Method for constructing syndrome decision-making model and syndrome decision-making method
By constructing a syndrome decision model, using the disease symptoms data and syndromes in the training data set, the problem of relying on physician experience in traditional Chinese medicine syndrome is solved, and the effect of improving clinical diagnosis is achieved.
Patent Information
- Application Number
- CN202411647723.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-18
- Publication Date
- 2025-03-11
AI Technical Summary
In the prior art, the determination of TCM syndrome depends on the experience of the physician, which makes it difficult to guarantee the consistency of clinical diagnosis, especially when primary doctors are not experienced.
By constructing a syndrome decision model, using the disease symptoms data and syndromes in the training data set, the symptoms of the first node and the symptoms of the second node are determined, the status data are divided until the syndrome is consistent, and the corresponding syndrome decision model is constructed.
It realizes automatic determination of syndromes based on the patient's symptom data, improves the consistency of clinical diagnosis, and helps doctors assist in diagnosis.
Smart Images
Figure CN119601214A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of auxiliary diagnosis technology, and more specifically, to the construction of a syndrome decision model and a syndrome decision method. Background Art
[0002] The "Syndrome" of TCM refers to the summary understanding of the pathophysiological differences at different stages of the disease, taking into account factors such as the nature of the disease and the location of the disease. Specific Chinese medicine prescriptions are prescribed according to the "Syndrome" exhibited by each patient. For example, according to TCM theory, depression can be classified into multiple subtypes ("Syndromes") based on the patient's symptoms / signs, including liver qi stagnation type, liver qi stagnation and spleen deficiency type, qi deficiency and blood stasis type, etc. Different treatment strategies are then provided based on the patient's specific "Syndrome", for example, the liver qi stagnation subtype uses a prescription to regulate qi, and the qi deficiency and blood stasis subtype uses a prescription to invigorate qi and activate blood.
[0003] Many modern studies have proven that TCM dialectical treatment is superior in terms of efficacy. However, the determination of "syndrome" depends largely on the experience of TCM physicians in extracting and analyzing the implicit information in symptoms / signs. Some doctors, especially inexperienced primary care physicians, lack long-term clinical guidance and experience, and have a significant gap in medical level and clinical experience with famous senior physicians. Therefore, how to improve the consistency of clinical diagnosis has become an urgent problem to be solved. Summary of the invention
[0004] In view of this, the purpose of the embodiments of the present application is to provide a syndrome decision model construction and a syndrome decision method to solve the technical problem of how to improve the consistency of clinical diagnosis.
[0005] In a first aspect, an embodiment of the present application provides a method for constructing a syndrome decision model, the method comprising:
[0006] Acquire a training data set; wherein the training data set includes disease symptom data and syndromes corresponding to at least one disease type; the disease symptom data includes at least two of the main symptoms, characteristic symptoms, main complications, secondary complications, tongue image and pulse image;
[0007] Determining a first node symptom in the disease symptom data;
[0008] Based on the multiple data states of the first node symptom, dividing the disease symptom data into multiple groups of state data;
[0009] Determine the second node symptoms in each of the state data respectively, and divide the state data based on the second node symptoms until the syndromes of the divided sub-state data are consistent;
[0010] A syndrome decision model for the disease type is constructed based on the first node symptom, the second node symptom in each of the state data, and the syndromes of the multiple sub-state data after division.
[0011] In the above implementation process, the syndrome decision model construction method obtains a training data set including disease symptom data and syndromes corresponding to at least one disease type, and determines the first node symptom in the disease symptom data; divides the disease symptom data into multiple groups of state data based on multiple data states of the first node symptom; respectively determines the second node symptom in each state data, and divides the state data based on the second node symptom until the syndromes of the multiple sub-state data after division are consistent; and constructs a syndrome decision model of the corresponding disease type according to the first node symptom, the second node symptom in each state data, and the syndromes of the multiple sub-state data after division. Since the constructed syndrome decision model includes the first node symptom, the second node symptom in each state data, and the syndromes of the multiple sub-state data after division, based on the constructed syndrome decision model, the syndrome corresponding to the symptom data of the target patient can be determined according to the data state corresponding to the symptom data of the target patient. Therefore, based on the above syndrome decision model construction method, a syndrome decision model that "can help doctors determine the syndrome corresponding to the patient's symptom data and realize auxiliary diagnosis" can be constructed, thereby improving the consistency of clinical diagnosis.
