System and method for detecting oral health

Abstract

The invention relates to the technical field of oral health detection, in particular to a system and method for detecting oral health, and the system comprises the following steps: an image acquisition module, a region detection module, a color analysis module, a model construction module and a chemical detection module. According to the invention, automatic adjustment of the focal length is realized through real-time analysis of image definition, the accuracy of oral health detection is improved, dental caries and pigmentation areas are accurately identified in combination with geometric data and color deviation analysis of tooth injury areas, the accuracy of early identification of oral diseases is improved, and the accuracy of oral health detection is improved. A three-dimensional model reconstruction method for dynamically adjusting the distribution point density is adopted, so that the oral cavity model has higher accuracy and detail degree, the abnormal data trend is monitored and detected in real time through combination of the oral cavity three-dimensional model and saliva chemical detection data, continuous monitoring and timely early warning are provided for oral cavity health, and the oral cavity quality is improved. The timeliness and accuracy of oral health management are effectively improved, and disease prediction and early diagnosis capabilities are enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of oral health detection, and particularly to a system and method for detecting oral health. Background Art

[0002] The technical field of oral health detection includes the collection, analysis, and evaluation of oral internal tissues, biochemical environments, and their related physiological states, aiming to achieve early identification and diagnosis of oral diseases. The core content involves image acquisition, image recognition, biological information detection, structural analysis, data analysis, and comparison. The overall technical system includes oral image acquisition devices, biosensing elements, detection and analysis modules, and human-computer interaction platforms, which are applied in oral medical institutions, home health management, and remote medical services. With the development of sensor performance, miniaturized imaging devices, and edge computing capabilities, diverse detection means and intelligent operation processes are realized, effectively supporting the continuity and timeliness of oral health management.

[0003] Among them, a system for detecting oral health refers to a technology that realizes the detection of oral health status by coordinating multiple information acquisition modules and data processing processes. The technical matters involved in the system include obtaining oral image data based on an optical imaging device, identifying tooth arrangement and gingival boundaries using image segmentation and region extraction methods, calculating the color change and caries area on the tooth surface by combining image comparison, analyzing the pH value and protein content in oral fluid by combining saliva collection and colorimetric detection elements, setting up a data fusion process to correlate and process structural information and chemical information, and displaying detection data and classification labels through a result output module. The system adopts a parallel data acquisition and process-based data processing method to form an integrated technical path to comprehensively detect oral health status.

[0004] Traditional oral health detection technologies mainly rely on static image acquisition and simple analysis methods in oral health detection. The clarity and focal length adjustment of images often rely on manual adjustment or fixed parameter settings, resulting in unstable image quality and affecting the early diagnosis and accurate identification of oral diseases. In the segmentation and damage detection of tooth regions, it is more dependent on rough texture analysis, easily ignoring the identification of tiny cracks and early caries, and unable to comprehensively evaluate all damaged regions in complex oral structures. The detection of color abnormalities in images usually adopts global analysis, failing to fully utilize the color change details in local regions, resulting in low detection accuracy for early caries and pigmentation. Dependent on manual analysis or insensitive devices, the analysis of fluctuations in the oral chemical environment is not fine enough, resulting in a lack of continuity and real-time monitoring of oral health, affecting the early detection and precise treatment of oral diseases. Summary of the Invention

[0005] To solve the technical problems existing in the prior art, an embodiment of the present invention provides a system and method for detecting oral health. The technical solutions are as follows:

[0006] On the one hand, a system for detecting oral health is provided. The system includes:

[0007] The image acquisition module uses an image acquisition device to collect and analyze the patient's oral images in real time, detect the edge sharpness, texture distribution, and gray scale fluctuation in the images in real time, calculate the deviation of the image focal length by analyzing the change trend of each sharpness parameter, adjust the focal length parameter, and obtain an oral image data set;

[0008] The area detection module calls the oral image data set, detects and segments the tooth area according to the texture features of the image, analyzes the texture density of the tooth area in the image, detects the damaged tooth area, and records the geometric data of the damaged area to generate a boundary structure recognition result;

[0009] The color analysis module calls the boundary structure recognition result, analyzes the color of the tooth area in the image, analyzes the deviation between various color parameters and the reference value, identifies the color abnormal position, detects the dental caries and pigment deposition areas, and obtains the tooth abnormal area;

[0010] The model construction module calls the tooth abnormal area, divides the tooth area into multiple sub-areas, adjusts the density of the distribution points according to the texture complexity and color change rate of each area, reconstructs the three-dimensional geometric model of the patient's teeth, and maps the dental caries and pigment deposition areas into the model to generate a patient oral model.

[0011] As a further solution of the present invention, the oral image data set includes a focal length adjustment parameter value, an image sharpness score result, and an image acquisition timestamp. The boundary structure recognition result is specifically a tooth area boundary line coordinate set, a damaged area position index, and geometric dimension parameters. The tooth abnormal area includes a color deviation point distribution map, an abnormal area type identifier, and a tooth surface color fluctuation range. The patient oral model specifically refers to a tooth three-dimensional structure lattice, a damaged area space mapping index, and a dental caries and deposition marked area.

[0012] As a further solution of the present invention, the image acquisition module includes:

[0013] The image sharpness detection module uses an image acquisition device to collect and analyze the patient's oral images in real time, detect the edge sharpness, texture distribution, and gray scale fluctuation in the images, detect the focal length deviation by monitoring the real-time change trend of each sharpness parameter in the image, and generate a focal length deviation judgment result;

[0014] Based on the result of the focal length offset judgment, the focal length adjustment calculation module calculates the focal length adjustment amplitude by analyzing the change amplitude of each clarity parameter, and generates an adjusted focal length parameter;

[0015] Based on the adjusted focal length parameter, the focal length adjustment execution module performs focal length adjustment, records the oral image data after the focal length parameter is adjusted, and obtains an oral image data set.

[0016] As a further solution of the present invention, the area detection module includes:

[0017] The tooth area recognition module calls the oral image data set, recognizes the boundary of the tooth area in the image, and segments the tooth area by detecting the continuity of the inner contour, the edge closure degree and the regional gray aggregation degree in the image, obtains the contour index and position identifier corresponding to each tooth structure, and generates a tooth area positioning parameter group;

[0018] Based on the tooth area positioning parameter group, the tooth damage detection module extracts the texture distribution in each tooth area, obtains the texture filling density, texture direction change rate, and texture continuous unit number in each area, and detects the tooth damage area by analyzing the texture density, and generates a tooth damage feature index set;

[0019] Based on the tooth damage feature index set, the structure and morphology extraction module extracts the size, contour extension direction, and position index value of the damage area, records the structure and morphology and spatial position of the damage area, and generates a boundary structure recognition result.

