Oral data processing method based on image processing

Abstract

The application provides an oral cavity data processing method based on image processing, comprising: in response to receiving an oral cavity diagnosis request sent by any user terminal, controlling an oral cavity collection plug-in to be loaded to the user terminal, so as to trigger the user terminal to collect images of the user's oral cavity based on the oral cavity collection plug-in, and obtain oral cavity images; extracting features from the oral cavity images, and determining a collection state corresponding to the oral cavity images based on obtained tooth features and tongue features; in response to the collection state being a qualified state, dividing the tongue features corresponding to different tongue parts based on the tooth features, and performing disease assessment based on each obtained tongue sub-feature, and obtaining an assessment result. The application at least improves diagnosis efficiency and convenience.

Description

Technical Field

[0001] This invention relates to data processing technology, and more particularly to a method for processing oral data based on image processing. Background Technology

[0002] Traditional Chinese medicine tongue diagnosis, a treasure of our nation's traditional medicine, infers the functional state of internal organs by observing the shape, color, and texture of the tongue, serving as an important basis for diagnosing oral and systemic diseases. With the increasing demand for health management and the widespread adoption of digital technology, combining traditional tongue diagnosis theory with modern image processing technology to achieve convenient diagnosis of oral diseases has become a trend, laying the technological foundation for providing users with a more efficient way to assess their health.

[0003] Currently, the diagnosis of oral diseases, especially tongue-based diagnoses, mainly relies on traditional offline methods: patients need to go to medical institutions where traditional Chinese medicine practitioners observe the tongue's characteristics with the naked eye and combine this with their experience to diagnose the condition, or use simple image acquisition devices to obtain tongue images for manual analysis. Although some medical scenarios have introduced digital auxiliary tools, these are mostly used within professional medical institutions and convenient remote diagnostic channels are not yet available to ordinary users. Online tongue diagnosis is still in the exploratory stage.

[0004] The aforementioned traditional offline model has significant limitations and cannot meet users' needs for online diagnosis of oral diseases based on tongue diagnosis. On the one hand, offline treatment requires users to spend time visiting medical institutions, which is greatly limited by location and time, reducing the convenience of diagnosis. On the other hand, the diagnostic process relies heavily on the personal experience of traditional Chinese medicine practitioners, and ordinary users cannot independently complete tongue image collection and feature interpretation. Furthermore, the lack of standardized online collection processes and automated analysis mechanisms makes it difficult for tongue diagnosis of oral diseases to break through the limitations of offline scenarios and achieve online, lightweight diagnostic services. Summary of the Invention

[0005] In view of the above problems, the present invention is proposed to provide an image processing-based oral data processing method that overcomes or at least partially solves the above problems.

[0006] According to one aspect of the present invention, an image processing-based oral cavity data processing method is provided, comprising:

[0007] In response to receiving an oral diagnosis request from any user terminal, the system controls the loading of the oral cavity acquisition plugin onto the user terminal, so as to trigger the user terminal to acquire images of the user's oral cavity based on the oral cavity acquisition plugin, thereby obtaining oral cavity images;

[0008] Feature extraction is performed on the oral cavity image, and the acquisition status of the corresponding oral cavity image is determined based on the obtained tooth and tongue features;

[0009] If the acquisition status is deemed satisfactory, the tongue features are divided into corresponding features for different tongue parts based on the tooth features, and a disease assessment is performed based on each obtained tongue sub-feature to obtain the assessment result.

[0010] Optionally, in the method according to the present invention, feature extraction is performed on the oral cavity image, and the acquisition state corresponding to the oral cavity image is determined based on the obtained tooth features and tongue features, including:

[0011] The oral cavity image is input into the first feature recognition model for image recognition, and the model responds by determining, based on the recognition result, that there is a tooth region indicating the tooth features. An image coordinate system corresponding to the oral cavity image is established with the image center point of the oral cavity image as the origin.

[0012] The coordinate points of each tooth constituting the tooth region are determined based on the image coordinate system.

[0013] If each tooth coordinate point is not located on the X-axis of the corresponding image coordinate system and is distributed on both sides of the X-axis, the tooth region is determined to have a uniform distribution attribute; otherwise, it is determined to have an off-distribution attribute.

[0014] The response has a uniform distribution attribute, and the oral cavity image is input into a second feature recognition model for image recognition;

[0015] The response determines the existence of a uvula region indicating uvula features based on the recognition results, performs morphological determination on the uvula region, and obtains the display morphology corresponding to the uvula region.

[0016] In response to the requirement that the display mode is a complete display, the oral cavity image is input into a third feature recognition model for image recognition, and the acquisition status of the corresponding oral cavity image is determined based on the obtained tongue region indicating tongue features and the tooth region.

[0017] Optionally, in the method according to the present invention, the oral cavity image is input into a first feature recognition model for image recognition, and in response to determining, based on the recognition result, that a tooth region indicating the tooth feature exists, an image coordinate system corresponding to the oral cavity image is established with the image center point corresponding to the oral cavity image as the origin, and then the method further includes:

[0018] The oral cavity image is sent to the user terminal;

[0019] The response receives the missing filling data sent by the user terminal based on the oral cavity image within a preset feedback period after the image transmission time, and determines the missing filling location of the corresponding oral cavity region based on the missing filling data;

[0020] Based on the tooth region, establish a vertical line of symmetry parallel to the Y-axis, and based on the vertical line of symmetry, determine a lateral symmetric position that has a vertical symmetric relationship with the missing filling position;

[0021] In response to the existence of a tooth sub-region corresponding to any tooth type at the lateral symmetrical position, the tooth sub-region is over-copied, and the resulting copied sub-region is filled into the missing filling position;

[0022] If there is no tooth sub-region corresponding to any tooth type at the lateral symmetrical position, the missing filling position is determined as the missing filling type, and the standard sub-region of the corresponding missing filling type located in the retrieved standard tooth distribution map is filled into the missing filling position.

[0023] Optionally, in the method according to the invention, the method further includes:

[0024] The response has a deviation distribution property. Based on the image coordinate system, the tooth coordinate point with the largest corresponding vertical coordinate value is determined as the maximum coordinate point, and the tooth coordinate point with the smallest corresponding vertical coordinate value is determined as the minimum coordinate point.

[0025] The difference between the maximum and minimum coordinate points is calculated based on the longitudinal coordinate values, and the resulting longitudinal difference of the teeth is then calculated as half to obtain a uniformly distributed value.

[0026] The vertical coordinate values ​​of the corresponding maximum and minimum coordinate points are processed by absolute value, and the difference between the first absolute value, the second absolute value and the uniform distribution value are calculated to obtain the first uniform difference and the second uniform difference.

[0027] The larger of the first uniform difference and the second uniform difference is determined as the adjusted distribution value, and the coordinate point corresponding to the adjusted distribution value is determined as the deviation coordinate point;

[0028] Determine the pointing coordinate point that has the same horizontal coordinate value as the origin and the same vertical coordinate value as the offset coordinate point, and generate an adjustment indicator line pointing to the center point of the image from the pointing coordinate point.

[0029] Retrieve the standard dental adjustment template and fill the adjustment distribution value into the adjustment indicator slot located on the standard dental adjustment template;

[0030] The system sends oral images to the user's device and responds to the user's request to receive them. Based on the oral image acquisition plugin, the system triggers the user's device to play a standard teeth adjustment template via voice.

[0031] Optionally, in the method according to the present invention, in response to determining, based on the recognition result, that a uvula region indicating uvular features exists, morphological determination is performed on the uvula region to obtain a display morphology corresponding to the uvula region, including:

[0032] Obtain the uvula contour corresponding to the uvula region, and determine the coordinate points of each uvula that make up the uvula contour based on the image coordinate system.

[0033] Determine the number of lateral coordinates for all uvula coordinate points corresponding to each identical longitudinal coordinate value, calculate the half value of the largest number of lateral coordinates, and multiply the obtained initial comparison number with the retrieved preset update coefficient to obtain the current comparison number;

[0034] The uvula coordinate point with the smallest corresponding vertical coordinate value is determined as the midpoint. Starting from the midpoint, along the uvula contour, the uvula coordinate points with corresponding horizontal coordinate values ​​greater than the midpoint and corresponding to the current comparison number are divided into the first trend comparison group. The uvula coordinate points with corresponding horizontal coordinate values ​​less than the midpoint and corresponding to the current comparison number are divided into the second trend comparison group.

[0035] The longitudinal trend corresponding to each uvula coordinate point located in the first trend comparison group and the second trend comparison group is determined;

[0036] If the vertical trend of the first trend comparison group is downward and the vertical trend of the second trend comparison group is upward, the display form of the corresponding uvula region will be determined as fully displayed; otherwise, it will be determined as partially displayed.

[0037] Optionally, in the method according to the present invention, determining the longitudinal trend of the corresponding trend comparison group based on the obtained longitudinal differences of each uvula includes:

[0038] The uvula coordinate point with the smallest corresponding horizontal coordinate value in the first trend comparison group and the uvula coordinate point with the largest corresponding horizontal coordinate value in the second trend comparison group are determined as the first end point and the second end point, respectively.

