An oral examination method and system based on big data analysis

Abstract

The application belongs to the technical field of oral health detection, and discloses an oral examination method and system based on big data analysis, which comprises the following steps: collecting oral image data and corresponding image evaluation labels, performing distortion conversion on the oral image data to obtain an oral correction image, performing reflection compensation on the oral correction image to obtain an oral compensation image, performing lesion segmentation on the oral compensation image to obtain a lesion sub-image, constructing an oral health evaluation model based on the lesion sub-image and the corresponding image evaluation labels, and realizing accurate evaluation of oral health based on the oral health evaluation model. The application greatly improves the accuracy of oral health evaluation of different patients and further promotes the realization of personalized oral care and health management.

Description

Technical Field

[0001] The present invention relates to the technical field of oral health detection, and more specifically, to an oral examination method and system based on big data analysis. Background Art

[0002] With the rapid development of modern dentistry, oral examinations are becoming increasingly important in disease prevention, early diagnosis, and personalized treatment. However, existing oral examination technologies still face many technical challenges. For example, the inner surface of the oral cavity has complex curves and irregular shapes, including the natural curvature and folding of teeth, tongue, gums, and other parts. These structures cause severe image distortion during imaging, making it difficult to accurately restore the position and shape of the lesion area, resulting in low correction accuracy and difficulty meeting clinical requirements. During oral examinations, the presence of saliva is inevitable. The saliva surface has strong light reflection properties, especially under high-brightness light sources, which easily produce strong reflective areas in the imaging. These reflective areas will obscure the details of the lesion area and reduce the recognition rate of the lesion. The color and texture of oral tissues vary little, especially the difference in imaging characteristics between the lesion area and the surrounding healthy tissue is not obvious. Due to the blurred boundaries, methods based on traditional edge detection or simple threshold segmentation often have difficulty in effectively segmenting the lesion area. In addition, complex lesions may overlap with each other, further increasing the difficulty of segmentation.

[0003] In view of this, the present invention proposes an oral examination method and system based on big data analysis to solve the above problems. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solution: an oral examination method based on big data analysis, characterized in that it includes:

[0005] S1. Collect oral image data and corresponding image evaluation labels, perform distortion transformation on the oral image data, and obtain an oral correction image;

[0006] S2. Perform reflection compensation on the oral correction image to obtain an oral compensation image;

[0007] S3, performing lesion segmentation on the oral compensation image to obtain a lesion sub-image;

[0008] S4. Construct an oral health assessment model based on the lesion sub-graph and the corresponding image assessment labels, and achieve accurate assessment of oral health based on the oral health assessment model.

[0009] Further, the acquisition method of the oral cavity image data comprises: using a high-definition intraoral camera equipped with multi-angle imaging function and macro lens, setting the angle and position of the device, combining with the flexible arm mechanical auxiliary system, using the geometric registration technology of multi-view images to splice different view images, and generating a panoramic oral cavity image as the oral cavity image data.

[0010] Further, the distortion conversion method of the oral cavity image data comprises:

[0011] The RGB weighted average method is used to gray the oral cavity image data to obtain an oral cavity grayscale image; for each pixel in the oral cavity grayscale image, the pixel at the lower left corner of the oral cavity grayscale image is taken as the pixel center, and the pixel height and the pixel width of the oral cavity grayscale image are taken as the coordinate scale to assign coordinates to each pixel to obtain the pixel coordinates of each pixel; the horizontal gradient operator in the edge detection operator is used to calculate the horizontal pixel gradient of the pixels in the oral cavity grayscale image, and the vertical gradient operator in the edge detection operator is used to calculate the vertical pixel gradient of the pixels in the oral cavity grayscale image; the pixel tensor of each pixel is constructed based on the horizontal pixel gradient and the vertical pixel gradient, and the distortion evaluation of each pixel is evaluated based on the pixel tensor to obtain the distortion evaluation value;

[0012] A preset distortion threshold is set, and the pixels with a distortion evaluation value greater than or equal to the distortion threshold are marked as distortion pixels; the distortion displacement of each pixel marked as a distortion pixel is explored to obtain a distortion displacement field; the pixel transfer of the distortion pixels is performed based on the distortion displacement field, the coordinates of the distortion pixels are calculated based on the horizontal distortion offset and the vertical distortion offset in the distortion displacement field to obtain offset coordinates, the distortion pixels are replaced with the pixels at the offset coordinates, the pixel value at the distortion pixel is filled with the pixel mean value of the neighboring pixels directly adjacent to the distortion pixel, and an oral cavity correction image is obtained.