[0020] As a further solution of the present invention, the color analysis module includes:

[0021] The color offset analysis module calls the boundary structure recognition result, extracts the brightness, hue, and saturation parameters of each pixel block in the tooth area, calculates the offset amplitude between the color parameters at multiple positions in the tooth area of the image and the set color reference value, and generates a color offset coefficient distribution value;

[0022] Based on the color offset coefficient distribution value, the color anomaly detection module detects the color anomaly position according to the offset coefficient, and extracts the contour coordinates, area, and color change trend of the anomaly position to obtain the anomaly area information;

[0023] Based on the color anomaly area information, the tooth anomaly position recognition module combines the recognized tooth damage area, recognizes the type of the color anomaly position, detects the caries and pigment deposition areas, and obtains the tooth anomaly area.

[0024] As a further solution of the present invention, the specific formula for calculating the offset amplitude between the color parameters at multiple positions in the tooth area of the image and the set color reference value is:

[0025]

[0026] Calculate the color offset coefficient;

[0027] wherein, L k′ is the brightness value of the k'-th pixel block, H k′ is the hue value of the k'-th pixel block, S k′ is the saturation value of the k'-th pixel block, L0 is the reference brightness value, H0 is the reference hue value, S0 is the reference saturation value, and ΔC k′ is the color offset amplitude of the k'-th pixel block, w L is the weight coefficient of brightness, w H is the weight coefficient of hue, w S is the weight coefficient of saturation.

[0028] As a further aspect of the present invention, the model construction module includes:

[0029] The tooth sub-region division module calls the tooth abnormal region, divides the tooth region into multiple sub-regions, extracts the texture distribution width and direction change frequency in each sub-region, and establishes a sub-region division index to generate a tooth sub-region feature index set;

[0030] The point density adjustment module extracts the texture complexity value and color change rate value in each sub-region based on the tooth sub-region feature index set, adjusts the point spacing of each sub-region according to the consistency threshold of texture distribution and the continuity threshold of color gradient, and generates a distribution point density adjustment result;

[0031] The oral cavity three-dimensional reconstruction module reconstructs the three-dimensional geometric model of the patient's teeth according to the distribution point density adjustment result, extracts the spatial mapping positions of the caries and pigment deposition regions, maps them into the model, and obtains the patient's oral cavity model.

[0032] As a further aspect of the present invention, the system further includes:

[0033] The chemical detection module calls the patient's oral cavity model, regularly collects the patient's saliva samples, detects the protein concentration and pH value, analyzes the fluctuations of the data in multiple cycles, detects the abnormal change trend of the data, and generates an oral cavity environment detection record;

[0034] The oral cavity environment detection record includes the protein concentration change amplitude, the pH value fluctuation range, and the abnormal change duration period.

[0035] As a further aspect of the present invention, the chemical detection module includes:

[0036] The saliva sample collection module calls the oral model of the patient, regularly collects the patient's saliva samples, detects the protein concentration and pH value of the patient's oral cavity in each cycle, and generates a basic data set of cycle samples;

[0037] Based on the basic data set of cycle samples, the fluctuation trend analysis module compares the change ranges of the protein concentration and pH value in each cycle, extracts the change direction and duration of each item of data in consecutive cycles, analyzes the relative change rate and the fluctuation duration period, and generates a saliva component fluctuation trend value;

[0038] The oral health scoring module detects and records the abnormal change trend of the data according to the saliva component fluctuation trend value, combines the abnormal tooth position information, calculates the oral health score of the patient, and obtains the oral environment detection record;

[0039] The specific formula for calculating the oral health score of the patient is as follows:

[0040]

[0041] Calculate the oral health score;

[0042] Among them, H is the oral health score, n' is the total number of data cycles, i' is the cycle index, is the protein concentration change rate in the i'-th cycle, is the pH value change rate in the i'-th cycle, W i′ is the weight coefficient of the i'-th cycle, is the protein concentration change rate in the (i'-1)-th cycle, is the pH value change rate in the (i'-1)-th cycle, A cav is the area of the caries area, A total is the total tooth area, W cav is the weight coefficient of the caries area.

[0043] On the other hand, a method for detecting oral health is provided. This method is applied to a system for detecting oral health, and this method includes:

[0044] S1: Based on the image acquisition device, collect and analyze the patient's oral image in real time. By detecting the edge sharpness, texture distribution, and gray scale fluctuation in the image, analyze the change trend of each clarity parameter, calculate the focal length deviation of the image, and adjust the focal length parameter to obtain an oral image data set;

[0045] S2: Based on the oral image data set, according to the texture characteristics of the image, perform the detection and segmentation of the tooth area, combine the texture density analysis, detect the tooth damage area in the image, and record the geometric data of the damage area to generate a boundary structure recognition result;

[0046] S3: Based on the recognition result of the boundary structure, analyze the color of the tooth area in the image. By calculating the deviation between each color parameter and the healthy benchmark value, identify the positions with abnormal colors, and combine with the tooth damage area information to detect dental caries and pigment deposition areas, generating tooth abnormal area data;

[0047] S4: Based on the tooth abnormal area data, divide the tooth area into multiple sub - regions. According to the texture complexity and color change rate of each region, adjust the density of the distribution points within the region, reconstruct the three - dimensional geometric model of the patient's teeth, and map the dental caries and pigment deposition areas onto the model to generate the three - dimensional model of the patient's oral cavity;

[0048] S5: Based on the three - dimensional model of the patient's oral cavity, regularly collect the patient's saliva samples, detect the protein concentration and pH value. By analyzing the fluctuations of the detection data over multiple cycles, identify the abnormal change trends of the data, generating the oral environment detection record.

[0049] The beneficial effects brought by the technical solution provided by the embodiments of the present invention at least include:

[0050] By analyzing the image clarity in real - time to achieve automatic adjustment of the focal length, the accuracy of oral health detection is improved. Combining the geometric data of the tooth damage area with color deviation analysis, dental caries and pigment deposition areas are accurately identified, improving the accuracy of early identification of oral diseases. Using a three - dimensional model reconstruction method with dynamically adjusted distribution point density makes the oral model more accurate and detailed. By combining the three - dimensional model of the oral cavity with saliva chemical detection data, abnormal data trends are monitored and detected in real - time, providing continuous monitoring and timely warning for oral health, effectively improving the timeliness and accuracy of oral health management, and enhancing the disease prediction and early diagnosis capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0052] Figure 1 It is the system flow chart of the present invention;

[0053] Figure 2 It is the schematic diagram of the system framework of the present invention;

[0054] Figure 3 It is the schematic diagram of the method steps of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0055] The technical solutions in the present invention will be described below with reference to the accompanying drawings.