[0039] Based on the uvula contour, a first trend direction and a second trend direction are determined from the first end point and the second end point toward the middle point;

[0040] The longitudinal coordinate value corresponding to each uvula coordinate point in the first trend comparison group gradually decreases along the first trend direction, and the longitudinal trend of the first trend comparison group is determined to be a downward trend.

[0041] The longitudinal coordinate value corresponding to each uvula coordinate point in the second trend comparison group gradually increases along the direction of the second trend, thus determining the longitudinal trend of the second trend comparison group as an upward trend.

[0042] Optionally, in the method according to the invention, determining the acquisition state of the corresponding oral cavity image based on the obtained tongue region indicating tongue features and the tooth region includes:

[0043] Connect the coordinates of each tooth that makes up the tooth region based on their adjacent positions to obtain the upper tooth region and the lower tooth region included in the tooth region.

[0044] All tooth coordinate points corresponding to the same horizontal coordinate value that make up the upper tooth region are grouped into the same upper tooth vertical group, and all tooth coordinate points corresponding to the same horizontal coordinate value that make up the lower tooth region are grouped into the same lower tooth vertical group.

[0045] The tooth coordinate point with the smallest vertical coordinate value in each vertical group of upper teeth is determined as the upper tooth minimum point, and the upper tooth minimum point with the largest vertical coordinate value is determined as the upper tooth open / close point.

[0046] The tooth with the largest vertical coordinate value in each lower tooth vertical group is determined as the lower tooth maximum point, and the lower tooth maximum point with the smallest vertical coordinate value is determined as the lower tooth open / close point.

[0047] The difference between the vertical coordinate values ​​of the upper and lower teeth opening points is calculated, and the tooth opening value obtained is greater than the preset opening value, and each lower tooth maximum point is adjacent to any tongue coordinate point that makes up the tongue region. The collection state is then determined to be a qualified state.

[0048] Optionally, in the method according to the present invention, in response to the acquisition state being a qualified state, the tongue features are segmented according to different tongue parts based on the tooth features, and a disease assessment is performed based on each obtained tongue sub-feature to obtain an assessment result, including:

[0049] In response to the acquisition status being in a qualified state, each tooth sub-region of the corresponding anterior tooth type and each tooth sub-region of the corresponding posterior tooth type are determined based on the lower tooth region.

[0050] Each pair of adjacent tooth sub-regions is defined as the same tooth division group, and a vertical line of symmetry parallel to the Y-axis is established based on the tooth regions.

[0051] In response to any tooth division group including any tooth sub-region of the corresponding posterior tooth type, a horizontal dividing line is generated based on the tooth division group, and two horizontal dividing lines with vertical symmetry are merged based on the vertical symmetry line.

[0052] In response to any tooth division group including two tooth sub-regions corresponding to the anterior tooth type, a longitudinal dividing line is generated based on the tooth division group.

[0053] The tongue region is divided into regions based on all the horizontal and vertical dividing lines to obtain tongue sub-regions with different division numbers.

[0054] Based on the traversal of the retrieved preset partitioning template, all partitioning numbers corresponding to the same tongue sub-feature are merged based on the tongue sub-region, and disease assessment is performed based on the obtained disease sub-regions corresponding to different tongue sub-features to obtain the assessment results.

[0055] Optionally, in the method according to the invention, disease assessment is performed based on the obtained disease sub-regions corresponding to different tongue features to obtain assessment results, including:

[0056] The oral cavity acquisition plugin retrieves the initial report interface corresponding to each tongue feature and fills the symptom sub-area corresponding to the tongue feature into the symptom indicator sub-interface located in the initial report interface.

[0057] The disease sub-region is compared with the health indicator area filled in the health comparison sub-interface of the initial report interface, and the health assessment value of the tongue sub-feature is determined based on the comparison result.

[0058] If the health assessment value falls within any preset assessment range, the preset assessment content corresponding to the preset assessment range will be filled into the symptom assessment sub-interface located in the initial report interface; otherwise, the initial report interface will be sent to the medical staff terminal, and the real-time assessment content obtained from the medical staff terminal will be filled into the symptom assessment sub-interface to obtain the current report interface.

[0059] In response to the current report interface that retrieves all tongue features, the evaluation results integrated from all current report interfaces are sent to the user terminal.

[0060] According to another aspect of the present invention, an image processing-based oral cavity data processing system is provided, comprising:

[0061] The image acquisition module is configured to respond to receiving an oral diagnosis request sent by any user terminal, control the loading of the oral acquisition plugin onto the user terminal, and trigger the user terminal to acquire images of the user's oral cavity based on the oral acquisition plugin to obtain oral images;

[0062] The state determination module is configured to extract features from the oral cavity image and determine the acquisition state of the corresponding oral cavity image based on the obtained tooth and tongue features.

[0063] The symptom assessment module is configured to respond to the acquisition status as a qualified state, divide the tongue features according to the tooth features into features corresponding to different tongue parts, and perform symptom assessment based on each obtained tongue sub-feature to obtain the assessment result.

[0064] According to the present invention, an oral image acquisition plugin enables users to independently acquire oral images. Combined with dental features, standardized regional division and disease assessment of tongue features are performed, effectively solving the problems of traditional offline tongue diagnosis relying on professionals and the lack of online diagnostic processes for users. By using dental features to assist in judging the acquisition status, image quality and diagnostic accuracy are significantly improved, ensuring that the tongue images acquired by users meet diagnostic requirements. Based on traditional Chinese medicine tongue diagnosis theory, multi-regional feature division of the tongue is performed, realizing intelligent assessment from a single tongue image to diseases related to multiple organs. This breaks through the dependence of traditional tongue diagnosis on the experience of professional physicians, enabling ordinary users to obtain standardized and automated preliminary oral disease diagnosis services online, greatly improving diagnostic efficiency and convenience. Furthermore, the standardized acquisition and analysis process established by this invention provides reliable technical support for telemedicine and health management, and has broad application prospects. Attached Figure Description

[0065] Figure 1 A flowchart of an image processing-based oral data processing method according to an embodiment of the present invention is shown;

[0066] Figure 2 An image of the oral cavity in a qualified state is shown in this embodiment;

[0067] Figure 3 This diagram illustrates the division of the tongue region in this embodiment.

[0068] Figure 4 A structural block diagram of an image processing-based oral data processing system according to another embodiment of the present invention is shown. Detailed Implementation

[0069] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0070] To address the problems existing in the prior art, the inventors proposed the solution of this invention. One embodiment of this invention provides an image processing-based oral cavity data processing method, which can be executed in a computing device, wherein the computing device can be understood as a terminal with data processing capabilities, such as a mobile phone or a computer.

[0071] Figure 1 A flowchart of an image processing-based oral data processing method according to an embodiment of the present invention is shown, as follows: Figure 1As shown, the method proposed in this embodiment begins with step S1, which includes the following:

[0072] In response to any oral diagnosis request sent by a user terminal, the system controls the loading of the oral cavity acquisition plugin onto the user terminal, so as to trigger the user terminal to acquire images of the user's oral cavity based on the oral cavity acquisition plugin, thereby obtaining oral cavity images.

[0073] For example, in this embodiment, when a user wants to have a corresponding oral disease diagnosis online, they can use a corresponding client to send a corresponding oral diagnosis request. Here, the client can be a terminal with data processing and communication functions, such as a mobile phone or computer. When the server receives the oral diagnosis request, the server can control the oral image acquisition plugin to be loaded onto the client. It can be explained that by loading the oral image acquisition plugin onto the client, the client can be directly triggered to perform image acquisition operations on the user's oral cavity, and finally obtain oral images. This allows users to complete the acquisition of oral images without having to go to a physical medical institution, simply by initiating a request online and cooperating with the client. This greatly breaks the limitations of time and space, allowing users to conveniently start the first step of oral diagnosis, providing basic data support for subsequent online disease diagnosis, and improving the efficiency of corresponding disease diagnosis.

[0074] Here, since the user terminal is a mobile phone or computer, a corresponding image acquisition unit can be pre-set to perform image acquisition operations based on the image acquisition unit, which can be a camera.

[0075] Step S2 includes the following:

[0076] Feature extraction is performed on the oral cavity image, and the acquisition status of the corresponding oral cavity image is determined based on the obtained tooth and tongue features.

[0077] For example, in this embodiment, after acquiring the corresponding oral cavity image based on the user's completion of the corresponding oral cavity image acquisition operation, in order to determine whether the oral cavity image meets the corresponding diagnostic requirements, feature extraction can be performed on the acquired oral cavity image to extract tooth features and tongue features. By accurately extracting the features of these two core parts of the oral cavity, specific analysis objects can be provided for subsequent status judgment and disease assessment, avoiding deviations in online diagnosis due to missing features, and ensuring that the diagnosis has a reliable basis. Furthermore, based on the extracted tooth and tongue features, the acquisition status of the corresponding oral cavity image can be determined. That is, by using tooth and tongue features as the core basis for judging the acquisition status, the online oral cavity image acquired by the user can be comprehensively evaluated to determine whether it meets the diagnostic requirements, effectively filtering out unqualified images, avoiding the impact of inadequate image acquisition on the accuracy of online diagnosis, and allowing subsequent disease assessment to be carried out on the basis of qualified images, thereby improving the reliability of the user's online diagnostic results.

