Tongue diagnosis auxiliary intelligent analysis system and method based on big data

Abstract

The invention relates to the technical field of tongue diagnosis auxiliary analysis, in particular to a tongue diagnosis auxiliary intelligent analysis system and method based on big data. A focus risk early warning analysis module in the system carries out focus relevance matching on tongue image features at different times in a health record of a corresponding patient according to a constructed tongue image-focus relevance data model; and generating a focus risk fluctuation map of the corresponding patient based on the time sequence information, and predicting concurrent focus risk early warning information of the patient at the current time based on the latest medical diagnosis information of the patient. According to the method, the reference standards of the tongue picture clustering features in the tongue picture associated data corresponding to the lesions are continuously optimized based on the clustering analysis mode, so that the precision of the constructed tongue picture-lesion associated analysis data model is ensured; meanwhile, in the tongue diagnosis auxiliary analysis process, medical diagnosis information of the patient is considered, effective prediction of the focus risk of the patient is achieved, and the accuracy of the tongue diagnosis auxiliary analysis result for the patient is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of tongue diagnosis auxiliary analysis, and specifically to a tongue diagnosis auxiliary intelligent analysis system and method based on big data. Background Art

[0002] Tongue diagnosis is an important method for Chinese medicine practitioners to judge the health status and disease type of patients by observing the color, shape, and coating of the tongue. However, traditional tongue diagnosis relies on the physician's experience and subjective judgment, and has the following problems:

[0003] Highly subjective: Different doctors may have different judgments on the same tongue image, which affects the accuracy and consistency of diagnosis;

[0004] Dependence on experience: High-level tongue diagnosis requires long-term accumulation, and young doctors may not have sufficient diagnostic capabilities;

[0005] Inconvenient data management: Traditional tongue diagnosis lacks systematic data recording and analysis tools, making it difficult to conduct large-scale case research and tracking.

[0006] With the rapid development of image analysis technology, tongue diagnosis instruments have gradually become an important auxiliary tool for Chinese medicine practitioners in disease diagnosis. The existing tongue diagnosis instruments have low analysis accuracy for the lesion analysis model of tongue diagnosis images, and in the process of tongue diagnosis auxiliary analysis, they cannot combine the patient's historical lesion diagnosis situation to achieve accurate prediction of the existing tongue diagnosis results. Therefore, the existing technology has major defects. Summary of the invention

[0007] The purpose of the present invention is to provide a tongue diagnosis auxiliary intelligent analysis system and method based on big data to solve the problems raised in the above background technology.

[0008] In order to solve the above technical problems, the present invention provides the following technical solution: a tongue diagnosis auxiliary intelligent analysis method based on big data, the method comprising the following steps:

[0009] S1. Obtain the tongue diagnosis image of the patient through the tongue diagnosis instrument, extract the tongue image features of the tongue diagnosis image collected by the tongue diagnosis instrument based on the image recognition technology; and combine the doctor's disease diagnosis information of the patient belonging to the corresponding tongue diagnosis image in the historical data, and construct a tongue image-lesion association data model through cluster analysis technology;

[0010] S2. Based on the time series analysis technology, the tongue image characteristics corresponding to the same patient at different times are summarized with the binding results of the corresponding multivariate detection information to construct the health record of the corresponding patient;

[0011] S3. Match the tongue image characteristics at different times in the health records of the corresponding patients according to the constructed tongue image-lesion association data model to generate a lesion risk fluctuation map of the corresponding patients based on time series information, and predict the concurrent lesion risk warning information of the patients at the current time based on the patients' most recent medical diagnosis information;

[0012] S4. Feedback the concurrent lesion risk warning information of the patient at the current time to the diagnosing doctor of the corresponding tongue diagnosis image at the current time, to assist the doctor in making diagnosis decisions for the patient.

[0013] Furthermore, in the process of extracting tongue features from the tongue diagnosis image collected by the tongue diagnosis instrument based on the image recognition technology, the S1 separates the tongue coating from the tongue quality by the chromaticity threshold method, and marks the tongue coating area and the tongue quality area in the tongue diagnosis image in different ways; the chromaticity threshold method identifies the tongue coating area and the tongue quality area in the tongue diagnosis image by querying the tongue coating color preset in the database;

[0014] The tongue image characteristics include tongue coating color, tongue quality color and tongue coating texture characteristics;

[0015] The tongue coating color represents the average value of the grayscale values ​​corresponding to each pixel point in the tongue coating area; the tongue quality color represents the average value of the grayscale values ​​corresponding to each pixel point in the tongue quality area; the tongue coating texture feature represents the texture pattern composed of each tongue coating texture node and the corresponding tongue coating texture node in the tongue coating area; the grayscale value corresponding to each pixel point in the tongue coating area is obtained, and the pixel points whose corresponding grayscale values ​​in the tongue coating area belong to the tongue coating texture grayscale threshold interval preset in the database are summarized and recorded as the tongue coating texture node candidate set; the grayscale value of each pixel point in the n×n pixel point area around each pixel point in the tongue coating area is obtained, and if the grayscale value of the center point pixel in the corresponding n×n pixel point area is less than or equal to the preset threshold value, the center point pixel in the corresponding n×n pixel point area is used as an element in the tongue coating texture node comparison set; each pixel point in the tongue coating texture node candidate set that does not belong to the tongue coating texture node comparison set is used as a tongue coating texture node.