[0056] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to represent examples, illustrations or explanations. Any embodiment or design solution described as an "example" in the present invention should not be construed as being more preferred or more advantageous than other embodiments or design solutions. Rather, the use of the word "example" is intended to present concepts in a specific manner. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one of the two can be selected.

[0057] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when the difference is not emphasized, the meanings they express are the same. "(of)", "corresponding", and "corresponding" can sometimes be used interchangeably. It should be noted that when the difference is not emphasized, the meanings they express are the same.

[0058] In the embodiments of the present invention, sometimes subscripts such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meanings they express are the same.

[0059] To make the technical problems to be solved, technical solutions and advantages of the present invention clearer, the following will be described in detail with reference to the accompanying drawings and specific embodiments.

[0060] The embodiments of the present invention provide a system for detecting oral health. Please refer to Figures 1 to 2 , the present invention provides a technical solution. A system for detecting oral health includes:

[0061] The image acquisition module uses an image acquisition device to collect and analyze the oral images of patients in real time, detect the edge sharpness, texture distribution, and gray-scale fluctuation in the images in real time, calculate the deviation of the image focal length by analyzing the change trend of each sharpness parameter, adjust the focal length parameter, and obtain an oral image data set;

[0062] The area detection module calls the oral image data set, detects and segments the tooth area according to the texture features of the image, detects the damaged tooth area by analyzing the texture density of the tooth area in the image, records the geometric data of the damaged area, and generates a boundary structure recognition result;

[0063] The color analysis module calls the boundary structure recognition result, analyzes the color of the tooth area in the image, analyzes the deviation between various color parameters and the reference value, identifies the position of abnormal color, detects the areas of dental caries and pigmentation, and obtains the abnormal tooth area;

[0064] The model construction module calls the abnormal tooth area, divides the tooth area into multiple sub-areas, adjusts the density of distribution points according to the texture complexity and color change rate of each area, reconstructs the three-dimensional geometric model of the patient's teeth, and maps the caries and pigment deposition areas into the model to generate the patient's oral cavity model;

[0065] The chemical detection module calls the patient's oral cavity model, regularly collects the patient's saliva samples, detects the protein concentration and pH value, analyzes the fluctuations of the data in multiple cycles, detects the abnormal change trend of the data, and generates the oral cavity environment detection record.

[0066] The oral cavity image dataset includes the focal length adjustment parameter value, the image clarity score result, the image acquisition timestamp, and the boundary structure recognition result is specifically the tooth area boundary line coordinate set, the damage area position index, and the geometric dimension parameter. The abnormal tooth area includes the color offset point distribution map, the abnormal area type identifier, and the tooth surface color fluctuation range. The patient's oral cavity model specifically refers to the tooth three-dimensional structure lattice, the damage area space mapping index, and the caries and deposition marked areas. The oral cavity environment detection record includes the protein concentration change range, the pH value fluctuation range, and the abnormal change duration period.

[0067] The image acquisition module includes:

[0068] The image clarity detection module uses the image acquisition device to collect and analyze the patient's oral cavity image in real time, detects the edge sharpness, texture distribution, and gray level fluctuation in the image, monitors the real-time change trend of each clarity parameter in the image, detects the focal length deviation, and generates the focal length deviation judgment result;

[0069] The image acquisition module collects the patient's oral cavity image in real time through the image acquisition device, and the image data is transmitted to the image processing system. In this process, first, the edge sharpness of the image is detected, and the gradient of the edge contrast in the image is used to calculate the edge sharpness value of the image. The specific method is to perform a neighborhood calculation on each pixel point in the image to obtain the sharpness change of the edge. If the sharpness change at the image edge is too large, it indicates that the focal length may deviate, resulting in a loss of image clarity. Then, the texture distribution in the image is extracted, and the texture density is analyzed to calculate the complexity and distribution of the texture within the image area. If the texture distribution is relatively sparse, it means that the image may not be in correct focus. Furthermore, the gray level fluctuation amplitude of the image is monitored, and the stability of the image is determined by measuring the frequency of the pixel gray level value change in the image. If the gray level fluctuation amplitude is large, it usually means that the focal length deviates. By analyzing these three clarity parameters (edge sharpness, texture distribution, gray level fluctuation) simultaneously, the system can calculate the deviation degree of the focal length in real time, and finally generate the focal length deviation judgment result, providing a basis for focal length adjustment. Using the formula:

[0070]

[0071] Calculate the amplitude of gray-scale fluctuation, where G i is the gray-scale value of the i-th pixel, n is the total number of pixels in the image, and ΔG is the amplitude of gray-scale fluctuation of the image. Assume there are 5 pixel points in the image, and the gray-scale values are 150, 155, 160, 170, 165 respectively, then:

[0072]

[0073] The calculation result shows that the amplitude of gray-scale fluctuation is 6.25. If it exceeds the set threshold (5%), the system will determine that the image needs to adjust the focal length.

[0074] Based on the result of focal length offset judgment, the focal length adjustment calculation module calculates the focal length adjustment amplitude by analyzing the change amplitude of each clarity parameter, and generates the focal length adjustment parameter;

[0075] Based on the result of focal length offset judgment, the focal length adjustment calculation module analyzes the change amplitude of each clarity parameter in the image. If it is found that the change amplitude of the edge sharpness, texture distribution or gray-scale fluctuation of the image exceeds the set normal range (for example, the edge sharpness change exceeds ±15%, the texture density change exceeds ±10%, and the gray-scale fluctuation amplitude exceeds ±5%), the system will start to calculate the focal length adjustment amplitude. The focal length adjustment amplitude is calculated by combining the offset of each parameter and a fixed adjustment factor to achieve the required image clarity. The greater the change amplitude of the edge sharpness, the greater the corresponding focal length adjustment amplitude; if the change amplitude of the texture distribution or gray-scale fluctuation is smaller, the focal length adjustment amplitude is smaller. During the calculation process, the system will perform weighted synthesis on the offset of each clarity parameter to obtain the final focal length adjustment amplitude. Assume the set adjustment factor is 1.2, and the formula is used:

[0076] ΔF = k × (ΔS + ΔT + ΔG);

[0077] Calculate the focal length adjustment amplitude, where ΔF is the focal length adjustment amplitude, k is the adjustment factor, ΔS is the change amplitude of edge sharpness, ΔT is the change amplitude of texture distribution, and ΔG is the amplitude of gray-scale fluctuation. Assume the change amplitude of edge sharpness is 10%, the change amplitude of texture distribution is 8%, and the change amplitude of gray-scale fluctuation is 6%, then:

[0078] ΔF = 1.2 × (0.10 + 0.08 + 0.06) = 1.2 × 0.24 = 0.288;

[0079] The calculation result shows that the focal length adjustment amplitude is 0.288, indicating that focal length correction is required.