[0078] Furthermore, in this embodiment, the aforementioned "extracting features from the oral cavity image and determining the acquisition state of the corresponding oral cavity image based on the obtained tooth and tongue features" may further include the following steps:

[0079] The oral cavity image is input into the first feature recognition model for image recognition, and the model responds by determining, based on the recognition result, that there is a tooth region indicating the tooth features. An image coordinate system corresponding to the oral cavity image is established with the image center point of the oral cavity image as the origin.

[0080] The coordinate points of each tooth constituting the tooth region are determined based on the image coordinate system.

[0081] If each tooth coordinate point is not located on the X-axis of the corresponding image coordinate system and is distributed on both sides of the X-axis, the tooth region is determined to have a uniform distribution attribute; otherwise, it is determined to have an off-distribution attribute.

[0082] The response has a uniform distribution attribute, and the oral cavity image is input into a second feature recognition model for image recognition;

[0083] The response determines the existence of a uvula region indicating uvula features based on the recognition results, performs morphological determination on the uvula region, and obtains the display morphology corresponding to the uvula region.

[0084] In response to the requirement that the display mode is a complete display, the oral cavity image is input into a third feature recognition model for image recognition, and the acquisition status of the corresponding oral cavity image is determined based on the obtained tongue region indicating tongue features and the tooth region.

[0085] For example, in this embodiment, the specific process of extracting features from oral images and determining the acquisition status can be described as follows:

[0086] First, the server can input the oral cavity image into the first feature recognition model for image recognition. If the recognition result determines that there is a tooth region indicating tooth features, then the corresponding image coordinate system is established with the image center point of the oral cavity image as the origin. That is, the tooth region is accurately identified through the first feature recognition model, providing a clear target area for subsequent analysis. The establishment of the image coordinate system provides a unified standard for the position description of various features, avoiding the confusion caused by inconsistent position references in online diagnosis, and ensuring that feature analysis has a reliable spatial benchmark.

[0087] Next, based on the above image coordinate system, the coordinate points of each tooth that makes up the tooth region can be determined. It can be explained that by clarifying the position of each tooth coordinate point in the coordinate system, corresponding data support can be provided for the subsequent judgment of the tooth distribution state, so as to determine whether there is regional deviation in the tooth region.

[0088] The server can then determine whether each tooth coordinate point is not located on the X-axis of the image coordinate system and is distributed on both sides of the X-axis to determine whether there is a regional deviation problem in the tooth area. If the condition is met, the tooth area can be determined to have a uniform distribution attribute (i.e., there is no regional deviation problem). Otherwise, it is determined to have a deviation distribution attribute (i.e., there is a regional deviation problem). It can be noted that the distribution attribute judgment based on coordinate position can quickly filter out images of the tooth area that meet the standard, avoid the tooth area deviation caused by improper online shooting angle of the user, and ensure that the images entering the next stage meet the basic requirements in terms of tooth distribution.

[0089] Next, it can be explained that since the subsequent disease diagnosis needs to be based on the tongue region, it is necessary to ensure that the oral image obtained by the user based on the image acquisition operation can display a relatively complete tongue region. Generally speaking, the greater the opening degree of the corresponding oral cavity, the more complete the uvula region of the corresponding oral image should be. Conversely, the smaller the opening degree of the corresponding oral cavity, the more incomplete or even non-existent the uvula region of the corresponding oral image should be. Therefore, when it is determined that the tooth region has a uniform distribution attribute, the server can further input the oral image into the second feature recognition model for image recognition to determine whether there is a corresponding uvula region, so that the tongue region of the corresponding oral image can be displayed relatively completely based on the uvula region.

[0090] Finally, if the recognition results indicate the presence of a uvula region indicating uvular features, further morphological determination of this region is necessary to obtain its displayed form. Clearly defining the uvula's morphology allows for assessment of whether the online images clearly present this crucial structure, preventing incomplete visualization of the oral cavity due to an incomplete uvula display. When the uvula is fully displayed, the server can input the oral cavity image into a third-feature recognition model for image recognition. Based on the obtained tongue and tooth regions indicating tongue features, the acquisition status of the oral cavity image is determined. This demonstrates that accurately extracting the tongue region using the third-feature recognition model and combining it with the tooth region to determine the corresponding acquisition status comprehensively assesses whether the online oral cavity images cover the core features required for diagnosis, ensuring a comprehensive and accurate assessment of the acquisition status. This provides a qualified image foundation for subsequent online disease diagnosis, improving the reliability of the diagnostic results.

[0091] It can be noted that the first feature recognition model, the second feature recognition model and the third feature recognition model mentioned in this embodiment can all be obtained based on the neural network learning model and machine learning model that exist in the prior art. The specific details of the model establishment and training process are not described in this embodiment.

[0092] Furthermore, since oral images are obtained based on user self-taken photos, and in some cases, teeth in certain users' mouths may be incomplete, directly using images with missing information for diagnosis may affect the accuracy of the diagnostic results and reduce the reliability of online diagnosis. Therefore, to solve this technical problem, in this embodiment, the aforementioned "inputting the oral image into a first feature recognition model for image recognition, and responding by determining, based on the recognition results, the existence of a tooth region indicating the tooth feature, and establishing an image coordinate system corresponding to the oral image with the image center point as the origin" may further include the following steps:

[0093] The oral cavity image is sent to the user terminal;

[0094] The response receives the missing filling data sent by the user terminal based on the oral cavity image within a preset feedback period after the image transmission time, and determines the missing filling location of the corresponding oral cavity region based on the missing filling data;

[0095] Based on the tooth region, establish a vertical line of symmetry parallel to the Y-axis, and based on the vertical line of symmetry, determine a lateral symmetric position that has a vertical symmetric relationship with the missing filling position;

[0096] In response to the existence of a tooth sub-region corresponding to any tooth type at the lateral symmetrical position, the tooth sub-region is over-copied, and the resulting copied sub-region is filled into the missing filling position;

[0097] If there is no tooth sub-region corresponding to any tooth type at the lateral symmetrical position, the missing filling position is determined as the missing filling type, and the standard sub-region of the corresponding missing filling type located in the retrieved standard tooth distribution map is filled into the missing filling position.

[0098] For example, in this embodiment, to address the technical problem that directly using images of missing teeth in some users' mouths might affect the accuracy of the diagnostic results, the following technical solution can be adopted to perform the corresponding filling work:

[0099] First, the server can send oral images to the user's end, allowing the user to directly view the oral images they have collected online. This facilitates the timely detection of missing teeth in the images, providing a basis for subsequent feedback on missing information, increasing user participation in the diagnostic process, and ensuring that missing issues are reported promptly.

[0100] Next, the server can obtain the image transmission time when the oral cavity image is sent to the user's terminal, and simultaneously start timing based on the image transmission time. When the server receives the missing filling data based on the oral cavity image sent by the user's terminal within the preset feedback period after the image transmission time, it indicates that the user has checked and uploaded the image based on the specific situation of their missing teeth after viewing the oral cavity image. At this time, the server can determine the missing filling position of the corresponding oral cavity area based on the missing filling data. It can be noted that the setting of the preset feedback period not only ensures that the user has enough time to check the image and provide feedback, but also avoids the impact of feedback delay on the overall efficiency of online diagnosis. Accurately determining the missing filling position clarifies the target for subsequent filling operations, ensuring that the filling work is accurate and in place.

[0101] Then, the server can establish a vertical symmetry line parallel to the Y-axis based on the tooth area, and determine the lateral symmetry position that has a vertical symmetry relationship with the missing filling position based on the vertical symmetry line. It can be explained that this embodiment utilizes the physiological characteristic that the teeth in the oral cavity are usually symmetrically distributed, and finds the lateral symmetry position through the vertical symmetry line, which can provide a scientific basis for filling the missing area, making the filling area more in line with the actual structure of the oral cavity, and improving the rationality and accuracy of the filling.

[0102] When a tooth sub-region corresponding to any tooth type exists in a horizontally symmetrical position, the tooth sub-region is copied and the copied sub-region is used to fill the missing filling position. The filling method based on symmetrical position copying can maximize the consistency of the filling area with the surrounding teeth in terms of shape and structure, avoid image distortion due to improper filling, and ensure that the oral image information on which the online diagnosis is based is complete and true.

