the target patients to update the patient satisfaction rating matrix (i.e., the matrix to be used).

[0143] The technical solution of this embodiment determines a cluster of patients to be processed associated with the target patient based on the target patient's basic information and symptom description information; determines at least one historical patient associated with the cluster of patients to be processed, and determines a matrix to be used corresponding to the cluster of patients to be processed based on the patient records of each historical patient and at least one pre-edited dimensional reference attribute; processes the matrix to be used based on the historical patient attributes of the target patient to determine the target feature value of each historical patient relative to the target patient; and designates historical patients whose target feature values ​​are higher than a preset feature value threshold as target patients of the target patient. This solves the problem in the prior art where the assessment of patients based on active evaluation is inaccurate, leading to the possibility that the recommended patients may not match the patient, and achieves the recommendation of more suitable patients, improving the effectiveness of patients in determining patients' patients.

[0144] Example 5

[0145] Figure 5The present invention provides a device for determining patients in a fifth embodiment, the device comprising: a cluster determination module 410 for patients to be processed, a matrix determination module 420 for a matrix to be used, a target feature value determination module 430, and a target patient determination module 440.

[0146] The pending medical user cluster determination module 410 is used to determine the pending medical user cluster associated with the target medical user based on the target medical user's basic information and symptom description information; wherein the pending medical user cluster includes multiple historical medical users.

[0147] The module 420 for determining the matrix to be used is used to determine at least one historical patient who is associated with the cluster of patients to be processed, and to determine the matrix to be used corresponding to the cluster of patients to be processed based on the patient record information of each historical patient and at least one pre-edited dimension reference attribute; wherein, each element in the matrix to be used represents the evaluation attribute value of each historical patient to the historical patient, and the elements in the same column of the matrix to be used correspond to the same patient.

[0148] The target feature value determination module 430 is used to process the matrix to be used based on the historical medical visit attributes of the target medical user, and determine the target feature value of each historical medical user relative to the target medical user.

[0149] The target patient identification module 440 is used to identify historical patients whose target feature values ​​are higher than a preset feature value threshold as target patients of the target patient.

[0150] The technical solution of this embodiment determines a cluster of patients to be processed associated with the target patient based on the target patient's basic information and symptom description information; determines at least one historical patient associated with the cluster of patients to be processed, and determines a matrix to be used corresponding to the cluster of patients to be processed based on the patient records of each historical patient and at least one pre-edited dimensional reference attribute; processes the matrix to be used based on the historical patient attributes of the target patient to determine the target feature value of each historical patient relative to the target patient; and designates historical patients whose target feature values ​​are higher than a preset feature value threshold as target patients of the target patient. This solves the problem in the prior art where the assessment of patients based on active evaluation is inaccurate, leading to the possibility that the recommended patients may not match the patient, and achieves the recommendation of more suitable patients, improving the effectiveness of patients in determining patients' patients.

[0151] Based on the above technical solution, optionally, the module for determining the cluster of patients awaiting treatment includes:

[0152] The information acquisition submodule is used to acquire the symptom description information and basic information in the symptom editing control on the target application; the keyword determination submodule is used to determine at least one keyword to be used based on the symptom description information and basic information; the pending medical user cluster determination submodule is used to determine the pending medical user cluster from the at least one historical medical user based on the at least one keyword to be used and the historical symptom descriptions of each historical medical user.

[0153] Based on the above technical solution, optionally, the matrix determination module includes:

[0154] The module for acquiring record data to be used is used to ensure that the patient visit record information is consistent with at least three dimensions of data. The at least three dimensions of data include patient visit fee information, treatment duration information, and treatment effect evaluation information. The at least one dimension reference attribute corresponds to the at least three dimensions of data in the patient visit record information. If the at least one dimension reference attribute is consistent with any one of the at least three dimensions of data in the patient visit record information, then the record data to be used that is consistent with the dimension reference attribute is acquired from the at least three dimensions of data. The module for determining vector values ​​is used to determine the actual medical data of the current historical patient visit user for each historical patient visit user, and to determine the vector value of the current historical patient visit user corresponding to each historical patient visit user based on the maximum and minimum values ​​in the actual medical data and the actual medical data of each historical patient visit user. The module for determining the matrix to be used is used to determine the matrix to be used based on the vector values ​​of each historical patient visit user. The elements in each column of the matrix to be used correspond to the same historical patient visit user.