[0080] Based on the focal length adjustment parameter, the focal length adjustment execution module performs focal length adjustment, records the oral image data after the focal length parameter adjustment, and obtains the oral image data set;

[0081] The focal length adjustment execution module performs a focal length adjustment operation based on the adjusted focal length parameter and records the adjusted oral image data. The focal length adjustment is performed by controlling the drive system of the optical lens. The optical lens changes the focusing distance from the oral image according to the adjusted focal length parameter to ensure image clarity. The adjusted image data is recorded in real time within the system and compared with the image before adjustment to verify the effect of the focal length adjustment. If the clarity of the image after the focal length adjustment is improved, the focal length adjustment operation is considered successful, and the system generates an oral image data set. The change in clarity after the focal length adjustment can be evaluated by the following formula:

[0082]

[0083] Calculate the change in image clarity, where ΔC is the percentage of clarity improvement, C new is the image clarity after the focal length adjustment, and C old is the image clarity before adjustment. Assume that the image clarity before adjustment is 75 and the image clarity after adjustment is 85, then:

[0084]

[0085] The calculation result shows that the image clarity has increased by 13.33%, the focal length adjustment is successful, and finally an oral image data set is generated.

[0086] The region detection module includes:

[0087] The tooth region recognition module calls the oral image data set, identifies the boundary of the tooth region in the image, and segments the tooth region by detecting the continuity of the inner contour, the closure of the edge, and the gray level aggregation degree in the image, obtains the contour index and position identifier corresponding to each tooth structure, and generates a tooth region positioning parameter group;

[0088] The tooth area recognition module first calls the oral image dataset and analyzes the edge information in the image to identify the boundary of the tooth area. In this process, an edge detection algorithm (such as the Canny algorithm) is used to calculate the gradient change around each pixel in the image and determine which areas in the image have significant edges. For each pixel in the image, the edge detection algorithm compares the brightness differences of adjacent pixels. If the brightness difference exceeds the set threshold, it indicates that the pixel is located at the edge of the image. Then, the system further analyzes the edge closure degree in the image. If the edges in the image can form a closed structure, the system considers that this area belongs to the tooth area. Next, the system calculates the gray-scale aggregation degree of the image area. The aggregation degree of gray-scale values is evaluated by measuring the uniformity of pixel gray-scale values within the image area. If the gray-scale change in this area is small, it indicates that the texture of this area is relatively stable, usually the tooth area. Assuming that the gray-scale value range in the image is from 0 to 255, if the average gray-scale value of a certain area is greater than 180 and the gray-scale fluctuation range is less than 10%, the system determines that this area is the tooth area. Finally, the system generates a set of tooth area positioning parameters and records the position and boundary information of each tooth area. Using the formula:

[0089]

[0090] Calculate the gray-scale aggregation degree, where G ig is the gray-scale value of the ig-th pixel, G avg is the average gray-scale value of the image area, ng is the number of pixels in the image, and ΔG is the gray-scale aggregation degree.

[0091] Assume that there are 5 pixels in the image area, and the gray-scale values are G1 = 180, G2 = 185, G3 = 190, G4 = 195, G5 = 200 respectively. Then:

[0092]

[0093] Then, calculate the gray-scale aggregation degree:

[0094]

[0095]

[0096] If the set threshold is 8, the gray-scale aggregation degree ΔG = 6 is less than the threshold, indicating that this area is the tooth area.

[0097] Based on the set of tooth area positioning parameters, the tooth damage detection module extracts the texture distribution within each tooth area, obtains the texture filling density, texture direction change rate, and number of texture continuous units within each area, and detects the tooth damage area by analyzing the texture density to generate a set of tooth damage characteristic indicators;

[0098] The tooth damage detection module extracts the texture distribution of each tooth area based on the tooth area positioning parameter set and calculates relevant damage feature indicators. First, the system extracts the texture density of each tooth area. The texture density reflects the number of textures per unit area. The higher the density, the finer the texture in that area, usually indicating that the area is more intact. By calculating the gray-scale difference and texture complexity of each pixel point in the area, the system can obtain the texture density value of that area. Next, the system calculates the change rate of the texture direction. This indicator measures the directionality of texture changes. If the texture direction changes significantly, there may be cracks in that area. The system also calculates the number of texture continuous units, which represents the continuity of the texture. The higher the number of continuous units, the more stable the texture in that area. If the number of units is small, there may be damage or fractures. Assuming the texture density is 30, the change rate of the texture direction is 20° / cm, and the number of texture continuous units is 3, the system will combine these indicators to determine that there may be damage in that area and generate a tooth damage feature indicator set for further analysis. Using the formula:

[0099]

[0100] Calculate the texture density, where T it is the texture value of the it-th pixel, nt is the number of pixels for texture calculation in the image, and T d is the texture density.

[0101] Assume there are 5 pixels in the image area, and the texture values are T1 = 10, T2 = 12, T3 = 15, T4 = 18, T5 = 20 respectively. Then:

[0102]

[0103] If the set texture density threshold is 20, then the texture density T d = 15 is less than the threshold, indicating that there is potential damage in that area.

[0104] The structure and morphology extraction module extracts the size, contour extension direction, and position index value of the damaged area based on the tooth damage feature indicator set, records the structure and morphology and spatial position of the damaged area, and generates the boundary structure recognition result;

[0105] The structural morphology extraction module extracts the size, contour extension direction, and position index value of the damaged area according to the tooth damage feature index set. First, the system calculates the pixel area of the damaged area based on the texture features of the damaged area. If the area of the damaged area exceeds a set threshold (e.g., 100 pixels), the area is considered severely damaged. Next, the system calculates the contour extension direction of the damaged area, which is determined by analyzing the angle between the long axis of the damaged area and the horizontal axis. If the angle is greater than 10°, it indicates that the damage direction is oblique. The position index value is obtained by recording the relative position of the damaged area in the image. Finally, the system generates the structural morphology information of the damaged area, records the size, shape, and position of the damaged area, and generates the boundary structure recognition result, providing a basis for subsequent oral health assessment. Using the formula:

[0106]

[0107] Calculate the area of the damaged area, where A is the area of the damaged area, and P ia represents whether the ia-th pixel belongs to the damaged area. If it belongs to the damaged area, P ia = 1, otherwise P ia = 0, and na is the total number of pixels in the image.