[0016] In the process of acquiring tongue image features, the present invention performs analysis from three perspectives: tongue coating color, tongue quality color and tongue coating texture features, thereby providing data support for the subsequent acquisition of the first tongue image cluster features and the second tongue image cluster features corresponding to each lesion; when acquiring the tongue coating texture features, a tongue coating texture node comparison set is obtained in order to screen the elements in the tongue coating texture node alternative set, and the elements in the tongue coating texture node comparison set are used as interference items of the tongue coating texture nodes.

[0017] Furthermore, in the process of constructing the tongue image-lesion association analysis data model by cluster analysis technology in S1, the doctor's disease diagnosis information for the patient to whom the corresponding tongue diagnosis image belongs in the historical data is obtained, and the patient's tongue diagnosis image is bound to the corresponding disease diagnosis information, the disease diagnosis information represents the diagnosis disease set of the patient by the corresponding doctor based on the tongue diagnosis image, and each element in the diagnosis disease set corresponds to a lesion; the tongue image features of the tongue diagnosis images corresponding to the same lesion in the historical data are summarized; the tongue image association data corresponding to the corresponding lesion type in the tongue image-lesion association data model is recorded as the tongue image clustering feature; the tongue image clustering feature includes the first tongue image clustering feature and the second tongue image clustering feature,

[0018] The steps for obtaining the first tongue image clustering feature are as follows:

[0019] S111, obtaining the grayscale difference between the tongue coating color and the tongue quality color corresponding to each tongue diagnosis image; summarizing the grayscale difference between the tongue coating color and the tongue quality color corresponding to each tongue diagnosis image corresponding to the same lesion in ascending order, and generating a first tongue image clustering feature analysis sequence of the corresponding lesion;

[0020] S112, calculating the average value of the grayscale difference corresponding to each element in the first tongue image clustering feature analysis sequence of the corresponding lesion, and recording it as the deviation mean of the corresponding lesion;

[0021] S113, calculating the standard deviation of the grayscale difference values ​​corresponding to each element in the first tongue image clustering feature analysis sequence of the corresponding lesion, recording it as the tongue image aggregation degree of the corresponding lesion, and comparing the tongue image aggregation degree of the corresponding lesion with a preset value,

[0022] If the tongue image aggregation degree of the corresponding lesion is less than or equal to the preset value, or the number of elements in the first tongue image clustering feature analysis sequence of the corresponding lesion is unique, the deviation mean of the corresponding lesion is transmitted to step S114;

[0023] If the tongue image aggregation degree of the corresponding lesion is greater than the preset value, the element with the largest absolute value of the difference between the corresponding grayscale difference and the deviation mean of the corresponding lesion in the first tongue image clustering feature analysis sequence of the corresponding lesion is removed to obtain a new first tongue image clustering feature analysis sequence of the corresponding lesion, and jump to step S112;

[0024] S114, using the received result as the first tongue image clustering feature corresponding to the corresponding lesion;

[0025] The steps for obtaining the second tongue image clustering feature are as follows:

[0026] S121, sequentially summarizing the tongue coating texture features corresponding to each tongue diagnosis image corresponding to the same lesion to generate a second tongue image clustering feature analysis sequence of the corresponding lesion;

[0027] S122, taking the first element in the second tongue image cluster feature analysis sequence of the corresponding lesion as the initial reference texture feature;

[0028] S123, obtaining the element corresponding to the current reference texture feature in the second tongue image cluster feature analysis sequence of the corresponding lesion and marking it, obtaining the first unmarked element in the second tongue image cluster feature analysis sequence of the corresponding lesion that is closest to the marked element, and recording it as a fusion element;

[0029] S124, adjusting the position of the fused element so that the overlap area between the adjusted fused element and the corresponding area of ​​the reference texture feature is maximized, saving the adjusted fused element position, and using the union area of ​​the adjusted fused element and the reference texture feature as a new reference texture feature;

[0030] If the fused element is the last element in the second tongue image cluster feature analysis sequence of the corresponding lesion, the new reference texture feature is transmitted to step S125;

[0031] If the fused element is not the last element in the second tongue image clustering feature analysis sequence of the corresponding lesion, the new reference texture feature is transmitted to step S123;

[0032] S125. Receive the reference texture features transmitted in step S124, analyze one by one the ratio of the number of overlapping areas between each pixel point position in the received reference texture features and the stored adjusted fusion element positions to the total number of elements in the second tongue image clustering feature analysis sequence of the corresponding lesion, and record it as the clustering index corresponding to the corresponding element in the received reference texture features; and record the set of pixel point positions in the received reference texture features whose corresponding clustering index is greater than the preset clustering index as the second tongue image clustering feature of the corresponding lesion.

[0033] In the present invention, when obtaining the first tongue image clustering feature and the second tongue image clustering feature in the tongue image clustering feature, the screening of different tongue image features is achieved through clustering; wherein, when obtaining the first tongue image clustering feature, the dynamic update of the elements in the first tongue image clustering feature analysis sequence of the lesion is achieved through iteration, thereby achieving continuous optimization of the deviation mean of the corresponding lesion, thereby ensuring the accuracy of the first tongue image clustering feature corresponding to the corresponding lesion.

[0034] Furthermore, in the binding results of the tongue image features corresponding to the same patient at different times and the corresponding multivariate detection information, the interval between the acquisition time of the tongue diagnosis image corresponding to the tongue image features and the detection time corresponding to the bound multivariate detection information is less than or equal to a preset value; the multivariate detection information includes a summary set of one or more corresponding detection results in various preset body detection items;

[0035] In the process of constructing the health record of the corresponding patient in S2, the acquisition time of the tongue diagnosis image corresponding to the tongue image features is used as a time reference, and the binding results of the tongue image features corresponding to each acquisition time, the corresponding multivariate detection information and the corresponding medical diagnosis information are used as a health record summary element. The health record summary elements are sorted and summarized in sequence according to the chronological order of the acquisition time to obtain the health record of the corresponding patient; the medical diagnosis information is the diagnosis result of the corresponding doctor on the type of patient lesions in the historical data based on the corresponding tongue image features and multivariate detection information.