[0103] In addition, when there is no tooth sub-region corresponding to any tooth type at the lateral symmetrical position, it is necessary to determine the missing filling type based on the missing filling position, and fill the missing filling position with the standard sub-region of the corresponding missing filling type located in the retrieved standard tooth distribution map. That is, in the absence of symmetrical reference, the filling can be carried out by the standard sub-region in the standard tooth distribution map. This can ensure that the filled area conforms to the conventional morphology and structural characteristics of the tooth type, and ensure that the oral images collected online are complete in information while maintaining a high degree of standardization. This provides a reliable image basis for subsequent accurate diagnosis, allowing users to obtain high-quality oral disease diagnosis services online.

[0104] Furthermore, when acquiring oral images online, the tooth area is prone to deviation from its intended distribution due to improper shooting angles. Users often struggle to accurately adjust their shooting posture to obtain satisfactory images, leading to repeated acquisitions without obtaining acceptable images. This negatively impacts the efficiency and user experience of online diagnosis. Therefore, to address this issue, this embodiment may further include the following steps:

[0105] The response has a deviation distribution property. Based on the image coordinate system, the tooth coordinate point with the largest corresponding vertical coordinate value is determined as the maximum coordinate point, and the tooth coordinate point with the smallest corresponding vertical coordinate value is determined as the minimum coordinate point.

[0106] The difference between the maximum and minimum coordinate points is calculated based on the longitudinal coordinate values, and the resulting longitudinal difference of the teeth is then calculated as half to obtain a uniformly distributed value.

[0107] The vertical coordinate values ​​of the corresponding maximum and minimum coordinate points are processed by absolute value, and the difference between the first absolute value, the second absolute value and the uniform distribution value are calculated to obtain the first uniform difference and the second uniform difference.

[0108] The larger of the first uniform difference and the second uniform difference is determined as the adjusted distribution value, and the coordinate point corresponding to the adjusted distribution value is determined as the deviation coordinate point;

[0109] Determine the pointing coordinate point that has the same horizontal coordinate value as the origin and the same vertical coordinate value as the offset coordinate point, and generate an adjustment indicator line pointing to the center point of the image from the pointing coordinate point.

[0110] Retrieve the standard dental adjustment template and fill the adjustment distribution value into the adjustment indicator slot located on the standard dental adjustment template;

[0111] The system sends oral images to the user's device and responds to the user's request to receive them. Based on the oral image acquisition plugin, the system triggers the user's device to play a standard teeth adjustment template via voice.

[0112] For example, in this embodiment, based on the deviation distribution attribute, this embodiment can instruct the user to make another adjustment for shooting based on the following technical solution:

[0113] First, since the tooth region has an off-distribution attribute, the server can determine the tooth coordinate point with the largest corresponding vertical coordinate value as the maximum coordinate point and the tooth coordinate point with the smallest corresponding vertical coordinate value as the minimum coordinate point based on the image coordinate system. By clearly defining these two extreme coordinate points, the server can accurately locate the key position of the tooth region's deviation in the vertical distribution, providing specific data for subsequent quantitative analysis of the degree of deviation. This allows the online system to have a clear basis for judging the deviation of the tooth distribution and avoids vague qualitative descriptions.

[0114] Next, the server can calculate the difference between the maximum and minimum coordinate points based on the longitudinal coordinate values, and then perform a half-value calculation on the obtained longitudinal difference of the teeth to obtain a uniform distribution value. Through the calculation process, the overall range of the longitudinal distribution of teeth can be transformed into a benchmark for measuring uniformity, so that the analysis of the degree of deviation has a reference standard, and the online assessment of deviation becomes concrete, improving the scientific nature of the analysis.

[0115] Then, the vertical coordinate values ​​of the corresponding maximum and minimum coordinate points are processed by absolute value to obtain the first absolute value and the second absolute value. The difference between these two absolute values ​​and the uniform distribution value is calculated to obtain the first uniform difference and the second uniform difference. It can be explained that by calculating these two differences, the specific degree of deviation of the two extreme coordinate points from the uniform distribution benchmark can be quantified, so as to clearly understand the primary and secondary deviations of the tooth distribution and point out the core direction for subsequent adjustments.

[0116] Then, the larger of the first uniform difference and the second uniform difference is determined as the adjustment distribution value, and the coordinate point of the corresponding adjustment distribution value is determined as the deviation coordinate point. By focusing on the coordinate point with the greater deviation, the main contradiction of tooth distribution deviation can be grasped, making subsequent adjustment measures more targeted, avoiding blind operation by users when adjusting online, and saving adjustment time.

[0117] Subsequently, a pointing coordinate point with the same horizontal coordinate value as the origin and the same vertical coordinate value as the offset coordinate point is determined. An adjustment indicator line pointing to the center point of the image is generated from the pointing coordinate point. Here, the intuitive adjustment indicator line allows users to clearly understand the direction in which the shooting angle needs to be adjusted, reducing the difficulty of online operation for users and making the adjustment process more efficient.

[0118] Simultaneously, the server can further retrieve a standard tooth adjustment template and fill the adjustment distribution values ​​into the adjustment indicator slots located in the standard tooth adjustment template. It can be explained that the combination of the standard template and specific adjustment values ​​provides users with standardized and quantitative adjustment references, allowing users to clearly understand the range of adjustment needed, ensuring that the adjusted image is closer to the qualified standard, and improving the pass rate of online image acquisition. The standard tooth adjustment template can be specifically "Please adjust the oral cavity according to the corresponding XX displacement", where "XX" can refer to the adjustment indicator slot.

[0119] Finally, the oral images are sent to the user's device. Upon receiving the images, the user's device is triggered by the oral image acquisition plugin to play a standard dental adjustment template via voice. This voice playback helps the user understand the adjustment requirements, which is especially helpful for users unfamiliar with online operations. It ensures that users can accurately follow the instructions to make adjustments, thereby quickly obtaining oral images that meet diagnostic requirements and improving the smoothness and success rate of online oral disease diagnosis.

[0120] Furthermore, in this embodiment, the aforementioned "responding to determine the existence of a uvula region indicating uvula features based on the recognition result, determining the morphology of the uvula region, and obtaining the display morphology corresponding to the uvula region" may further include the following steps:

[0121] Obtain the uvula contour corresponding to the uvula region, and determine the coordinate points of each uvula that make up the uvula contour based on the image coordinate system.

[0122] Determine the number of lateral coordinates for all uvula coordinate points corresponding to each identical longitudinal coordinate value, calculate the half value of the largest number of lateral coordinates, and multiply the obtained initial comparison number with the retrieved preset update coefficient to obtain the current comparison number;

[0123] The uvula coordinate point with the smallest corresponding vertical coordinate value is determined as the midpoint. Starting from the midpoint, along the uvula contour, the uvula coordinate points with corresponding horizontal coordinate values ​​greater than the midpoint and corresponding to the current comparison number are divided into the first trend comparison group. The uvula coordinate points with corresponding horizontal coordinate values ​​less than the midpoint and corresponding to the current comparison number are divided into the second trend comparison group.

[0124] The longitudinal trend corresponding to each uvula coordinate point located in the first trend comparison group and the second trend comparison group is determined;

[0125] If the vertical trend of the first trend comparison group is downward and the vertical trend of the second trend comparison group is upward, the display form of the corresponding uvula region will be determined as fully displayed; otherwise, it will be determined as partially displayed.

[0126] For example, in this embodiment, the identification and morphological determination of the uvula region can be based on the following technical solution:

[0127] First, the server can obtain the corresponding uvula contour based on the uvula region, and determine the coordinate points of each uvula that make up the uvula contour based on the established image coordinate system. Here, by transforming the uvula contour into specific coordinate points, the abstract morphological judgment can be transformed into a quantifiable coordinate analysis, providing a data basis for subsequent accurate judgment, avoiding deviations caused by subjective visual judgment in online diagnosis, and improving the objectivity of morphological determination.

[0128] Next, the number of lateral coordinates of all uvula coordinate points corresponding to each identical longitudinal coordinate value is determined. The largest number of lateral coordinates is calculated by half. Then, the initial number of comparisons is multiplied by the preset update coefficient to obtain the current number of comparisons. It can be explained that by determining the current number of comparisons through this calculation method, the number of analysis samples can be dynamically adjusted in combination with the actual morphological characteristics of the uvula, ensuring the scientific nature of subsequent grouping and making the online analysis of uvula morphology more consistent with its true structure.

[0129] Then, the uvula coordinate point with the smallest vertical coordinate value is determined as the midpoint. Starting from this midpoint, along the uvula contour, the uvula coordinate points with horizontal coordinate values ​​greater than the midpoint and corresponding to the current comparison number are divided into the first trend comparison group, and the uvula coordinate points with horizontal coordinate values ​​less than the midpoint and corresponding to the current comparison number are divided into the second trend comparison group. Here, based on the grouping method based on coordinate values ​​and the current comparison number, different regions of the uvula can be accurately segmented, clarifying the objects for subsequent trend analysis and ensuring that the online morphological analysis of different parts of the uvula is more targeted.

[0130] Subsequently, based on the coordinate points of each uvula in the first trend comparison group and the second trend comparison group, the longitudinal trends of the two groups were determined. It can be seen that by analyzing the longitudinal trends of the two groups of coordinate points, the shape of the uvula can be judged from the data level, avoiding misjudgment caused by visual observation alone, and improving the accuracy of online judgment of uvula shape.