[0013] Further, the formula for evaluating the distortion of each pixel is:

[0014] ; wherein, represents the distortion evaluation value of the pixel at the pixel coordinate , represents the modulus of the pixel tensor of the pixel at the pixel coordinate , represents the horizontal pixel gradient, the vertical pixel gradient, represents the horizontal axis value of the pixel coordinate, represents the vertical axis value of the pixel coordinate.

[0015] Further, the method for exploring the distortion displacement of each pixel marked as a distortion pixel comprises:

[0016] Initialize the distortion offset group A, initialize the optimal offset group to a null value, the distortion offset group includes the horizontal distortion offset and the vertical distortion offset, and construct the distortion energy field function based on the distortion offset group. The expression of the distortion energy field function is:

[0017] ;in, represents the distortion energy, represents the horizontal weight coefficient, represents the horizontal offset gradient, represents the vertical weight coefficient, represents the vertical offset gradient, represents the pixel distortion parameter, The pixel coordinates are The pixel grayscale of the pixel at The pixel coordinates are The pixel grayscale of the pixel at represents the horizontal distortion offset, represents the vertical distortion offset, Represents the integration of the horizontal axis value. represents the integration of the vertical axis value; the distortion energy of each distortion offset group is calculated based on the distortion energy field function, and the distortion offset group is initially selected using the tournament selection algorithm based on the distortion energy to obtain the initial distortion offset group;

[0018] The initial distortion offset groups are combined in pairs to obtain offset pairs, and the horizontal distortion offset and vertical distortion offset in each offset pair are recombined to obtain a child combination different from the initial distortion offset group; the distortion energy of each child combination is calculated by the distortion energy field function, and the child combination with a distortion energy less than the distortion energy of the offset pair is regarded as the new generation; a mutation scale is preset, and the mutation scale is an integer greater than zero. Based on the mutation scale, the horizontal distortion offset or vertical distortion offset of the new generation is randomly mutated, and the horizontal distortion offset is randomly increased or decreased by one mutation scale, and the vertical distortion offset is randomly increased or decreased by one mutation scale. The scale is changed to obtain a mutation combination; the distortion energy of each mutation combination is calculated through the distortion energy field function, and the mutation combination with a distortion energy less than the distortion energy of the new generation is selected as the preferred mutation, and the preferred mutation with the smallest distortion energy is selected as the candidate combination. When the optimal offset group is null, the candidate combination is used as the new optimal offset group. When the optimal offset group is not null, if the distortion energy of the candidate combination is less than the distortion energy of the optimal offset group, the candidate combination is used as the new optimal offset group; the preferred mutation is used as the new initial distortion offset group, and the process is repeated until the distortion energy of the optimal offset group converges, and the optimal offset group at this time is output as the distortion displacement field.

[0019] Furthermore, the method of performing reflection compensation on the oral correction image includes:

[0020] A set of scale components is preset, and illumination estimation is performed on each pixel in the oral correction image based on the set of scale components. The formula for illumination estimation for each pixel in the oral correction image is:

[0021] ;in, The pixel coordinates are The illumination component of the pixel at represents the size of the scale component set, Represents the scale component set The component weights of the scale components satisfy the component constraints, which are: , Represents the scale component set A scale component, represents the convolution function; based on the illumination component, the reflection of each pixel in the oral correction image is evaluated to obtain the reflection component; based on the reflection component, the adaptive compensation estimation is performed on each pixel in the oral correction image to obtain the compensation factor; based on the compensation factor, light compensation is performed on each pixel in the oral correction image to obtain the compensated pixel, and all compensated pixels constitute the oral compensation map.

[0022] Furthermore, the formula for performing reflection evaluation on each pixel in the oral correction image is:

[0023] ;in, The pixel coordinates are The reflection component of the pixel at , The pixel coordinates are The pixel grayscale of the pixel at represents the offset constant;

[0024] The formula for adaptive compensation estimation for each pixel in the oral correction image is:

[0025] ;in, The pixel coordinates are The compensation factor for the pixel at , represents the compensation control coefficient, The pixel coordinates are The reflection gradient of the reflection component of the pixel at .

[0026] Furthermore, the method of performing lesion segmentation on the oral compensation image includes:

[0027] A pixel search radius is preset. For pixel B in the oral compensation image, the pixel distance between other pixels and pixel B is calculated using the Euclidean distance formula. The pixel neighborhood set of pixel B is composed of other pixels whose coordinate distance to pixel B is less than the pixel search radius. Based on the pixel neighborhood, a density evaluation is performed on each pixel in the oral compensation image to obtain the pixel grayscale density. Based on the pixel grayscale density, a distance measurement is performed on the pixels in the oral compensation image to obtain the pixel distance.