[0036] Furthermore, in the process of generating the lesion risk fluctuation map of the corresponding patient based on the time series information in S3, the tongue image features at different times in the health file of the corresponding patient are obtained, and the grayscale difference between the tongue coating color and the tongue quality color in the tongue image features at time t in the health file of the corresponding patient is recorded as At, and the tongue coating texture features in the tongue image features at time t in the health file of the corresponding patient are recorded as Bt; (At, Bt) are respectively input into the tongue image-lesion association data model corresponding to each lesion, and the lesion risk value corresponding to the mth lesion based on (At, Bt) is obtained, which is recorded as F (m，t) , the calculation formula is as follows:

[0037] F (m，t) =P1 (m，t) +P2 (m，t)

[0038] Among them, P1 (m，t) represents the similarity between At and the first tongue image clustering feature in the tongue image-lesion association data model corresponding to the mth lesion; the P1 (m，t) =1-|At-AC m | / AC m , A.C. m P2 represents the first tongue image clustering feature in the tongue image-lesion association data model corresponding to the mth lesion; (m，t) P2 represents the similarity between Bt and the second tongue image clustering feature in the tongue image-lesion association data model corresponding to the mth lesion; (m，t) The value of is equal to the maximum value of the overlapping pixels between the area corresponding to Bt and the area corresponding to the second tongue clustering feature in the tongue image-lesion association data model of the mth lesion after the translation transformation of the area corresponding to Bt divided by the quotient of the total number of pixels in the area corresponding to Bt, and the total number of pixels in the area corresponding to Bt is greater than 0; if the total number of pixels in the area corresponding to Bt is equal to 0, then P2 is determined. (m，t) The value of is equal to 0;

[0039] Construct lesion risk association data pairs (t, F (m，t) ), and the corresponding (t, F (m，t)) are marked on the coordinate system corresponding to the time and the lesion risk value, and the adjacent marked points are connected in the coordinate system in chronological order. The resulting line graph is the m-th lesion risk fluctuation graph of the corresponding patient based on the time series information.

[0040] Furthermore, the concurrent lesion risk warning information of the patient at the current time in S3 includes a predicted set of concurrent lesions and a risk prediction value of each concurrent lesion;

[0041] Obtain the lesion diagnosis set in the patient's most recent medical diagnosis information, where each element in the lesion diagnosis set corresponds to a type of lesion diagnosed by a doctor; analyze the lesion matching evaluation value of the patient based on the mth lesion at the current time in combination with the patient's most recent medical diagnosis information, and the calculation formula is as follows:

[0042] Em=F (m，) ·(1+g·μ)

[0043] Where Em represents the lesion matching evaluation value of the patient based on the mth lesion at the current time; F (m，) represents the lesion risk value corresponding to the mth lesion based on the tongue diagnosis image of the patient at the current time; μ represents the preset conversion coefficient; g represents the weight coefficient. If the lesion diagnosis set in the patient's most recent medical diagnosis information contains the mth lesion, g = 1; otherwise, g = 0;

[0044] The predicted set of concurrent lesions represents a set of all lesion types whose corresponding values ​​are greater than or equal to the preset matching evaluation values ​​in each Em corresponding to different values ​​of m; the risk prediction value of each concurrent lesion is equal to the lesion matching evaluation value of the corresponding lesion type in the predicted set of concurrent lesions of the patient at the current time.

[0045] A tongue diagnosis auxiliary intelligent analysis system based on big data, the system includes the following modules:

[0046] A lesion association model building module, wherein the lesion association model building module obtains the tongue diagnosis image of the patient corresponding to the tongue diagnosis instrument, extracts the tongue image features of the tongue diagnosis image collected by the tongue diagnosis instrument based on the image recognition technology; and combines the doctor's disease diagnosis information of the patient belonging to the corresponding tongue diagnosis image in the historical data, and builds a tongue image-lesion association data model through the cluster analysis technology;

[0047] A patient file construction module, which, based on the time series analysis technology, aggregates the binding results of the tongue image characteristics corresponding to the same patient at different times and the corresponding multivariate detection information to construct a health file of the corresponding patient;

[0048] A lesion risk warning analysis module, which performs lesion correlation matching on the tongue image features at different times in the health records of the corresponding patient according to the constructed tongue image-lesion association data model, generates a lesion risk fluctuation diagram of the corresponding patient based on time series information, and predicts the concurrent lesion risk warning information of the patient at the current time based on the patient's most recent medical diagnosis information;

[0049] A feedback management module feeds back the concurrent lesion risk warning information of the patient at the current time to the diagnosing doctor of the corresponding tongue diagnosis image at the current time, to assist the doctor in making a diagnosis decision for the patient.

[0050] Furthermore, the lesion risk warning analysis module includes a lesion risk fluctuation map construction unit and a risk warning information generation unit.

[0051] The lesion risk fluctuation map construction unit performs lesion correlation matching on the tongue image features at different times in the health records of the corresponding patient according to the constructed tongue image-lesion correlation data model, and generates a lesion risk fluctuation map of the corresponding patient based on time series information;

[0052] The risk warning information generating unit predicts the concurrent lesion risk warning information of the patient at the current time based on the patient's most recent medical diagnosis information.

