User diagnosis and treatment information processing method and device, equipment and storage medium

Abstract

The embodiment of the invention provides a user diagnosis and treatment information processing method and device, equipment and a storage medium, and the method comprises the steps: obtaining a predicted disease and a predicted syndrome type through a target prediction strategy based on user diagnosis and treatment information; wherein the user diagnosis and treatment information comprises one or more of tongue diagnosis information, face diagnosis information, pulse diagnosis information and palm diagnosis information; the target prediction strategy comprises a first prediction strategy, and the first prediction strategy comprises: determining a plurality of target syndrome elements corresponding to the user diagnosis and treatment information according to a pre-established corresponding relationship between the diagnosis and treatment information and the syndrome elements; according to a pre-established corresponding relationship between diseases and syndrome elements, determining predicted diseases matched with at least part of the target syndrome elements; and determining a predicted syndrome type comprising at least part of the target syndrome elements from all syndrome types included in the predicted disease, wherein the predicted disease and the predicted syndrome type are used for providing auxiliary reference information for a physician. Therefore, the disease syndrome type prediction is realized through a lingual surface pulse palm-syndrome element-disease-syndrome type prediction link.

Description

Technical Field

[0001] The present application relates to the technical field of TCM-assisted diagnosis and treatment, and more specifically, to a method, device, equipment and storage medium for processing user diagnosis and treatment information. Background Art

[0002] In the field of TCM dialectics, doctors often need to combine observation, auscultation, questioning and palpation to make a diagnosis. With the rapid development of artificial intelligence technology, artificial intelligence technology is gradually being used to assist doctors in dialectics. Specifically, based on the patient's symptom information, combined with relevant TCM diagnostic knowledge graphs, neural network models and other technologies to predict the disease, and then output the predicted disease to the doctor to assist the doctor in dialectics.

[0003] In the related art, the use of computers to predict diseases relies heavily on symptoms. However, if it is difficult to accurately know the symptoms, the accuracy of disease prediction will be difficult to guarantee. Therefore, how to reduce the computer's reliance on symptoms when predicting diseases and improve the accuracy of disease prediction is a technical problem that needs to be solved in this field. Summary of the invention

[0004] The purpose of the embodiments of the present application is to provide a method, device, equipment and storage medium for processing user diagnosis and treatment information, so as to achieve the technical effect of reducing the dependence of disease prediction on disease symptoms while improving the accuracy of disease prediction.

[0005] In a first aspect, an embodiment of the present application provides a method for processing user diagnosis and treatment information, the method comprising: obtaining predicted diseases and predicted syndromes based on user diagnosis and treatment information through a target prediction strategy; wherein the user diagnosis and treatment information comprises one or more of tongue diagnosis information, face diagnosis information, pulse diagnosis information and palm diagnosis information; the target prediction strategy comprises a first prediction strategy, and the first prediction strategy comprises:

[0006] Determine a plurality of target syndromes corresponding to the user's diagnosis and treatment information according to a pre-established correspondence relationship between diagnosis and treatment information and syndromes;

[0007] Determining predicted diseases that match at least part of the target syndrome according to a pre-established correspondence between the disease and the syndrome;

[0008] From all syndrome types included in the predicted disease, a predicted syndrome type including at least part of the target syndrome is determined; wherein the predicted disease and the predicted syndrome type are used to provide auxiliary reference information to doctors.

[0009] In the above implementation process, the prediction of disease and syndrome is achieved through the prediction link of "tongue, face, pulse and palm-syndrome factor-disease-syndrome type". The whole prediction process does not rely on the user's obvious physical symptoms, and can achieve the technical effect of accurately predicting the disease syndrome before the user has obvious physical symptoms.

[0010] Further, determining a predicted disease that matches at least some of the target syndrome elements according to the pre-established correspondence between diseases and syndrome elements includes:

[0011] Determining a first candidate disease that includes one or more of the target syndrome elements according to the correspondence between the diseases and the syndrome elements;

[0012] According to the first quantity of the target syndrome elements included in each of the first candidate diseases, sequentially determining a preset number of second candidate diseases from the multiple first candidate diseases in descending order of the first quantity;

[0013] For each of the second candidate diseases, determining a syndrome element matching score of the second candidate disease according to the scores of the target syndrome elements included in the second candidate disease;

[0014] For each of the second candidate diseases, performing a weighted process on the corresponding syndrome element matching score based on the incidence rate of the second candidate disease to obtain a first prediction score; wherein, the incidence rate is positively correlated with the first prediction score;

[0015] Determining the predicted disease according to the first prediction scores of all the second candidate diseases.

[0016] In the above implementation process, by screening out a preset number of second candidate diseases that include one or more target syndrome elements, and then comprehensively considering the incidence rate of the diseases and the scores of each target syndrome element, estimating the first prediction score of each second candidate disease, so as to estimate the disease that the user is most likely to suffer from based on the first prediction score, the process of completing disease prediction only by using the user's diagnosis and treatment information is realized, which does not depend on the user's physical symptoms and ensures the accuracy of disease prediction at the same time.

[0017] Further, there are multiple pieces of the user's diagnosis and treatment information, and different pieces of the user's diagnosis and treatment information correspond to the same target syndrome element or different target syndrome elements; before determining the syndrome element matching score of the second candidate disease according to the scores of the target syndrome elements included in the second candidate disease, the method further includes:

[0018] Determining the target syndrome element corresponding to each piece of the user's diagnosis and treatment information respectively to obtain a plurality of target syndrome elements;

[0019] For each of the target syndrome elements, counting the number of occurrences of the target syndrome element and determining the score of the target syndrome element based on the number of occurrences; wherein, the number of occurrences is positively correlated with the score.

[0020] In the above implementation process, the number of occurrences of the target syndrome elements in the user's diagnosis and treatment information is combined to dynamically adjust the scores of each target syndrome element, so that diseases with a high number of occurrences can obtain a higher syndrome element matching score, improving the prediction accuracy of predicting diseases.

[0021] Further, the user's diagnosis and treatment information carries user attribute information; the user attribute information includes one or more of gender information, age information, and geographical location information; before weighting the corresponding syndrome element matching score based on the incidence rate of the second candidate disease, the method further includes:

[0022] Based on the user attribute information carried in the user's diagnosis and treatment information, determine the incidence rate of each second candidate disease.

[0023] In the above implementation process, determining the disease incidence rate of the user population based on the attributes of different users and comprehensively predicting the diseases that the user may suffer from in combination with the disease incidence rates of different populations can further improve the disease prediction accuracy.

[0024] Further, determining the predicted syndrome type including at least some of the target syndrome elements from all the syndrome types included in the predicted disease includes:

[0025] Determine the first syndrome type obtained by combining multiple target syndrome elements to obtain a first syndrome type set including all the first syndrome types, and determine the second syndrome type included in the predicted disease to obtain a second syndrome type set including all the second syndrome types;

[0026] According to the intersection of the first syndrome type set and the second syndrome type set, determine the first candidate syndrome type;

[0027] For each first candidate syndrome type, determine the syndrome element composite score of the first candidate syndrome type relative to multiple target syndrome elements, and weight the syndrome element composite score using the incidence rate of the first candidate syndrome type under the predicted disease to obtain a second predicted score; wherein, the syndrome element compliance score of the first candidate syndrome type is positively correlated with the second quantity of the target syndrome elements included in the first candidate syndrome type;

[0028] Based on the second predicted scores of all the first candidate syndrome types, determine the predicted syndrome type.

[0029] In the above implementation process, the first syndrome type set is obtained by combining the target syndrome elements, and then the intersection of the first syndrome type set and the second syndrome type set included in the predicted disease is taken as the first candidate syndrome type. At the same time, considering the incidence rate of the first candidate syndrome type under the predicted disease and the number of target syndrome elements included in the first candidate syndrome type, the second prediction score of each first candidate syndrome type is estimated, so as to estimate the syndrome type that the user is most likely to suffer from based on the second prediction score, realizing the prediction process of diseases and syndrome types only using the user's diagnosis and treatment information, ensuring the accuracy of disease prediction without relying on the user's physical symptoms.