[0028] Based on the pixel grayscale density, a clustering algorithm is used to cluster the pixels in the oral compensation image to obtain pixel density clusters; a density threshold is preset, the density mean of each pixel density cluster is calculated, and the pixel density clusters with a density mean greater than the density threshold are marked as density lesion areas; based on the pixel distance, a clustering algorithm is used to cluster the pixels in the oral compensation image to obtain pixel distance clusters; a distance threshold is preset, the distance mean of each pixel distance cluster is calculated, and the pixel distance clusters with a distance mean greater than the distance threshold are marked as distance lesion areas; a regional overlap comparison is performed on the density lesion area and the distance lesion area, and pixels that are both density lesion areas and distance lesion areas are screened out as lesion pixels. The area where lesion pixels are clustered in the oral compensation image is cropped to obtain a lesion sub-image.

[0029] Furthermore, the formula for evaluating the grayscale density of each pixel in the oral compensation image is:

[0030] ;in, Represents the oral compensation diagram The pixel grayscale density of pixels, represents the size of the pixel neighborhood set, Represents the oral compensation diagram The pixel grayscale of pixels, Represents the first pixel in the neighborhood The pixel grayscale of pixels, Represents the grayscale control parameter, Represents the oral compensation diagram pixel and the pixel neighborhood set The pixel distance of pixels, represents the pixel search radius;

[0031] The formula for measuring the distance between pixels in the oral compensation image is:

[0032] ;in, Represents the oral compensation diagram The pixel distance of pixels, Represents the oral compensation diagram The pixel grayscale density of pixels, a pixel gray density representing a pixel in a pixel neighborhood set, a pixel gray density representing a pixel in a pixel neighborhood set, a density weight representing a density weight, a filter function representing a filter function.

[0033] An oral examination system based on big data analysis comprises:

[0034] A data acquisition processing module: acquiring oral image data and corresponding image evaluation labels, performing distortion conversion on the oral image data to obtain an oral correction image;

[0035] An image compensation module: performing anti-reflection compensation on the oral correction image to obtain an oral compensation image;

[0036] A region segmentation module: performing lesion segmentation on the oral compensation image to obtain a lesion sub-image;

[0037] A model construction module: constructing an oral health evaluation model based on the lesion sub-image and the corresponding image evaluation labels, and realizing accurate evaluation of oral health based on the oral health evaluation model.

[0038] The technical effects and advantages of the oral examination method and system based on big data analysis of the present application are as follows:

[0039] The present application effectively identifies and corrects the distortion area in the oral image by performing distortion conversion on the oral image data, ensures the accuracy and detail retention of the image, especially when processing the corners, curved surfaces or reflection areas in the oral cavity, effectively reduces the influence of deformation and improves the image quality; by performing anti-reflection compensation on the oral correction image, the image deviation caused by uneven illumination can be eliminated, especially under high-intensity light source, the saliva surface has strong light reflection characteristics, which is easy to produce strong reflection area in imaging, the compensated image can better display the lesion area, so that the subsequent lesion identification is more accurate; by performing lesion segmentation on the oral compensation image, the lesion area in the image is effectively identified, the non-lesion area is effectively filtered out, the accuracy of segmentation is improved, the lesion area is accurately positioned, and human error is reduced; through the combination of big data analysis and deep learning model, the oral health of different patients can be accurately evaluated, so as to promote the realization of personalized oral care and health management. BRIEF DESCRIPTION OF DRAWINGS

[0040] Figure 1 It is a schematic diagram of an oral examination method based on big data analysis of the present application;

[0041] Figure 2 It is a schematic diagram of an oral examination system based on big data analysis of the present application. DETAILED DESCRIPTION

[0042] With reference to the accompanying drawings, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0043] Embodiment 1

[0044] Please refer to Figure 1 The oral examination method based on big data analysis comprises:

[0045] S1, collecting oral image data and corresponding image evaluation labels, performing distortion conversion on the oral image data to obtain an oral correction image;

[0046] S2, performing light compensation on the oral correction image to obtain an oral compensation image;

[0047] S3, performing lesion segmentation on the oral compensation image to obtain a lesion sub-image;

[0048] S4, constructing an oral health evaluation model based on the lesion sub-image and the corresponding image evaluation labels, and realizing accurate evaluation of oral health based on the oral health evaluation model.

[0049] The acquisition method of the oral image data comprises: using a high-definition oral endoscope equipped with multi-angle imaging function and a macro lens, setting the angle and position of the device, combining a flexible arm mechanical auxiliary system, making the image cover the key details of different areas of the oral cavity, using multi-view image geometric registration technology to splice different view images, and generating a panoramic oral image as the oral image data.