[0053] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention utilizes image recognition technology to extract the tongue image features of tongue diagnosis images in historical data, and continuously optimizes the reference standards of tongue image clustering features in tongue image associated data corresponding to each lesion based on cluster analysis, thereby ensuring the accuracy of the constructed tongue image-lesion association analysis data model; at the same time, in the process of tongue diagnosis auxiliary analysis, the patient's own medical diagnosis information is taken into account to achieve effective prediction of the patient's lesion risk, thereby ensuring the accuracy of the tongue diagnosis auxiliary analysis results for the patient. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0055] Figure 1 It is a structural schematic diagram of a tongue diagnosis auxiliary intelligent analysis system based on big data of the present invention;

[0056] Figure 2 It is a flow chart of a tongue diagnosis auxiliary intelligent analysis method based on big data of the present invention. DETAILED DESCRIPTION

[0057] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] See also Figure 1 The present invention provides a technical solution: a tongue diagnosis auxiliary intelligent analysis system based on big data, the system includes the following modules:

[0059] A lesion association model building module, wherein the lesion association model building module obtains the tongue diagnosis image of the patient corresponding to the tongue diagnosis instrument, extracts the tongue image features of the tongue diagnosis image collected by the tongue diagnosis instrument based on the image recognition technology; and combines the doctor's disease diagnosis information of the patient belonging to the corresponding tongue diagnosis image in the historical data, and builds a tongue image-lesion association data model through the cluster analysis technology;

[0060] A patient file construction module, which, based on the time series analysis technology, aggregates the binding results of the tongue image characteristics corresponding to the same patient at different times and the corresponding multivariate detection information to construct a health file of the corresponding patient;

[0061] A lesion risk warning analysis module, wherein the lesion risk warning analysis module comprises a lesion risk fluctuation map construction unit and a risk warning information generation unit,

[0062] The lesion risk fluctuation map construction unit performs lesion correlation matching on the tongue image features at different times in the health records of the corresponding patient according to the constructed tongue image-lesion correlation data model, and generates a lesion risk fluctuation map of the corresponding patient based on time series information;

[0063] The risk warning information generating unit predicts the concurrent lesion risk warning information of the patient at the current time based on the patient's most recent medical diagnosis information;

[0064] A feedback management module feeds back the concurrent lesion risk warning information of the patient at the current time to the diagnosing doctor of the corresponding tongue diagnosis image at the current time, to assist the doctor in making a diagnosis decision for the patient.

[0065] like Figure 2 As shown, a tongue diagnosis auxiliary intelligent analysis method based on big data, the method comprises the following steps:

[0066] S1. Obtain the tongue diagnosis image of the patient through the tongue diagnosis instrument, extract the tongue image features of the tongue diagnosis image collected by the tongue diagnosis instrument based on the image recognition technology; and combine the doctor's disease diagnosis information of the patient belonging to the corresponding tongue diagnosis image in the historical data, and construct a tongue image-lesion association data model through cluster analysis technology;

[0067] In the process of extracting tongue features from the tongue diagnosis image collected by the tongue diagnosis instrument based on the image recognition technology, the tongue coating area and the tongue quality area in the tongue diagnosis image are separated by the chromaticity threshold method, and the tongue coating area and the tongue quality area in the tongue diagnosis image are marked in different ways respectively; the chromaticity threshold method identifies the tongue coating area and the tongue quality area in the tongue diagnosis image respectively by querying the tongue coating color preset in the database;

[0068] The tongue image characteristics include tongue coating color, tongue quality color and tongue coating texture characteristics;

[0069] The tongue coating color represents the average value of the grayscale values ​​corresponding to each pixel point in the tongue coating area; the tongue quality color represents the average value of the grayscale values ​​corresponding to each pixel point in the tongue quality area; the tongue coating texture feature represents the texture pattern composed of each tongue coating texture node and the corresponding tongue coating texture node in the tongue coating area; the grayscale value corresponding to each pixel point in the tongue coating area is obtained, and the pixel points whose corresponding grayscale values ​​in the tongue coating area belong to the tongue coating texture grayscale threshold interval preset in the database are summarized and recorded as the tongue coating texture node candidate set; the grayscale value of each pixel point in the n×n pixel point area around each pixel point in the tongue coating area is obtained, and if the grayscale value of the center point pixel in the corresponding n×n pixel point area is less than or equal to the preset threshold value, the center point pixel in the corresponding n×n pixel point area is used as an element in the tongue coating texture node comparison set; each pixel point in the tongue coating texture node candidate set that does not belong to the tongue coating texture node comparison set is used as a tongue coating texture node.

[0070] In this embodiment, the elements in the default tongue coating texture node comparison set are interference pixels corresponding to the tongue coating texture nodes with relatively discrete distribution.

[0071] In the process of constructing the tongue image-lesion association analysis data model by cluster analysis technology in S1, the disease diagnosis information of the doctor for the patient to whom the corresponding tongue diagnosis image belongs in the historical data is obtained, and the tongue diagnosis image of the patient is bound to the corresponding disease diagnosis information, wherein the disease diagnosis information represents the diagnosis disease set of the patient by the corresponding doctor based on the tongue diagnosis image, and each element in the diagnosis disease set corresponds to a lesion; the tongue image features of the tongue diagnosis images corresponding to the same lesion in the historical data are summarized; the tongue image association data corresponding to the corresponding lesion type in the tongue image-lesion association data model is recorded as the tongue image clustering feature; the tongue image clustering feature includes the first tongue image clustering feature and the second tongue image clustering feature,

[0072] The steps for obtaining the first tongue image clustering feature are as follows:

[0073] S111, obtaining the grayscale difference between the tongue coating color and the tongue quality color corresponding to each tongue diagnosis image; summarizing the grayscale difference between the tongue coating color and the tongue quality color corresponding to each tongue diagnosis image corresponding to the same lesion in ascending order, and generating a first tongue image clustering feature analysis sequence of the corresponding lesion;

[0074] S112, calculating the average value of the grayscale difference corresponding to each element in the first tongue image clustering feature analysis sequence of the corresponding lesion, and recording it as the deviation mean of the corresponding lesion;

[0075] S113, calculating the standard deviation of the grayscale difference values ​​corresponding to each element in the first tongue image clustering feature analysis sequence of the corresponding lesion, recording it as the tongue image aggregation degree of the corresponding lesion, and comparing the tongue image aggregation degree of the corresponding lesion with a preset value,

[0076] If the tongue image aggregation degree of the corresponding lesion is less than or equal to the preset value, or the number of elements in the first tongue image clustering feature analysis sequence of the corresponding lesion is unique, the deviation mean of the corresponding lesion is transmitted to step S114;

[0077] If the tongue image aggregation degree of the corresponding lesion is greater than the preset value, the element with the largest absolute value of the difference between the corresponding grayscale difference and the deviation mean of the corresponding lesion in the first tongue image clustering feature analysis sequence of the corresponding lesion is removed to obtain a new first tongue image clustering feature analysis sequence of the corresponding lesion, and jump to step S112;

[0078] S114, using the received result as the first tongue image clustering feature corresponding to the corresponding lesion;

[0079] In this embodiment, if the first tongue image clustering feature analysis sequence of the lesion contains 4 elements, which are recorded as r1, r2, r3 and r4, and r1<r2<r3<r4;

[0080] If the mean deviation of the corresponding lesion is r5, and r1<r2<r3<r5<r4;

[0081] If the tongue image aggregation degree of the corresponding lesion is r6,

[0082] but

[0083] If r6 is greater than a preset value, and r6-r1>r4-r6, r1 in the first tongue image clustering feature analysis sequence of the lesion is deleted, and the elements in the new first tongue image clustering feature analysis sequence of the corresponding lesion are r2, r3 and r4;

[0084] The steps for obtaining the second tongue image clustering feature are as follows:

[0085] S121, sequentially summarizing the tongue coating texture features corresponding to each tongue diagnosis image corresponding to the same lesion to generate a second tongue image clustering feature analysis sequence of the corresponding lesion;

[0086] S122, taking the first element in the second tongue image cluster feature analysis sequence of the corresponding lesion as the initial reference texture feature;

[0087] S123, obtaining the element corresponding to the current reference texture feature in the second tongue image cluster feature analysis sequence of the corresponding lesion and marking it, obtaining the first unmarked element in the second tongue image cluster feature analysis sequence of the corresponding lesion that is closest to the marked element, and recording it as a fusion element;

[0088] S124, adjusting the position of the fused element so that the overlap area between the adjusted fused element and the corresponding area of ​​the reference texture feature is maximized, saving the adjusted fused element position, and using the union area of ​​the adjusted fused element and the reference texture feature as a new reference texture feature;

[0089] If the fused element is the last element in the second tongue image cluster feature analysis sequence of the corresponding lesion, the new reference texture feature is transmitted to step S125;

[0090] If the fused element is not the last element in the second tongue image clustering feature analysis sequence of the corresponding lesion, the new reference texture feature is transmitted to step S123;

[0091] S125. Receive the reference texture features transmitted in step S124, analyze one by one the ratio of the number of overlapping areas between each pixel point position in the received reference texture features and the stored adjusted fusion element positions to the total number of elements in the second tongue image clustering feature analysis sequence of the corresponding lesion, and record it as the clustering index corresponding to the corresponding element in the received reference texture features; and record the set of pixel point positions in the received reference texture features whose corresponding clustering index is greater than the preset clustering index as the second tongue image clustering feature of the corresponding lesion.

[0092] S2. Based on the time series analysis technology, the tongue image characteristics corresponding to the same patient at different times are summarized with the binding results of the corresponding multivariate detection information to construct the health record of the corresponding patient;

[0093] In the binding results of the tongue image features corresponding to the same patient at different times and the corresponding multivariate detection information, the interval between the acquisition time of the tongue diagnosis image corresponding to the tongue image features and the detection time corresponding to the bound multivariate detection information is less than or equal to a preset value; the multivariate detection information includes a summary set of one or more corresponding detection results in various preset body detection items;

[0094] In the process of constructing the health record of the corresponding patient in S2, the acquisition time of the tongue diagnosis image corresponding to the tongue image features is used as a time reference, and the binding results of the tongue image features corresponding to each acquisition time, the corresponding multivariate detection information and the corresponding medical diagnosis information are used as a health record summary element. The health record summary elements are sorted and summarized in sequence according to the chronological order of the acquisition time to obtain the health record of the corresponding patient; the medical diagnosis information is the diagnosis result of the corresponding doctor on the type of patient lesions in the historical data based on the corresponding tongue image features and multivariate detection information.