[0131] Finally, since the end of the common uvula region is generally conical, if the vertical trend of the first trend comparison group is downward and the vertical trend of the second trend comparison group is upward, the corresponding uvula region can be determined as fully displayed; otherwise, it can be determined as partially displayed. This allows for a clear distinction between whether the uvula is fully displayed, ensuring that the uvula region in the online oral images meets diagnostic requirements, providing a reliable basis for subsequent accurate diagnosis, and improving the effectiveness of online oral disease diagnosis for users.

[0132] Furthermore, in this embodiment, the aforementioned "determining the longitudinal trend of the corresponding trend comparison group based on the obtained longitudinal differences of each uvula" may further include the following steps:

[0133] The uvula coordinate point with the smallest corresponding horizontal coordinate value in the first trend comparison group and the uvula coordinate point with the largest corresponding horizontal coordinate value in the second trend comparison group are determined as the first end point and the second end point, respectively.

[0134] Based on the uvula contour, a first trend direction and a second trend direction are determined from the first end point and the second end point toward the middle point;

[0135] The longitudinal coordinate value corresponding to each uvula coordinate point in the first trend comparison group gradually decreases along the first trend direction, and the longitudinal trend of the first trend comparison group is determined to be a downward trend.

[0136] The longitudinal coordinate value corresponding to each uvula coordinate point in the second trend comparison group gradually increases along the direction of the second trend, thus determining the longitudinal trend of the second trend comparison group as an upward trend.

[0137] For example, in this embodiment, the process of determining the longitudinal trend of the corresponding trend comparison group based on the obtained longitudinal differences of each uvula can be specifically as follows:

[0138] First, the server can determine the uvula coordinate point with the smallest corresponding horizontal coordinate value in the first trend comparison group and the uvula coordinate point with the largest corresponding horizontal coordinate value in the second trend comparison group as the first end point and the second end point. By clarifying these two end points, the horizontal distribution direction of the corresponding uvula region can be determined, so that the trend analysis has a clear reference object and ensures that the online judgment of the trend direction is more targeted.

[0139] Next, based on the uvula contour, the first trend direction and the second trend direction from the first end point and the second end point to the middle point are determined. That is, the trend direction is determined based on the uvula contour, which can fit the actual morphological characteristics of the uvula and make the determined trend direction more consistent with the natural direction of the uvula, providing a reliable directional basis for subsequent longitudinal trend judgment.

[0140] Then, in response to the gradual decrease of the longitudinal coordinate value of each uvula coordinate point in the first trend comparison group along the first trend direction, the longitudinal trend of the first trend comparison group is determined to be a downward trend. It can be explained that the trend judgment method based on the change of coordinate value can accurately quantify the longitudinal direction of the first trend comparison group, avoid the deviation caused by subjective visual judgment, and ensure that the online judgment of the longitudinal trend of this group is accurate.

[0141] Finally, in response to the gradual increase of the longitudinal coordinate value corresponding to each uvula coordinate point in the second trend comparison group along the direction of the second trend, the longitudinal trend of the second trend comparison group is determined to be an upward trend. This shows that by judging the longitudinal trend of the second trend comparison group based on the same coordinate value change method, the consistency of the trend judgment criteria of the two groups can be guaranteed, improving the scientificity and accuracy of the overall trend judgment. This provides a reliable basis for subsequently determining the display morphology of the uvula region, ensuring that the oral images collected online meet the diagnostic requirements, thereby improving the effectiveness of users completing oral disease diagnosis online.

[0142] Furthermore, in this embodiment, the aforementioned "determining the acquisition state of the corresponding oral cavity image based on the obtained tongue region indicating tongue features and the tooth region" may further include the following steps:

[0143] Connect the coordinates of each tooth that makes up the tooth region based on their adjacent positions to obtain the upper tooth region and the lower tooth region included in the tooth region.

[0144] All tooth coordinate points corresponding to the same horizontal coordinate value that make up the upper tooth region are grouped into the same upper tooth vertical group, and all tooth coordinate points corresponding to the same horizontal coordinate value that make up the lower tooth region are grouped into the same lower tooth vertical group.

[0145] The tooth coordinate point with the smallest vertical coordinate value in each vertical group of upper teeth is determined as the upper tooth minimum point, and the upper tooth minimum point with the largest vertical coordinate value is determined as the upper tooth open / close point.

[0146] The tooth with the largest vertical coordinate value in each lower tooth vertical group is determined as the lower tooth maximum point, and the lower tooth maximum point with the smallest vertical coordinate value is determined as the lower tooth open / close point.

[0147] The difference between the vertical coordinate values ​​of the upper and lower teeth opening points is calculated, and the tooth opening value obtained is greater than the preset opening value, and each lower tooth maximum point is adjacent to any tongue coordinate point that makes up the tongue region. The collection state is then determined to be a qualified state.

[0148] For example, as can be seen from the above, the relative positional relationship between the tongue region and the teeth region is complex. If it is not possible to scientifically determine whether the two meet the acquisition criteria required for diagnosis, unqualified images may easily enter the subsequent diagnostic process, affecting the accuracy of online diagnostic results. Therefore, in this embodiment, the determination of the corresponding acquisition status can be achieved based on the following technical solution:

[0149] First, based on the oral cavity image, the server can connect the coordinates of each tooth that makes up the tooth region based on the coordinates of adjacent positions to obtain the upper tooth region and the lower tooth region included in the tooth region. Here, the upper tooth region and the lower tooth region are distinguished by the coordinate connection method, which can clearly define the range of the upper and lower teeth in the oral cavity. This provides a clear partitioning basis for subsequent analysis of the positional relationship between teeth and tongue, ensuring that the online division of the tooth region is accurate and conforms to the actual oral cavity structure.

[0150] Next, all tooth coordinate points corresponding to the same horizontal coordinate value that make up the upper tooth region can be grouped into the same upper tooth vertical group, and all tooth coordinate points corresponding to the same horizontal coordinate value that make up the lower tooth region can be grouped into the same lower tooth vertical group. It can be explained that vertical grouping according to horizontal coordinate value can refine the upper and lower tooth regions into multiple vertical analysis units, making the subsequent analysis of tooth coordinate points more targeted, avoiding the omission of details caused by overall analysis, and improving the detail of online analysis.

[0151] Then, the tooth coordinate point with the smallest vertical coordinate value in each upper tooth vertical group is determined as the upper tooth minimum point, and the upper tooth minimum point with the largest vertical coordinate value is determined as the upper tooth open / close point; similarly, the tooth coordinate point with the largest vertical coordinate value in each lower tooth vertical group is determined as the lower tooth maximum point, and the lower tooth maximum point with the smallest vertical coordinate value is determined as the lower tooth open / close point. This embodiment can accurately locate the key positions affecting the open / close state in the upper and lower tooth regions by determining the upper tooth minimum point, upper tooth open / close point, lower tooth maximum point, and lower tooth open / close point, providing specific data support for judging the degree of tooth open / close and ensuring that the online assessment of the tooth open / close state has a clear basis;

[0152] Finally, the difference between the vertical coordinate values ​​of the upper and lower teeth open bite points can be calculated. If the obtained tooth open bite value is greater than the preset open bite value, and each maximum point of the lower teeth is adjacent to any tongue coordinate point that makes up the tongue region, the acquisition status of the oral image can be determined as qualified. Otherwise, it can be determined as rejected. It can be explained that by quantifying the tooth open bite value and combining it with the adjacent relationship between the tongue and the lower teeth to judge the acquisition status, we can comprehensively evaluate whether the online acquired images clearly present the key positional relationship between the tongue and teeth, ensure that the images entering subsequent diagnosis meet the requirements, and improve the reliability and accuracy of online oral disease diagnosis for users.

[0153] It should be noted that since subsequent diagnosis of the disease will be based on the tongue area, it is necessary to ensure that the user's mouth has a certain degree of openness to expose a large area of ​​the tongue area, and that the tongue is placed close to the teeth.

[0154] For example, Figure 2 This embodiment shows an image of a mouth in a qualified state, wherein... Figure 2 It can be seen that the end of the uvula is conical, the mouth opens to a large degree, and the tongue is placed close to the teeth.

[0155] Step S3 includes the following:

[0156] If the acquisition status is deemed satisfactory, the tongue features are divided into corresponding features for different tongue parts based on the tooth features, and a disease assessment is performed based on each obtained tongue sub-feature to obtain the assessment result.

[0157] For example, in this embodiment, using tooth features as a reference to divide different parts of the tongue can leverage the relatively fixed and easily identifiable nature of teeth to provide a clear and stable benchmark for tongue division. This ensures that the divided tongue parts are accurate and consistent, avoiding confusion caused by the variable shape of the tongue in online images. This makes subsequent disease assessments more targeted. Furthermore, after obtaining the corresponding features of each tongue sub-feature, disease assessments are performed separately for each sub-feature. This allows for a focus on the health status of each specific part of the tongue, in-depth analysis of potential problems in each part, and avoids the limitations of overall assessment. This enables online diagnosis to accurately locate diseases in each part of the tongue, making the assessment results more detailed and accurate. This provides users with more valuable online diagnostic information, effectively meeting their needs for online oral disease diagnosis and increasing user acceptance of online diagnostic results.