[0012] Optionally, in an embodiment of the present application, determining the first node symptom in the disease symptom data includes: calculating the information gain corresponding to each symptom data in the disease symptom data; and determining the symptom data with the largest information gain as the first node symptom.
[0013] In the above implementation process, symptom data with large information gain can better distinguish different categories. By determining the symptom data with the largest information gain as the first-node symptom, it can help build a more accurate and efficient syndrome decision model, thereby improving the accuracy of the syndrome decision results determined based on the constructed syndrome decision model.
[0014] Optionally, in an embodiment of the present application, the calculation of the information gain corresponding to each type of symptom data in the disease symptom data includes: calculating the information entropy corresponding to each type of symptom data in the disease symptom data; determining the information gain of the symptom data based on the information entropy of the symptom data and the information entropy of multiple groups of symptom classification data; wherein the disease symptom data is divided into multiple groups of symptom classification data by the symptom data.
[0015] In the above implementation process, information entropy can be used to measure the uncertainty of symptom data or symptom classification data. The larger the information entropy, the worse the classification effect. Therefore, the information gain that can characterize the classification effect can be determined based on the information entropy of the symptom data and the information entropy of multiple groups of symptom classification data. According to the information gain of the symptom data, the first node symptom with a better classification effect can be determined, thereby improving the decision accuracy of the constructed syndrome decision model.
[0016] Optionally, in an embodiment of the present application, in the process of constructing the syndrome decision model, the first node symptom is used as the root node of the syndrome decision model, the second node symptom is used as the internal node of the syndrome decision model, and the syndrome of the multiple sub-state data after division is used as the leaf node of the syndrome decision model.
[0017] In the above implementation process, since the syndromes of the multiple sub-state data after division are consistent, the accuracy of the syndrome decision results determined based on the constructed syndrome decision model can be improved by using the syndromes of the multiple sub-state data after division as leaf nodes of the syndrome decision model.
[0018] In a second aspect, an embodiment of the present application provides a syndrome decision-making method, the syndrome decision-making method comprising:
[0019] Determine the disease type based on the target patient's presenting symptoms;
[0020] The symptom data of the target patient is input into the syndrome decision model corresponding to the disease type to obtain the syndrome decision result output by the syndrome decision model; wherein the symptom data includes at least two of the chief symptom, characteristic symptoms, major concomitant symptoms, minor concomitant symptoms, tongue image and pulse image; and the syndrome decision model is constructed based on the syndrome decision model construction method as described in any one of the first aspects.
[0021] In the above-mentioned implementation process, since the syndrome decision model construction method based on the first aspect can construct a syndrome decision model that "can help doctors determine the syndromes corresponding to the patient's symptom data and achieve auxiliary diagnosis", by inputting the target patient's symptom data into the constructed syndrome decision model, the syndrome corresponding to the target patient's symptom data can be determined based on the syndrome decision model, thereby improving the consistency of clinical diagnosis.
[0022] Optionally, in an embodiment of the present application, after obtaining the syndrome decision result output by the syndrome decision model, the method further includes: obtaining a recommended prescription for the target patient based on the syndrome decision result.
[0023] In the above implementation process, the syndrome corresponding to the symptom data of the target patient is determined based on the syndrome decision model, and then the corresponding recommended prescription is determined according to the syndrome decision result, thereby improving the consistency of clinical diagnosis.
[0024] Optionally, in an embodiment of the present application, obtaining the recommended prescription for the target patient based on the syndrome decision result includes: determining the preliminary prescription for the target patient based on the syndrome decision result; adjusting the preliminary prescription based on the personalized data of the target patient to obtain the recommended prescription; wherein the personalized data includes current medical history, past medical history, family history, allergies, historical medication and / or financial needs.
