Medical data processing method, device, electronic device and storage medium

Abstract

The present application discloses a medical data processing method, device, electronic device and storage medium; including: obtaining current text data of a target user; performing label extraction processing on the current text data to obtain a target label corresponding to the target user; obtaining a negative state recovery process corresponding to the target user according to the correspondence between the label and the state recovery process, and the target label, wherein the negative state recovery process is a process for assisting the target user to recover the negative state; according to a preset sending rule, sending first data related to the negative state recovery process to a target user terminal to assist the target user to recover the negative state, wherein the target user terminal is a terminal corresponding to the target user. Compared with the prior art, the embodiments of the present application can quickly and timely pay attention to and help the negative physiological or psychological states of ordinary people.

Description

Technical Field

[0001] The present application relates to the field of intelligent medical care, and specifically to a medical data processing method, device, electronic device and storage medium. Background Art

[0002] With the improvement of living standards, the general public is paying more and more attention to their physical health, and is paying more and more attention to their own physiological or psychological sub-health status.

[0003] However, the current growth rate of medical resources cannot match the general public's demand for physical health, resulting in the general public's negative physical or psychological conditions often not receiving quick and timely attention and help. Summary of the invention

[0004] The present application provides a medical data processing method, device, electronic device and storage medium, which can improve the problem in the prior art that it is impossible to timely intervene in the negative physiological or psychological conditions of the general public.

[0005] An embodiment of the present application provides a medical data processing method, which includes: acquiring current text data of a target user; performing label extraction processing on the current text data to obtain a target label corresponding to the target user; acquiring a negative state recovery process corresponding to the target user based on the correspondence between the label and the state recovery process, and the target label, wherein the negative state recovery process is a process for assisting the target user to recover from a negative state; and sending first data related to the negative state recovery process to a target user terminal according to a pre-set sending rule to assist the target user to recover from the negative state, wherein the target user terminal is a terminal corresponding to the target user.

[0006] The present application also provides a medical data processing device, the device comprising:

[0007] A text data acquisition unit, used to acquire current text data of a target user;

[0008] A label extraction unit, used to perform label extraction processing on the current text data to obtain a target label corresponding to the target user;

[0009] A process acquisition unit, configured to acquire a negative state recovery process corresponding to the target user according to a correspondence between a tag and a state recovery process and the target tag, wherein the negative state recovery process is a process for assisting the target user to recover from a negative state;

[0010] The data sending unit is used to send the first data related to the negative state recovery process to the target user terminal according to a preset sending rule to assist the target user in recovering the negative state, wherein the target user terminal is a terminal corresponding to the target user.

[0011] In some embodiments, the label extraction unit includes:

[0012] A first encoding subunit, configured to perform a first encoding process on a first number of characters included in the current text data to obtain a first encoding result;

[0013] A letter representation subunit, used for obtaining a letter representation corresponding to each character included in the current text data;

[0014] A second encoding subunit is used to perform a second encoding process on the first number of letter representation characters to obtain a second encoding result;

[0015] a merging result subunit, configured to merge the first encoding result and the second encoding result to obtain a merged result;

[0016] An interaction result subunit, used for obtaining the interaction result corresponding to the merged result according to the attention mechanism;

[0017] a fully connected subunit, configured to perform a fully connected transformation on the interaction result to obtain a first number of output vectors, each of which has a second number of dimensions, wherein the first number of output vectors corresponds one-to-one to the first number of characters, the second number is the total number of labels, and the second number of dimensions corresponds one-to-one to the second number of labels;

[0018] The target subunit is used to obtain, for each output vector of the first number of output vectors, at least one dimension whose score value exceeds a preset score value from the second number of dimensions, wherein the label corresponding to at least one of the dimensions is a target label of the character corresponding to the dimension.

[0019] In some embodiments, the label extraction unit is specifically used to perform label extraction processing on the current text data using a trained label extraction model to obtain a target label corresponding to the target user.

[0020] In some embodiments, the label extraction model includes a first feature extraction model, a second feature extraction model, an attention layer, and a fully connected layer;

[0021] A first encoding subunit, specifically configured to perform a first encoding process on the first number of characters using the first feature extraction model to obtain a first encoding result;

[0022] A second encoding subunit is specifically used to perform a second encoding process on the first number of letter representation symbols using the second feature extraction model to obtain a second encoding result;

[0023] An interaction result subunit, specifically used to obtain an interaction result corresponding to the merged result using the attention layer;

[0024] The fully connected subunit is specifically used to use the fully connected layer to perform a fully connected transformation on the interaction result to obtain a first number of output vectors.

[0025] In some embodiments, the data sending unit includes:

[0026] An inquiry information subunit, configured to send a plurality of inquiry information to the target user terminal at a first time node;

[0027] A feedback information receiving subunit, configured to receive a plurality of feedback information returned by the target user terminal;

[0028] A solution generation subunit, used to generate a targeted repair solution according to the plurality of feedback information;

[0029] The solution sending subunit is used to send the targeted repair solution to the target user terminal.

[0030] In some embodiments, the apparatus further comprises:

[0031] The popular science sending unit is used to send target popular science knowledge to the target user terminal at a second time node in the process execution cycle, wherein the process execution cycle is the execution cycle of the negative state recovery process, and the target popular science knowledge is popular science knowledge associated with the negative state.

[0032] In some embodiments, the apparatus further comprises:

[0033] An inquiry information sending unit, configured to send indicator inquiry information to the target user terminal at a third time node of the process execution cycle, wherein the indicator inquiry information is used to inquire about multiple indicator parameters of the target user;

[0034] An indicator parameter receiving unit, configured to receive the multiple indicator parameters returned by the target user terminal;

[0035] The early warning display unit is used to generate and display an early warning signal when there is an indicator parameter whose parameter value exceeds its corresponding normal range value among the multiple indicator parameters.