[0158] For example, from the perspective of traditional Chinese medicine tongue diagnosis theory, different areas of the tongue are generally considered to correspond to different internal organs, and therefore can be used as a reference for diagnosing diseases.

[0159] The tip of the tongue: This area is mostly related to the heart and lungs. If the tip of the tongue is red, it may indicate heat in the heart and lungs, commonly seen in cases of fever due to external pathogens, sore throat, etc.; if there are petechiae on the tip of the tongue, it may be related to stagnation of the heart meridian.

[0160] The middle region of the tongue: This is mostly related to the spleen and stomach. If the coating in the middle of the tongue is thick and greasy, it may indicate damp-heat in the spleen and stomach or food stagnation; if cracks appear in the middle of the tongue, it may be related to yin deficiency of the spleen and stomach and insufficient body fluids.

[0161] The root of the tongue: This area is mostly related to the kidneys and bladder. If the root of the tongue is yellow and greasy, it may indicate damp-heat in the lower burner, such as urinary tract infections; if the root of the tongue has little and dry coating, it may be related to kidney yin deficiency.

[0162] The sides of the tongue: These areas are mostly related to the liver and gallbladder. If the sides of the tongue are red or have prickles, it may indicate excessive liver and gallbladder fire, commonly seen in symptoms such as dizziness, irritability, and bitter taste in the mouth; if there are ecchymoses on the sides of the tongue, it may be related to liver stagnation and blood stasis.

[0163] Therefore, based on the above-mentioned TCM tongue diagnosis theory, corresponding disease diagnoses can be made according to different tongue characteristics.

[0164] Furthermore, in this embodiment, the aforementioned "responding to the acquisition state being in a qualified state, dividing the tongue features according to the tooth features into features corresponding to different tongue parts, and performing disease assessment based on each obtained tongue sub-feature to obtain an assessment result" may also include the following steps:

[0165] In response to the acquisition status being in a qualified state, each tooth sub-region of the corresponding anterior tooth type and each tooth sub-region of the corresponding posterior tooth type are determined based on the lower tooth region.

[0166] Each pair of adjacent tooth sub-regions is defined as the same tooth division group, and a vertical line of symmetry parallel to the Y-axis is established based on the tooth regions.

[0167] In response to any tooth division group including any tooth sub-region of the corresponding posterior tooth type, a horizontal dividing line is generated based on the tooth division group, and two horizontal dividing lines with vertical symmetry are merged based on the vertical symmetry line.

[0168] In response to any tooth division group including two tooth sub-regions corresponding to the anterior tooth type, a longitudinal dividing line is generated based on the tooth division group.

[0169] The tongue region is divided into regions based on all the horizontal and vertical dividing lines to obtain tongue sub-regions with different division numbers.

[0170] Based on the traversal of the retrieved preset partitioning template, all partitioning numbers corresponding to the same tongue sub-feature are merged based on the tongue sub-region, and disease assessment is performed based on the obtained disease sub-regions corresponding to different tongue sub-features to obtain the assessment results.

[0171] For example, in this embodiment, when the response acquisition status is in the qualified state, the process of segmenting and evaluating tongue features based on tooth features is as follows:

[0172] First, based on oral images, the server can determine the sub-regions of each tooth for the corresponding anterior tooth type and the sub-regions of each tooth for the corresponding posterior tooth type based on the lower tooth region. It can be explained that by distinguishing the sub-regions of the anterior and posterior teeth, the fixed position characteristics of different types of teeth in the oral cavity can be used to provide an accurate reference for the division of the tongue, ensuring that the subsequent division of the tongue sub-regions is consistent with the actual anatomical structure of the oral cavity, and improving the scientific nature of the online division.

[0173] Next, every two adjacent tooth sub-regions are identified as the same tooth division group, and a vertical symmetry line parallel to the Y-axis is established based on the tooth region. Here, the division group composed of adjacent tooth sub-regions and the establishment of the vertical symmetry line provide a structured reference framework for the subsequent generation of division lines, making the generation of division lines more in line with the characteristics of the symmetrical distribution of the oral cavity, and ensuring that the division of the tongue region is uniform and reasonable.

[0174] Specifically, when any tooth division group includes any tooth sub-region of the corresponding posterior tooth type, a horizontal dividing line is generated based on the tooth division group, and two horizontal dividing lines with vertical symmetry are merged based on the vertical symmetry line. That is, a horizontal dividing line is generated for the posterior tooth type and the symmetry line is merged. This can adapt to the horizontal distribution characteristics of posterior teeth in the oral cavity, so that the dividing line accurately separates the posterior region of the tongue, avoids the division deviation caused by the position of posterior teeth, and improves the accuracy of the division of the posterior region of the tongue.

[0175] In addition, when any tooth division group includes two tooth sub-regions corresponding to the anterior tooth type, a longitudinal dividing line can be generated based on the tooth division group, that is, a longitudinal dividing line can be generated for the anterior tooth type. This can match the longitudinal distribution characteristics of the anterior teeth in the oral cavity, accurately divide the anterior region of the tongue, ensure that the division of the anterior region of the tongue corresponds to the position of the anterior teeth, and improve the targeting of the division.

[0176] After obtaining all horizontal and vertical dividing lines, the tongue region can be divided based on these lines to obtain tongue sub-regions with different division numbers. In other words, by combining horizontal and vertical dividing lines, the tongue region can be divided into multiple independent sub-regions, each with a clear division number, making the boundaries of each part of the tongue clearly distinguishable. This provides specific objects for subsequent targeted assessments and avoids regional confusion during online assessments. Here, the division number can be set based on the region location corresponding to each sub-region.

[0177] Finally, by traversing the retrieved preset partitioning template, all partitioning numbers corresponding to the same tongue sub-feature can be merged based on the tongue sub-region. Then, based on the obtained disease sub-regions corresponding to different tongue sub-features, disease assessment is performed to obtain the assessment results. It should be noted that the preset partitioning template can be pre-stored in the oral cavity acquisition plugin. The preset partitioning template includes all sub-regions corresponding to each tongue sub-feature, which ensures the standardization of region merging and enables the accurate integration of sub-regions of the same tongue sub-feature. The assessment of each disease sub-region enables accurate diagnosis of different feature parts of the tongue, making the online assessment results more detailed and more valuable, effectively improving the user's experience and effectiveness in completing oral disease diagnosis online.

[0178] For example Figure 3 As shown, Figure 3 This diagram illustrates the division of the tongue region into areas in this embodiment. Figure 3 Including multiple horizontal and vertical dividing lines, it can be explained that anterior teeth are a medical term published in 1992, referring to the teeth located at the front of the oral cavity, including three classes: central incisors, lateral incisors, and canines, totaling 12 permanent teeth, which are mainly responsible for cutting food; while posterior teeth refer to the teeth at the back of the oral cavity that are mainly responsible for chewing, including premolars (bicuspids) and molars.

[0179] It should be noted that the specific method for generating horizontal and vertical dividing lines based on the tooth division group can be, for example, by merging two tooth sub-regions located in the same tooth division group, and then establishing a dividing line passing through the center point of the merged sub-region based on the merged sub-region.

[0180] Furthermore, in this embodiment, the aforementioned "assessing the disease based on the obtained disease sub-regions corresponding to different tongue features and obtaining the assessment results" may further include the following steps:

[0181] The oral cavity acquisition plugin retrieves the initial report interface corresponding to each tongue feature and fills the symptom sub-area corresponding to the tongue feature into the symptom indicator sub-interface located in the initial report interface.

[0182] The disease sub-region is compared with the health indicator area filled in the health comparison sub-interface of the initial report interface, and the health assessment value of the tongue sub-feature is determined based on the comparison result.

[0183] If the health assessment value falls within any preset assessment range, the preset assessment content corresponding to the preset assessment range will be filled into the symptom assessment sub-interface located in the initial report interface; otherwise, the initial report interface will be sent to the medical staff terminal, and the real-time assessment content obtained from the medical staff terminal will be filled into the symptom assessment sub-interface to obtain the current report interface.

[0184] In response to the current report interface that retrieves all tongue features, the evaluation results integrated from all current report interfaces are sent to the user terminal.