[0025] In the above implementation process, the initial prescription determined based on the syndrome decision results is adjusted according to personalized data such as the target patient's current medical history, past medical history, family history, allergies, historical medication and / or economic needs, so as to obtain a recommended prescription that is more suitable for the target patient.
[0026] In a third aspect, an embodiment of the present application further provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the method described in any one of the first aspect or the second aspect.
[0027] In a fourth aspect, an embodiment of the present application further provides an electronic device; the electronic device includes:
[0028] Memory;
[0029] processor;
[0030] The memory stores a computer program executable by the processor, and when the computer program is executed by the processor, the method described in any one of the first aspect or the second aspect is performed.
[0031] In a fifth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the method described in any one of the first aspect or the second aspect is executed.
[0032] In a sixth aspect, an embodiment of the present application provides a syndrome decision model construction device, the device comprising:
[0033] A data acquisition module, used to acquire a training data set; wherein the training data set includes disease symptom data and syndromes corresponding to at least one disease type; the disease symptom data includes at least two of the main symptoms, specific symptoms, major complications, minor complications, tongue image and pulse image;
[0034] A first node determination module, used to determine a first node symptom in the disease symptom data;
[0035] A data division module, configured to divide the disease symptom data into a plurality of groups of state data based on the plurality of data states of the first node symptom;
[0036] A second node determination module, used to respectively determine the second node symptoms in each of the state data, and divide the state data based on the second node symptoms until the syndromes of the divided sub-state data are consistent;
[0037] The model building module is used to build a syndrome decision model for the disease type based on the first node symptoms, the second node symptoms in each state data, and the syndromes of the multiple sub-state data after division.
[0038] In a seventh aspect, an embodiment of the present application provides a syndrome decision-making device, the device comprising:
[0039] A disease type determination module, used to determine the disease type based on the main complaint symptoms of the target patient;
[0040] A decision result acquisition module is used to input the symptom data of the target patient into the syndrome decision model corresponding to the disease type, and obtain the syndrome decision result output by the syndrome decision model; wherein the symptom data includes at least two of the main symptoms, characteristic symptoms, major complications, minor complications, tongue conditions and pulse conditions; and the syndrome decision model is constructed based on the syndrome decision model construction method as described in any one of the first aspects.
[0041] The beneficial effects of the present application include at least: the above syndrome decision model construction method obtains a training data set including disease symptom data and syndromes corresponding to at least one disease type, and determines the first node symptom in the disease symptom data; based on multiple data states of the first node symptom, the disease symptom data is divided into multiple groups of state data; the second node symptom in each state data is determined respectively, and based on the second node symptom, the state data is divided until the syndromes of the multiple sub-state data after the division are consistent; according to the first node symptom, the second node symptom in each state data, and the syndromes of the multiple sub-state data after the division, a syndrome decision model of the corresponding disease type is constructed. Since the constructed syndrome decision model includes the first node symptom, the second node symptom in each state data, and the syndromes of the multiple sub-state data after the division, based on the constructed syndrome decision model, the syndrome corresponding to the symptom data of the target patient can be determined according to the data state corresponding to the symptom data of the target patient. Therefore, based on the above syndrome decision model construction method, a syndrome decision model that "can help doctors determine the syndrome corresponding to the patient's symptom data and realize auxiliary diagnosis" can be constructed, thereby improving the consistency of clinical diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.
[0043] Figure 1 A schematic diagram of a process for constructing a syndrome decision model provided in an embodiment of the present application;
[0044] Figure 2 A schematic diagram of a process flow of a syndrome decision-making method provided in an embodiment of the present application;
[0045] Figure 3 A flowchart of another syndrome decision-making method provided in an embodiment of the present application;
[0046] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0047] The following embodiments of the technical solution of the present application are described in detail in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present application, and are therefore only used as examples, and cannot be used to limit the scope of protection of the present application.
[0048] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application.