[0036] In some embodiments, the apparatus further comprises:

[0037] An indicator parameter acquisition unit, configured to acquire a plurality of indicator parameters corresponding to the third time node of a preset number of process execution cycles after the negative state recovery process has run through a preset number of process execution cycles;

[0038] A new solution generating unit, configured to generate a new targeted repair solution according to a plurality of indicator parameters corresponding to the third time node of the preset number of process execution cycles;

[0039] The new solution sending unit is used to send the new targeted repair solution to the target user terminal.

[0040] In the medical data processing method provided in the embodiment of the present application, the current text data of the target user can be obtained, and then the label extraction processing is performed on the current text data to obtain one or more target labels corresponding to the target user. Then, based on the correspondence between the label and the state recovery process, and one or more target labels, one or more negative state recovery processes corresponding to the target user are obtained; wherein the negative state recovery process can assist the target user in recovering the negative state. According to the sending rules, the first data is sent to the target user terminal corresponding to the target user, so as to cooperate with the target user to recover his own negative state; wherein the first data is associated with the negative state recovery process.

[0041] In the present application, the target tag of the target user can be obtained based on the current text data of the target user, and then the corresponding negative state recovery process can be obtained according to the target tag, and then the first data related to the negative state recovery process can be sent to the target user terminal according to the preset sending rules, so that the target user corresponding to the target user terminal can recover his own negative state according to the first data. Compared with the existing technology, the negative physiological or psychological states of the general public can be paid attention to and helped quickly and timely. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0043] Figure 1 is a scenario diagram of a medical data processing method provided in an embodiment of the present application;

[0044] Figure 2 is a flowchart of a medical data processing method provided by an embodiment of the present application;

[0045] Figure 3A schematic diagram of a model showing a specific implementation of a medical data processing model in the application stage;

[0046] Figure 4 Shows Figure 3 A schematic diagram of a specific implementation method of the step of "obtaining the tags existing in the current text" in the application stage;

[0047] Figure 5 is a structural schematic diagram of a medical data processing device provided by an embodiment of the present application;

[0048] Figure 6 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0050] Embodiments of the present application provide a medical data processing method, device, electronic device and storage medium.

[0051] The medical data processing device may be integrated into an electronic device, which may be a terminal, a server, or other device. The terminal may be a mobile phone, a tablet computer, a smart Bluetooth device, a laptop, or a personal computer (PC). The server may be a single server or a server cluster consisting of multiple servers.

[0052] In some embodiments, the medical data processing device may also be integrated into multiple electronic devices. For example, the medical data processing device may be integrated into multiple servers, and the medical data processing method of the present application may be implemented by multiple servers.

[0053] In some embodiments, the server may also be implemented in the form of a terminal.

[0054] For example, see Figure 1The electronic device can execute the following method: obtaining current text data of a target user; performing label extraction processing on the current text data to obtain a target label corresponding to the target user; obtaining a negative state recovery process corresponding to the target user according to the correspondence between the label and the state recovery process, and the target label, wherein the negative state recovery process is a process for assisting the target user to recover the negative state; according to a preset sending rule, sending first data related to the negative state recovery process to a target user terminal to assist the target user to recover the negative state, wherein the target user terminal is a terminal corresponding to the target user.

[0055] The medical data processing method provided in the embodiment of the present application can be used to restore the negative state of the target user. The negative state is the physiological or psychological state of the target user that is unfavorable to the health condition. The negative state can be a bad psychological emotion, such as anxiety, depression, mania, etc.; it can also be a sub-healthy state of the body, such as dizziness, headache, palpitations, chest tightness, sleep disorders, loss of appetite, etc.; it can also be a state of suffering from a disease.

[0056] It should be noted that the serial numbers of the following embodiments are not intended to limit the preferred order of the embodiments.

[0057] In this embodiment, a medical data processing method is provided. Figure 2 As shown, the medical data processing method is applied in a terminal, and the specific process of the method may include the following steps 110 to 140:

[0058] 110. Obtain current text data of the target user.

[0059] The current text data is text information related to the target user and belonging to a certain medical field. The current text data can be a psychological test text result that represents the psychological state of the target user; it can also be text data that represents the physiological state of the target user, such as the target user's physical examination text result; it can also be text data that represents the target user's disease recovery status, such as the target user's hospital admission record, first course of illness record, surgical data, discharge record, etc. For ease of description, the following may be explained by taking the current text as text data that represents the target user's disease recovery status as an example.

[0060] 120. Perform label extraction processing on the current text data to obtain a target label corresponding to the target user.

[0061] The target tag is a tag used to reflect the user characteristics summarized in the current text data of the target user. Multiple target tags can build a user profile of the target user.

[0062] Continuing with the example above, the target label can be specific information such as gender, age, diagnosis, surgery name, drug name, discharge date, smoking history, etc. Among them, some labels of the target user, such as gender and age, are well structured in the text and usually appear in specific locations in the text. They can be directly obtained based on preset rules; however, for labels whose content is embedded in a piece of text, such as diagnosis, surgery name, drug name, discharge date, smoking history, etc., they can be extracted through the label extraction model. For details, please refer to Figure 3 , the label extraction model may include a first feature extraction model, a second feature extraction model, an attention layer, and a fully connected layer.

[0063] Optionally, in a specific implementation, step 120 may specifically include the following steps 121 to 127:

[0064] 121. Perform a first encoding process on a first number of characters included in the current text data to obtain a first encoding result.

[0065] The value range of the first quantity is a positive integer, and the specific value of the first quantity should not be understood as a limitation on the present application. The first encoding result includes a first number of d-dimensional vectors, and the first number of d-dimensional vectors correspond one-to-one to the first number of characters, and the first encoding result is a matrix of the first number × d. Among them, the value of d is a positive integer. For example, d can take values ​​such as 50, 80, 100, etc., and its specific value should not be understood as a limitation on the present application.