[0030] Further, determining the predicted syndrome type based on the second prediction scores of each of the first candidate syndrome types includes:

[0031] Based on the maximum value among all the second prediction scores and a preset score difference, determine the score error range;

[0032] From all the first candidate syndrome types, determine the second candidate syndrome types that fall within the score error range;

[0033] From all the second candidate syndrome types, determine the second candidate syndrome type with the highest composite score of syndrome elements as the predicted syndrome type.

[0034] In the above implementation process, when the second prediction scores of multiple first candidate syndrome types are not much different, the number of target syndrome elements included in the syndrome type is preferentially considered to select the predicted syndrome type, improving the accuracy of syndrome type prediction.

[0035] Further, the target prediction strategy includes a second prediction strategy, and the second prediction strategy includes:

[0036] According to the pre-established correspondence relationship between the diagnosis and treatment information and the syndrome type, determine the predicted syndrome type corresponding to at least one piece of the user's diagnosis and treatment information;

[0037] Determine the predicted disease corresponding to the predicted syndrome type, wherein the incidence rate of the predicted syndrome type under the predicted disease meets the preset high-incidence condition.

[0038] In the above implementation process, the prediction of diseases and syndrome types is realized through the prediction link of "tongue, face, pulse, palm - syndrome type - disease". Compared with the first prediction strategy, the second prediction strategy omits the process of matching syndrome elements, directly establishes the correspondence relationship between the diagnosis and treatment information and the syndrome type, and then inversely predicts one or more diseases based on the high-incidence syndrome type. Since there is no need to match syndrome elements, the prediction efficiency can be improved.

[0039] In a second aspect of the embodiments of the present application, a user diagnosis and treatment information processing device is provided. The device is configured to obtain a predicted disease and a predicted syndrome type based on user diagnosis and treatment information through a target prediction strategy. Among them, the user diagnosis and treatment information includes one or more of tongue diagnosis information, face diagnosis information, pulse diagnosis information, and palm diagnosis information. The target prediction strategy includes a first prediction strategy. Specifically, the device includes:

[0040] A syndrome element module, configured to determine a plurality of target syndrome elements corresponding to the user diagnosis and treatment information according to the pre-established correspondence between diagnosis and treatment information and syndrome elements;

[0041] A disease module, configured to determine a predicted disease that matches at least part of the target syndrome elements according to the pre-established correspondence between diseases and syndrome elements;

[0042] A syndrome type module, configured to determine a predicted syndrome type that includes at least part of the target syndrome elements from all the syndrome types included in the predicted disease. Among them, the predicted disease and the predicted syndrome type are used to provide auxiliary reference information to physicians.

[0043] In a third aspect of the embodiments of the present application, an electronic device is provided. The electronic device includes:

[0044] A processor;

[0045] A memory for storing executable instructions of the processor;

[0046] Among them, when the processor calls the executable instructions, the operations of any of the methods in the first aspect are implemented.

[0047] In a fourth aspect of the embodiments of the present application, a computer-readable storage medium is provided, on which computer instructions are stored. When the computer instructions are executed by a processor, the steps of any of the methods in the first aspect are implemented. Description of the Drawings

[0048] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings required to be used in the embodiments of the present application. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0049] Figure 1 It is a schematic flowchart of a user diagnosis and treatment information processing method provided by an embodiment of the present application;

[0050] Figures 2 - 8 It is a schematic flowchart of another user diagnosis and treatment information processing method provided by an embodiment of the present application;

[0051] Figure 9This is a hardware structure diagram of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0052] Next, the technical solutions in the embodiments of the present application will be described with reference to the accompanying drawings in the embodiments of the present application.

[0053] It should be noted that similar reference numerals and letters denote similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of the present application, terms such as "first" and "second" are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0054] In some cases, it is difficult to accurately know the symptoms of a disease. For example, for a disease with latent onset, patients do not perceive obvious symptoms in the early stage of the disease. Another example is that the symptoms of some diseases are not obvious, or the patient's perception of the symptoms is not obvious. Another example is that the patient's language expression is unclear, or the doctor's understanding of the symptoms is not in place. All these situations will make it difficult for a computer to accurately obtain the symptoms of a disease and even form a wrong description of the symptoms.

[0055] Currently, the prediction of a disease highly depends on the symptoms of the disease. When it is difficult for a computer to accurately know the symptoms of a disease, it is often difficult to accurately predict the disease suffered. In addition, for a situation where the symptoms are very obvious, "predicting" the disease suffered at this moment based on the current symptoms actually belongs to the process of disease differentiation, rather than prediction. The so-called prediction should be able to detect the disease suffered by a patient before obvious symptoms appear or even before somatic symptoms appear. Obviously, such a prediction process is more conducive to controlling the course of the disease and preventing the disease from deteriorating.

[0056] For this reason, the present application proposes a method for processing user diagnosis and treatment information. Among them, all steps included in this method are implemented by an electronic device with computing capabilities such as a computer. The computer processes the user diagnosis and treatment information by executing all steps of this method to obtain one or more intermediate processing results. And the final result output by the computer is used as reference information for doctors to refer to. Specifically, the method for processing user diagnosis and treatment information includes: predicting the predicted disease and the predicted syndrome type based on the user diagnosis and treatment information through a target prediction strategy.

[0057] Among them, the predicted disease refers to the disease that the user is predicted to have. The so-called syndrome type refers to the syndromes that are relatively common, typical, with standardized or established syndrome names in clinical practice. Regarding the relationship between diseases and syndrome types, one disease can be divided into multiple syndrome types, and the symptoms and treatment methods of different syndrome types under the same disease are different. Also, the same syndrome type can correspond to multiple diseases. For example, the syndrome type of "syndrome of spleen-yang deficiency" can correspond to multiple diseases such as diarrhea, constipation, and vomiting. In this embodiment, the predicted syndrome type refers to the syndrome type predicted corresponding to the disease that the user is predicted to have.

[0058] It should be noted that the predicted disease and predicted syndrome type obtained in this embodiment should play an auxiliary role in disease diagnosis. After the predicted disease and predicted syndrome type are output to the physician, the physician can choose to adopt or refer to the predicted disease and predicted syndrome type to determine the final diagnosis. At the same time, the physician can also choose not to adopt or refer to the predicted disease and predicted syndrome type. The predicted disease and predicted syndrome type should not be directly used as the diagnosis result to replace the physician's final diagnosis.

[0059] The user's diagnosis and treatment information includes one or more of the user's tongue diagnosis information, face diagnosis information, pulse diagnosis information, and palm diagnosis information. The tongue diagnosis information, for example, can be obtained by performing image recognition on the user's tongue image. The tongue diagnosis information includes, for example, but is not limited to, information on the tongue coating, tongue body, and tongue bottom that can be used for diagnosis. The face diagnosis information, for example, can be obtained by performing image recognition on the user's face image. The face diagnosis information includes, for example, but is not limited to, information on the facial color, acne, and spots that can be used for diagnosis. The pulse diagnosis information, for example, can be obtained from the collected data of a pulse sensor. The pulse diagnosis information includes, for example, but is not limited to, the cun pulse information, guan pulse information, and chi pulse information that can be used for diagnosis. The palm diagnosis information, for example, can be obtained by performing image recognition on the user's palm. The palm diagnosis information includes, for example, but is not limited to, the palm shape and palm color that can be used for diagnosis.

[0060] It can be understood that the tongue diagnosis information, face diagnosis information, pulse diagnosis information, and palm diagnosis information in the user's diagnosis and treatment information are different from the symptoms mentioned above. The symptoms mentioned above usually refer to the symptoms that bring discomfort to the patient, such as obvious somatic symptoms like pain, diarrhea, cough, drowsiness, and nightmares. When the patient has these obvious somatic symptoms, they will go to see a doctor. However, user diagnosis and treatment information such as the tongue, face, pulse, and palm are not obvious somatic symptoms and may not even attract the user's attention. For example, a user will not go to see a doctor because the palm color is relatively red. But from the perspective of traditional Chinese medicine, these diagnosis and treatment information on the tongue, face, pulse, and palm often imply disease information, indicating the disease the patient has and the course of the disease. Therefore, predicting the possible diseases and syndrome types of the user based on this diagnosis and treatment information that does not attract the user's attention and revealing the diseases before the user has obvious somatic symptoms is of great significance for controlling the course of the disease and preventing the deterioration of the disease.