[0049] In the description of the embodiments of the present application, the technical terms "first", "second", etc. are only used to distinguish different objects, and cannot be understood as indicating or implying relative importance or implicitly indicating the number, specific order or primary and secondary relationship of the indicated technical features. In the description of the embodiments of the present application, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.
[0050] See also Figure 1 The flowchart of a syndrome decision model construction method provided by an embodiment of the present application is shown. The syndrome decision model construction method may include the following steps:
[0051] S101, obtaining a training data set; wherein the training data set includes disease symptom data and syndromes corresponding to at least one disease type; the disease symptom data includes at least two of the main symptoms, specific symptoms, major complications, minor complications, tongue image and pulse image;
[0052] S102, determining a first node symptom in the disease symptom data;
[0053] S103, dividing the disease symptom data into multiple groups of state data based on the multiple data states of the first node symptoms;
[0054] S104, respectively determining the second node symptoms in each of the state data, and dividing the state data based on the second node symptoms until the syndromes of the divided sub-state data are consistent;
[0055] S105. Construct a syndrome decision model for the disease type according to the first node symptom, the second node symptom in each state data, and the syndromes of the divided multiple sub-state data.
[0056] Among them, in step S101, the training data set may include historical syndrome diagnosis data of medical experts, and may also include syndrome diagnosis data in syndrome diagnosis-related books such as "Treatise on Febrile and Miscellaneous Diseases", "Chinese Medicine Syndrome", "Chinese Medicine Internal Medicine", "Chinese Medicine Surgery", "Chinese Medicine Pediatrics", "Internal Medicine of Integrated Traditional and Western Medicine", "Surgery of Integrated Traditional and Western Medicine", "Pediatrics of Integrated Traditional and Western Medicine", etc., and may also include syndrome diagnosis data in relevant guidelines and standards such as "Guidelines for Diagnosis and Treatment of Common Diseases in Chinese Medicine Internal Medicine" and "Standards for Diagnosis and Efficacy of Diseases and Syndromes", and may also include historical syndrome diagnosis data of medical experts, syndrome diagnosis data in relevant books, guidelines and standards. The main complaint symptoms refer to the main symptoms or signs of the patient and the duration of the duration, and the type of disease of the patient can be judged based on the main complaint symptoms. Specific symptoms refer to symptoms specific to different syndromes. The main concomitant symptoms refer to the necessary dialectical judgment symptoms among several concomitant symptoms other than the main complaint symptoms. Secondary concomitant symptoms refer to several concomitant symptoms other than the main complaint symptoms and the main concomitant symptoms. Tongue image refers to the overall changes and morphology of the tongue, which can be used to assist in dialectics. Pulse refers to the speed, strength, depth, etc. of the pulse, which can be used to assist in dialectics. The training data set may include disease symptom data and syndromes corresponding to one disease type, and may also include disease symptom data and syndromes corresponding to multiple disease types. Taking the disease type of "insomnia / insomnia" as an example, the disease symptom data may include "insomnia and difficulty falling asleep (main complaint symptom)", "fatigue of limbs, fatigue (specific symptoms)", "forgetfulness, poor appetite, abdominal distension (main concurrent symptom)", "dizziness, vertigo, palpitations (secondary concurrent symptom)", "pale tongue, thin fur (tongue image)" and "thin pulse, weak pulse (pulse image)" at least two of them. The training data set may specifically include multiple disease symptom data corresponding to the disease type of "insomnia / insomnia", as well as the syndromes of each disease symptom data.
[0057] Wherein, in step S102, the first node symptom may be any one of the main complaint symptom, the characteristic symptom, the main concurrent symptom, the secondary concurrent symptom, the tongue image or the pulse image, and the first node symptom corresponding to different disease types may be the same or different. According to the division effect of the main complaint symptom, the characteristic symptom, the main concurrent symptom, the secondary concurrent symptom, the tongue image or the pulse image, the disease symptom with the best division effect may be determined as the first node symptom. The disease symptom with the best division effect may be determined based on information gain, information gain ratio, Gini index or covariance. Before determining the first node symptom in the disease symptom data, the disease symptom data may be standardized. Taking the main complaint symptoms including the three states of "insomnia and difficulty falling asleep", "insomnia and dreaminess" and "insomnia and upset" as an example, the first data may be used to represent insomnia, the second data may be used to represent difficulty falling asleep, the third data may be used to represent dreaminess, and the fourth data may be used to represent upset, then "1100" means "insomnia and difficulty falling asleep", and "1001" means "insomnia and upset". In this way, the disease symptom data can be converted into multi-bit binary data, thereby reducing the difficulty of determining the first node symptom.