[0066] Optionally, step 121 may be Figure 3 Correspondingly, step 121 may specifically utilize the first feature extraction model in the label extraction model to perform a first encoding process on the first number of characters to obtain a first encoding result.

[0067] For more information, please see Figure 3 , let's assume that the current text data is: "Ticagrelor (90mg / tablet) twice a day, one tablet each time", then the first quantity of the current text data specifically corresponds to 16, and the 16 characters are: ticagrelor, 90mg, / , tablet, every, day, two, times, every, times, one, tablet. By using the first feature extraction model in the label extraction model to perform the first encoding processing on the above 16 characters, 16 d-dimensional vectors can be obtained.

[0068] The first feature extraction model may be a BERT model, that is, the first feature extraction model may be a first BERT model; the first feature extraction model may also be a CNN model, that is, the first feature extraction model may be a first CNN model. The specific model structure of the first feature extraction model should not be understood as a limitation to the present application.

[0069] 122. Obtain a letter representation corresponding to each character included in the current text data.

[0070] The letter representation is the letter representation of each Chinese character. The letter representation is used to enhance the error correction capability of the label extraction model. The letter representation can be the pinyin letter corresponding to each Chinese character. For details, see Figure 3 ; It can also be the English letters corresponding to each Chinese character. Since the letter representation symbol corresponds to the character one by one, the number of the letter representation symbols is also the first number.

[0071] There is a preset correspondence between characters and letter representations, and the correspondence can be stored in a database of a server, and the letter representation corresponding to each character can be obtained by searching the database.

[0072] For more information, please see Figure 3 , we may assume that the letter representation is the pinyin corresponding to the character, and continue with the above example. The 16 letter representations are: ti, ge, rui, luo, 90, mg, mei, pian, mei, tian, liang, ci, mei, ci, yi, pian.

[0073] 123. Perform a second encoding process on the first number of letter representation characters to obtain a second encoding result.

[0074] The second encoding result includes a first number of e-dimensional vectors, the first number of e-dimensional vectors correspond one-to-one to the first number of letter representation symbols, and the second encoding result is a matrix of the first number × e. Wherein, the value of e is a positive integer; the value of e may be the same as or different from the value of d, and the value range of e should not be understood as a limitation on the present application.

[0075] For more information, please see Figure 3 In step 123, the second feature extraction model may be used to perform a second encoding process on the first number of letter representation symbols to obtain a second encoding result.

[0076] The second feature extraction model may be a BERT model, that is, the second feature extraction model may be a second BERT model; the second feature extraction model may also be a CNN model, that is, the second feature extraction model may be a second CNN model. The specific model structure of the second feature extraction model should not be understood as a limitation to the present application.

[0077] 124. Merge the first encoding result and the second encoding result to obtain a merged result.

[0078] In one implementation, each d-dimensional vector may be concatenated with the corresponding e-dimensional vector to obtain a first number of (d+e)-dimensional vectors. In this implementation, the value of d may be the same as or different from that of e.

[0079] In another implementation, assuming that the value of d is the same as the value of e, the first encoding result and the second encoding result can be directly concatenated to obtain twice the first number of d-dimensional vectors (or e-dimensional vectors).

[0080] It should be understood that the specific way of merging the first encoding result and the second encoding result should not be understood as a limitation to the present application.

[0081] 125. Obtain an interaction result corresponding to the merged result according to the attention mechanism.

[0082] Step 125 can specifically utilize the attention layer to obtain the interaction result corresponding to the merged result.

[0083] The calculation process of the attention layer will be different depending on the form of the merged result. For the sake of description, let's take the merged result as the first number of (d+e)-dimensional vectors as an example, and the first number is represented by m. For each of the m (d+e)-dimensional vectors, the following steps can be performed:

[0084] Let us take the first (d+e)-dimensional vector as an example for explanation. The first (d+e)-dimensional vector is any (d+e)-dimensional vector among the m (d+e)-dimensional vectors.

[0085] Calculate the inner product of the first (d+e)-dimensional vector and m (d+e) vectors, and get a total of m products.

[0086] By performing softmax transformation on these m products, we can get the weight values ​​corresponding to the first (d+e)-dimensional vector and all the m (d+e)-dimensional vectors.

[0087] Then, the weighted sum of the m (d+e)-dimensional vectors is calculated, and the weighted sum is the interaction result of the first (d+e)-dimensional vector.

[0088] In the above manner, the interaction results of each (d + e)-dimensional vector among all m (d + e)-dimensional vectors can be obtained.

[0089] 126. Perform a fully-connected transformation on the interaction results to obtain a first number of output vectors, where the dimension of each output vector is a second number. Among them, the first number of output vectors correspond one-to-one with the first number of characters, the second number is the total number of labels, and the second number of dimensions correspond one-to-one with the second number of labels.

[0090] For details, please refer to Figure 3 In step 126, specifically, the fully-connected layer can be used to perform a fully-connected transformation on the interaction results to obtain a first number of output vectors.

[0091] Input the interaction results into the fully-connected layer of the label extraction model, and the second number of dimensional vectors corresponding to each of the 16 characters "Ti, Ge, Rui, Luo, 90, mg, / , Pian, Mei, Tian, Liang, Ci, Mei, Ci, Yi, Pian" can be obtained. Among them, the second number is the number of labels that the label extraction model can recognize, which may be denoted as n, and n is a positive integer; therefore, the vectors corresponding to each character are all n-dimensional vectors, and among them, the n-dimensional vectors corresponding to each character all have their respective score values.

[0092] 127. For each of the first number of output vectors, obtain at least one dimension from the second number of dimensions whose score value exceeds a preset score value, where the label corresponding to at least one of the dimensions is the target label of the character corresponding to the dimension.

[0093] For each of multiple characters, it can be calculated whether there is one or more dimensions in the corresponding n-dimensional vector whose score value exceeds the preset score value. If so, the target label of the character corresponding to the above one or more dimensions can be considered as the target label of the character.