[0061] In some embodiments, the target prediction strategy includes a first prediction strategy. Thus, a method for processing user diagnosis and treatment information provided in this application may include steps 110 - 140 as Figure 1 shown.

[0062] Step 110: Obtain user diagnosis and treatment information. Among them, the user diagnosis and treatment information includes one or more of tongue diagnosis information, face diagnosis information, pulse diagnosis information, and palm diagnosis information.

[0063] For the description of user diagnosis and treatment information, refer to the above text and will not be elaborated here.

[0064] Step 120: Determine a plurality of target syndrome elements corresponding to the user diagnosis and treatment information according to the pre - established correspondence between diagnosis and treatment information and syndrome elements.

[0065] Before executing the method for processing user diagnosis and treatment information, a traditional Chinese medicine expert team may pre - establish the correspondence between diagnosis and treatment information and syndrome elements, including but not limited to the correspondence between tongue diagnosis information and syndrome elements, the correspondence between face diagnosis information and syndrome elements, the correspondence between pulse diagnosis information and syndrome elements, and the correspondence between palm diagnosis information and syndrome elements. And store these correspondences in a computer or server, so that when the computer executes the steps of this method, it can call the corresponding correspondence from the local or from the server.

[0066] Among them, the so - called syndrome element refers to the element of syndrome, including the disease location syndrome element and the disease nature syndrome element. And the syndrome type is composed of syndrome elements. Specifically, the syndrome type can be composed of the disease location syndrome element and the disease nature syndrome element. For example, the syndrome type "syndrome of spleen - kidney yang deficiency" is composed of the disease location syndrome element "spleen", the disease location syndrome element "kidney", and the disease nature syndrome element "yang deficiency". In addition, the syndrome type can also be composed of multiple disease nature syndrome elements. For example, the syndrome type "syndrome of deficiency of both qi and blood" is composed of the disease nature syndrome element "qi deficiency" and the disease nature syndrome element "blood deficiency".

[0067] In addition, the relationship between diagnosis and treatment information and syndrome elements can be one - to - one. For example, the tongue diagnosis information "yellow and greasy tongue coating" corresponds to the syndrome element "damp - heat". The relationship between diagnosis and treatment information and syndrome elements can also be one - to - many. For example, the pulse diagnosis information "thin and deep on the right chi position" corresponds to the syndrome elements "qi deficiency" and "yin deficiency".

[0068] Thus, by searching for user diagnosis and treatment information in the pre - established correspondence between diagnosis and treatment information and syndrome elements, the corresponding target syndrome elements can be mapped, and there are multiple target syndrome elements.

[0069] Step 130: Determine the predicted diseases that match at least part of the target syndrome elements according to the pre - established correspondence between diseases and syndrome elements.

[0070] Before executing the user diagnosis and treatment information processing method, a correspondence relationship between diseases and syndrome elements can be established in advance by a team of traditional Chinese medicine experts. Among them, the relationship between diseases and syndrome elements is one-to-many. For example, the syndrome elements corresponding to the disease "diarrhea" include the location syndrome element "spleen", the location syndrome element "kidney", the location syndrome element "liver", the nature syndrome element "qi deficiency", the nature syndrome element "yang deficiency", the nature syndrome element "damp-heat", the nature syndrome element "dampness", and the nature syndrome element "qi stagnation", etc.

[0071] In this way, by searching for target syndrome elements in the correspondence relationship between diseases and syndrome elements, a predicted disease that matches at least some of the target syndrome elements is mapped. The so-called predicted disease that matches at least some of the target syndrome elements means that at least some of the target syndrome elements are included in the multiple syndrome elements corresponding to the predicted disease.

[0072] Step 140: Determine a predicted syndrome type that includes at least some of the target syndrome elements from all the syndrome types included in the predicted disease.

[0073] As described above, a disease can be divided into multiple syndrome types. Therefore, after predicting the predicted disease through step 130, it is possible to predict which specific syndrome type under the predicted disease the user may have by executing step 140. Specifically, all the syndrome types included in the predicted disease can be determined first. At the same time, since the syndrome type is composed of syndrome elements, a syndrome type that includes at least some of the target syndrome elements is then determined from all the syndrome types included in the predicted disease as the predicted syndrome type. Finally, the predicted disease and the predicted syndrome type can be output to complete the prediction process of the disease and the syndrome type based on the user diagnosis and treatment information.

[0074] It can be seen that a user diagnosis and treatment information processing method provided in this embodiment realizes the prediction of diseases and syndrome types through the prediction link of "tongue, face, pulse, palm - syndrome element - disease - syndrome type". The entire prediction process does not depend on the obvious physical symptoms of the user, and can achieve the technical effect of accurately predicting the disease syndrome type before the user shows obvious physical symptoms.

[0075] The following provides a detailed introduction to steps 110 - 140.

[0076] Regarding the prediction process of the predicted disease in step 130, in some embodiments, it may include steps 131 - 133 as Figure 2 shown.

[0077] Step 131: Determine a preset number of second candidate diseases according to the correspondence relationship between the diseases and the syndrome elements. Among them, the second candidate diseases include one or more of the target syndrome elements.

[0078] First, a preset number of second candidate diseases are screened out from all diseases, and the second candidate diseases include one or more target syndrome elements. Among them, the preset number can be determined according to actual needs, for example, it is 5.

[0079] Step 132: For each of the second candidate diseases, determine the first prediction score of the second candidate disease according to the scores of the target syndrome elements included in the second candidate disease and the incidence rate of the second candidate disease.

[0080] Next, estimate the first prediction score for each second candidate disease. The first prediction score is used to screen out the final predicted diseases. Specifically, scores can be assigned to each target syndrome element, and the score assignment rules will be introduced below. Then, in combination with the incidence rate of each second candidate disease and the scores of the target syndrome elements included in the second candidate disease, the first prediction score is comprehensively calculated.

[0081] Step 133: Determine the predicted disease according to the first prediction scores of all the second candidate diseases.

[0082] Finally, according to the first prediction scores of all the second candidate diseases, screen out the predicted diseases from all the second candidate diseases. For example, the second candidate disease with the highest first prediction score can be determined as the predicted disease.

[0083] In some embodiments, as Figure 3 shown, step 131 may specifically include steps 1311 - 1312, and step 132 may specifically include steps 1321 - 1322. Thus, regarding the prediction process of the predicted disease in step 130, it specifically includes steps 1311 - 133 as Figure 3 shown.

[0084] Step 1311: Determine the first candidate diseases including one or more of the target syndrome elements according to the correspondence relationship between the diseases and the syndrome elements.

[0085] Exemplarily, first, according to the correspondence relationship between the diseases and the syndrome elements, the diseases including one or more target syndrome elements are screened out as the first candidate diseases. That is, traverse the correspondence relationship between the diseases and the syndrome elements. For each traversed disease, if the syndrome elements corresponding to the disease hit one or more target syndrome elements, then determine the disease as the first candidate disease. If none of the target syndrome elements are hit by the syndrome elements corresponding to a certain disease, then exclude the possibility of that disease.

[0086] Thus, in step 1311, all the diseases that hit one or more target syndrome elements are first screened out.

[0087] Step 1312: According to the first quantity of the target syndrome elements included in each of the first candidate diseases, sequentially determine a preset number of second candidate diseases from the multiple first candidate diseases in descending order of the first quantity.

[0088] Subsequently, the first quantity of the target syndrome elements included in each first candidate disease can be counted. Then, the first candidate diseases are sorted in descending order according to the first quantity, and a preset number of second candidate diseases are sequentially determined therefrom. Thus, the process of determining the preset number of second candidate diseases in step 131 above is completed.