[0058] Among them, in step S103, taking the disease type of "insomnia / drowsiness" as an example, the main symptoms may include "insomnia and difficulty falling asleep", "insomnia and dreaminess" or "insomnia and irritability" and other states. Special symptoms may include "dizziness, head swelling, red eyes", "yellow and thick sputum, phlegm in the throat", "fatigue of limbs, fatigue" or "shortness of breath, spontaneous sweating, fatigue and weakness" and other states. The main concomitant symptoms may include "easy to get irritable, dry and bitter mouth", "chest tightness, dizziness, heavy head", "forgetfulness, less food, abdominal distension", "weak waist and knees, irregular menstruation" or "timid palpitations" and other states. Secondary concomitant symptoms may include "dizziness, vertigo, palpitations", "dreams, tinnitus" or "no appetite, tinnitus" and other states. Tongue image may include red tongue, pale tongue, thin fur, thick fur, yellow fur, white fur and other states. Pulse image may include deep pulse, stringy pulse, slippery pulse, slow pulse and other states. Taking the first node symptom as the chief complaint symptom as an example, the disease symptom data can be divided into multiple groups of state data such as "insomnia and difficulty falling asleep", "insomnia and frequent dreams" and "insomnia and irritability" according to the chief complaint symptom in the disease symptom data.
[0059] Wherein, in step S104, S105, the second node symptom can be any one of the main complaint symptom, characteristic symptom, main concurrent symptom, secondary concurrent symptom, tongue image or pulse image except the first node symptom, and the second node symptom corresponding to different disease types can be the same or different. According to the division effect of the main complaint symptom, characteristic symptom, main concurrent symptom, secondary concurrent symptom, tongue image or pulse image except the first node symptom, the disease symptom with the best division effect can be determined as the second node symptom. The syndromes of the divided multiple sub-state data are consistent, indicating that after the division based on the current node symptom, the syndrome uniquely corresponding to the disease symptom data can be determined. Accordingly, based on the first node symptom, the second node symptom in each state data and the syndrome of the divided multiple sub-state data, the syndrome decision model of the disease type constructed can determine the unique syndrome corresponding to the different disease symptom data to improve the consistency of clinical diagnosis. It should be noted that when the disease symptom data includes chief symptoms, characteristic symptoms, major complications, minor complications, tongue patterns and pulse patterns, the number of nodes in the constructed syndrome decision model can be 2, 3 or 6, and the number of nodes included in the syndrome decision models of different disease types may be different.
[0060] It can be seen that the syndrome decision model construction method provided in the embodiment of the present application obtains a training data set including disease symptom data and syndromes corresponding to at least one disease type, and determines the first node symptom in the disease symptom data; based on the multiple data states of the first node symptom, the disease symptom data is divided into multiple groups of state data; the second node symptom in each state data is determined respectively, and based on the second node symptom, the state data is divided until the syndromes of the multiple sub-state data after the division are consistent; according to the first node symptom, the second node symptom in each state data, and the syndromes of the multiple sub-state data after the division, the syndrome decision model of the corresponding disease type is constructed. Since the constructed syndrome decision model includes the first node symptom, the second node symptom in each state data, and the syndromes of the multiple sub-state data after the division, based on the constructed syndrome decision model, the syndrome corresponding to the symptom data of the target patient can be determined according to the data state corresponding to the symptom data of the target patient. Therefore, based on the above syndrome decision model construction method, a syndrome decision model that "can help doctors determine the syndrome corresponding to the patient's symptom data and realize auxiliary diagnosis" can be constructed, thereby improving the consistency of clinical diagnosis.