[0108] Assume there are 100 pixels in the image area, and 50 of them belong to the damaged area, then:

[0109]

[0110] If the set area threshold is 100 pixels and the area A of the damaged area = 50 is less than the threshold, it indicates that the damage to this area is relatively light.

[0111] The color analysis module includes:

[0112] The color offset analysis module calls the boundary structure recognition result, extracts the brightness, hue, and saturation parameters of each pixel block within the tooth area, calculates the offset amplitude between the color parameters at multiple positions in the tooth area of the image and the set color reference value, and generates the color offset coefficient distribution value;

[0113] The specific formula for calculating the offset amplitude between the color parameters at multiple positions in the tooth area of the image and the set color reference value is:

[0114]

[0115] Calculate the color offset coefficient;

[0116] where L k′ is the brightness value of the k'-th pixel block, H k′ is the hue value of the k'-th pixel block, S k′is the saturation value of the k'-th pixel block, L0 is the reference luminance value, H0 is the reference hue value, S0 is the reference saturation value, and ΔC k′ is the color offset amplitude of the k'-th pixel block, w L is the weight coefficient of luminance, w H is the weight coefficient of hue, w S is the weight coefficient of saturation.

[0117] Formula:

[0118]

[0119] Detailed explanation of the formula and the derivation process of formula calculation:

[0120] The formula is used to calculate the color offset coefficient, and the result is used to evaluate the color change amplitude of each pixel block. The result quantifies the color deviation in the tooth area of the image to further analyze the abnormal area;

[0121] Meaning and setting values of parameters:

[0122] L k′ is the luminance value of the k'-th pixel block, representing the luminance of the pixel, and is set to 50;

[0123] H k′ is the hue value of the k'-th pixel block, indicating the hue of the pixel, and is set to 30;

[0124] S k′ is the saturation value of the k'-th pixel block, representing the color saturation of the pixel, and is set to 0.75;

[0125] L0 is the reference luminance value, representing the standard luminance of the image, and is set to 55;

[0126] H0 is the reference hue value, representing the standard hue of the image, and is set to 28;

[0127] S0 is the reference saturation value, representing the standard saturation of the image, and is set to 0.7;

[0128] w L is the luminance weight coefficient, set to 0.4, indicating the weight of luminance in the color offset calculation;

[0129] w H is the hue weight coefficient, set to 0.35, indicating the weight of hue in the color offset calculation;

[0130] w S is the saturation weight coefficient, set to 0.25, indicating the weight of saturation in the color offset calculation.

[0131] Substitute the parameters into the formula for calculation:

[0132]

[0133]

[0134]

[0135]

[0136] ΔC k′ ≈3.37;

[0137] The result 3.37 indicates the color offset amplitude of the k'-th pixel block. The higher the value, the more significant the color change. The result is used for subsequent color anomaly detection and tooth region recognition to determine the abnormal regions in the image.

[0138] The color anomaly detection module detects the color anomaly positions according to the color offset coefficient distribution values and the offset coefficients, and extracts the contour coordinates, area, and color change trend of the anomaly positions to obtain the abnormal region information.

[0139] The color anomaly detection module analyzes and detects the color anomaly positions in the image according to the color offset coefficient distribution values. The system first sorts the color offset coefficients to find the regions with larger offset amplitudes, indicating that the color changes in these regions may exceed the normal range. Then, the system determines whether there is a color anomaly based on the magnitude of the offset amplitude. If the offset amplitude exceeds the set threshold, the system considers that there is a color anomaly in this region. The system further extracts the contour coordinates of the anomaly position, which are extracted through an image segmentation algorithm and represent the boundary of the abnormal region. Then, the system calculates the area of the abnormal region, which is obtained by calculating the total number of pixels in the region. Finally, the system analyzes the color change trend by comparing the color change amplitude in this region to determine whether the color change is temporary or continuous. Suppose the offset amplitude of the abnormal region is 12 and the set threshold is 10, then this region is determined as a color abnormal region. The system will record the contour coordinates and color change trend of this region to generate the abnormal region information. Using the formula:

[0140]

[0141] Calculate the area of the abnormal region, where AY is the area of the abnormal region, and P iy is whether the iy-th pixel in the image belongs to the abnormal region. If it belongs to the abnormal region, P iy = 1, otherwise P iy = 0, and ny is the total number of pixels in the image.

[0142] Suppose there are 100 pixels in the image region, and 60 of them belong to the color abnormal region, then:

[0143]

[0144] If the set area threshold is 80 pixels and the area AY of the abnormal area is 60, which is less than the threshold, it indicates that the degree of abnormality in this area is relatively low.

[0145] Based on the information of the color abnormal area and combined with the already identified tooth damage area, the tooth abnormal position recognition module identifies the type of the color abnormal position, detects the caries and pigment deposition areas, and obtains the tooth abnormal area.

[0146] Based on the information of the color abnormal area and combined with the already identified tooth damage area, the tooth abnormal position recognition module further detects the type of the color abnormal area. First, the system classifies the color abnormal area. According to its offset amplitude, contour shape and position index, it judges whether the area is caries or pigment deposition. If the color offset amplitude is large and the area contour is irregular, the system judges that the area is caries. If the color offset is small and the contour is regular, the system determines that the area is pigment deposition. Then, the system combines the already identified tooth damage area for comprehensive analysis. If the color abnormal area is close to or overlaps with the known damage area in position, the system further confirms that this area may be caries or early damage. Finally, the system generates the tooth abnormal area and outputs the type information of this area. This process is achieved by comparing the geometric features and relative positions of the color abnormal area and the damage area to ensure the accuracy of recognition. Using the formula:

[0147]

[0148] Calculate the overlapping area between the tooth abnormal area and the damage area, where R is the overlapping area, and P ir represents whether the ir-th pixel in the image belongs to the abnormal area and overlaps with the damage area. If it belongs to the overlapping area, P ir = 1, otherwise P ir = 0, and nr is the total number of pixels in the image.

[0149] Suppose there are 100 pixels in the image area, and 40 of them belong to the abnormal area and overlap with the damage area, then:

[0150]

[0151] If the set overlapping area threshold is 50 pixels and the overlapping area R = 40 is less than the threshold, it indicates that the damage in this area is relatively low.