[0155] Based on the above technical solution, optionally, the matrix determination module includes:

[0156] The system includes a data acquisition unit for acquiring record data to be used, configured to acquire record data to be used that matches the at least two dimensions of the at least three dimensions of the patient reception record information if the at least one dimension reference attribute matches at least two dimensions of the at least three dimensions of the patient reception record information; a dimension vector determination unit, configured to determine the actual medical data of the current historical patient reception user relative to each historical patient reception user for each historical patient reception user, and determine the dimension vector of the current historical patient reception user corresponding to each historical patient reception user based on the maximum and minimum values ​​in the actual medical data and the actual medical data of each historical patient reception user; a feature vector to be spliced ​​unit, configured to process the dimension vectors of at least two dimension reference attributes corresponding to the current historical patient reception user according to the weight values ​​corresponding to the pre-edited dimension reference attributes for each historical patient reception user, and determine the feature vector to be spliced ​​relative to each historical patient reception user for each historical patient reception user; and a matrix to be used unit, configured to determine the matrix to be used based on the feature vector to be spliced ​​for each historical patient reception user.

[0157] Based on the above technical solution, optionally, the target feature value determination module includes:

[0158] The target feature value determination submodule is used to determine the mean value of the elements in the same column of the matrix to be used, and to determine the target feature value of each historical patient if the historical patient visit attributes do not include the participation in patient visit attributes.

[0159] Based on the above technical solution, optionally, the target feature value determination submodule includes:

[0160] The unit for determining the feature vector to be registered is used to obtain row data corresponding to the target patient from the matrix to be used if the historical patient visit attributes include participation in patient visit attributes, and use this data as the feature vector to be registered; the similarity determination unit is used to determine the similarity between the feature vector to be registered and the row vector to be called corresponding to each row element in the matrix to be used; the target row vector determination unit is used to use the row vector to be called with a similarity value higher than a preset similarity threshold as the target row vector; and the target feature value determination unit is used to determine the target feature value of each historical patient visit user relative to the target patient visit user based on the target row vector.

[0161] Based on the above technical solution, optionally, the target feature value determination unit includes:

[0162] The target prediction matrix determination subunit is used to determine the target prediction matrix based on each target row vector; the target feature value determination subunit is used to determine the target feature value of each historical patient relative to the target patient by performing prediction processing on the target prediction matrix.

[0163] The device for determining the patient provided in the embodiments of the present invention can execute the method for determining the patient provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of executing the method.

[0164] It is worth noting that the various units and modules included in the above-mentioned device are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the protection scope of the embodiments of the present invention.

[0165] Example 6

[0166] Figure 6 This is a schematic diagram of the structure of an electronic device provided in Embodiment Six of the present invention. Figure 6 A block diagram is shown of an exemplary electronic device 40 suitable for implementing embodiments of the present invention. Figure 6 The electronic device 40 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of the present invention.

[0167] like Figure 6 As shown, electronic device 40 is represented in the form of a general-purpose computing device. The components of electronic device 40 may include, but are not limited to: one or more processors or processing units 401, system memory 402, and bus 403 connecting different system components (including system memory 402 and processing unit 401).

[0168] Bus 403 represents one or more of several bus architectures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of the various bus architectures. For example, these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect (PCI) bus.

[0169] Electronic device 40 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by electronic device 40, including volatile and non-volatile media, removable and non-removable media.

[0170] System memory 402 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 404 and / or cache memory 405. Electronic device 40 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 406 may be used to read and write non-removable, non-volatile magnetic media (… Figure 6Not shown; usually referred to as a "hard drive"). Although Figure 6 Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk") and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 403 via one or more data media interfaces. Memory 402 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of the present invention.

[0171] A program / utility 408 having a set (at least one) of program modules 407 may be stored, for example, in memory 402. Such program modules 407 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment. Program modules 407 typically perform the functions and / or methods described in the embodiments of the present invention.