[0095] S3. Match the tongue image characteristics at different times in the health records of the corresponding patients according to the constructed tongue image-lesion association data model to generate a lesion risk fluctuation map of the corresponding patients based on time series information, and predict the concurrent lesion risk warning information of the patients at the current time based on the patients' most recent medical diagnosis information;

[0096] In the process of generating the lesion risk fluctuation map of the corresponding patient based on the time series information in S3, the tongue image features at different times in the health file of the corresponding patient are obtained, and the grayscale difference between the tongue coating color and the tongue quality color in the tongue image features at time t in the health file of the corresponding patient is recorded as At, and the tongue coating texture features in the tongue image features at time t in the health file of the corresponding patient are recorded as Bt; (At, Bt) are respectively input into the tongue image-lesion association data model corresponding to each lesion, and the lesion risk value corresponding to the mth lesion based on (At, Bt) is obtained, which is recorded as F (m，t) , the calculation formula is as follows:

[0097] F (m，t) =P1 (m，t) +P2 (m，t)

[0098] Among them, P1 (m，t) represents the similarity between At and the first tongue image clustering feature in the tongue image-lesion association data model corresponding to the mth lesion; the P1 (m，t) =1-|At-AC m | / AC m , A.C. m P2 represents the first tongue image clustering feature in the tongue image-lesion association data model corresponding to the mth lesion; (m，t) P2 represents the similarity between Bt and the second tongue image clustering feature in the tongue image-lesion association data model corresponding to the mth lesion; (m，t) The value of is equal to the maximum value of the overlapping pixels between the area corresponding to Bt and the area corresponding to the second tongue clustering feature in the tongue image-lesion association data model of the mth lesion after the translation transformation of the area corresponding to Bt divided by the quotient of the total number of pixels in the area corresponding to Bt, and the total number of pixels in the area corresponding to Bt is greater than 0; if the total number of pixels in the area corresponding to Bt is equal to 0, then P2 is determined.(m，t) The value of is equal to 0;

[0099] Construct lesion risk association data pairs (t, F (m，t) ), and the corresponding (t, F (m，t) ) are marked on the coordinate system corresponding to the time and the lesion risk value, and the adjacent marked points are connected in the coordinate system in chronological order. The resulting line graph is the m-th lesion risk fluctuation graph of the corresponding patient based on the time series information.

[0100] In this embodiment, the coordinate system corresponding to time and lesion risk value is constructed with o as the origin, time as the x-axis and lesion risk value as the y-axis; the health records of the same patient have differences in lesion risk fluctuation graphs based on different lesions;

[0101] The concurrent lesion risk warning information of the patient at the current time in S3 includes a predicted set of concurrent lesions and a risk prediction value of each concurrent lesion;

[0102] Obtain the lesion diagnosis set in the patient's most recent medical diagnosis information, where each element in the lesion diagnosis set corresponds to a type of lesion diagnosed by a doctor; analyze the lesion matching evaluation value of the patient based on the mth lesion at the current time in combination with the patient's most recent medical diagnosis information, and the calculation formula is as follows:

[0103] Em=F (m，) ·(1+g·μ)

[0104] Where Em represents the lesion matching evaluation value of the patient based on the mth lesion at the current time; F (m，) represents the lesion risk value corresponding to the mth lesion based on the tongue diagnosis image of the patient at the current time; μ represents the preset conversion coefficient; g represents the weight coefficient. If the lesion diagnosis set in the patient's most recent medical diagnosis information contains the mth lesion, g = 1; otherwise, g = 0;

[0105] The predicted set of concurrent lesions represents a set of all lesion types whose corresponding values ​​are greater than or equal to the preset matching evaluation values ​​in each Em corresponding to different values ​​of m; the risk prediction value of each concurrent lesion is equal to the lesion matching evaluation value of the corresponding lesion type in the predicted set of concurrent lesions of the patient at the current time.

[0106] S4. Feedback the concurrent lesion risk warning information of the patient at the current time to the diagnosing doctor of the corresponding tongue diagnosis image at the current time, to assist the doctor in making diagnosis decisions for the patient.

[0107] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0108] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A tongue diagnosis auxiliary intelligent analysis method based on big data, characterized in that: The method comprises the following steps: S1. Obtain the tongue diagnosis image of the patient through the tongue diagnosis instrument, extract the tongue image features of the tongue diagnosis image collected by the tongue diagnosis instrument based on the image recognition technology; and combine the doctor's disease diagnosis information of the patient belonging to the corresponding tongue diagnosis image in the historical data, and construct a tongue image-lesion association data model through cluster analysis technology; S2. Based on the time series analysis technology, the tongue image characteristics corresponding to the same patient at different times are summarized with the binding results of the corresponding multivariate detection information to construct the health record of the corresponding patient; S3. Match the tongue image characteristics at different times in the health records of the corresponding patients according to the constructed tongue image-lesion association data model to generate a lesion risk fluctuation map of the corresponding patients based on time series information, and predict the concurrent lesion risk warning information of the patients at the current time based on the patients' most recent medical diagnosis information; S4. Feedback the concurrent lesion risk warning information of the patient at the current time to the diagnosing doctor of the corresponding tongue diagnosis image at the current time, to assist the doctor in making diagnosis decisions for the patient.

2. The tongue diagnosis auxiliary intelligent analysis method based on big data according to claim 1 is characterized by: In the process of extracting tongue features from the tongue diagnosis image collected by the tongue diagnosis instrument based on the image recognition technology, S1 separates the tongue coating from the tongue diagnosis image by the chromaticity threshold method, and marks the tongue coating area and the tongue quality area in the tongue diagnosis image in different ways; The chromaticity threshold method identifies the tongue coating area and tongue quality area in the tongue diagnosis image by querying the tongue coating color preset in the database; The tongue image characteristics include tongue coating color, tongue quality color and tongue coating texture characteristics; The tongue coating color represents the average value of the grayscale values ​​corresponding to each pixel point in the tongue coating area; the tongue quality color represents the average value of the grayscale values ​​corresponding to each pixel point in the tongue quality area; the tongue coating texture feature represents the texture pattern composed of each tongue coating texture node and the corresponding tongue coating texture node in the tongue coating area; the grayscale value corresponding to each pixel point in the tongue coating area is obtained, and the pixel points whose corresponding grayscale values ​​in the tongue coating area belong to the tongue coating texture grayscale threshold interval preset in the database are summarized and recorded as the tongue coating texture node candidate set; the grayscale value of each pixel point in the n×n pixel point area around each pixel point in the tongue coating area is obtained, and if the grayscale value of the center point pixel in the corresponding n×n pixel point area is less than or equal to the preset threshold value, the center point pixel in the corresponding n×n pixel point area is used as an element in the tongue coating texture node comparison set; each pixel point in the tongue coating texture node candidate set that does not belong to the tongue coating texture node comparison set is used as a tongue coating texture node.