[0061] In some optional embodiments, S102, determining the first node symptom in the disease symptom data, includes: calculating the information gain corresponding to each symptom data in the disease symptom data; and determining the symptom data with the largest information gain as the first node symptom.
[0062] Among them, information gain refers to the reduction in information entropy before and after division under the condition of given features (symptom data). The information gain corresponding to the corresponding symptom data can be determined based on the reduction in information entropy before and after the disease symptom data is divided based on the corresponding symptom data. Symptom data with large information gain can better distinguish different categories. By determining the symptom data with the largest information gain as the first node symptom, it can help build a more accurate and efficient syndrome decision model and improve the accuracy of the syndrome decision results determined based on the constructed syndrome decision model.
[0063] In some optional embodiments, S104, respectively determining the second node symptom in each of the state data, and dividing the state data based on the second node symptom until the syndromes of the multiple sub-state data after division are consistent, may include: calculating the information gain corresponding to each symptom data in each state data, and determining the symptom data with the largest information gain (except the first node symptom) as the second node symptom.
[0064] Among them, the second node symptoms can specifically include the second root node symptoms, the second internal node symptoms and the second leaf node symptoms. The information gain corresponding to each symptom data in each state data can be calculated first, and the symptom data with the largest information gain (except the first node symptoms) can be determined as the second root node symptoms. Based on the various data states of the second root node symptoms, the state data of the corresponding group is divided into multiple groups of second root node state data; the information gain corresponding to each symptom data in the second root node state data is calculated, and the symptom data with the largest information gain (except the first node symptoms and the second root node symptoms) is determined as the second internal node symptoms. And the symptom data that can divide the state data of the corresponding group into multiple sub-state data with consistent syndromes is determined as the second leaf node symptom.
[0065] In some optional embodiments, the calculation of the information gain corresponding to each type of symptom data in the disease symptom data includes: calculating the information entropy corresponding to each type of symptom data in the disease symptom data; determining the information gain of the symptom data based on the information entropy of the symptom data and the information entropy of multiple groups of symptom classification data; wherein the disease symptom data is divided into multiple groups of symptom classification data by the symptom data.
[0066] Among them, it can be based on Calculate the information entropy E corresponding to each symptom data; p(x i ) represents the proportion of the corresponding symptom data in the disease symptom data. Determine the information gain G(D,c) of the corresponding symptom data; D represents the disease symptom data, c represents the corresponding symptom data (which can be the main symptom, special symptom, main comorbidity, secondary comorbidity, tongue image or pulse image), and v represents the number of states of the corresponding symptom data (that is, the corresponding symptom data c has v different values / symptom states). Based on the different states of the corresponding symptom data c, the disease symptom data D can be divided into multiple groups of symptom division data D 1 , D 2 ,…,D v Information entropy can be used to measure the uncertainty of symptom data or symptom classification data. The larger the information entropy, the worse the classification effect. Therefore, we can use the information entropy E(D) of disease symptom data and the information entropy E(D) of multiple groups of symptom classification data to calculate the uncertainty of symptom data. v ), determine the information gain that can characterize the division effect of the corresponding symptom data c. According to the information gain of each symptom data, the first node symptom with better division effect can be determined, thereby improving the decision accuracy of the constructed syndrome decision model.
[0067] In some optional embodiments, in the process of constructing the syndrome decision model, the first node symptom is used as the root node of the syndrome decision model, the second node symptom is used as the internal node of the syndrome decision model, and the syndrome of the multiple sub-state data after division is used as the leaf node of the syndrome decision model.
[0068] The number of internal nodes may be 1, 2 or other reasonable values. Since the syndromes of the divided sub-state data are consistent, the accuracy of the syndrome decision results determined based on the constructed syndrome decision model can be improved by using the syndromes of the divided sub-state data as leaf nodes of the syndrome decision model.