[0152] The model construction module includes:

[0153] The tooth sub-region division module calls the abnormal tooth region, divides the tooth region into multiple sub-regions, extracts the texture distribution width and direction change frequency within each sub-region, and establishes a sub-region division index to generate a tooth sub-region feature index set;

[0154] The tooth sub-region division module calls the abnormal tooth region and divides the tooth region into multiple sub-regions. First, the system divides the tooth region into multiple small blocks through an image segmentation algorithm, and each small block represents a sub-region. The boundary of each sub-region is determined by detecting the texture continuity and contour closure in the image. Texture continuity refers to whether the texture change between adjacent pixels is smooth, and closure indicates whether the contour of the region is completely closed. Then, the system extracts the texture distribution width and direction change frequency of each sub-region. The texture distribution width refers to the distribution range of the texture within the sub-region, and the direction change frequency refers to the frequency of texture direction change. Through these features, the system can judge the structural complexity and morphology of each sub-region. Assuming that the texture width of the sub-region is W = 5 pixels and the texture direction change frequency is F = 3 times / unit region, the system can generate the sub-region feature index set of this region. Using the formula:

[0155]

[0156] Calculate the texture density of the sub-region, where T is the texture density, W is the texture width, and F is the texture direction change frequency.

[0157] Assume W = 5 and F = 3, then:

[0158]

[0159] The texture density of this sub-region is 1.67, indicating that the texture distribution in this region is relatively extensive. The system uses this information as part of the sub-region feature index set for further subsequent processing.

[0160] The point density adjustment module, based on the tooth sub-region feature index set, extracts the texture complexity value and color change rate value within each sub-region, and adjusts the point spacing of each sub-region according to the consistency threshold of texture distribution and the continuity threshold of color gradient to generate a distribution point density adjustment result;

[0161] The dot density adjustment module extracts the texture complexity value and the color change rate value within each sub-region based on the tooth sub-region feature index set. The texture complexity value represents the diversity of textures within the region, and the color change rate value refers to the rate of color change within the region. The system first calculates the texture complexity of each sub-region. A region with a high texture complexity indicates that there are more texture details within the region. Next, the system calculates the color change rate of each sub-region. The color change rate is measured by comparing the color changes of the pixels within the region. If the color change in a certain region is relatively drastic, then its color change rate is high. The system further analyzes the consistency of the texture distribution and the continuity of the color gradient. These two indicators are used to determine the adjustment range of the dot spacing within the region. Specifically, if the texture distribution of a certain sub-region is relatively consistent and the color change is relatively stable, the system will reduce the dot spacing in this region to reduce the computational amount. If the texture distribution is inconsistent and the color change is drastic, the system will increase the dot spacing to capture more detailed information. Suppose the texture complexity of a certain region is C = 4 and the color change rate is R = 2. The system will determine the density adjustment of the dots in this region according to the set threshold. Using the formula:

[0162]

[0163] Calculate the dot density, where D is the dot density, C is the texture complexity, and R is the color change rate.

[0164] Suppose C = 4, R = 2, then:

[0165]

[0166] The dot density of this region is 2, indicating that the system will perform a moderate dot density adjustment on this region to ensure the retention of details.

[0167] The oral three-dimensional reconstruction module reconstructs the three-dimensional geometric model of the patient's teeth according to the distribution dot density adjustment result, extracts the spatial mapping positions of the dental caries and pigment deposition regions, maps them into the model, and obtains the patient's oral model;

[0168] The oral three-dimensional reconstruction module reconstructs the three-dimensional geometric model of the patient's teeth according to the adjusted distribution point density. First, the system calculates the coordinates of each sub-region in three-dimensional space based on the point density of the region. For regions with a high point density, the system accurately reconstructs the shape and structure of the region by increasing the number of points; for regions with a low point density, the system reduces the number of points to reduce the computational complexity. In this way, the system can dynamically adjust the point density in different regions to maintain the details and accuracy of the model. Next, the system marks the spatial positions of the identified caries and pigment deposition regions in the reconstructed three-dimensional model. The caries region usually shows damage or depression in the local structure, while the pigment deposition region shows a region with obvious color changes. The system further confirms the positions of these regions by comparing the color offsets and texture features of these regions and maps them to the three-dimensional model. The specific positions and shapes of these regions are automatically identified and labeled for subsequent diagnosis and treatment plan formulation. Finally, the system generates a complete oral model of the patient, which not only includes the spatial positions and shapes of each tooth but also the regional information of caries and pigment deposition. The accuracy and integrity of this oral model provide very important reference for dentists and help with further health management and treatment decisions.

[0169] The chemical detection module includes:

[0170] The saliva sample collection module calls the patient's oral model to regularly collect the patient's saliva samples, detect the protein concentration and pH value in the patient's oral cavity in each cycle, and generate a cycle sample basic data set;

[0171] The saliva sample collection module calls the patient's oral model to regularly collect the patient's saliva samples. First, the system determines the optimal sampling time point according to the oral model data of the patient, usually collecting saliva once a month. During each collection, the system records the protein concentration and pH value in the saliva. The detection of the protein concentration is carried out by a dedicated sensor, which generates a corresponding signal intensity by reacting with the protein in the saliva sample, and then calculates the protein concentration. The pH value is detected by a pH sensor for the saliva sample. When collecting data, the system stores the protein concentration and pH value of each cycle collected in the database and generates a cycle sample basic data set. Suppose that in a certain cycle, the collected protein concentration is 3.5 mg / dL and the pH value is 6.8. By analyzing these data, the system generates a cycle sample basic data set for subsequent fluctuation trend analysis.

[0172] The fluctuation trend analysis module compares the change amplitudes of protein concentration and pH value in each cycle based on the periodic sample basic data set, extracts the change directions and durations of each item of data in consecutive cycles, analyzes the relative change rate and the fluctuation duration period, and generates the saliva component fluctuation trend value;

[0173] The fluctuation trend analysis module compares the change amplitudes of protein concentration and pH value in each cycle based on the periodic sample basic data set. The system first calculates the difference between the protein concentration and pH value in each cycle to obtain their change amplitudes in different cycles. Then, the system extracts the relative change rate of each item of data, that is, the change rate of concentration or pH value within a certain cycle. The change rate calculation formula is as follows:

[0174]

[0175] where R is the change rate, C t+1 and C t are the protein concentration or pH value at two consecutive time points respectively, and T is the cycle time length. Suppose that within two cycles, the protein concentrations are 3.5 mg / dL and 3.8 mg / dL respectively, and the cycle is 30 days, then:

[0176]

[0177] The calculated change rate is 0.01 mg / dL per day. The system then records this rate and analyzes its fluctuation duration period, that is, analyzes the change trend of concentration or pH value in multiple cycles. The system collates these data into the saliva component fluctuation trend value for subsequent oral health score analysis.