3. The tongue diagnosis auxiliary intelligent analysis method based on big data according to claim 1 is characterized by: In the process of constructing the tongue image-lesion association analysis data model by cluster analysis technology in S1, the disease diagnosis information of the doctor for the patient to whom the corresponding tongue diagnosis image belongs in the historical data is obtained, and the tongue diagnosis image of the patient is bound to the corresponding disease diagnosis information, wherein the disease diagnosis information represents the diagnosis disease set of the patient by the corresponding doctor based on the tongue diagnosis image, and each element in the diagnosis disease set corresponds to a lesion; the tongue image features of the tongue diagnosis images corresponding to the same lesion in the historical data are summarized; the tongue image association data corresponding to the corresponding lesion type in the tongue image-lesion association data model is recorded as the tongue image clustering feature; the tongue image clustering feature includes the first tongue image clustering feature and the second tongue image clustering feature, The steps for obtaining the first tongue image clustering feature are as follows: S111, obtaining the grayscale difference between the tongue coating color and the tongue quality color corresponding to each tongue diagnosis image; summarizing the grayscale difference between the tongue coating color and the tongue quality color corresponding to each tongue diagnosis image corresponding to the same lesion in ascending order, and generating a first tongue image clustering feature analysis sequence of the corresponding lesion; S112, calculating the average value of the grayscale difference corresponding to each element in the first tongue image clustering feature analysis sequence of the corresponding lesion, and recording it as the deviation mean of the corresponding lesion; S113, calculating the standard deviation of the grayscale difference values ​​corresponding to each element in the first tongue image clustering feature analysis sequence of the corresponding lesion, recording it as the tongue image aggregation degree of the corresponding lesion, and comparing the tongue image aggregation degree of the corresponding lesion with a preset value, If the tongue image aggregation degree of the corresponding lesion is less than or equal to the preset value, or the number of elements in the first tongue image clustering feature analysis sequence of the corresponding lesion is unique, the deviation mean of the corresponding lesion is transmitted to step S114; If the tongue image aggregation degree of the corresponding lesion is greater than the preset value, the element with the largest absolute value of the difference between the corresponding grayscale difference and the deviation mean of the corresponding lesion in the first tongue image clustering feature analysis sequence of the corresponding lesion is removed to obtain a new first tongue image clustering feature analysis sequence of the corresponding lesion, and jump to step S112; S114, using the received result as the first tongue image clustering feature corresponding to the corresponding lesion; The steps for obtaining the second tongue image clustering feature are as follows: S121, sequentially summarizing the tongue coating texture features corresponding to each tongue diagnosis image corresponding to the same lesion to generate a second tongue image clustering feature analysis sequence of the corresponding lesion; S122, taking the first element in the second tongue image cluster feature analysis sequence of the corresponding lesion as the initial reference texture feature; S123, obtaining the element corresponding to the current reference texture feature in the second tongue image cluster feature analysis sequence of the corresponding lesion and marking it, obtaining the first unmarked element in the second tongue image cluster feature analysis sequence of the corresponding lesion that is closest to the marked element, and recording it as a fusion element; S124, adjusting the position of the fused element so that the overlap area between the adjusted fused element and the corresponding area of ​​the reference texture feature is maximized, saving the adjusted fused element position, and using the union area of ​​the adjusted fused element and the reference texture feature as a new reference texture feature; If the fused element is the last element in the second tongue image cluster feature analysis sequence of the corresponding lesion, the new reference texture feature is transmitted to step S125; If the fused element is not the last element in the second tongue image clustering feature analysis sequence of the corresponding lesion, the new reference texture feature is transmitted to step S123; S125. Receive the reference texture features transmitted in step S124, analyze one by one the ratio of the number of overlapping areas between each pixel point position in the received reference texture features and the stored adjusted fusion element positions to the total number of elements in the second tongue image clustering feature analysis sequence of the corresponding lesion, and record it as the clustering index corresponding to the corresponding element in the received reference texture features; and record the set of pixel point positions in the received reference texture features whose corresponding clustering index is greater than the preset clustering index as the second tongue image clustering feature of the corresponding lesion.

4. The tongue diagnosis auxiliary intelligent analysis method based on big data according to claim 2 is characterized by: In the binding results of the tongue image features corresponding to the same patient at different times and the corresponding multivariate detection information, the interval between the acquisition time of the tongue diagnosis image corresponding to the tongue image features and the detection time corresponding to the bound multivariate detection information is less than or equal to a preset value; the multivariate detection information includes a summary set of one or more corresponding detection results in various preset body detection items; In the process of constructing the health record of the corresponding patient in S2, the acquisition time of the tongue diagnosis image corresponding to the tongue image features is used as a time reference, and the binding results of the tongue image features corresponding to each acquisition time, the corresponding multivariate detection information and the corresponding medical diagnosis information are used as a health record summary element. The health record summary elements are sorted and summarized in sequence according to the chronological order of the acquisition time to obtain the health record of the corresponding patient; the medical diagnosis information is the diagnosis result of the corresponding doctor on the type of patient lesions in the historical data based on the corresponding tongue image features and multivariate detection information.