[0089] Step 1321: For each of the second candidate diseases, determine the syndrome element matching score of the second candidate disease according to the scores of the target syndrome elements included in the second candidate disease.

[0090] For each of the preset number of second candidate diseases screened out, according to the scores of the target syndrome elements included in the second candidate disease, the syndrome element matching score of the second candidate disease can be determined. For example, if the second candidate disease A includes the target syndrome element a and the target syndrome element b, and the scores of both target syndrome elements are 5 points, then the syndrome element matching score of the second candidate disease A is 10 points, and so on.

[0091] Step 1322: For each of the second candidate diseases, perform weighted processing on the corresponding syndrome element matching score based on the incidence rate of the second candidate disease to obtain the first prediction score. Wherein, the incidence rate is positively correlated with the first prediction score.

[0092] After obtaining the syndrome element matching score of each second candidate disease, then obtain the incidence rate of each second candidate disease, and perform weighted processing on the syndrome element matching score by using the incidence rate to obtain the first prediction score. As an example, the product of the incidence rate and the syndrome element matching score can be determined as the first prediction score. As an example, a weight coefficient can be determined based on the incidence rate, and the product of the weight coefficient and the syndrome element matching score can be determined as the first prediction score. Wherein, the incidence rate is positively correlated with the weight coefficient. For example, when the incidence rate is between 1 / 100000 and 5 / 100000, the weight coefficient is 1; when the incidence rate is between 6 / 100000 and 1 / 10000, the weight coefficient is 3 points, and so on.

[0093] Step 133: Determine the predicted disease according to the first prediction scores of all the second candidate diseases.

[0094] It can be seen that in this embodiment, by screening out a preset number of second candidate diseases including one or more target syndrome elements, and then comprehensively considering the incidence rate of the diseases and the scores of each target syndrome element, the first prediction score of each second candidate disease is estimated, so as to estimate the disease that the user is most likely to suffer from based on the first prediction score, realizing the process of disease prediction only by using the user's diagnosis and treatment information, ensuring the accuracy of disease prediction without relying on the user's physical symptoms.

[0095] In addition, regarding the scores of the target syndrome elements, in some embodiments, the scores of each target syndrome element can be the same, for example, all are 5 points.

[0096] Regarding the scores of the target syndrome elements, in some embodiments, different scores can be assigned to different target syndrome elements. The score of a target syndrome element is positively correlated with the number of occurrences of the target syndrome element in the user's diagnosis and treatment information. Specifically, since the user's diagnosis and treatment information includes multiple pieces, and different user's diagnosis and treatment information can correspond to the same target syndrome element or different target syndrome elements. Therefore, there may be multiple pieces of user's diagnosis and treatment information that are all mapped to the same target syndrome element, that is, a target syndrome element appears multiple times in the user's diagnosis and treatment information. Thus, in this embodiment, the user's diagnosis and treatment information includes multiple pieces, and before performing step 1321, for example, when performing step 120, the steps 410 - step 420 as shown in Figure 4 can be executed.

[0097] Step 410: Determine the target syndrome element corresponding to each piece of the user's diagnosis and treatment information respectively, and obtain multiple target syndrome elements.

[0098] Exemplarily, for each piece of user's diagnosis and treatment information, search for the user's diagnosis and treatment information in the pre - established correspondence relationship between the diagnosis and treatment information and the syndrome elements, so as to map out the corresponding target syndrome element, and obtain the target syndrome element corresponding to each piece of user's diagnosis and treatment information.

[0099] Step 420: For each of the target syndrome elements, count the number of occurrences of the target syndrome element, and determine the score of the target syndrome element based on the number of occurrences. Wherein, the number of occurrences is positively correlated with the score.

[0100] Exemplarily, the so - called counting the number of occurrences of each target syndrome element means counting how many pieces of user's diagnosis and treatment information are mapped to each target syndrome element respectively. Then, determine the score of each target syndrome element based on the number of occurrences. And the number of occurrences is positively correlated with the score.

[0101] It can be understood that if a certain target syndrome element is mapped from multiple pieces of user's diagnosis and treatment information, and this target syndrome element is a disease location syndrome element, it can be considered that the user is very likely to have a lesion in the part indicated by the disease location syndrome element; if this target syndrome element is a disease nature syndrome element, it can be considered that the user is very likely to have the disease nature indicated by the disease nature syndrome element. Therefore, different scores can be assigned to different target syndrome elements according to the number of occurrences of each target syndrome element.

[0102] In addition, it can also be determined that the score of the disease nature syndrome element is higher than the score of the disease location syndrome element, or the score of the disease location syndrome element is higher than the score of the disease nature syndrome element.

[0103] It can be seen that in this embodiment, the number of occurrences of the target syndrome elements in the user's diagnosis and treatment information is combined to dynamically adjust the scores of each target syndrome element, so that the diseases that hit the target syndrome elements with a high number of occurrences can obtain a higher syndrome element matching score, improving the prediction accuracy of predicting diseases.

[0104] In addition, regarding the incidence of the second candidate diseases, in some embodiments, the incidence of each second candidate disease can be determined by referring to the data published by the relevant industry. The same disease can adopt the same incidence in different populations. In other embodiments, in order to improve the disease prediction accuracy, considering that the same disease may have different incidences in different populations. The disease incidence is often related to age, gender, and region. Therefore, on the basis of any of the above embodiments, the user's diagnosis and treatment information can carry user attribute information. And the user attribute information includes one or more of gender information, age information, and regional information where the user is located. Based on this, before performing step 1322, the method further includes the steps:

[0105] Based on the user attribute information carried by the user's diagnosis and treatment information, determine the incidence of each of the second candidate diseases.

[0106] Exemplarily, based on one or more of the gender information, age information, and regional information carried by the user's diagnosis and treatment information, determine the incidence of each second candidate disease relative to the user. Then, perform a weighted process on the syndrome element matching score based on the incidence.

[0107] Thus, in this embodiment, determining the disease incidence of the user's population based on the attributes of different users and comprehensively predicting the diseases that the user may suffer from in combination with the disease incidences of different populations can further improve the disease prediction accuracy.

[0108] To better understand the prediction process of predicting diseases, the following example is used for illustration.

[0109] First, if after mapping out 5 target syndrome elements of "spleen", "kidney", "qi deficiency", "blood deficiency", and "yang deficiency" according to the user's diagnosis and treatment information, determine the first candidate diseases by looking up the corresponding relationship between diseases and syndrome elements. At this time, usually a large number of first candidate diseases will be determined. Therefore, according to the first quantity of the target syndrome elements included in each first candidate disease, then screen out the first five first candidate diseases corresponding to the first quantity as the second candidate diseases. For example, the second candidate diseases screened out include "diarrhea", "irritable bowel syndrome", "edema", "forgetfulness", and "spontaneous sweating".

[0110] And from the corresponding relationship between diseases and syndrome elements, it can be known that the syndrome elements corresponding to the above 5 second candidate diseases are:

[0111] "Diarrhea" includes syndrome elements: spleen, kidney, liver, qi deficiency, yang deficiency, damp-heat, dampness, qi stagnation...

[0112] "Edema" includes syndrome elements: spleen, kidney, lung, qi deficiency, yang deficiency, damp-heat, dampness, blood stasis...

[0113] "Forgetfulness" includes syndrome elements: heart, kidney, spleen, liver, yin deficiency, blood deficiency, phlegm...

[0114] "Spontaneous sweating" includes syndrome elements: lung, qi deficiency, yang deficiency, wind, dampness, summerheat, damp-heat...

[0115] "Irritable bowel syndrome" includes syndrome elements: liver, spleen, kidney, dampness, food, qi deficiency, yang deficiency, blood deficiency...

[0116] Assuming that each target syndrome element is 5 points, it can be seen that the second candidate disease "diarrhea" hits 4 target syndrome elements: "spleen", "kidney", "qi deficiency" and "yang deficiency". Therefore, the syndrome element matching score of "diarrhea" is 20 points. At the same time, it is known that the incidence rate of "diarrhea" is 80 / 10000, and the corresponding weight coefficient is 5. Therefore, the first predicted score of "diarrhea" is 100 points.