[0094] For details, please refer to Figure 4, continuing with the above example for illustration, the first quantity takes a value of 16. For each of the 16 output vectors corresponding to the 16 characters "替, 格, 瑞, 洛, 90, mg, / , 片, 每, 天, 两, 次, 每, 次, 一, 片", namely "output vector 1, output vector 2, output vector 3, output vector 4, output vector 5, output vector 6, output vector 7, output vector 8, output vector 9, output vector 10, output vector 11, output vector 12, output vector 13, output vector 14, output vector 15, output vector 16", for each of the n dimensions of each output vector, compare the score value of each dimension with a preset score value, obtain one or more dimensions whose score values exceed the preset score value, and use the label corresponding to the above dimensions as the target label of the character corresponding to the output vector. Suppose the score value of the 3rd dimension of output vector 4 exceeds the preset score value, then obtain the character "洛" corresponding to output vector 4, and the target label corresponding to this character is label 3.

[0095] 130. According to the corresponding relationship between the label and the status recovery process, and the target label, obtain the negative status recovery process corresponding to the target user, where the negative status recovery process is a process to assist the target user in recovering from a negative status.

[0096] The status recovery process is a process to guide the user to adjust and recover from their corresponding negative status. The corresponding relationship between the label and the status recovery process can be preset. Optionally, the corresponding relationship between the label and the status recovery process can be a one-to-one correspondence. For example, label 1 corresponds to status recovery process B, label 3 corresponds to status recovery process D, etc.; the corresponding relationship between the label and the status recovery process can also be a many-to-one correspondence. For example, the three labels 1, 3, and 4 correspond to status recovery process A, and the two labels 2 and 3 correspond to status recovery process C, etc.; there can also be both a one-to-one correspondence and a one-to-many correspondence.

[0097] Continuing with the above example for illustration, suppose after being processed by step 120, the target labels corresponding to the target user are: label 1, label 2, and label 3. Then, according to the corresponding relationship between the label and the status recovery process described above, the negative status recovery processes corresponding to the target user can be obtained, including: status recovery process B (corresponding to label 1), status recovery process D (corresponding to label 3), and status recovery process C (corresponding to labels 2 and 3). It should be understood that since the target labels corresponding to the target user can be multiple, the negative status recovery processes determined according to step 130 can also be multiple.

[0098] Let label 1 be a diagnosis of coronary heart disease, label 2 be an age no older than 80 years old, and label 3 be a discharge date within 3 months. The corresponding state recovery process B can be a medication process, state recovery process D can be a smoking cessation process, and state recovery process C can be a food nutrition process.

[0099] 140. According to a preset sending rule, first data related to the negative status recovery process is sent to a target user terminal to assist the target user in recovering the negative status.

[0100] The target user terminal is a terminal corresponding to the target user.

[0101] The first data is data that assists the target user in recovering from a negative state. The first data may include data that helps to understand the current state of the target user, such as inquiry information; it may also include data on specific measures taken against the negative state, such as targeted repair plans.

[0102] The target user terminal may be a terminal device held by or owned by the target user, for example, a mobile phone, a tablet computer, a smart Bluetooth device, a laptop computer, or a personal computer (PC); the target user terminal may also be a terminal device bound to the target user.

[0103] Optionally, in a specific implementation, step 140 may specifically include the following steps 141 to 144:

[0104] 141. At a first time point, send multiple inquiry information to the target user terminal.

[0105] The first time node can be determined according to the time node when a specific behavior of the target user occurs. There is a fixed relative relationship between the first time node and the time node when the specific behavior occurs. For example, the first time node can be the same as the time node when the specific behavior occurs, or the first time node can be the next day after the time node when the specific behavior occurs. The time node when a specific behavior of the target user occurs can be obtained from the database.

[0106] Depending on the negative state of the target user, the type of specific behavior will also change. For example, if the negative state is a negative psychological emotion, the specific behavior can be that the target user starts psychological counseling; if the negative state is a sub-healthy state of the body, the specific behavior can be that the target user obtains the results of a physical examination; if the negative state is a state of suffering from a disease, the specific behavior can be that the target user is discharged from the hospital. For ease of description, let's take the example of a negative state being a state of suffering from a disease and the specific behavior being that the target user is discharged from the hospital.

[0107] The inquiry information is question information related to the negative status; for example, it can be the target user's physical index parameters and diet. Physical index parameters include weight, blood pressure, blood sugar, etc.; diet includes whether to drink alcohol, whether to eat fried food, and the frequency of eating fried food (such as how many times a month).

[0108] 142. Receive multiple feedback information returned by the target user terminal.

[0109] After receiving the multiple inquiry information, the target user terminal displays the multiple inquiry information to the target user corresponding to the target user terminal so that the target user can answer the inquiry information one by one. After the target user completes the answer, the target user terminal sends the target user's answer as feedback information to the server.

[0110] 143. Generate a targeted repair plan based on the multiple feedback information.

[0111] The targeted repair plan is a plan to carry out targeted repairs on the negative status of the target user.

[0112] The server can generate a targeted repair plan based on the multiple feedback information received. The process of generating a targeted repair plan is as follows:

[0113] Obtain the specific values ​​of the target user's physical index parameters and dietary conditions, and obtain the numerical range corresponding to the specific values; adjust the numerical values ​​in the targeted repair plan according to the numerical range. The targeted repair plan includes a pre-set template text and a numerical value reflecting the number of behaviors represented by the text. For example, let's assume that the targeted repair plan is a targeted dietary plan, then the template text can be:

[0114] Breakfast: carbohydrates __g, fat __g, protein __g;

[0115] Lunch: Carbohydrates __g, fat __g, protein __g;

[0116] Dinner: Carbohydrates __g, fat __g, protein __g.

[0117] The “__” part in the template text is the numerical value of the number of behaviors represented above.