[0069] Please refer to Figure 2 , Figure 2 A flow chart of a syndrome decision-making method provided in an embodiment of the present application. The syndrome decision-making method includes:
[0070] S201, determining the disease type based on the main symptoms of the target patient;
[0071] S202. Input the symptom data of the target patient into the syndrome decision model corresponding to the disease type, and obtain the syndrome decision result output by the syndrome decision model; wherein the symptom data includes at least two of the chief symptom, characteristic symptoms, major concomitant symptoms, minor concomitant symptoms, tongue image and pulse image; and the syndrome decision model is constructed based on the syndrome decision model construction method as described in any one of the first aspects.
[0072] The syndrome decision result may include the syndrome corresponding to the symptom data of the target patient. By determining the disease type based on the main symptoms of the target patient and inputting the symptom data into the corresponding syndrome decision model according to the determined disease type, the accuracy of the syndrome decision result obtained based on the syndrome decision method can be ensured.
[0073] It should be noted that, since the syndrome decision model construction method described in the first aspect can construct a syndrome decision model that "can help doctors determine the syndrome corresponding to the patient's symptom data and achieve auxiliary diagnosis", by inputting the symptom data of the target patient into the constructed syndrome decision model, the syndrome corresponding to the symptom data of the target patient can be determined based on the syndrome decision model, thereby improving the consistency of clinical diagnosis. For the specific implementation of the syndrome decision model in the syndrome decision method, please refer to the description of the syndrome decision model construction method above. To avoid repetition, the detailed description is appropriately omitted here.
[0074] In some optional embodiments, after obtaining the syndrome decision result output by the syndrome decision model, the method further includes: S203, obtaining a recommended prescription for the target patient based on the syndrome decision result.
[0075] Among them, the recommended prescription for the target patient can be determined based on the correspondence between the syndrome and the Chinese medicine prescription, combined with the syndrome decision result. The recommended prescription can include multiple Chinese medicine prescriptions, and the doctor can select the appropriate Chinese medicine prescription as the diagnostic prescription for the target patient according to the actual needs of the patient. The recommended prescription suitable for the target patient can also be determined based on the recommended prescription corresponding to the syndrome decision result.
[0076] Please refer to Figure 3 , Figure 3 A flowchart of another syndrome decision method provided for an embodiment of the present application. In some optional embodiments, S203, i.e., obtaining a recommended prescription for the target patient based on the syndrome decision result, includes: S2031, determining a preliminary prescription for the target patient based on the syndrome decision result; S2032, adjusting the preliminary prescription based on the personalized data of the target patient to obtain the recommended prescription; wherein the personalized data includes current medical history, past medical history, family history, allergies, historical medications and / or financial needs.
[0077] Among them, the personalized data of the target patient may include personalized data related to the patient himself, as well as personalized data related to the patient's condition. Personalized data may also include incompatibility taboos, toxicity of medicinal materials, and population taboos. The preliminary prescription can be adjusted by combining, adding or adjusting the prescription in combination with the "principle of monarch, minister, assistant and envoy", "treatment based on syndrome differentiation", "simplification of prescriptions and medicines" or "prescription and medicine compatibility". According to the personalized data of the target patient's current medical history, past medical history, family history, allergies, historical medication and / or economic needs, the preliminary prescription determined based on the syndrome decision-making results can be adjusted to obtain a recommended prescription that is more suitable for the target patient.
[0078] An embodiment of the present application also provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the syndrome decision model construction method as described in any one of the first aspects or the syndrome decision method as described in any one of the second aspects.
[0079] Please refer to Figure 4 , Figure 4 The present invention provides a schematic diagram of the structure of an electronic device 300 provided in an embodiment of the present application. The electronic device 300 includes: a memory 302 and a processor 301; the memory 302 stores a computer program executable by the processor 301, and when the computer program is executed by the processor 301, the syndrome decision model construction method described in any one of the first aspects or the syndrome decision method described in any one of the second aspects is executed.
[0080] The memory 302 and the processor 301 can be interconnected and communicate with each other through a communication bus 303 and / or other forms of connection mechanisms (not shown). The memory 302 stores a computer program executable by the processor 301. When the computer program is executed by the processor 301, the syndrome decision model construction method described in any one of the first aspects or the syndrome decision method described in any one of the second aspects is executed.