[0117] The second candidate disease "edema" hits 4 target syndrome elements: "spleen", "kidney", "qi deficiency" and "yang deficiency". Therefore, the syndrome element matching score of "diarrhea" is 20 points. At the same time, it is known that the incidence rate of "edema" is 3 / 10000, and the corresponding weight coefficient is 3. Therefore, the first predicted score of "edema" is 60 points.

[0118] The second candidate disease "forgetfulness" hits 3 target syndrome elements: "spleen", "kidney" and "blood deficiency". Therefore, the syndrome element matching score of "forgetfulness" is 15 points. At the same time, it is known that the incidence rate of "forgetfulness" is 8 / 10000, and the corresponding weight coefficient is 3. Therefore, the first predicted score of "forgetfulness" is 45 points.

[0119] The second candidate disease "spontaneous sweating" hits 2 target syndrome elements: "qi deficiency" and "yang deficiency". Therefore, the syndrome element matching score of "spontaneous sweating" is 10 points. At the same time, it is known that the incidence rate of "spontaneous sweating" is 9 / 100000, and the corresponding weight coefficient is 2. Therefore, the first predicted score of "spontaneous sweating" is 20 points.

[0120] The second candidate disease "irritable bowel syndrome" hits 5 target syndrome elements: "spleen", "kidney", "qi deficiency", "blood deficiency" and "yang deficiency". Therefore, the syndrome element matching score of "irritable bowel syndrome" is 25 points. At the same time, it is known that the incidence rate of "irritable bowel syndrome" is 35 / 10000, and the corresponding weight coefficient is 5. Therefore, the first predicted score of "irritable bowel syndrome" is 125 points.

[0121] By synthesizing the first predicted scores of each second candidate disease, it can be determined that the "irritable bowel syndrome" with the highest first predicted score is the predicted disease that the user may suffer from.

[0122] After determining the predicted disease, the predicted syndrome type can be further predicted. Based on any of the above embodiments, regarding the prediction process of the predicted syndrome type in step 140, it may include steps 141-step 143 as shown in Figure 5 below.

[0123] Step 141: Based on the multiple target syndrome elements and all syndrome types included in the predicted disease, determine the first candidate syndrome type.

[0124] As described above, since a syndrome type is composed of multiple syndrome elements, the first candidate syndrome type can be first determined from all syndrome types included in the predicted disease based on the multiple target syndrome elements. Exemplarily, the first candidate syndrome type includes at least some of the target syndrome elements. For example, the first candidate syndrome type only includes at least some of the target syndrome elements. That is, the syndrome elements that make up the first candidate syndrome type are all target syndrome elements.

[0125] Step 142: For each first candidate syndrome type, based on the incidence rate of the first candidate syndrome type under the predicted disease and the second quantity of the target syndrome elements included in the first candidate syndrome type, determine the second prediction score of the first candidate syndrome type.

[0126] Next, estimate the second prediction score for each first candidate syndrome type. The second prediction score is used to screen out the predicted syndrome type from all the first candidate syndrome types. Specifically, the second prediction score can be determined by combining the second quantity of the target syndrome elements included in each first candidate syndrome type and the incidence rate of each first candidate syndrome type under the predicted disease. As described above, a disease can be further divided into multiple syndrome types, and the same syndrome type can correspond to multiple diseases. Therefore, different syndrome types under a disease will have different incidence rates, and the same syndrome type may also have different incidence rates under different diseases.

[0127] Step 143: Based on the second prediction scores of all the first candidate syndrome types, determine the predicted syndrome type.

[0128] Finally, according to the second prediction scores of all the first candidate syndrome types, screen out the predicted syndrome type from all the first candidate syndrome types. For example, the first candidate syndrome type with the highest second prediction score can be determined as the predicted syndrome type.

[0129] In some embodiments, as shown in Figure 6 below, step 141 may specifically include steps 1411-step 1412, and step 142 may specifically include step 1421. Thus, regarding the prediction process of the predicted syndrome type in step 140, it may specifically include steps 1411-step 143 as shown in Figure 6 below.

[0130] Step 1411: Determine the first syndrome type obtained by combining multiple target syndrome elements, obtain a first syndrome type set including all the first syndrome types, and determine the second syndrome types included in the predicted disease, obtain a second syndrome type set including all the second syndrome types.

[0131] Exemplarily, first, multiple target syndrome elements can be combined to obtain the first syndrome type. Among them, the syndrome elements include disease location syndrome elements and disease nature syndrome elements. Then, the disease nature syndrome elements and disease location syndrome elements in the target syndrome elements can be combined to obtain the first syndrome type; or multiple disease nature syndrome elements can be combined to obtain the first syndrome type. Thus, a first syndrome type set including multiple first syndrome types is obtained.

[0132] And, all the syndrome types subdivided under the predicted disease can be determined as the second syndrome types, and a second syndrome type set including all the second syndrome types is obtained.

[0133] Step 1412: Determine the first candidate syndrome type according to the intersection of the first syndrome type set and the second syndrome type set.

[0134] Take the intersection of the first syndrome type set and the second syndrome type set, and the elements in the intersection are the first candidate syndrome types.

[0135] Step 1421: For each first candidate syndrome type, determine the syndrome element composite score of the first candidate syndrome type relative to multiple target syndrome elements, and use the incidence rate of the first candidate syndrome type in the predicted disease to weight the syndrome element composite score to obtain a second prediction score. Among them, the syndrome element compliance score of the first candidate syndrome type is positively correlated with the second quantity of the target syndrome elements included in the first candidate syndrome type.

[0136] Exemplarily, for each first candidate syndrome type, the second quantity of the target syndrome elements included in the first candidate syndrome type can be counted, and the syndrome element composite score of the first candidate syndrome type can be determined based on the second quantity. Optionally, the second quantity can be directly determined as the syndrome element composite score. At this time, the syndrome element compliance score is used to represent the quantity of the target syndrome elements included in the first candidate syndrome type. Optionally, it can be determined to use a preset mapping relationship to determine the syndrome element composite score corresponding to the second quantity.

[0137] After obtaining the syndrome element composite score of each first candidate syndrome type, the incidence rate of each first candidate syndrome type in the predicted disease is obtained, and the incidence rate is used to weight the syndrome element composite score to obtain the second prediction score. As an example, all first candidate syndrome types can be sorted based on the incidence rate from high to low, and the weight corresponding to each first candidate syndrome type in each order is determined. For example, the weight of the first place is 3, the weight of the second place is 2, the weight of the third place is 1, and so on. As another example, the weight coefficient can be determined based on the incidence rate of the first candidate syndrome type, and the incidence rate is positively correlated with the weight coefficient. Subsequently, the product of the weight coefficient and the syndrome element composite score is determined as the second prediction score.

[0138] In addition, regarding the incidence rate of the first candidate syndrome type in the predicted disease, optionally, it can be determined that the incidence rates of the first candidate syndrome type in different populations in the predicted disease are the same. Optionally, considering that the incidence rates of the same disease syndrome type may be different in different populations, the incidence rate of the first candidate syndrome type in the predicted disease can be determined based on the user attribute information (such as age, gender, region) carried in the user diagnosis and treatment information.

[0139] Step 143: Determine the predicted syndrome type based on the second prediction scores of all the first candidate syndrome types.

[0140] It can be seen that in this embodiment, the first syndrome type set is obtained by combining the target syndrome elements, and then the intersection of the first syndrome type set and the second syndrome type set included in the predicted disease is taken as the first candidate syndrome type. At the same time, the incidence rate of the first candidate syndrome type in the predicted disease and the number of target syndrome elements included in the first candidate syndrome type are comprehensively considered, and the second prediction score of each first candidate syndrome type is estimated. Thus, based on the second prediction score, the syndrome type that the user is most likely to suffer from is estimated, realizing the prediction process of disease and syndrome type only by using the user diagnosis and treatment information, ensuring the accuracy of disease prediction without relying on the user's physical symptoms.