[0050] The image evaluation label is evaluation information of the oral image data, for example, gum health, gum bleeding and gum inflammation of gum type, initial caries (early tooth cavity mark), severe caries (deep tooth cavity or decay), periodontal health, periodontitis, periodontal damage, etc. of oral type.

[0051] The distortion conversion method of the oral image data comprises:

[0052] The RGB weighted average method is used to gray the oral cavity image data to obtain an oral cavity grayscale image; for each pixel in the oral cavity grayscale image, the pixel at the lower left corner of the oral cavity grayscale image is taken as a pixel center, and the pixel height and the pixel width of the oral cavity grayscale image are taken as coordinate scales to perform coordinate assignment on each pixel (each horizontal increase of a pixel increases the horizontal axis value of the pixel by one scale, and each vertical increase of a pixel increases the vertical axis value of the pixel by one scale) to obtain the pixel coordinates of each pixel; a horizontal gradient operator in an edge detection operator is used to perform horizontal gradient calculation on the pixels in the oral cavity grayscale image to obtain horizontal pixel gradients, and a vertical gradient operator in the edge detection operator is used to perform vertical gradient calculation on the pixels in the oral cavity grayscale image to obtain vertical pixel gradients; common edge detection operators include Sobel operators and Prewitt operators; a pixel tensor of each pixel is constructed based on the horizontal pixel gradients and the vertical pixel gradients; the expression of the pixel tensor is ; wherein, represents the pixel tensor of the pixel with the pixel coordinates represents the horizontal pixel gradient, the vertical pixel gradient, represents the horizontal axis value of the pixel coordinates, represents the vertical axis value of the pixel coordinates; distortion evaluation is performed on each pixel based on the pixel tensor, and the formula for the distortion evaluation on each pixel is:

[0053] ; wherein, represents the distortion evaluation value of the pixel with the pixel coordinates represents the modulus of the pixel tensor of the pixel with the pixel coordinates The distortion evaluation value of each pixel is obtained by performing distortion evaluation on each pixel, and the distortion region in the image can be identified based on the distortion evaluation value; the greater the distortion evaluation value of a pixel, the stronger the deformation of the pixel; for example, the corners, curved surfaces or reflective regions in the oral cavity image may cause large gradient changes, thereby generating a large distortion evaluation value.

[0054] A preset distortion threshold is set by a person skilled in the art based on actual conditions; pixels with a distortion evaluation value greater than or equal to the distortion threshold are marked as distortion pixels based on the distortion threshold; distortion displacement exploration is performed on each pixel marked as a distortion pixel, a group A distortion offset group is initialized, and the best offset group is initialized as a null value; the distortion offset group includes a horizontal distortion offset and a vertical distortion offset; a distortion energy field function is constructed based on the distortion offset group, and the expression of the distortion energy field function is:

[0055] ; wherein, represents the distortion energy, ​​Represents the horizontal weight coefficient, which is used to control the degree of influence of horizontal smoothness. represents the horizontal offset gradient, Represents the vertical weight coefficient, which is used to control the degree of influence of vertical smoothness. represents the vertical offset gradient, Represents the pixel distortion parameter, which is used to control the influence of pixel error. The pixel coordinates are The pixel grayscale of the pixel at The pixel coordinates are The pixel grayscale of the pixel at represents the horizontal distortion offset, represents the vertical distortion offset, Represents the integration of the horizontal axis value. Represents the integration of the vertical axis value;

[0056] The distortion energy of each distortion offset group is calculated based on the distortion energy field function. The distortion offset group is initially selected based on the distortion energy using the tournament selection algorithm to obtain the initial distortion offset group. The initial distortion offset groups are combined in pairs to obtain offset pairs. The horizontal distortion offset and the vertical distortion offset in each offset pair are recombined to obtain a child combination different from the initial distortion offset group. The distortion energy of each child combination is calculated using the distortion energy field function. The child combination with a distortion energy less than that of the offset pair is taken as the new generation. A mutation scale is preset, which is an integer greater than zero. The horizontal distortion offset or vertical distortion offset of the new generation is randomly mutated based on the mutation scale. The horizontal distortion offset is Randomly increase or decrease one mutation scale, and randomly increase or decrease one mutation scale for the vertical distortion offset to obtain a mutation combination; calculate the distortion energy of each mutation combination through the distortion energy field function, and select the mutation combination with a distortion energy less than the distortion energy of the new generation as the preferred mutation, and select the preferred mutation with the smallest distortion energy as the candidate combination. When the optimal offset group is null, the candidate combination is used as the new optimal offset group. When the optimal offset group is not null, if the distortion energy of the candidate combination is less than the distortion energy of the optimal offset group, the candidate combination is used as the new optimal offset group; use the preferred mutation as the new initial distortion offset group, repeat until the distortion energy of the optimal offset group converges, and output the optimal offset group at this time as the distortion displacement field;