[0118] For example, if the target user's weight value x1 is within the normal weight range, the generated targeted repair plan can be:

[0119] Breakfast: Carbohydrates A1 Grams, Fat B1 Grams, protein C1 gram;

[0120] Lunch: Carbohydrates A2 Grams, Fat B2 Grams, protein C2 gram;

[0121] Dinner: Carbohydrates A3 Grams, Fat B3 Grams, protein C3 gram.

[0122] For example, if the target user's weight value x2 is in the overweight range, the generated targeted repair plan can be:

[0123] Breakfast: Carbohydrates A1-a1 Grams, Fat B1-b1 Grams, protein C1+c1 gram;

[0124] Lunch: Carbohydrates A2-a2 Grams, Fat B2-b2 Grams, protein C2+c2 gram;

[0125] Dinner: Carbohydrates A3-a3 Grams, Fat B3-b3 Grams, protein C3+c3 gram.

[0126] Among them, the correspondence between the above-mentioned numerical range and the numerical value in the adjustment targeted repair solution can be stored in a preset state repair knowledge graph.

[0127] In addition to the above-mentioned targeted repair solution generation rules, the state repair knowledge graph can also include the following corresponding to the negative state of the target user: pre-set template text, inquiry information, nutrition knowledge, humanistic care knowledge, indicator inquiry information, risk determination rules, etc. The state repair knowledge graph can be sorted out by the staff.

[0128] 144. Send the targeted repair solution to the target user terminal.

[0129] The server may send the targeted repair solution to the target user terminal, so that the target user corresponding to the target user terminal may adjust his / her living habits according to the targeted repair solution.

[0130] Targeted repair solutions can be presented in a variety of forms on the target user terminal, such as video, picture, text, voice, etc., so as to meet the target user's need to know the targeted repair solutions as much as possible and improve the target user's user experience.

[0131] Optionally, in a specific embodiment, the method further comprises:

[0132] At a second time point in the process execution cycle, target popular science knowledge is sent to the target user terminal.

[0133] Among them, the process execution cycle is the execution cycle of the negative state recovery process, and the execution cycle can be 7 days or 10 days. The specific time length of the process execution cycle should not be understood as a limitation on this application.

[0134] The target popular science knowledge is the popular science knowledge related to the negative state. Specifically, it can be the nutrition knowledge and humanistic care knowledge in the state repair knowledge graph mentioned above; among which, the humanistic care knowledge can be the solar term diet reminder information.

[0135] The second time node is a time node that is relatively fixed to the process execution cycle, that is, each process execution cycle has a second time node. For example, assuming that the process execution cycle is one week, the second time node may be the middle of the week, such as Wednesday or Thursday.

[0136] In the above implementation, the target user's awareness of his / her own negative state can be improved by sending target popular science knowledge to the target user terminal, thereby making it easier for the target user to repair his / her own negative state.

[0137] Optionally, in a specific implementation, the method may further include the following steps S1 to S3:

[0138] S1. At a third time point in a process execution cycle, indicator query information is sent to the target user terminal.

[0139] The indicator query information is used to query the target user for multiple indicator parameters. Continuing with the above example, the indicator parameters can be the target user's physical indicators, which include indicators that can be measured by the target user at home, such as weight, blood pressure, blood sugar, etc., and indicators that require professional equipment to be tested, such as international normalized ratio (INR).

[0140] The third time node is a time node that is relatively fixed to the process execution cycle, that is, each process execution cycle has a third time node. For example, assuming that the process execution cycle is one week, the third time node may be a weekend of the week, such as Saturday or Sunday.

[0141] S2. Receive the multiple indicator parameters returned by the target user terminal.

[0142] After receiving the indicator query information, the target user terminal displays the indicator query information to the target user corresponding to the target user terminal, so that the target user can fill in the multiple indicators included in the indicator query information one by one. After the target user completes filling in the indicator parameters, the target user terminal returns the indicator parameters filled in by the target user to the server.

[0143] S3. If, among the multiple indicator parameters, there are indicator parameters whose parameter values ​​exceed their corresponding normal range values, a warning signal is generated and displayed.

[0144] For each of the multiple indicator parameters, the indicator parameter can be compared with its corresponding normal range value. If there is an indicator parameter whose parameter value exceeds its corresponding normal range value among the multiple indicator parameters, a warning signal is generated and displayed. Among them, the warning signal may include a warning signal for the target user, and may also include a warning signal for the medical staff corresponding to the target user. The warning signal for the target user may be a warning signal to remind the target user to seek medical treatment in time, and the warning signal for the medical staff corresponding to the target user may be a warning signal to remind the medical staff to treat the target user in time.

[0145] For example, suppose that among multiple indicator parameters, one indicator parameter is INR, and its corresponding value is 3.33, but the normal value range of INR is 0.8 to 1.2; therefore, it can be determined that the INR parameter is abnormal, and a warning signal can be generated and displayed.

[0146] In the above implementation, an indicator query message can be sent to the target user terminal at the third time node of each process execution cycle to obtain the indicator parameters of the target user; and each indicator parameter can be compared with the corresponding normal range value. If there is an indicator parameter that exceeds the normal range value, an early warning signal can be issued to remind the target user and the corresponding medical staff, so as to provide protection for the target user as much as possible. For the target user, this method can perform status repair management on the target user online, minimize the frequency of the target user's daily visits to medical institutions, and save the target user's medical costs; for medical staff, this method can help medical staff to conduct rehabilitation management of patient users after diagnosis, making it possible for a doctor to guide the rehabilitation of hundreds or thousands of patient users at the same time, saving social medical resources.

[0147] Optionally, in a specific implementation, the method may further include the following steps A1 to A3:

[0148] A1. After the negative state recovery process has run through a preset number of process execution cycles, a plurality of indicator parameters corresponding to the third time node of the preset number of process execution cycles are obtained.