[0081] An embodiment of the present application also provides a computer-readable storage medium, on which computer program instructions are stored. When the computer program instructions are executed by the processor 301, the syndrome decision model construction method described in any one of the first aspects or the syndrome decision method described in any one of the second aspects is executed.
[0082] Among them, the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable red-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.
[0083] In several embodiments provided by the embodiments of the present application, it should be understood that the disclosed device / system and method can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to the multiple embodiments of the embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and a part of the module, program segment or code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.
[0084] In addition, the functional modules in each embodiment of the present application can be integrated together to form an independent part, or each module can exist separately, or two or more modules can be integrated to form an independent part.
[0085] The above description is only an optional implementation manner of the embodiments of the present application, but the protection scope of the embodiments of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or replacements within the technical scope disclosed in the embodiments of the present application, which should be covered within the protection scope of the embodiments of the present application.
Claims
1. A method for constructing a syndrome decision model, characterized in that: The method comprises: Acquire a training data set; wherein the training data set includes disease symptom data and syndromes corresponding to at least one disease type; the disease symptom data includes at least two of the main symptoms, characteristic symptoms, main complications, secondary complications, tongue image and pulse image; Determining a first node symptom in the disease symptom data; Based on the multiple data states of the first node symptom, dividing the disease symptom data into multiple groups of state data; Determine the second node symptoms in each of the state data respectively, and divide the state data based on the second node symptoms until the syndromes of the divided sub-state data are consistent; A syndrome decision model for the disease type is constructed based on the first node symptom, the second node symptom in each of the state data, and the syndromes of the multiple sub-state data after division.
2. The method according to claim 1, characterized in that The determining of the first node symptom in the disease symptom data comprises: Calculating the information gain corresponding to each symptom data in the disease symptom data; The symptom data with the largest information gain is determined as the first node symptom.
3. The method according to claim 2, characterized in that The calculating the information gain corresponding to each symptom data in the disease symptom data includes: Calculating the information entropy corresponding to each symptom data in the disease symptom data; The information gain of the symptom data is determined based on the information entropy of the symptom data and the information entropy of multiple groups of symptom classification data; wherein the disease symptom data is divided into multiple groups of symptom classification data by the symptom data.
4. The method according to any one of claims 1 to 3, characterized in that: in, In the process of constructing the syndrome decision model, the first node symptom is used as the root node of the syndrome decision model, the second node symptom is used as the internal node of the syndrome decision model, and the syndrome of the multiple sub-state data after division is used as the leaf node of the syndrome decision model.
5. A syndrome decision-making method, characterized in that: The method comprises: Determine the disease type based on the target patient's presenting symptoms; The symptom data of the target patient is input into the syndrome decision model corresponding to the disease type to obtain the syndrome decision result output by the syndrome decision model; wherein the symptom data includes at least two of the chief symptom, characteristic symptoms, major concomitant symptoms, minor concomitant symptoms, tongue image and pulse image; the syndrome decision model is constructed based on the syndrome decision model construction method as described in any one of claims 1 to 4.
6. The method according to claim 5, characterized in that After obtaining the syndrome decision result output by the syndrome decision model, the method further includes: According to the syndrome decision result, a recommended prescription for the target patient is obtained.
7. The method according to claim 6, characterized in that The step of obtaining a recommended prescription for the target patient according to the syndrome decision result includes: Determine the preliminary prescription for the target patient according to the syndrome decision result; The preliminary prescription is adjusted according to the personalized data of the target patient to obtain the recommended prescription; wherein the personalized data includes current medical history, past medical history, family history, allergies, historical medications and / or financial needs.
8. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the method according to any one of claims 1 to 7 is implemented.
9. An electronic device, characterized in that: The electronic device comprises: Memory; processor; The memory stores a computer program executable by the processor, and when the computer program is executed by the processor, the method according to any one of claims 1 to 7 is performed.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the method according to any one of claims 1 to 7 is executed.