5. The tongue diagnosis auxiliary intelligent analysis method based on big data according to claim 4 is characterized by: In the process of generating the lesion risk fluctuation map of the corresponding patient based on the time series information in S3, the tongue image features at different times in the health file of the corresponding patient are obtained, and the grayscale difference between the tongue coating color and the tongue quality color in the tongue image features at time t in the health file of the corresponding patient is recorded as At, and the tongue coating texture features in the tongue image features at time t in the health file of the corresponding patient are recorded as Bt; (At, Bt) are respectively input into the tongue image-lesion association data model corresponding to each lesion, and the lesion risk value corresponding to the mth lesion based on (At, Bt) is obtained, which is recorded as F (m，t) , the calculation formula is as follows: F (m，t) =P1 (m，t) +P2 (m，t) Among them, P1 (m，t) represents the similarity between At and the first tongue image clustering feature in the tongue image-lesion association data model corresponding to the mth lesion; the P1 (m，t) =1-|At-AC m | / AC m , A.C. m P2 represents the first tongue image clustering feature in the tongue image-lesion association data model corresponding to the mth lesion; (m，t) P2 represents the similarity between Bt and the second tongue image clustering feature in the tongue image-lesion association data model corresponding to the mth lesion; (m，t) The value of is equal to the maximum value of the overlapping pixels between the area corresponding to Bt and the area corresponding to the second tongue clustering feature in the tongue image-lesion association data model of the mth lesion after the translation transformation of the area corresponding to Bt divided by the quotient of the total number of pixels in the area corresponding to Bt, and the total number of pixels in the area corresponding to Bt is greater than 0; if the total number of pixels in the area corresponding to Bt is equal to 0, then P2 is determined. (m，t) The value of is equal to 0; Construct lesion risk association data pairs (t, F (m，t) ), and the corresponding (t, F (m，t) ) are marked on the coordinate system corresponding to the time and the lesion risk value, and the adjacent marked points are connected in the coordinate system in chronological order. The resulting line graph is the m-th lesion risk fluctuation graph of the corresponding patient based on the time series information.

6. The tongue diagnosis auxiliary intelligent analysis method based on big data according to claim 5, characterized in that: The concurrent lesion risk warning information of the patient at the current time in S3 includes a predicted set of concurrent lesions and a risk prediction value of each concurrent lesion; Obtain the lesion diagnosis set in the patient's most recent medical diagnosis information, where each element in the lesion diagnosis set corresponds to a type of lesion diagnosed by a doctor; analyze the lesion matching evaluation value of the patient based on the mth lesion at the current time in combination with the patient's most recent medical diagnosis information, and the calculation formula is as follows: Em=F (m，) ·(1+g·μ) Where Em represents the lesion matching evaluation value of the patient based on the mth lesion at the current time; F (m，) represents the lesion risk value corresponding to the mth lesion based on the tongue diagnosis image of the patient at the current time; μ represents the preset conversion coefficient; g represents the weight coefficient. If the lesion diagnosis set in the patient's most recent medical diagnosis information contains the mth lesion, g = 1; otherwise, g = 0; The predicted set of concurrent lesions represents a set of all lesion types whose corresponding values ​​are greater than or equal to the preset matching evaluation values ​​in each Em corresponding to different values ​​of m; the risk prediction value of each concurrent lesion is equal to the lesion matching evaluation value of the corresponding lesion type in the predicted set of concurrent lesions of the patient at the current time.

7. A tongue diagnosis auxiliary intelligent analysis system based on big data, characterized in that: The system includes the following modules: A lesion association model building module, wherein the lesion association model building module obtains the tongue diagnosis image of the patient corresponding to the tongue diagnosis instrument, extracts the tongue image features of the tongue diagnosis image collected by the tongue diagnosis instrument based on the image recognition technology; and combines the doctor's disease diagnosis information of the patient belonging to the corresponding tongue diagnosis image in the historical data, and builds a tongue image-lesion association data model through the cluster analysis technology; A patient file construction module, which, based on the time series analysis technology, aggregates the binding results of the tongue image characteristics corresponding to the same patient at different times and the corresponding multivariate detection information to construct a health file of the corresponding patient; A lesion risk warning analysis module, which performs lesion correlation matching on the tongue image features at different times in the health records of the corresponding patient according to the constructed tongue image-lesion association data model, generates a lesion risk fluctuation diagram of the corresponding patient based on time series information, and predicts the concurrent lesion risk warning information of the patient at the current time based on the patient's most recent medical diagnosis information; A feedback management module feeds back the concurrent lesion risk warning information of the patient at the current time to the diagnosing doctor of the corresponding tongue diagnosis image at the current time, to assist the doctor in making a diagnosis decision for the patient.

8. The tongue diagnosis auxiliary intelligent analysis system based on big data according to claim 7 is characterized by: The lesion risk warning analysis module includes a lesion risk fluctuation map construction unit and a risk warning information generation unit. The lesion risk fluctuation map construction unit performs lesion correlation matching on the tongue image features at different times in the health records of the corresponding patient according to the constructed tongue image-lesion correlation data model, and generates a lesion risk fluctuation map of the corresponding patient based on time series information; The risk warning information generating unit predicts the concurrent lesion risk warning information of the patient at the current time based on the patient's most recent medical diagnosis information.

Images