[0149] The value of the preset number is a positive integer, and the specific value of the preset number should not be understood as a limitation of the present application. In a specific implementation, the preset number can be 4, that is, after 4 process execution cycles, the indicator parameter corresponding to the third time node of the 4th process execution cycle can be obtained.

[0150] A2. Generate a new targeted repair plan based on multiple indicator parameters corresponding to the third time node of the preset number of process execution cycles.

[0151] The specific method of generating a new targeted repair plan based on multiple indicator parameters is the same as the specific method of generating a targeted repair plan in step 143, which will not be repeated here.

[0152] A3. Send the new targeted repair solution to the target user terminal.

[0153] In the above-mentioned implementation, after the negative state recovery process is executed for a period of time (such as a preset number of process execution cycles), the targeted repair plan can be adjusted according to the newly acquired indicator parameters, so as to ensure that the targeted repair plan is a plan for the newer data of the target user, thereby helping to improve the accuracy and timeliness of the targeted repair plan.

[0154] In the medical data processing method provided in the embodiment of the present application, the current text data of the target user can be obtained, and then the label extraction processing is performed on the current text data to obtain one or more target labels corresponding to the target user. Then, based on the correspondence between the label and the state recovery process, and one or more target labels, one or more negative state recovery processes corresponding to the target user are obtained; wherein the negative state recovery process can assist the target user in recovering the negative state. According to the sending rules, the first data is sent to the target user terminal corresponding to the target user, so as to cooperate with the target user to recover his own negative state; wherein the first data is associated with the negative state recovery process.

[0155] In the present application, the target tag of the target user can be obtained based on the current text data of the target user, and then the corresponding negative state recovery process can be obtained according to the target tag, and then the first data related to the negative state recovery process can be sent to the target user terminal according to the preset sending rules, so that the target user corresponding to the target user terminal can recover his own negative state according to the first data. Compared with the existing technology, the negative physiological or psychological states of the general public can be paid attention to and helped quickly and timely.

[0156] In order to better implement the above method, the embodiment of the present application also provides a medical data processing device, which can be integrated in an electronic device, and the electronic device can be a terminal, a server, etc. Among them, the terminal can be a mobile phone, a tablet computer, a smart Bluetooth device, a laptop, or a personal computer (Personal Computer, PC) and other devices; the server can be a single server or a server cluster composed of multiple servers. For example, Figure 5 As shown, the medical data processing device may include:

[0157] A text data acquisition unit 501 is used to acquire current text data of a target user;

[0158] The tag extraction unit 502 is used to perform tag extraction processing on the current text data to obtain a target tag corresponding to the target user;

[0159] A process acquisition unit 503 is used to acquire the negative state recovery process corresponding to the target user according to the corresponding relationship between the tag and the state recovery process and the target tag, wherein the negative state recovery process is a process for assisting the target user to recover from the negative state;

[0160] The data sending unit 504 is used to send the first data related to the negative status recovery process to the target user terminal according to a preset sending rule, so as to assist the target user to recover from the negative status.

[0161] In some embodiments, the tag extraction unit 502 includes:

[0162] A first encoding subunit, configured to perform a first encoding process on a first number of characters included in the current text data to obtain a first encoding result;

[0163] A letter representation subunit, used for obtaining a letter representation corresponding to each character included in the current text data;

[0164] A second encoding subunit is used to perform a second encoding process on the first number of letter representation characters to obtain a second encoding result;

[0165] a merging result subunit, configured to merge the first encoding result and the second encoding result to obtain a merged result;

[0166] An interaction result subunit, used for obtaining the interaction result corresponding to the merged result according to the attention mechanism;

[0167] a fully connected subunit, configured to perform a fully connected transformation on the interaction result to obtain a first number of output vectors, each of which has a second number of dimensions, wherein the first number of output vectors corresponds one-to-one to the first number of characters, the second number is the total number of labels, and the second number of dimensions corresponds one-to-one to the second number of labels;

[0168] The target subunit is used to obtain, for each output vector of the first number of output vectors, at least one dimension whose score value exceeds a preset score value from the second number of dimensions, wherein the label corresponding to at least one of the dimensions is a target label of the character corresponding to the dimension.

[0169] In some embodiments, the label extraction unit 502 is specifically configured to perform label extraction processing on the current text data using a trained label extraction model to obtain a target label corresponding to the target user.

[0170] In some embodiments, the label extraction model includes a first feature extraction model, a second feature extraction model, an attention layer, and a fully connected layer;

[0171] A first encoding subunit, specifically configured to perform a first encoding process on the first number of characters using the first feature extraction model to obtain a first encoding result;

[0172] A second encoding subunit is specifically used to perform a second encoding process on the first number of letter representation symbols using the second feature extraction model to obtain a second encoding result;

[0173] An interaction result subunit, specifically used to obtain an interaction result corresponding to the merged result using the attention layer;

[0174] The fully connected subunit is specifically used to use the fully connected layer to perform a fully connected transformation on the interaction result to obtain a first number of output vectors.

[0175] In some embodiments, the data sending unit 504 includes:

[0176] An inquiry information subunit, configured to send a plurality of inquiry information to the target user terminal at a first time node;

[0177] A feedback information receiving subunit, configured to receive a plurality of feedback information returned by the target user terminal;

[0178] A solution generation subunit, used to generate a targeted repair solution according to the plurality of feedback information;

[0179] The solution sending subunit is used to send the targeted repair solution to the target user terminal.

[0180] In some embodiments, the apparatus further comprises:

[0181] The popular science sending unit is used to send target popular science knowledge to the target user terminal at a second time node in the process execution cycle, wherein the process execution cycle is the execution cycle of the negative state recovery process, and the target popular science knowledge is popular science knowledge associated with the negative state.