[0172] Electronic device 40 can also communicate with one or more external devices 409 (e.g., keyboard, pointing device, display, etc.), and with one or more devices that enable a user to interact with electronic device 40, and / or with any device that enables electronic device 40 to communicate with one or more other computing devices (e.g., network card, modem, etc.). This communication can be performed via input / output (I / O) interface 411. Furthermore, electronic device 40 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 412. As shown, network adapter 412 communicates with other modules of electronic device 40 via bus 403. It should be understood that, although... Figure 6 Not shown, other hardware and / or software modules may be used in conjunction with electronic device 40, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0173] The processing unit 401 executes various functional applications and data processing by running programs stored in the system memory 402, such as implementing the method for determining patients provided in the embodiments of the present invention.

[0174] Example 7

[0175] Embodiment 7 of the present invention also provides a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform a method for determining a patient receiving user. The method includes: determining a cluster of patients awaiting treatment associated with the target patient based on basic information and symptom description information of the target patient; wherein the cluster of patients awaiting treatment includes multiple historical patients; determining at least one historical patient receiving user associated with the cluster of patients awaiting treatment, and determining a matrix to be used corresponding to the cluster of patients awaiting treatment based on the patient reception record information of each historical patient receiving user and at least one pre-edited dimension reference attribute; wherein each element in the matrix to be used represents the evaluation attribute value of each historical patient receiving user for that historical patient, and elements in the same column of the matrix to be used correspond to the same patient receiving user; processing the matrix to be used based on the historical patient reception attributes of the target patient to determine a target feature value for each historical patient receiving user relative to the target patient; and designating historical patients receiving users whose target feature values ​​are higher than a preset feature value threshold as target patients receiving the target patient.

[0176] The computer storage medium of this invention can be any combination of one or more computer-readable media. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0177] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of sending, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.

[0178] Program code contained on a computer-readable medium may be transmitted using any suitable medium, including—but not limited to—wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0179] Computer program code for performing the operations of embodiments of the present invention can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0180] Note that the above description is merely a preferred embodiment of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of the present invention, the scope of which is determined by the scope of the appended claims.

Claims

1. A method for identifying patients, characterized in that, include: Based on the basic information and symptom description of the target patient, a cluster of patients awaiting treatment associated with the target patient is identified; wherein, the cluster of patients awaiting treatment includes multiple historical patients. Identify at least one historical patient who is associated with the cluster of patients to be processed, and determine the matrix to be used corresponding to the cluster of patients to be processed based on the patient record information of each historical patient and at least one pre-edited dimension reference attribute; wherein, each element in the matrix to be used represents the evaluation attribute value of each historical patient to the historical patient, and the elements in the same column of the matrix to be used correspond to the same patient. Based on the historical medical visit attributes of the target patient, the matrix to be used is processed to determine the target feature value of each historical patient relative to the target patient. Historical patients whose target feature values ​​are higher than a preset feature value threshold are considered as target patients of the target patient. The patient reception record information must be consistent across at least three dimensions, including patient reception fee information, treatment duration information, and treatment effectiveness evaluation information. The at least one dimension reference attribute corresponds to the at least three dimensions of the patient reception record information. Determining the matrix to be used corresponding to the patient cluster to be processed, based on the patient reception record information of each historical patient and the pre-edited at least one dimension reference attribute, includes: If the at least one dimension reference attribute is consistent with any one of the at least three dimensions of the patient reception record information, then the record data to be used that is consistent with the dimension reference attribute is obtained from the at least three dimensions of the data. For each historical patient, determine the actual medical data of the current historical patient to each historical patient, and determine the vector value of the current historical patient corresponding to each historical patient based on the maximum and minimum values ​​in the actual medical data and the actual medical data of each historical patient. The matrix to be used is determined based on the vector values ​​of each historical patient. In this context, each column element in the matrix to be used corresponds to the same historical patient.

2. The method according to claim 1, characterized in that, The step of determining the cluster of patients to be processed associated with the target patient based on the target patient's basic information and symptom description information includes: Retrieve the symptom description and basic information from the symptom editing control on the target application; Based on the description of the symptoms and basic information, at least one keyword to be used is determined; Based on the at least one keyword to be used and the historical symptom descriptions of each historical patient, a cluster of patients to be processed is identified from the at least one historical patient.