[0057] The distorted pixels are transferred based on the distortion displacement field, and the coordinates of the distorted pixels are calculated using the horizontal distortion offset and the vertical distortion offset in the distortion displacement field (the horizontal axis value is summed with the horizontal distortion offset, and the vertical axis value is summed with the vertical distortion offset) to obtain the offset coordinates. The distorted pixels are replaced by the pixels at the offset coordinates, and the pixels directly adjacent to the distorted pixels are used as neighborhood pixels. The pixel values ​​at the distorted pixels are filled with the pixel mean of the neighborhood pixels to obtain the oral correction image. The filling based on the pixel mean can better preserve the details of the image and avoid excessive smoothing of the image, ensuring that the quality of the corrected image reaches a high level, especially in terms of preserving details and textures.

[0058] Methods for compensating for reflections in orthodontic images include:

[0059] A set of scale components is preset, and illumination estimation is performed on each pixel in the oral correction image based on the set of scale components. The formula for illumination estimation for each pixel in the oral correction image is:

[0060] ;in, The pixel coordinates are The illumination component of the pixel at represents the size of the scale component set, Represents the scale component set The component weights of the scale components are used to control the influence of different scale components on the illumination component. The component weights satisfy the component constraints, which are: , Represents the scale component set A scale component, Represents the convolution function, which is used to represent the scale component The kernel function for pixel coordinates is The pixels at are convolved. Common kernel functions include one-dimensional Gaussian kernel function and two-dimensional Gaussian kernel function. The reflection evaluation of each pixel in the oral correction image is performed based on the illumination component. The formula for the reflection evaluation of each pixel in the oral correction image is:

[0061] ;in, The pixel coordinates are The reflection component of the pixel at , The pixel coordinates are The pixel grayscale of the pixel at Represents an offset constant, which is used to smooth the reflection component. Based on the reflection component, adaptive compensation estimation is performed on each pixel in the oral correction image. The formula for adaptive compensation estimation for each pixel in the oral correction image is:

[0062] ;in, The pixel coordinates are The compensation factor for the pixel at , Represents the compensation control coefficient, which is used to adjust the size of the compensation factor. It is set by technicians in this field based on actual conditions. The larger the compensation control coefficient, the more obvious the compensation effect for illumination changes. When the compensation control coefficient is small, the image adjustment is more stable. The pixel coordinates are The reflection gradient of the reflection component of the pixel at ; light compensation is performed on each pixel in the oral correction image based on the compensation factor, and the formula for light compensation for each pixel in the oral correction image is:

[0063] ;in, The pixel coordinates are The compensation pixels of the pixels at , all the compensation pixels constitute the oral compensation map;

[0064] The compensation factor is used to compensate the oral correction image so that the edge and texture areas remain clear, avoiding excessive smoothing that causes loss of image details. The shadow and reflective areas are compensated for brightness, making the lesion area more obvious and facilitating subsequent lesion identification. The scale component set is the scale of the image convolution. Common convolution scales are: scale, Scale and Scale, etc.; the preferred scale component set in this embodiment is scale, Scale and scale.

[0065] Methods for segmenting lesions in oral compensation images include:

[0066] A pixel search radius is preset. For pixel B in the oral compensation map, the Euclidean distance formula is used to calculate the pixel distance between other pixels and pixel B. The pixel neighborhood set of pixel B is composed of other pixels whose coordinate distance to pixel B is less than the pixel search radius. Density evaluation is performed on each pixel in the oral compensation map based on the pixel neighborhood. The formula for grayscale density evaluation of each pixel in the oral compensation map is: ;in, Represents the oral compensation diagram The pixel grayscale density of pixels, represents the size of the pixel neighborhood set, Represents the oral compensation diagram The pixel grayscale of pixels, Represents the first pixel in the neighborhood The pixel grayscale of pixels, Represents the grayscale control parameter, which is used to control the grayscale difference. Represents the oral compensation diagram pixel and the pixel neighborhood set The pixel distance of pixels, Represents the pixel search radius; the distance measurement of pixels in the oral compensation image is performed based on the pixel grayscale density. The formula for the distance measurement of pixels in the oral compensation image is: ;in, Represents the oral compensation diagram The pixel distance of pixels, Represents the oral compensation diagram The pixel grayscale density of pixels, Represents the first pixel in the neighborhood The pixel grayscale density of pixels, Represents the density weight, which is used to smooth the effect of density on pixel distance. Represents the filter function, which means that only Greater than The pixels involved in the calculation;