[0182] In some embodiments, the apparatus further comprises:

[0183] An inquiry information sending unit, configured to send indicator inquiry information to the target user terminal at a third time node of the process execution cycle, wherein the indicator inquiry information is used to inquire about multiple indicator parameters of the target user;

[0184] An indicator parameter receiving unit, configured to receive the multiple indicator parameters returned by the target user terminal;

[0185] The early warning display unit is used to generate and display an early warning signal when there is an indicator parameter whose parameter value exceeds its corresponding normal range value among the multiple indicator parameters.

[0186] In some embodiments, the apparatus further comprises:

[0187] An indicator parameter acquisition unit, configured to acquire a plurality of indicator parameters corresponding to the third time node of a preset number of process execution cycles after the negative state recovery process has run through a preset number of process execution cycles;

[0188] A new solution generating unit, configured to generate a new targeted repair solution according to a plurality of indicator parameters corresponding to the third time node of the preset number of process execution cycles;

[0189] The new solution sending unit is used to send the new targeted repair solution to the target user terminal.

[0190] In specific implementation, the above units can be implemented as independent entities, or can be arbitrarily combined to be implemented as the same or several entities. The specific implementation of the above units can refer to the previous method embodiments, which will not be repeated here.

[0191] In the present application, the target tag of the target user can be obtained based on the current text data of the target user, and then the corresponding negative status recovery process can be obtained according to the target tag, and then the first data related to the negative status recovery process can be sent to the target user terminal according to the pre-set sending rules, so that the target user corresponding to the target user terminal can recover his own negative status according to the first data.

[0192] In the present application, compared with the prior art, negative physiological or psychological conditions of ordinary people can be quickly and promptly paid attention to and helped.

[0193] The embodiment of the present application also provides an electronic device, which can be a terminal, a server, etc. The terminal can be a mobile phone, a tablet computer, a smart Bluetooth device, a laptop, a personal computer, etc. The server can be a single server or a server cluster composed of multiple servers, etc.

[0194] In some embodiments, the medical data processing device may also be integrated into multiple electronic devices. For example, the medical data processing device may be integrated into multiple servers, and the medical data processing method of the present application may be implemented by multiple servers.

[0195] In this embodiment, the electronic device of this embodiment is an electronic device as an example for detailed description, for example, Figure 6 As shown, it shows a schematic diagram of the structure of the electronic device involved in the embodiment of the present application, specifically:

[0196] The electronic device may include one or more processors 601 of processing cores, one or more computer-readable storage media memories 602, a power supply 603, an input module 604, and a communication module 605. Those skilled in the art will appreciate that Figure 6 The electronic device structure shown in the figure does not constitute a limitation on the electronic device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0197] The processor 601 is the control center of the electronic device. It uses various interfaces and lines to connect various parts of the entire electronic device. By running or executing software programs and / or modules stored in the memory 602 and calling data stored in the memory 602, it performs various functions of the electronic device and processes data, thereby monitoring the electronic device as a whole. In some embodiments, the processor 601 may include one or more processing cores; in some embodiments, the processor 601 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application programs, and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 601.

[0198] The memory 602 can be used to store software programs and modules. The processor 601 executes various functional applications and medical data processing by running the software programs and modules stored in the memory 602. The memory 602 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of the electronic device, etc. In addition, the memory 602 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the memory 602 may also include a memory controller to provide the processor 601 with access to the memory 602.

[0199] The electronic device also includes a power supply 603 for supplying power to various components. In some embodiments, the power supply 603 can be logically connected to the processor 601 through a power management system, so as to manage charging, discharging, and power consumption through the power management system. The power supply 603 can also include any components such as one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, and power status indicators.

[0200] The electronic device may further include an input module 604, which may be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control.

[0201] The electronic device may further include a communication module 605. In some embodiments, the communication module 605 may include a wireless module. The electronic device may perform short-range wireless transmission through the wireless module of the communication module 605, thereby providing the user with wireless broadband Internet access. For example, the communication module 605 may be used to help the user send and receive emails, browse web pages, and access streaming media.

[0202] Although not shown, the electronic device may further include a display unit, etc., which will not be described in detail herein. Specifically in this embodiment, the processor 601 in the electronic device will load the executable files corresponding to the processes of one or more application programs into the memory 602 according to the following instructions, and the processor 601 will run the application programs stored in the memory 602, thereby realizing various functions, as follows:

[0203] Acquire current text data of a target user; perform label extraction processing on the current text data to obtain a target label corresponding to the target user; acquire a negative state recovery process corresponding to the target user based on a correspondence between the label and the state recovery process, and the target label, wherein the negative state recovery process is a process for assisting the target user to recover from a negative state; according to a pre-set sending rule, send first data related to the negative state recovery process to a target user terminal to assist the target user to recover from the negative state, wherein the target user terminal is a terminal corresponding to the target user.

[0204] The specific implementation of the above operations can be found in the previous embodiments, which will not be described in detail here.

[0205] A person of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0206] To this end, an embodiment of the present application provides a computer-readable storage medium, in which a plurality of instructions are stored, and the instructions can be loaded by a processor to execute the steps in any medical data processing method provided in the embodiment of the present application. For example, the instructions can execute the following steps:

[0207] Acquire current text data of a target user; perform label extraction processing on the current text data to obtain a target label corresponding to the target user; acquire a negative state recovery process corresponding to the target user based on a correspondence between the label and the state recovery process, and the target label, wherein the negative state recovery process is a process for assisting the target user to recover from a negative state; according to a pre-set sending rule, send first data related to the negative state recovery process to a target user terminal to assist the target user to recover from the negative state, wherein the target user terminal is a terminal corresponding to the target user.

[0208] The storage medium may include: a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0209] According to one aspect of the present application, a computer program product or a computer program is provided, the computer program product or the computer program including computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the methods provided in the various optional implementations provided in the above embodiments.

[0210] Since the instructions stored in the storage medium can execute the steps in any medical data processing method provided in the embodiments of the present application, the beneficial effects that can be achieved by any medical data processing method provided in the embodiments of the present application can be achieved. Please refer to the previous embodiments for details and will not be repeated here.