[0178] The oral health score module detects and records the abnormal change trends of the data according to the saliva component fluctuation trend value, combines the abnormal tooth position information, calculates the oral health score of the patient, and obtains the oral environment detection record;

[0179] The specific formula for calculating the oral health score of the patient is:

[0180]

[0181] Calculate the oral health score;

[0182] where H is the oral health score, n′ is the total number of data cycles, i′ is the cycle index, R pi′ is the protein concentration change rate in the i′-th cycle, R phi′ is the pH value change rate in the i′-th cycle, W i′ is the weight coefficient of the i′-th cycle, is the protein concentration change rate in the (i′ - 1)-th cycle, is the rate of change of pH value in the (i'-1)th period, A cav is the area of the caries region, A total is the total tooth area, W cav is the weight coefficient of the caries region.

[0183] Formula:

[0184]

[0185] Detailed explanation of the formula and the derivation process of formula calculation:

[0186] The formula is used to calculate the oral health score of the patient and evaluate the oral health status of the patient;

[0187] Meaning and setting values of parameters:

[0188] H is the oral health score, indicating the oral health status of the patient;

[0189] n' = 5 is the total number of data periods, indicating the number of data collection times in 5 periods;

[0190] i' is the period index, indicating the i'th data collection period, ranging from 1 to n';

[0191] is the rate of change of protein concentration in the i'th period. The data for periods 1 to 5 are set as:

[0192]

[0193] is the rate of change of pH value in the i'th period. The data for periods 1 to 5 are set as:

[0194] W i′ is the weight coefficient in the i'th period, assumed to be 0.2, indicating the weight of each period in the total score;

[0195] and are the protein concentration and the rate of change of pH value in the previous period (i'-1), set as:

[0196] A cav is the area of the caries region, assumed to be 2 cm 2 ;

[0197] A total is the total tooth area, assumed to be 35 cm 2 ;

[0198] W cavis the weight coefficient for the dental caries area, assumed to be 0.5;

[0199] Substitute the parameters into the formula for calculation:

[0200]

[0201] The calculation result is the comprehensive score of the patient's oral health status. This value indicates that the patient's oral health is relatively normal, but it is still necessary to continue monitoring the changes in saliva components and the development of dental caries to further evaluate the oral health risks.

[0202] Please refer to Figure 3 , which provides a method for detecting oral health. This method is applied to a system for detecting oral health, and the method includes:

[0203] S1: Based on the image acquisition device, collect and analyze the patient's oral images in real time. By detecting the edge sharpness, texture distribution, and gray level fluctuations in the images, analyze the change trends of each clarity parameter, calculate the focal length deviation of the images, and adjust the focal length parameters to obtain an oral image dataset;

[0204] S2: Based on the oral image dataset, detect and segment the tooth areas according to the texture features of the images. Combine texture density analysis to detect the tooth damage areas in the images, and record the geometric data of the damage areas to generate a boundary structure recognition result;

[0205] S3: Based on the boundary structure recognition result, analyze the colors of the tooth areas in the images. By calculating the deviations between each color parameter and the healthy reference value, identify the positions with abnormal colors, and combine the tooth damage area information to detect the dental caries and pigment deposition areas to generate tooth abnormal area data;

[0206] S4: Based on the tooth abnormal area data, divide the tooth areas into multiple sub-areas. According to the texture complexity and color change rate of each area, adjust the density of the distribution points within the area, reconstruct the three-dimensional geometric model of the patient's teeth, and map the dental caries and pigment deposition areas to the model to generate a three-dimensional model of the patient's oral cavity;

[0207] S5: Based on the three-dimensional model of the patient's oral cavity, regularly collect the patient's saliva samples, detect the protein concentration and pH value, and by analyzing the fluctuations of the detection data in multiple cycles, identify the abnormal change trends of the data to generate an oral environment detection record.

[0208] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions according to the embodiments of the present invention are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wired (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more collections of available media. The available media can be magnetic media (such as floppy disks, hard disks, magnetic tapes), optical media (such as DVDs), or semiconductor media. The semiconductor media can be a solid-state drive.

[0209] It should be understood that the term "and / or" in this document is merely a description of the association relationship between associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. Here, A and B can be singular or plural. In addition, the character " / " in this document generally represents an "or" relationship between the associated objects before and after, but it may also represent an "and / or" relationship, which can be specifically understood by referring to the context before and after.

[0210] In the present invention, "at least one" means one or more, and "a plurality" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can represent: a, b, c, a - b, a - c, b - c, or a - b - c, where a, b, and c can be single or multiple.

[0211] It should be understood that in various embodiments of the present invention, the magnitudes of the sequence numbers of the above processes do not mean the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0212] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0213] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described devices, apparatuses, and units can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0214] In several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms.

[0215] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0216] In addition, the functional units in each embodiment of the present invention can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit.

[0217] When the above-mentioned function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or a part of this technical solution can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs.

[0218] As described above, the above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A system for detecting oral health, characterized in that, The system includes: The image acquisition module uses an image acquisition device to collect and analyze the patient's oral cavity images in real time, detect the edge sharpness, texture distribution, and gray scale fluctuation in the images in real time, calculate the deviation of the image focal length by analyzing the change trend of each sharpness parameter, adjust the focal length parameter, and obtain an oral cavity image data set; The region detection module calls the oral cavity image data set, detects and segments the tooth region according to the texture features of the image, detects the tooth damage region by analyzing the texture density in the tooth region of the image, records the geometric data of the damage region, and generates a boundary structure recognition result; The color analysis module calls the boundary structure recognition result, analyzes the color of the tooth region in the image, analyzes the deviation between various color parameters and the reference value, identifies the color abnormal position, detects the dental caries and pigment deposition regions, and obtains the tooth abnormal region; The model construction module calls the tooth abnormal region, divides the tooth region into multiple sub-regions, adjusts the density of the distribution points according to the texture complexity and color change rate of each region, reconstructs the three-dimensional geometric model of the patient's teeth, and maps the dental caries and pigment deposition regions into the model to generate a patient oral cavity model.

2. The system for detecting oral health according to claim 1, wherein The oral cavity image data set includes the focal length adjustment parameter value, the image sharpness score result, and the image acquisition timestamp. The boundary structure recognition result is specifically the tooth region boundary line coordinate set, the damage region position index, and the geometric dimension parameter. The tooth abnormal region includes the color offset point distribution map, the abnormal region type identifier, and the tooth surface color fluctuation range. The patient oral cavity model specifically refers to the tooth three-dimensional structure dot matrix, the damage region space mapping index, and the dental caries and deposition marked regions.