[0067] Based on the pixel grayscale density, a clustering algorithm is used to cluster the pixels in the oral compensation image to obtain pixel density clusters; a density threshold is preset, the density mean of each pixel density cluster is calculated, and the pixel density clusters with a density mean greater than the density threshold are marked as density lesion areas; lesion areas usually appear as areas with higher local density, which may be formed by irregular tissue structures, masses or lesions. These areas are relatively dense and have fewer surrounding pixels; based on pixel distance, a clustering algorithm is used to cluster the pixels in the oral compensation image to obtain pixel distance clusters; a distance threshold is preset, the density mean of each pixel density cluster is calculated, and the pixel density clusters with a density mean greater than the density threshold are marked as density lesion areas. The distance mean of each pixel from the cluster is calculated, and the pixel distance cluster with a distance mean greater than the distance threshold is marked as the distance lesion area; the lesion area is usually far away from the normal area (non-lesion area) in space, and far away from other normal areas with low density; commonly used clustering algorithms include K-Means clustering algorithm and hierarchical clustering algorithm; the density lesion area and the distance lesion area are compared for regional overlap, and pixels that are both density lesion areas and distance lesion areas are screened as lesion pixels. The area where lesion pixels are concentrated in the oral compensation image is cropped to obtain the lesion sub-image.

[0068] The construction methods of oral health assessment model include:

[0069] Based on the lesion sub-images and the corresponding image evaluation labels, the CNN model is used as the initial model of the oral health evaluation model, the lesion sub-images and the corresponding image evaluation labels are used as training data, and the training data is used as the training sample set. The CNN model is trained using the training sample set, and the lesion sub-images and the corresponding image evaluation labels are used as the input data of the oral health evaluation model, and the predicted evaluation labels are used as the output data of the oral health evaluation model; minimizing the error between the actual image evaluation labels and the evaluation labels predicted by the oral health evaluation model is used as the training goal, and the recall rate function is used as the loss function of the oral health evaluation model. When the loss function reaches convergence, the training is stopped to obtain the oral health evaluation model.

[0070] This embodiment effectively identifies and corrects distorted areas in oral images by performing distortion conversion on oral image data, ensuring image accuracy and detail retention, especially when processing corners, curved surfaces or reflective areas in the oral cavity, effectively reducing the impact of deformation and improving image quality; by performing reflection compensation on oral correction images, image deviations caused by uneven lighting can be eliminated, especially under high-brightness light sources. The saliva surface has strong light reflection characteristics and easily produces strong reflective areas in the imaging. The compensated image can better display the lesion area, making subsequent lesion identification more accurate; by performing lesion segmentation on the oral compensation image, the lesion area in the image is effectively identified, and non-lesion areas are effectively filtered out, thereby improving the accuracy of segmentation, accurately locating the lesion area, and reducing human errors; through the combination of big data analysis and deep learning models, the oral health of different patients can be accurately assessed, thereby promoting the realization of personalized oral care and health management.

[0071] Example 2

[0072] See also Figure 2 As shown, for parts not described in detail in this embodiment, please refer to the description of Example 1. An oral examination system based on big data analysis is provided, including:

[0073] Data acquisition and processing module: collects oral image data and corresponding image evaluation labels, performs distortion conversion on the oral image data, and obtains oral correction images;

[0074] Image compensation module: performs reflection compensation on the oral correction image to obtain the oral compensation image;

[0075] Region segmentation module: performs lesion segmentation on the oral compensation image to obtain lesion sub-images;

[0076] Model building module: Build an oral health assessment model based on lesion sub-graphs and corresponding image assessment labels, and achieve accurate assessment of oral health based on the oral health assessment model;

[0077] The modules are connected via wired and / or wireless means to achieve data transmission between modules.

[0078] Example 3

[0079] This embodiment discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the operation mode of the oral examination method based on big data analysis provided above is implemented.

[0080] Since the electronic device introduced in this embodiment is an electronic device used to implement an oral examination method based on big data analysis in the embodiment of this application, based on the oral examination method based on big data analysis introduced in the embodiment of this application, those skilled in the art can understand the specific implementation of the electronic device of this embodiment and its various variations, so how the electronic device implements the method in the embodiment of this application will not be described in detail here. As long as those skilled in the art implement the electronic device used in the oral examination method based on big data analysis in the embodiment of this application, it falls within the scope of protection to be provided by this application.

[0081] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters and thresholds in the formulas are set by technicians in this field according to actual conditions.