[0141] In addition, regarding determining the predicted syndrome type based on the second prediction score in step 143, in some embodiments, the first candidate syndrome type with the highest second prediction score can be determined as the predicted syndrome type. In other embodiments, step 143 may specifically include steps 1431-step 1433 as Figure 7 shown.

[0142] Step 1431: Determine the scoring error range based on the maximum value among all the second prediction scores and a preset scoring difference.

[0143] As an example, the maximum value among the second prediction scores can be determined as the upper limit of the scoring error range, the difference between the maximum value and the scoring difference is the lower limit of the scoring error range, and the scoring error range is a closed interval.

[0144] As an example, it can be determined that the sum of the maximum value and the scoring difference is the upper limit of the scoring error range, the difference between the maximum value and the scoring difference is the lower limit of the scoring error range, and the scoring error range is an open interval.

[0145] Step 1432: From all the first candidate syndrome types, determine the second candidate syndromes that fall within the scoring error range.

[0146] After determining the scoring error range, it can be determined that all the first candidate syndrome types that fall within the scoring error range are the second candidate syndrome types. It can be seen that the first candidate syndrome type with the highest second predicted score must fall within the scoring error range. If there is only one first candidate syndrome type with the highest second predicted score within the scoring error range, then this first candidate syndrome type can be directly determined as the predicted syndrome type. If the scoring error range includes multiple second candidate syndrome types, which means that the difference between the second predicted scores of the multiple second candidate syndrome types does not exceed the preset scoring difference, then perform step 1433.

[0147] Step 1433: Determine the second candidate syndrome type with the highest syndrome element composite score among all the second candidate syndrome types as the predicted syndrome type.

[0148] If the difference between the second predicted scores of multiple second candidate syndrome types does not exceed the preset scoring difference, it can be considered that the second predicted scores of these second candidate syndrome types do not differ much. At this time, the syndrome element composite score is given priority, and the second candidate syndrome type with the highest syndrome element composite score, that is, the second candidate syndrome type including the most target syndrome elements, is determined as the predicted syndrome type.

[0149] It can be seen that in this embodiment, when the difference between the second predicted scores of multiple first candidate syndrome types is small, the number of target syndrome elements included in the syndrome type is given priority to select the predicted syndrome type, which improves the accuracy of syndrome type prediction.

[0150] To better understand the prediction process of the predicted syndrome type, the above example is continued for illustration.

[0151] In the above example, the target syndrome elements mapped by the user's diagnosis and treatment information include 5 target syndrome elements: "spleen", "kidney", "qi deficiency", "blood deficiency" and "yang deficiency". These 5 target syndrome elements can be combined to obtain 10 first syndrome types: "syndrome of spleen-kidney yang deficiency", "syndrome of spleen-kidney qi deficiency", "syndrome of kidney yang deficiency", "syndrome of spleen yang deficiency", "syndrome of spleen qi deficiency", "syndrome of kidney qi deficiency", "syndrome of qi and blood deficiency", "syndrome of qi deficiency", "syndrome of yang deficiency" and "syndrome of blood deficiency", obtaining the first syndrome type set.

[0152] And in the above example, "irritable bowel syndrome" was determined as the predicted disease that the user may suffer from. The second syndrome types included in "irritable bowel syndrome" are "kidney-yang deficiency syndrome", "spleen-qi deficiency syndrome", "spleen-kidney yang deficiency syndrome", "dampness-heat in large intestine syndrome", "liver depression and spleen deficiency syndrome", "mixture of cold and heat syndrome", "spleen deficiency and dampness accumulation syndrome", and "food retention in the intestines syndrome", obtaining the second syndrome type set.

[0153] Take the intersection of the first syndrome type set and the second syndrome type set, and the first candidate syndrome types obtained include "kidney-yang deficiency syndrome", "spleen-qi deficiency syndrome", and "spleen-kidney yang deficiency syndrome".

[0154] For the first candidate syndrome type "kidney-yang deficiency syndrome", which includes 2 target syndrome elements of "kidney" and "yang deficiency", the composite syndrome element score is determined to be 2. For the first candidate syndrome type "spleen-qi deficiency syndrome", which includes 2 target syndrome elements of "spleen" and "qi deficiency", the composite syndrome element score is determined to be 2. For the first candidate syndrome type "spleen-kidney yang deficiency syndrome", which includes 3 target syndrome elements of "spleen", "kidney", and "yang deficiency", the composite syndrome element score is determined to be 3.

[0155] Subsequently, the composite syndrome element score is weighted using the syndrome type incidence rate to obtain the second prediction score, and the predicted syndrome type is determined using the second prediction score.

[0156] As an example, the incidence rates of the three first candidate syndrome types under the predicted disease "irritable bowel syndrome" can be sorted, and then the weights corresponding to the first candidate syndrome types in each order are determined. It is known that the incidence rates of "spleen-qi deficiency syndrome", "kidney-yang deficiency syndrome", and "spleen-kidney yang deficiency syndrome" under "irritable bowel syndrome" are 23%, 17%, and 15% respectively. The weight of the first candidate syndrome type in the first order is 3, that is, the weight of "spleen-qi deficiency syndrome" is 3, and after weighting the composite syndrome element score (2 points), the second prediction score is 6 points. The weight of the first candidate syndrome type in the second order is 2, that is, the weight of "kidney-yang deficiency syndrome" is 2, and after weighting the composite syndrome element score (2 points), the second prediction score is 4 points. The weight of the first candidate syndrome type in the third order is 1, that is, the weight of "spleen-kidney yang deficiency syndrome" is 1, and after weighting the composite syndrome element score (3 points), the second prediction score is 3 points.

[0157] Subsequently, it is known that the preset score difference is 3 points, and the maximum value of all the second prediction scores is 6 points, so the score error range can be determined to be [6, 3]. It can be seen that the above three first candidate syndromes are all second candidate syndromes that fall into the score error range. At this time, it is believed that the second prediction scores of the three second candidate syndromes are not much different, so the number of target syndrome elements included in the three second candidate syndromes is given priority. Specifically, the syndrome with the highest syndrome element composite score is determined from the three second candidate syndromes, that is, "spleen and kidney yang deficiency syndrome" is the predicted syndrome. So far, the predicted disease "irritable bowel syndrome" and the predicted syndrome "spleen and kidney yang deficiency syndrome" predicted based on the user's diagnosis and treatment information are output.

[0158] As another example, for each first candidate syndrome type, a weight coefficient can be determined based on the incidence of the first candidate syndrome type under the predicted disease, and then the product of the weight coefficient and the syndrome factor composite score is determined as the second prediction score. Specifically, for the first candidate syndrome type "Spleen Qi Deficiency Syndrome", its incidence under "Irritable Bowel Syndrome" is 23%, and the corresponding weight coefficient can be determined to be 23, so the second prediction score is 46 points (23*2). For the first candidate syndrome type "Kidney Yang Deficiency Syndrome", its incidence under "Irritable Bowel Syndrome" is 17%, and the corresponding weight coefficient can be determined to be 17, so the second prediction score is 34 points (17*2). For the first candidate syndrome type "Spleen and Kidney Yang Deficiency Syndrome", its incidence under "Irritable Bowel Syndrome" is 15%, and the corresponding weight coefficient can be determined to be 15, so the second prediction score is 45 points (15*3).

[0159] Subsequently, it is known that the preset score difference is 10 points, and the maximum value of all second prediction scores is 46 points, so the score error range can be determined to be [46, 36]. It can be seen that "Spleen Qi Deficiency Syndrome" and "Spleen and Kidney Yang Deficiency Syndrome" are second candidate syndromes that fall into the score error range. At this time, it is believed that the second prediction scores of these two second candidate syndromes are not much different, so the number of target syndrome elements included in the two second candidate syndromes is given priority. Specifically, the syndrome with the highest syndrome element composite score is determined from all second candidate syndromes, that is, "Spleen and Kidney Yang Deficiency Syndrome" is the predicted syndrome. So far, the predicted disease "Irritable Bowel Syndrome" and the predicted syndrome "Spleen and Kidney Yang Deficiency Syndrome" predicted based on the user's diagnosis and treatment information are output.