3. The system for detecting oral health according to claim 1, wherein The image acquisition module includes: The image sharpness detection module uses an image acquisition device to collect and analyze the patient's oral cavity images in real time, detect the edge sharpness, texture distribution, and gray scale fluctuation in the images, detect the focal length offset by monitoring the real-time change trend of each sharpness parameter in the image, and generate a focal length offset judgment result; The focal length adjustment calculation module calculates the focal length adjustment amplitude by analyzing the change amplitude of each sharpness parameter according to the focal length offset judgment result, and generates an adjusted focal length parameter; The focal length adjustment execution module performs the focal length adjustment based on the adjusted focal length parameter, records the oral cavity image data after the focal length parameter adjustment, and obtains an oral cavity image data set.

4. The system for detecting oral health according to claim 3, wherein, The region detection module includes: The tooth region recognition module calls the oral cavity image data set, identifies the boundary of the tooth region in the image, segments the tooth region by detecting the inner contour continuity, edge closure degree, and region gray scale aggregation degree in the image, obtains the contour index and position identifier corresponding to each tooth structure, and generates a tooth region positioning parameter group; The tooth damage detection module extracts the texture distribution in each tooth region based on the tooth region positioning parameter group, obtains the texture filling density, texture direction change rate, and texture continuous unit number in each region, detects the tooth damage region by analyzing the texture density, and generates a tooth damage feature index set; The structural morphology extraction module extracts the size, contour extension direction, and position index value of the damaged area according to the tooth damage feature index set, records the structural morphology and spatial position of the damaged area, and generates a boundary structure recognition result.

5. The system for detecting oral health according to claim 4, wherein The color analysis module includes: The color offset analysis module calls the boundary structure recognition result, extracts the brightness, hue, and saturation parameters of each pixel block in the tooth area, calculates the offset amplitude between the color parameters at multiple positions in the tooth area of the image and the set color reference value, and generates a color offset coefficient distribution value; The color anomaly detection module detects the color anomaly positions according to the color offset coefficient distribution value and the offset coefficient, and extracts the contour coordinates, area, and color change trend of the anomaly positions to obtain the anomaly area information; The tooth anomaly position recognition module identifies the type of the color anomaly positions based on the color anomaly area information, combines with the already recognized tooth damage areas, detects the caries and pigment deposition areas, and obtains the tooth anomaly areas.

6. The system for detecting oral health according to claim 5, characterized in that, The specific formula for calculating the offset amplitude between the color parameters at multiple positions in the tooth area of the image and the set color reference value is: Calculate the color offset coefficient; Among them, L k′ is the luminance value of the k'-th pixel block, H k′ is the hue value of the k'-th pixel block, S k′ is the saturation value of the k'-th pixel block, L0 is the reference luminance value, H0 is the reference hue value, S0 is the reference saturation value, ΔC k′ is the color offset amplitude of the k'-th pixel block, w L is the weight coefficient of luminance, w H is the weight coefficient of hue, w S is the weight coefficient of saturation.

7. The system for detecting oral health according to claim 5, wherein The model construction module includes: The tooth sub-region division module calls the tooth anomaly areas, divides the tooth area into multiple sub-regions, extracts the texture distribution width and direction change frequency in each sub-region, and establishes a sub-region division index to generate a tooth sub-region feature index set; The point density adjustment module extracts the texture complexity value and color change rate value in each sub-region based on the tooth sub-region feature index set, adjusts the point spacing of each sub-region according to the consistency threshold of the texture distribution and the continuity threshold of the color gradient, and generates a distribution point density adjustment result; The oral cavity three-dimensional reconstruction module reconstructs the three-dimensional geometric model of the patient's teeth according to the distribution point density adjustment result, extracts the spatial mapping positions of the caries and pigment deposition areas, maps them into the model, and obtains the patient's oral cavity model.

8. The system for detecting oral health according to claim 1, characterized in that, The system further includes: The chemical detection module calls the patient's oral cavity model, regularly collects the patient's saliva samples, detects the protein concentration and pH value, analyzes the fluctuations of the data in multiple cycles, detects the abnormal change trend of the data, and generates an oral cavity environment detection record; The oral cavity environment detection record includes the protein concentration change amplitude, pH value fluctuation range, and abnormal change duration period.

9. The system for detecting oral health according to claim 8, wherein The chemical detection module includes: The saliva sample collection module calls the patient's oral cavity model, regularly collects the patient's saliva samples, detects the protein concentration and pH value of the patient's oral cavity in each cycle, and generates a cycle sample basic data set; The fluctuation trend analysis module compares the change amplitudes of the protein concentration and pH value in each cycle based on the cycle sample basic data set, extracts the change direction and duration of each item of data in consecutive cycles, analyzes the relative change rate and fluctuation duration period, and generates a saliva component fluctuation trend value; The oral health scoring module detects and records the abnormal change trend of data according to the salivary component fluctuation trend value, combines the abnormal tooth position information, calculates the oral health score of the patient, and obtains the oral environment detection record; The specific formula for calculating the oral health score of the patient is: Calculate the oral health score; Where H is the oral health score, n′ is the total number of data cycles, and i′ is the cycle index. is the rate of change of protein concentration in the i′-th cycle. is the rate of change of pH value in the i′-th cycle, and W i′ is the weight coefficient of cycle i′. is the rate of change of protein concentration in cycle i′ - 1. is the rate of change of pH value in cycle i′ - 1, and A cav is the area of the caries region, and A total is the total tooth area, and W cav is the weight coefficient of the caries region.

10. A method for detecting oral health, characterized in that, According to the system for detecting oral health according to any one of claims 1-9, the method includes: S1: Based on the image acquisition device, collect and analyze the patient's oral image in real time. By detecting the edge sharpness, texture distribution, and gray scale fluctuation in the image, analyze the change trend of each clarity parameter, calculate the focal length deviation of the image, and adjust the focal length parameter to obtain the oral image data set; S2: Based on the oral image data set, detect and segment the tooth area according to the texture features of the image. Combine texture density analysis to detect the tooth damage area in the image, and record the geometric data of the damage area to generate the boundary structure recognition result; S3: Based on the boundary structure recognition result, analyze the color of the tooth area in the image. By calculating the deviation between each color parameter and the healthy reference value, identify the abnormal color position, and combine the tooth damage area information to detect the caries and pigment deposition areas to generate the tooth abnormal area data; S4: Based on the tooth abnormal area data, divide the tooth area into multiple sub-areas. According to the texture complexity and color change rate of each area, adjust the density of the distribution points in the area, reconstruct the three-dimensional geometric model of the patient's teeth, and map the caries and pigment deposition areas to the model to generate the patient's oral three-dimensional model; S5: Based on the patient's oral three-dimensional model, regularly collect the patient's saliva samples, detect the protein concentration and pH value. By analyzing the fluctuation of the detection data in multiple cycles, identify the abnormal change trend of the data and generate the oral environment detection record.

Images