[0082] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiment. All technical solutions based on the principles of the present invention are within the scope of protection of the present invention. It should be noted that for users of ordinary skill in the art, various improvements and modifications that do not depart from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. An oral examination method based on big data analysis, characterized in that: include: S1. Collect oral image data and corresponding image evaluation labels, perform distortion transformation on the oral image data, and obtain an oral correction image; S2. Perform reflection compensation on the oral correction image to obtain an oral compensation image; S3, performing lesion segmentation on the oral compensation image to obtain a lesion sub-image; S4. Build an oral health assessment model based on the lesion sub-graphs and the corresponding image assessment labels, and achieve accurate assessment of oral health based on the oral health assessment model; The method of performing distortion conversion on the oral image data includes: The oral image data is grayscaled using the RGB weighted average method to obtain an oral grayscale image. For each pixel in the oral grayscale image, the pixel in the lower left corner of the oral grayscale image is used as the pixel center, and the pixel height and pixel width of the oral grayscale image are used as the coordinate scale to assign coordinates to each pixel to obtain the pixel coordinates of each pixel. The horizontal gradient operator in the edge detection operator is used to calculate the horizontal gradient of the pixels in the oral grayscale image to obtain the horizontal pixel gradient, and the vertical gradient operator in the edge detection operator is used to calculate the vertical gradient of the pixels in the oral grayscale image to obtain the vertical pixel gradient. A pixel tensor for each pixel is constructed based on the horizontal pixel gradient and the vertical pixel gradient. The distortion of each pixel is evaluated based on the pixel tensor to obtain a distortion estimate. A distortion threshold is preset, and pixels whose distortion estimate is greater than or equal to the distortion threshold are marked as distorted pixels; a distortion displacement is explored for each pixel marked as a distorted pixel to obtain a distortion displacement field; pixel shifting is performed on the distorted pixels based on the distortion displacement field, and the coordinates of the distorted pixels are calculated using the horizontal distortion offset and the vertical distortion offset in the distortion displacement field to obtain offset coordinates. The pixels at the offset coordinates are replaced by the distorted pixels, and the pixels directly adjacent to the distorted pixels are used as neighboring pixels. The pixel value at the distorted pixel is filled with the pixel mean of the neighboring pixels to obtain a dental correction image; The formula for evaluating the distortion of each pixel is: ;in, The pixel coordinates are The distortion estimate of the pixel at , The pixel coordinates are The modulus of the pixel tensor at pixel , represents the horizontal pixel gradient, vertical pixel gradient, The horizontal axis value represents the pixel coordinate. The vertical axis value representing the pixel coordinate; The method of performing distortion displacement exploration on each pixel marked as a distorted pixel includes: Initialize the distortion offset group A, initialize the optimal offset group to a null value, the distortion offset group includes the horizontal distortion offset and the vertical distortion offset, and construct the distortion energy field function based on the distortion offset group. The expression of the distortion energy field function is: ;in, represents the distortion energy, represents the horizontal weight coefficient, represents the horizontal offset gradient, represents the vertical weight coefficient, represents the vertical offset gradient, represents the pixel distortion parameter, The pixel coordinates are The pixel grayscale of the pixel at The pixel coordinates are The pixel grayscale of the pixel at represents the horizontal distortion offset, represents the vertical distortion offset, Represents the integration of the horizontal axis value. represents the integration of the vertical axis value; the distortion energy of each distortion offset group is calculated based on the distortion energy field function, and the distortion offset group is initially selected using the tournament selection algorithm based on the distortion energy to obtain the initial distortion offset group; The initial distortion offset groups are combined in pairs to obtain offset pairs, and the horizontal distortion offset and vertical distortion offset in each offset pair are recombined to obtain a child combination different from the initial distortion offset group; the distortion energy of each child combination is calculated by the distortion energy field function, and the child combination with a distortion energy less than the distortion energy of the offset pair is regarded as the new generation; a mutation scale is preset, and the mutation scale is an integer greater than zero. Based on the mutation scale, the horizontal distortion offset or vertical distortion offset of the new generation is randomly mutated, and the horizontal distortion offset is randomly increased or decreased by one mutation scale, and the vertical distortion offset is randomly increased or decreased by one mutation scale. The scale is changed to obtain a mutation combination; the distortion energy of each mutation combination is calculated through the distortion energy field function, and the mutation combination with a distortion energy less than the distortion energy of the new generation is selected as the preferred mutation, and the preferred mutation with the smallest distortion energy is selected as the candidate combination. When the optimal offset group is null, the candidate combination is used as the new optimal offset group. When the optimal offset group is not null, if the distortion energy of the candidate combination is less than the distortion energy of the optimal offset group, the candidate combination is used as the new optimal offset group; the preferred mutation is used as the new initial distortion offset group, and the process is repeated until the distortion energy of the optimal offset group converges, and the optimal offset group at this time is output as the distortion displacement field.