[0160] In addition, in addition to the first prediction strategy described in any of the above embodiments, in some embodiments, the target prediction strategy also includes a second prediction strategy. Figure 8 Steps 810 - 830 are shown.

[0161] Step 810: Obtain user diagnosis information, wherein the user diagnosis information includes one or more of tongue diagnosis information, face diagnosis information, pulse diagnosis information and palm diagnosis information.

[0162] For the description of the user's diagnosis and treatment information, please refer to the above text and will not be elaborated here.

[0163] Step 820: Determine at least one predicted syndrome type corresponding to the user's diagnosis and treatment information according to the pre-established corresponding relationship between the diagnosis and treatment information and the syndrome type.

[0164] Before executing the user diagnosis and treatment information processing method, the corresponding relationship between the diagnosis and treatment information and the syndrome type can be pre-established by a team of traditional Chinese medicine experts in advance. For example, the corresponding relationship between the diagnosis and treatment information and the syndrome type can be that "the tongue is pale red, the tongue coating is thin, white and slippery, the tongue body is enlarged, and there are tooth marks on the sides" corresponds to "syndrome of yang deficiency of spleen and kidney"; "the tip of the tongue is red with prickles" corresponds to "syndrome of exuberant heart fire"; "the sublingual collaterals are thick, tortuous and varicose" corresponds to "syndrome of qi stagnation and blood stasis", and so on. In this way, after obtaining the user's diagnosis and treatment information, one or more user diagnosis and treatment information can be used to map out the corresponding predicted syndrome type.

[0165] Step 830: Determine the predicted disease corresponding to the predicted syndrome type, where the incidence rate of the predicted syndrome type under the predicted disease meets a preset high-incidence condition.

[0166] After predicting the predicted syndrome type that the user may suffer from, the disease that the user may suffer from can be further predicted. Exemplarily, since each disease has a corresponding high-incidence syndrome type, that is, the incidence rate described above, after determining the predicted syndrome type, by looking up which diseases the predicted syndrome type is the high-incidence syndrome type of, the predicted disease can be deduced backwards. That is, the incidence rate of the predicted syndrome type under the predicted disease meets a preset high-incidence condition. The high-incidence condition includes, for example, that the incidence rate of the predicted syndrome type under the predicted disease is higher than a preset threshold. The high-incidence condition also includes, for example, that the predicted syndrome type is the syndrome type with the highest incidence rate among all the syndrome types included in the predicted disease. It can be understood that some syndrome types may correspond to multiple diseases that all meet the high-incidence condition, then multiple predicted diseases may be found, and multiple predicted diseases can be output.

[0167] It can be seen that a user diagnosis and treatment information processing method provided in this embodiment realizes the prediction of diseases and syndrome types through the prediction link of "tongue, face, pulse, palm - syndrome type - disease". Compared with the first prediction strategy, the second prediction strategy omits the matching process of syndrome elements, directly establishes the corresponding relationship between the diagnosis and treatment information and the syndrome type, and then inversely deduces and predicts one or more diseases based on the high-incidence syndrome type. Since there is no need to match syndrome elements, the prediction efficiency can be improved. Similarly, the entire prediction process does not rely on the obvious physical symptoms of the user, and the technical effect of accurately predicting the disease syndrome type before the user shows obvious physical symptoms can be achieved.

[0168] In the actual prediction process, the first prediction strategy or the second prediction strategy can be selected for predicting diseases and syndrome types. Or, the first prediction strategy and the second prediction strategy can be adopted simultaneously for predicting diseases and syndrome types. If the predicted diseases and predicted syndrome types output by the two strategies are consistent, it indicates that the reliability of the prediction result is relatively high. If the predicted diseases or predicted syndrome types output by the two strategies are inconsistent, it can be known that the output is sent to the traditional Chinese medicine expert team for discrimination. It is emphasized again that all the steps recorded in any of the above embodiments are executed by an electronic device with computing capabilities such as a computer. And regardless of the prediction strategy used, the prediction result should not interfere with the final diagnosis result of the physician, that is, the physician can discard the prediction result and give a different diagnosis result of the disease syndrome type, and the prediction result only serves as a reference and auxiliary role.

[0169] Based on a user diagnosis and treatment information processing method described in any of the above embodiments, the present application also provides a computer program product, which includes one or more computer programs or instructions. The computer program or instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. When the computer program is executed by a processor, it implements a user diagnosis and treatment information processing method described in any of the above embodiments.

[0170] Based on a user diagnosis and treatment information processing method described in any of the above embodiments, the present application also provides a user diagnosis and treatment information processing device, which is used to obtain a predicted disease and a predicted syndrome type through a target prediction strategy based on user diagnosis and treatment information; wherein, the user diagnosis and treatment information includes one or more of tongue diagnosis information, face diagnosis information, pulse diagnosis information, and palm diagnosis information; the target prediction strategy includes a first prediction strategy, and the device specifically includes:

[0171] A syndrome element module, configured to determine a plurality of target syndrome elements corresponding to the user diagnosis and treatment information according to a pre-established correspondence between diagnosis and treatment information and syndrome elements;

[0172] A disease module, configured to determine a predicted disease that matches at least part of the target syndrome elements according to a pre-established correspondence between diseases and syndrome elements;

[0173] A syndrome type module, configured to determine a predicted syndrome type that includes at least part of the target syndrome elements from all the syndrome types included in the predicted disease; wherein, the predicted disease and the predicted syndrome type are used to provide auxiliary reference information to the physician.

[0174] In some embodiments, the disease module is specifically configured to: determine a first candidate disease that includes one or more of the target syndrome elements according to the correspondence between the diseases and the syndrome elements;

[0175] According to the first quantity of target syndrome elements included in each of the first candidate diseases, a preset number of second candidate diseases are sequentially determined from the multiple first candidate diseases in descending order of the first quantity;

[0176] For each of the second candidate diseases, according to the scores of the target syndrome elements included in the second candidate disease, determine the syndrome element matching score of the second candidate disease;

[0177] For each of the second candidate diseases, based on the incidence rate of the second candidate disease, perform a weighted process on the corresponding syndrome element matching score to obtain a first prediction score; wherein, the incidence rate is positively correlated with the first prediction score;

[0178] According to the first prediction scores of all the second candidate diseases, determine the predicted disease. In some embodiments, there are multiple pieces of the user diagnosis and treatment information, and different pieces of the user diagnosis and treatment information correspond to the same target syndrome element or different target syndrome elements;

[0179] The syndrome element module is specifically configured to: determine the target syndrome elements respectively corresponding to each piece of the user diagnosis and treatment information to obtain a plurality of target syndrome elements;

[0180] For each of the target syndrome elements, count the occurrence times of the target syndrome element, and determine the score of the target syndrome element based on the occurrence times; wherein, the occurrence times are positively correlated with the score.

[0181] In some embodiments, the user diagnosis and treatment information carries user attribute information; the user attribute information includes one or more of gender information, age information, and geographical location information;

[0182] The disease module is further configured to: based on the user attribute information carried by the user diagnosis and treatment information, determine the incidence rate of each of the second candidate diseases.

[0183] In some embodiments, the syndrome type module is specifically configured to: determine a first syndrome type obtained by combining multiple target syndrome elements to obtain a first syndrome type set including all the first syndrome types, and determine the second syndrome type included in the predicted disease to obtain a second syndrome type set including all the second syndrome types;

[0184] According to the intersection of the first syndrome type set and the second syndrome type set, determine a first candidate syndrome type;

[0185] For each first candidate syndrome type, determine the syndrome element composite score of the first candidate syndrome type relative to a plurality of the target syndrome elements, and use the incidence rate of the first candidate syndrome type in the predicted disease to weight the syndrome element composite score to obtain a second prediction score; wherein, the syndrome element compliance score of the first candidate syndrome type is positively correlated with the second quantity of the target syndrome elements included in the first candidate syndrome type;

[0186] Based on the second prediction scores of all the first candidate syndrome types, determine the predicted syndrome type.