[0185] For example, based on the above, according to the theory of tongue diagnosis in Traditional Chinese Medicine, different areas of the tongue are generally considered to correspond to different internal organs. Therefore, they can be used as a reference for diagnosing diseases. Based on this, in this embodiment, disease assessment is performed based on the disease sub-regions corresponding to different tongue sub-features. The process for obtaining the assessment results is as follows:

[0186] First, the server can control the oral cavity acquisition plugin to retrieve the initial report interface corresponding to each tongue sub-feature, and fill the symptom sub-region of the corresponding tongue sub-feature into the symptom indication sub-interface located in the initial report interface. It can be explained that by matching a unique initial report interface for each tongue sub-feature and accurately displaying the corresponding symptom sub-region in the symptom indication sub-interface, the symptom location of each tongue sub-feature can be presented intuitively, avoiding confusion of symptom information of different sub-features, improving the clarity of online assessment reports, and making it easier for users to quickly locate specific problems. Furthermore, it can be explained that the initial report interface can be pre-stored in the oral cavity acquisition plugin for easy retrieval and use.

[0187] Next, the server compares the symptom sub-region with the health indicator region filled in the health comparison sub-interface of the initial report interface, and determines the health assessment value of the tongue sub-feature based on the comparison results. By comparing with the health indicator region, the health status of the tongue sub-feature can be quantified, transforming the abstract symptom manifestation into a specific health assessment value, providing an objective basis for the generation of subsequent assessment content, and ensuring the scientific and standardized nature of online assessment. It can be noted that the greater the regional difference between the symptom sub-region and the health indicator region, the smaller the corresponding health assessment value.

[0188] Specifically, when the health assessment value falls within any preset assessment range, the preset assessment content corresponding to that range is filled into the symptom assessment sub-interface on the initial report interface. Conversely, if the value falls outside the preset range, the initial report interface is sent to the healthcare provider (i.e., the initial report interface with a low health assessment value is sent to the healthcare provider for real-time assessment), and the real-time assessment content obtained from the healthcare provider is filled into the symptom assessment sub-interface to obtain the current report interface. Here, the preset assessment content can quickly address common health conditions and improve the efficiency of online assessment. For cases exceeding the preset range, the introduction of real-time assessment from the healthcare provider ensures the professionalism and accuracy of the assessment, balancing the efficiency of online diagnosis with the proper handling of complex situations, and enhancing users' trust in the assessment results.

[0189] Finally, in response to the current report interface that retrieves all tongue features, the evaluation results integrated from all current report interfaces are sent to the user. In other words, by integrating the evaluation information of all tongue features, comprehensive and systematic diagnostic results can be provided to users, avoiding information fragmentation and allowing users to clearly understand the overall health status of the tongue. This meets users' needs for efficient and accurate online diagnosis of oral diseases and improves their online diagnostic experience.

[0190] In summary, this invention enables users to independently acquire oral images, and combines dental features with standardized regional division and disease assessment of tongue features. This effectively solves the problems of traditional offline tongue diagnosis relying on professionals and the lack of online diagnostic processes for users. By using dental features to assist in judging the acquisition status, image quality and diagnostic accuracy are significantly improved, ensuring that the tongue images acquired by users meet diagnostic requirements. Based on the theory of traditional Chinese medicine tongue diagnosis, multi-regional feature division of the tongue is performed, realizing intelligent assessment from a single tongue image to diseases related to multiple organs. This breaks through the dependence of traditional tongue diagnosis on the experience of professional physicians, allowing ordinary users to obtain standardized and automated preliminary diagnosis services for oral diseases online, greatly improving diagnostic efficiency and convenience. In addition, the standardized acquisition and analysis process established by this invention provides reliable technical support for telemedicine and health management, and has broad application prospects.

[0191] Another embodiment of the present invention provides an image processing-based oral cavity data processing system. Figure 4 Its corresponding system block diagram, such as Figure 4 As shown, the system includes:

[0192] The image acquisition module is configured to respond to receiving an oral diagnosis request sent by any user terminal, control the loading of the oral acquisition plugin onto the user terminal, and trigger the user terminal to acquire images of the user's oral cavity based on the oral acquisition plugin to obtain oral images;

[0193] The state determination module is configured to extract features from the oral cavity image and determine the acquisition state of the corresponding oral cavity image based on the obtained tooth and tongue features.

[0194] The symptom assessment module is configured to respond to the acquisition status as a qualified state, divide the tongue features according to the tooth features into features corresponding to different tongue parts, and perform symptom assessment based on each obtained tongue sub-feature to obtain the assessment result.

[0195] In the specification provided herein, the algorithms and displays are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used with the examples of this invention. The required structure for constructing such systems is apparent from the above description. Furthermore, this invention is not directed to any particular programming language. It should be understood that the contents of the invention described herein can be implemented using various programming languages, and the above description of specific languages ​​is for the purpose of disclosing preferred embodiments of the invention.

[0196] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0197] Similarly, it should be understood that, in order to streamline this disclosure and aid in understanding one or more of the various aspects of the invention, in the above description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof.

[0198] Those skilled in the art will understand that modules, units, or components of the devices disclosed in the examples herein can be arranged in the devices described in this embodiment, or alternatively, can be located in one or more devices different from the devices in this example. The modules in the foregoing examples can be combined into a single module or, in addition, can be divided into multiple sub-modules.

[0199] Those skilled in the art will understand that the modules in the device of the embodiment can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiment can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components.

[0200] Furthermore, those skilled in the art will understand that although some embodiments described herein include certain features included in other embodiments but not others, combinations of features from different embodiments are meant to be within the scope of the invention and form different embodiments.

[0201] Furthermore, some of the embodiments described herein are methods or combinations of method elements that can be implemented by a processor of a computer system or by other means of performing the functions. Therefore, a processor having the necessary instructions for implementing the methods or method elements forms means for implementing the methods or method elements. Furthermore, the elements described herein in the apparatus embodiments are examples of means for implementing the functions performed by elements for the purposes of carrying out the invention.

[0202] As used herein, unless otherwise specified, the use of ordinal numbers such as “first,” “second,” “third,” etc., to describe ordinary objects merely indicates different instances of similar objects and is not intended to imply that the objects being described must have a given order in time, space, ordering, or any other manner.

[0203] Although the invention has been described with respect to a limited number of embodiments, those skilled in the art will understand from the foregoing description that other embodiments are conceivable within the scope of the invention described herein. Furthermore, it should be noted that the language used in this specification has been chosen primarily for readability and edibility purposes, and not for the purpose of explaining or limiting the subject matter of the invention.

Claims

1. A method for processing oral cavity data based on image processing, characterized in that, include: In response to receiving an oral diagnosis request from any user terminal, the system controls the loading of the oral cavity acquisition plugin onto the user terminal, so as to trigger the user terminal to acquire images of the user's oral cavity based on the oral cavity acquisition plugin, thereby obtaining oral cavity images; Feature extraction is performed on the oral cavity image, and the acquisition status of the corresponding oral cavity image is determined based on the obtained tooth and tongue features; The process includes extracting features from the oral cavity image and determining the acquisition state of the corresponding oral cavity image based on the obtained tooth and tongue features, including: The oral cavity image is input into the first feature recognition model for image recognition, and the model responds by determining, based on the recognition result, that there is a tooth region indicating the tooth features. An image coordinate system corresponding to the oral cavity image is established with the image center point of the oral cavity image as the origin. The coordinate points of each tooth constituting the tooth region are determined based on the image coordinate system. If each tooth coordinate point is not located on the X-axis of the corresponding image coordinate system and is distributed on both sides of the X-axis, the tooth region is determined to have a uniform distribution attribute; otherwise, it is determined to have an off-distribution attribute. The response has a uniform distribution property, and the oral cavity image is input into a second feature recognition model for image recognition; Based on the recognition results, the system determines the existence of a uvula region that indicates uvula features, performs morphological determination on the uvula region, and obtains the display morphology of the corresponding uvula region. In response to the display mode being a complete display, the oral cavity image is input into a third feature recognition model for image recognition, and the acquisition status of the corresponding oral cavity image is determined based on the obtained tongue region indicating tongue features and the tooth region. If the acquisition status is deemed satisfactory, the tongue features are divided into corresponding features for different tongue parts based on the tooth features, and a disease assessment is performed based on each obtained tongue sub-feature to obtain the assessment result.

2. The method according to claim 1, characterized in that, The oral cavity image is input into a first feature recognition model for image recognition, and based on the recognition result, a tooth region indicating the tooth feature is determined. An image coordinate system corresponding to the oral cavity image is established with the image center point of the oral cavity image as the origin. The process then includes: The oral cavity image is sent to the user terminal; The response receives the missing filling data sent by the user terminal based on the oral cavity image within a preset feedback period after the image transmission time, and determines the missing filling location of the corresponding oral cavity region based on the missing filling data; Based on the tooth region, establish a vertical line of symmetry parallel to the Y-axis, and based on the vertical line of symmetry, determine a lateral symmetric position that has a vertical symmetric relationship with the missing filling position; In response to the existence of a tooth sub-region corresponding to any tooth type at the lateral symmetrical position, the tooth sub-region is over-copied, and the resulting copied sub-region is filled into the missing filling position; If there is no tooth sub-region corresponding to any tooth type at the lateral symmetrical position, the missing filling position is determined as the missing filling type, and the standard sub-region of the corresponding missing filling type located in the retrieved standard tooth distribution map is filled into the missing filling position.