2. The oral examination method based on big data analysis according to claim 1, characterized in that: The method for acquiring the oral image data includes: using a high-definition oral endoscope equipped with a multi-angle imaging function and a macro lens, setting the angle and position of the device, combining a flexible arm mechanical assistance system, and utilizing the geometric alignment technology of multi-view images to splice images from different perspectives. The generated panoramic oral image is the oral image data.

3. The oral examination method based on big data analysis according to claim 2, characterized in that: The method of performing reflection compensation on the oral correction image includes: A set of scale components is preset, and illumination estimation is performed on each pixel in the oral correction image based on the set of scale components. The formula for illumination estimation for each pixel in the oral correction image is: ;in, The pixel coordinates are The illumination component of the pixel at represents the size of the scale component set, Represents the scale component set The component weights of the scale components satisfy the component constraints, which are: , Represents the scale component set A scale component, represents the convolution function; based on the illumination component, the reflection of each pixel in the oral correction image is evaluated to obtain the reflection component; based on the reflection component, the adaptive compensation estimation is performed on each pixel in the oral correction image to obtain the compensation factor; based on the compensation factor, light compensation is performed on each pixel in the oral correction image to obtain the compensated pixel, and all compensated pixels constitute the oral compensation map.

4. The oral examination method based on big data analysis according to claim 3, characterized in that: The formula for evaluating the reflection of each pixel in the orthodontic image is: ;in, The pixel coordinates are The reflection component of the pixel at , The pixel coordinates are The pixel grayscale of the pixel at represents the offset constant; The formula for adaptive compensation estimation for each pixel in the oral correction image is: ;in, The pixel coordinates are The compensation factor for the pixel at , represents the compensation control coefficient, The pixel coordinates are The reflection gradient of the reflection component of the pixel at .

5. The oral examination method based on big data analysis according to claim 4, characterized in that: The method of performing lesion segmentation on the oral compensation image includes: A pixel search radius is preset. For pixel B in the oral compensation image, the pixel distance between other pixels and pixel B is calculated using the Euclidean distance formula. The pixel neighborhood set of pixel B is composed of other pixels whose coordinate distance to pixel B is less than the pixel search radius. Based on the pixel neighborhood, a density evaluation is performed on each pixel in the oral compensation image to obtain the pixel grayscale density. Based on the pixel grayscale density, a distance measurement is performed on the pixels in the oral compensation image to obtain the pixel distance. Based on the pixel grayscale density, a clustering algorithm is used to cluster the pixels in the oral compensation image to obtain pixel density clusters; a density threshold is preset, the density mean of each pixel density cluster is calculated, and the pixel density clusters with a density mean greater than the density threshold are marked as density lesion areas; based on the pixel distance, a clustering algorithm is used to cluster the pixels in the oral compensation image to obtain pixel distance clusters; a distance threshold is preset, the distance mean of each pixel distance cluster is calculated, and the pixel distance clusters with a distance mean greater than the distance threshold are marked as distance lesion areas; a regional overlap comparison is performed on the density lesion area and the distance lesion area, and pixels that are both density lesion areas and distance lesion areas are screened out as lesion pixels. The area where lesion pixels are clustered in the oral compensation image is cropped to obtain a lesion sub-image.

6. The oral examination method based on big data analysis according to claim 5, characterized in that: The formula for evaluating the grayscale density of each pixel in the oral compensation image is: ;in, Represents the oral compensation diagram The pixel grayscale density of pixels, represents the size of the pixel neighborhood set, Represents the oral compensation diagram The pixel grayscale of pixels, Represents the first pixel in the neighborhood The pixel grayscale of pixels, Represents the grayscale control parameter, Represents the oral compensation diagram pixel and the pixel neighborhood set The pixel distance of pixels, represents the pixel search radius; The formula for measuring the distance between pixels in the oral compensation image is: ;in, Represents the oral compensation diagram The pixel distance of pixels, Represents the oral compensation diagram The pixel grayscale density of pixels, Represents the first pixel in the neighborhood The pixel grayscale density of pixels, represents the density weight, Represents a filter function.

7. An oral examination system based on big data analysis, which is used to implement the oral examination method based on big data analysis according to any one of claims 1 to 6, characterized in that: include: Data acquisition and processing module: collects oral image data and corresponding image evaluation labels, performs distortion conversion on the oral image data, and obtains oral correction images; Image compensation module: performs reflection compensation on the oral correction image to obtain the oral compensation image; Region segmentation module: performs lesion segmentation on the oral compensation image to obtain lesion sub-images; Model building module: Build an oral health assessment model based on the lesion sub-graph and the corresponding image assessment labels, and achieve accurate assessment of oral health based on the oral health assessment model.

Images