[0187] In some embodiments, the syndrome type module is specifically configured to: determine a scoring error range based on the maximum value among all the second prediction scores and a preset scoring difference;

[0188] From all the first candidate syndrome types, determine second candidate syndrome types that fall within the scoring error range;

[0189] Determine the second candidate syndrome type with the highest syndrome element composite score among all the second candidate syndrome types as the predicted syndrome type.

[0190] In some embodiments, the target prediction strategy includes a second prediction strategy,

[0191] The syndrome type module is further configured to: determine the predicted syndrome type corresponding to at least one piece of diagnosis and treatment information of the user according to the correspondence relationship established in advance between the diagnosis and treatment information and the syndrome type;

[0192] The disease module is further configured to: determine the predicted disease corresponding to the predicted syndrome type, wherein the incidence rate of the predicted syndrome type in the predicted disease satisfies a preset high-incidence condition.

[0193] Based on the method for processing user diagnosis and treatment information according to any of the above embodiments, the present application further provides a structural schematic diagram of an electronic device as shown in Figure 9 As shown in Figure 9 , at the hardware level, the electronic device includes a processor, an internal bus, a network interface, a memory, and a non-volatile memory. Of course, there may also be other hardware required for other services. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to implement the method for processing user diagnosis and treatment information according to any of the above embodiments.

[0194] The present application further provides a computer storage medium, and the storage medium stores a computer program. When the computer program is executed by a processor, it can be used to execute the method for processing user diagnosis and treatment information according to any of the above embodiments.

[0195] In several embodiments provided in this application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architectures, functions, and operations of devices, methods, and computer program products according to multiple embodiments of this application. In this regard, each block in the flowchart or block diagram can represent a module, a program segment, or a part of code, and the module, program segment, or part of 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 blocks may occur in a different order than that marked in the accompanying drawings. For example, two consecutive blocks can actually be executed substantially in parallel, and they can sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, as well as the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system that performs the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0196] In addition, each functional module in various embodiments of this application can be integrated together to form an independent part, or each module can exist alone, or two or more modules can be integrated to form an independent part.

[0197] If the described function is implemented in the form of a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application. The aforementioned storage medium includes: USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs, etc., which can store program codes.

[0198] The above are only embodiments of the present application and are not intended to limit the protection scope of the present application. For those skilled in the art, various changes and modifications can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the protection scope of the present application. It should be noted that similar reference numerals and letters indicate similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0199] As described above, the above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present application, and all should be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

[0200] It should be noted that in this text, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article or device including the said element.

Claims

1. A method for processing user diagnosis and treatment information, characterized in that: The method comprises: obtaining predicted diseases and predicted syndromes based on user diagnosis and treatment information through a target prediction strategy; wherein the user diagnosis and treatment information comprises one or more of tongue diagnosis information, face diagnosis information, pulse diagnosis information and palm diagnosis information; the target prediction strategy comprises a first prediction strategy, and the first prediction strategy comprises: Determine a plurality of target syndromes corresponding to the user's diagnosis and treatment information according to a pre-established correspondence relationship between diagnosis and treatment information and syndromes; Determining predicted diseases that match at least part of the target syndrome according to a pre-established correspondence between the disease and the syndrome; From all syndrome types included in the predicted disease, a predicted syndrome type including at least part of the target syndrome is determined; wherein the predicted disease and the predicted syndrome type are used to provide auxiliary reference information to doctors.

2. The method according to claim 1, characterized in that The step of determining a predicted disease matching at least part of the target syndrome according to the pre-established correspondence between the disease and syndrome factors comprises: Determining a first candidate disease including one or more target syndromes according to the correspondence between the disease and syndromes; According to the first number of target syndromes included in each of the first candidate diseases, determining a preset number of second candidate diseases from a plurality of the first candidate diseases in descending order of the first number; For each of the second candidate diseases, determining a syndrome matching score for the second candidate disease according to the score of the target syndrome included in the second candidate disease; For each of the second candidate diseases, weighting the corresponding syndrome-factor matching score based on the incidence rate of the second candidate disease to obtain a first prediction score; wherein the incidence rate is positively correlated with the first prediction score; The predicted disease is determined according to the first prediction scores of all the second candidate diseases.

3. The method according to claim 2, characterized in that The user diagnosis and treatment information includes a plurality of pieces, and different pieces of the user diagnosis and treatment information correspond to the same target syndrome or correspond to different target syndromes; before determining the syndrome matching score of the second candidate disease according to the score of the target syndrome included in the second candidate disease, the method further includes: Determine the target syndrome element corresponding to each piece of the user diagnosis and treatment information, and obtain multiple target syndrome elements; For each of the target tokens, the number of occurrences of the target token is counted, and the score of the target token is determined based on the number of occurrences; wherein the number of occurrences is positively correlated with the score.

4. The method according to claim 2 or 3, characterized in that: The user diagnosis and treatment information carries user attribute information; the user attribute information includes one or more of gender information, age information, and geographical location information; before weighting the corresponding syndrome matching score based on the incidence rate of the second candidate disease, the method further includes: Based on the user attribute information carried in the user diagnosis and treatment information, the incidence rate of each of the second candidate diseases is determined.

5. The method according to claim 1, characterized in that Determining a predicted syndrome type including at least part of the target syndrome factor from all syndrome types included in the predicted disease comprises: Determine a first syndrome type obtained by combining multiple target syndrome factors to obtain a first syndrome type set including all of the first syndrome types, and determine a second syndrome type included in the predicted disease to obtain a second syndrome type set including all of the second syndrome types; Determining a first candidate syndrome type according to the intersection of the first syndrome type set and the second syndrome type set; For each first candidate syndrome type, determine the syndrome composite score of the first candidate syndrome type relative to the plurality of target syndromes, and weight the syndrome composite score using the incidence of the first candidate syndrome type under the predicted disease to obtain a second prediction score; wherein the syndrome conformity score of the first candidate syndrome type is positively correlated with the second number of target syndromes included in the first candidate syndrome type; The predicted syndrome type is determined based on the second prediction scores of all of the first candidate syndrome types.

6. The method according to claim 5, characterized in that The step of determining the predicted syndrome type based on the second prediction scores of all the first candidate syndrome types comprises: Determine a score error range based on a maximum value of all second prediction scores and a preset score difference; Determine, from all first candidate syndromes, a second candidate syndrome that falls within the scoring error range; The second candidate syndrome type with the highest syndrome factor composite score is determined from all the second candidate syndrome types as the predicted syndrome type.

7. The method according to claim 1, characterized in that The target prediction strategy includes a second prediction strategy, and the second prediction strategy includes: Determining a predicted syndrome type corresponding to at least one piece of diagnosis and treatment information of the user according to a pre-established correspondence relationship between diagnosis and treatment information and syndrome types; The predicted disease corresponding to the predicted syndrome type is determined, wherein the incidence rate of the predicted syndrome type under the predicted disease meets a preset high incidence condition.

8. A user diagnosis and treatment information processing device, characterized in that: The device is used to obtain predicted diseases and predicted syndromes based on user diagnosis and treatment information through a target prediction strategy; wherein the user diagnosis and treatment information includes one or more of tongue diagnosis information, face diagnosis information, pulse diagnosis information and palm diagnosis information; the target prediction strategy includes a first prediction strategy, and the device specifically includes: A certificate element module, used for determining a plurality of target certificate elements corresponding to the user's diagnosis and treatment information according to a pre-established correspondence relationship between diagnosis and treatment information and certificate elements; A disease module, used to determine a predicted disease matching at least part of the target syndrome according to a pre-established correspondence between the disease and syndrome; The syndrome type module is used to determine the predicted syndrome type including at least part of the target syndrome from all the syndrome types included in the predicted disease; wherein the predicted disease and the predicted syndrome type are used to provide auxiliary reference information to the physician.

9. An electronic device, characterized in that: The electronic device comprises: processor; a memory for storing processor-executable instructions; Wherein, when the processor calls the executable instructions, the operation of any method described in claims 1-7 is implemented.

10. A computer-readable storage medium, characterized in that: Computer instructions are stored thereon, and when the computer instructions are executed by a processor, the steps of any method described in claims 1-7 are implemented.

Images