3. The method according to claim 1, characterized in that, The method further includes: The response has a deviation distribution property. Based on the image coordinate system, the tooth coordinate point with the largest corresponding vertical coordinate value is determined as the maximum coordinate point, and the tooth coordinate point with the smallest corresponding vertical coordinate value is determined as the minimum coordinate point. The difference between the maximum and minimum coordinate points is calculated based on the longitudinal coordinate values, and the resulting longitudinal difference of the teeth is then calculated as half to obtain a uniformly distributed value. The vertical coordinate values ​​of the corresponding maximum and minimum coordinate points are processed by absolute value, and the difference between the first absolute value, the second absolute value and the uniform distribution value are calculated to obtain the first uniform difference and the second uniform difference. The larger of the first uniform difference and the second uniform difference is determined as the adjusted distribution value, and the coordinate point corresponding to the adjusted distribution value is determined as the deviation coordinate point; Determine the pointing coordinate point that has the same horizontal coordinate value as the origin and the same vertical coordinate value as the offset coordinate point, and generate an adjustment indicator line pointing to the center point of the image from the pointing coordinate point. Retrieve the standard dental adjustment template and fill the adjustment distribution value into the adjustment indicator slot located on the standard dental adjustment template; The system sends oral images to the user's device and responds to the user's request to receive them. Based on the oral image acquisition plugin, the system triggers the user's device to play a standard teeth adjustment template via voice.

4. The method according to claim 1, characterized in that, The response determines, based on the recognition results, the existence of a uvula region indicating uvular features, performs morphological determination on the uvula region, and obtains the corresponding display morphology of the uvula region, including: Obtain the uvula contour corresponding to the uvula region, and determine the uvula coordinate points that make up the uvula contour based on the image coordinate system. Determine the number of lateral coordinates for all uvula coordinate points corresponding to each identical longitudinal coordinate value, calculate the half value of the largest number of lateral coordinates, and multiply the obtained initial comparison number with the retrieved preset update coefficient to obtain the current comparison number; The uvula coordinate point with the smallest corresponding vertical coordinate value is determined as the midpoint. Starting from the midpoint, along the uvula contour, the uvula coordinate points with corresponding horizontal coordinate values ​​greater than the midpoint and corresponding to the current comparison number are divided into the first trend comparison group. The uvula coordinate points with corresponding horizontal coordinate values ​​less than the midpoint and corresponding to the current comparison number are divided into the second trend comparison group. The longitudinal trend corresponding to the first trend comparison group and the second trend comparison group is determined based on the uvula coordinate points located in the first trend comparison group and the second trend comparison group; If the vertical trend of the first trend comparison group is downward and the vertical trend of the second trend comparison group is upward, the display form of the corresponding uvula region will be determined as fully displayed; otherwise, it will be determined as partially displayed.

5. The method according to claim 4, characterized in that, Based on the uvula coordinates of each point located in the first trend comparison group and the second trend comparison group, the longitudinal trend corresponding to the first trend comparison group and the second trend comparison group is determined, including: The uvula coordinate point with the smallest corresponding horizontal coordinate value in the first trend comparison group and the uvula coordinate point with the largest corresponding horizontal coordinate value in the second trend comparison group are determined as the first end point and the second end point, respectively. Based on the uvula contour, a first trend direction and a second trend direction are determined from the first end point and the second end point toward the middle point; The longitudinal coordinate value corresponding to each uvula coordinate point in the first trend comparison group gradually decreases along the first trend direction, and the longitudinal trend of the first trend comparison group is determined to be a downward trend. The longitudinal coordinate value corresponding to each uvula coordinate point in the second trend comparison group gradually increases along the direction of the second trend, thus determining the longitudinal trend of the second trend comparison group as an upward trend.

6. The method according to claim 1, characterized in that, Determining the acquisition state of the corresponding oral cavity image based on the obtained tongue region indicating tongue features and the tooth region includes: Connect the coordinates of each tooth that makes up the tooth region based on their adjacent positions to obtain the upper tooth region and the lower tooth region included in the tooth region. All tooth coordinate points corresponding to the same horizontal coordinate value that make up the upper tooth region are grouped into the same upper tooth vertical group, and all tooth coordinate points corresponding to the same horizontal coordinate value that make up the lower tooth region are grouped into the same lower tooth vertical group. The tooth coordinate point with the smallest corresponding vertical coordinate value in each vertical group of upper teeth is determined as the upper tooth minimum point, and the upper tooth minimum point with the largest corresponding vertical coordinate value is determined as the upper tooth open / close point. The tooth with the largest vertical coordinate value in each lower tooth vertical group is determined as the lower tooth maximum point, and the lower tooth maximum point with the smallest vertical coordinate value is determined as the lower tooth open / close point. The difference between the vertical coordinate values ​​of the upper and lower teeth opening points is calculated, and the tooth opening value obtained is greater than the preset opening value, and each lower tooth maximum point is adjacent to any tongue coordinate point that makes up the tongue region. The collection state is then determined to be a qualified state.

7. The method according to claim 6, characterized in that, In response to the acquisition status being deemed satisfactory, the tongue features are segmented according to the dental features, corresponding to different tongue locations. Based on each obtained tongue sub-feature, a disease assessment is performed to obtain the assessment results, including: In response to the acquisition status being in a qualified state, each tooth sub-region of the corresponding anterior tooth type and each tooth sub-region of the corresponding posterior tooth type are determined based on the lower tooth region. Each pair of adjacent tooth sub-regions is defined as the same tooth division group, and a vertical line of symmetry parallel to the Y-axis is established based on the tooth regions. In response to any tooth division group including any tooth sub-region of the corresponding posterior tooth type, a horizontal dividing line is generated based on the tooth division group, and two horizontal dividing lines with vertical symmetry are merged based on the vertical symmetry line. In response to any tooth division group including two tooth sub-regions corresponding to the anterior tooth type, a longitudinal dividing line is generated based on the tooth division group. The tongue region is divided into regions based on all the horizontal and vertical dividing lines to obtain tongue sub-regions with different division numbers. Based on the traversal of the retrieved preset partitioning template, all partitioning numbers corresponding to the same tongue sub-feature are merged based on the tongue sub-region, and disease assessment is performed based on the obtained disease sub-regions corresponding to different tongue sub-features to obtain the assessment results.

8. The method according to claim 7, characterized in that, Based on the obtained disease sub-regions corresponding to different tongue features, disease assessment is performed to obtain assessment results, including: The oral cavity acquisition plugin retrieves the initial report interface corresponding to each tongue feature and fills the symptom sub-area corresponding to the tongue feature into the symptom indicator sub-interface located in the initial report interface. The disease sub-region is compared with the health indicator area filled in the health comparison sub-interface of the initial report interface, and the health assessment value of the tongue sub-feature is determined based on the comparison result. If the health assessment value falls within any preset assessment range, the preset assessment content corresponding to the preset assessment range will be filled into the symptom assessment sub-interface located in the initial report interface; otherwise, the initial report interface will be sent to the medical staff terminal, and the real-time assessment content obtained from the medical staff terminal will be filled into the symptom assessment sub-interface to obtain the current report interface. In response to the current report interface that retrieves all tongue features, the evaluation results integrated from all current report interfaces are sent to the user terminal.

9. A dental data processing system based on image processing, characterized in that, include: The image acquisition module is configured to respond to receiving an oral diagnosis request sent by any user terminal, control the loading of the oral acquisition plugin onto the user terminal, and trigger the user terminal to acquire images of the user's oral cavity based on the oral acquisition plugin to obtain oral images; The state determination module is configured to extract features from the oral cavity image and determine the acquisition state of the corresponding oral cavity image based on the obtained tooth and tongue features. The process includes extracting features from the oral cavity image and determining the acquisition state of the corresponding oral cavity image based on the obtained tooth and tongue features, including: The oral cavity image is input into the first feature recognition model for image recognition, and the model responds by determining, based on the recognition result, that there is a tooth region indicating the tooth features. An image coordinate system corresponding to the oral cavity image is established with the image center point of the oral cavity image as the origin. The coordinate points of each tooth constituting the tooth region are determined based on the image coordinate system. If each tooth coordinate point is not located on the X-axis of the corresponding image coordinate system and is distributed on both sides of the X-axis, the tooth region is determined to have a uniform distribution attribute; otherwise, it is determined to have an off-distribution attribute. The response has a uniform distribution property, and the oral cavity image is input into a second feature recognition model for image recognition; Based on the recognition results, the system determines the existence of a uvula region that indicates uvula features, performs morphological determination on the uvula region, and obtains the display morphology of the corresponding uvula region. In response to the display mode being a complete display, the oral cavity image is input into a third feature recognition model for image recognition, and the acquisition status of the corresponding oral cavity image is determined based on the obtained tongue region indicating tongue features and the tooth region. The symptom assessment module is configured to respond to the acquisition status as a qualified state, divide the tongue features according to the tooth features into features corresponding to different tongue parts, and perform symptom assessment based on each obtained tongue sub-feature to obtain the assessment result.

Images