[0211] The above is a detailed introduction to a medical data processing method, device, electronic device and computer-readable storage medium provided in the embodiments of the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for technical personnel in this field, according to the ideas of the present application, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A medical data processing method, characterized in that: The method comprises: Get the current text data of the target user; Performing label extraction processing on the current text data to obtain a target label corresponding to the target user; According to the correspondence between the tag and the state recovery process and the target tag, the negative state recovery process corresponding to the target user is obtained, wherein the negative state recovery process is a process for assisting the target user to recover from the negative state; According to a preset sending rule, first data related to the negative status recovery process is sent to a target user terminal to assist the target user in recovering the negative status; The step of performing label extraction processing on the current text data to obtain a target label corresponding to the target user includes: Performing a first encoding process on a first number of characters included in the current text data to obtain a first encoding result; Obtaining a letter representation corresponding to each character included in the current text data, wherein the letter representation is a letter representation of the character; Performing a second encoding process on the first number of letter representation characters to obtain a second encoding result; Merging the first encoding result and the second encoding result to obtain a merged result; Obtaining the interaction result corresponding to the merged result according to the attention mechanism; Performing a full connection transformation on the interaction result to obtain a first number of output vectors, each of which has a second number of dimensions, wherein the first number of output vectors corresponds one-to-one to the first number of characters, the second number is the total number of labels, and the second number of dimensions corresponds one-to-one to the second number of labels; For each output vector of the first number of output vectors, at least one dimension whose score value exceeds a preset score value is obtained from the second number of dimensions, wherein a label corresponding to at least one of the dimensions is a target label of the character corresponding to the dimension.

2. The method according to claim 1, characterized in that The performing label extraction processing on the current text data to obtain a target label corresponding to the target user includes: The trained label extraction model is used to perform label extraction processing on the current text data to obtain a target label corresponding to the target user.

3. The method according to claim 2, characterized in that The label extraction model includes a first feature extraction model, a second feature extraction model, an attention layer and a fully connected layer; The performing a first encoding process on a first number of characters included in the current text data to obtain a first encoding result includes: Using the first feature extraction model to perform a first encoding process on the first number of characters to obtain a first encoding result; The performing a second encoding process on the first number of letter representation characters to obtain a second encoding result includes: Using the second feature extraction model, performing a second encoding process on the first number of letter representations to obtain a second encoding result; The obtaining the interaction result corresponding to the merged result according to the attention mechanism includes: Using the attention layer to obtain an interaction result corresponding to the merged result; The performing a full connection transformation on the interaction result to obtain a first number of output vectors includes: The fully connected layer is used to perform a fully connected transformation on the interaction result to obtain a first number of output vectors.

4. The method according to claim 1, characterized in that The step of sending the first data related to the negative status recovery process to the target user terminal according to a preset sending rule to assist the target user in recovering the negative status includes: At a first time point, sending multiple inquiry messages to the target user terminal; receiving a plurality of feedback information returned by the target user terminal; Generate a targeted repair plan based on the plurality of feedback information; The targeted repair solution is sent to the target user terminal.

5. The method according to claim 4, characterized in that The method further comprises: At a second time point in the process execution cycle, target popular science knowledge is sent to the target user terminal, wherein the process execution cycle is the execution cycle of the negative state recovery process, and the target popular science knowledge is popular science knowledge associated with the negative state.

6. The method according to claim 4, characterized in that The method further comprises: At a third time point in the process execution cycle, sending indicator query information to the target user terminal, wherein the indicator query information is used to query multiple indicator parameters of the target user; Receiving the multiple indicator parameters returned by the target user terminal; If there are indicator parameters among the multiple indicator parameters whose parameter values ​​exceed their corresponding normal range values, a warning signal is generated and displayed.

7. The method according to claim 6, characterized in that The method further comprises: After the negative state recovery process has run through a preset number of process execution cycles, obtaining a plurality of indicator parameters corresponding to the third time node of the preset number of process execution cycles; Generate a new targeted repair plan according to the multiple indicator parameters corresponding to the third time node of the preset number of process execution cycles; The new targeted repair solution is sent to the target user terminal.

8. A medical data processing device, characterized in that: The device comprises: A text data acquisition unit, used to acquire current text data of a target user; A label extraction unit, used to perform label extraction processing on the current text data to obtain a target label corresponding to the target user; A process acquisition unit, configured to acquire a negative state recovery process corresponding to the target user according to a correspondence between a tag and a state recovery process and the target tag, wherein the negative state recovery process is a process for assisting the target user to recover from a negative state; A data sending unit, configured to send first data related to the negative status recovery process to a target user terminal according to a preset sending rule, so as to assist the target user in recovering the negative status; The label extraction unit is specifically used for: Performing a first encoding process on a first number of characters included in the current text data to obtain a first encoding result; Obtaining a letter representation corresponding to each character included in the current text data, wherein the letter representation is a letter representation of the character; Performing a second encoding process on the first number of letter representation characters to obtain a second encoding result; Merging the first encoding result and the second encoding result to obtain a merged result; Obtaining the interaction result corresponding to the merged result according to the attention mechanism; Performing a full connection transformation on the interaction result to obtain a first number of output vectors, each of which has a second number of dimensions, wherein the first number of output vectors corresponds one-to-one to the first number of characters, the second number is the total number of labels, and the second number of dimensions corresponds one-to-one to the second number of labels; For each output vector of the first number of output vectors, at least one dimension whose score value exceeds a preset score value is obtained from the second number of dimensions, wherein a label corresponding to at least one of the dimensions is a target label of the character corresponding to the dimension.

9. An electronic device, characterized in that: It comprises a processor and a memory, wherein the memory stores a plurality of instructions; the processor loads instructions from the memory to execute the steps in the medical data processing method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor to execute the steps in the medical data processing method according to any one of claims 1 to 7.

Images