3. The method according to claim 1, characterized in that, The step of determining the matrix to be used corresponding to the cluster of patients to be processed, based on the patient reception record information of each historical patient and at least one pre-edited dimension reference attribute, includes: If the at least one dimension reference attribute is consistent with at least two of the at least three dimensions of the patient reception record information, then the record data to be used that is consistent with the dimension reference attribute is obtained from the at least three dimensions of the data. For each historical patient, determine the actual medical data of the current historical patient to each historical patient, and determine the dimension vector of the current historical patient corresponding to each historical patient based on the maximum and minimum values ​​in the actual medical data and the actual medical data of each historical patient. For each historical patient, based on the weight values ​​corresponding to the pre-edited dimensional reference attributes, the dimension vectors of at least two dimensional reference attributes corresponding to the current historical patient are processed to determine the feature vector to be concatenated for the current historical patient relative to each historical patient. The matrix to be used is determined based on the feature vectors to be spliced ​​from the historical medical records of each user.

4. The method according to claim 1, characterized in that, The step of processing the matrix to be used based on the historical medical visit attributes of the target patient to determine the target feature value of each historical patient relative to the target patient includes: If the historical medical visit attributes do not include the participation in medical visits attribute, determine the mean of the elements in the same column of the matrix to be used, and determine the target feature value of each historical patient.

5. The method according to claim 4, characterized in that, The step of processing the matrix to be used based on the historical medical visit attributes of the target patient to determine the target feature value of each historical patient relative to the target patient includes: If the historical medical visit attributes include participation in medical visits, then the row data corresponding to the target medical visit user is obtained from the matrix to be used as the feature vector to be registered. Determine the similarity between the feature vector to be registered and the row vector to be called corresponding to each row element in the matrix to be used; The row vectors to be called with similarity values ​​higher than a preset similarity threshold are used as target row vectors; Based on the target row vector, determine the target feature value of each historical patient relative to the target patient.

6. The method according to claim 5, characterized in that, The step of determining the target feature value of each historical patient relative to the target patient based on the target row vector includes: Based on each target row vector, a target prediction matrix is ​​determined; by predicting and processing the target prediction matrix, the target feature value of each historical patient relative to the target patient is determined.

7. A device for identifying patients, characterized in that, include: The pending medical user cluster determination module is used to determine the pending medical user cluster associated with the target medical user based on the target medical user's basic information and symptom description information; wherein, the pending medical user cluster includes multiple historical medical users; The module for determining the matrix to be used is used to determine at least one historical patient who is associated with the cluster of patients to be processed, and to determine the matrix to be used corresponding to the cluster of patients to be processed based on the patient record information of each historical patient and at least one pre-edited dimension reference attribute; wherein, each element in the matrix to be used represents the evaluation attribute value of each historical patient to the historical patient, and the elements in the same column of the matrix to be used correspond to the same patient. The target feature value determination module is used to process the matrix to be used based on the historical medical visit attributes of the target medical user, and determine the target feature value of each historical medical user relative to the target medical user. The target patient identification module is used to identify historical patients whose target feature values ​​are higher than a preset feature value threshold as target patients of the target patient. The patient reception record information is consistent across at least three dimensions, including patient reception fee information, treatment duration information, and treatment effectiveness evaluation information. The at least one dimension reference attribute corresponds to the at least three dimensions of the patient reception record information. The matrix determination module includes: The data acquisition submodule is used to acquire data to be used that matches the dimension reference attribute from the at least three dimension data if the at least one dimension reference attribute matches any one of the at least three dimension data of the patient reception record information. The vector value determination submodule is used to determine the actual medical data of the current historical patient for each historical patient, and to determine the vector value of the current historical patient corresponding to each historical patient based on the maximum and minimum values ​​in the actual medical data and the actual medical data of each historical patient. The module for determining the matrix to be used is used to determine the matrix to be used based on the vector values ​​of each historical patient; wherein each column element in the matrix to be used corresponds to the same historical patient.

8. An electronic device, characterized in that, The electronic device includes: One or more processors; Storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method for determining the receiving user as described in any one of claims 1-6.

9. A storage medium containing computer-executable instructions, characterized in that, The computer-executable instructions, when executed by a computer processor, are used to perform the method for determining a patient as described in any one of claims 1-6.

Images