Detection risk assessment method for decayed teeth of children
By using image processing technology to collect, identify, and segment teeth, combined with color and morphology analysis, the problems of long culture time and human interference in existing caries detection methods have been solved, thus achieving accuracy and objectivity in caries detection.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-29
- Publication Date
- 2026-04-03
AI Technical Summary
Existing methods for detecting dental caries involve long culture times, high risks of culture medium infection, inconsistent results due to visual observation, and human interference, which affect the accuracy of caries risk assessment.
Image processing technology is used to acquire, identify and divide tooth images into regions. Comparative tooth images are randomly selected for color judgment and difference analysis. By comparing the color tones of caries samples, abnormal areas are screened out. The morphological judgment is made based on the principle of caries formation, thereby improving the accuracy of detection.
This method achieves accuracy and objectivity in dental caries detection, reduces human interference, improves the effectiveness and accuracy of dental caries diagnosis, and ensures the reliability of the results.
Smart Images
Figure CN121789994A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of image processing, and in particular to a method for risk assessment of dental caries detection in children. Background Technology
[0002] Dental caries is a progressive lesion of the hard tissues of the teeth caused by the combined effects of multiple factors in the oral cavity. It manifests as demineralization of inorganic matter and decomposition of organic matter, evolving from color changes to the formation of substantial lesions as the disease progresses.
[0003] In related technologies, the Cariostat caries detection method is used to predict the condition of a patient's teeth. Its core principle lies in simulating the acid-producing process of cariogenic bacteria in the oral cavity. By culturing bacteria samples on the tooth surface in vitro and observing their acid-producing capacity, the likelihood of an individual developing new cavities within a certain period can be determined. The risk of caries is assessed by visually observing changes in the color of the reagent. The color change is directly related to the bacteria's acid-producing capacity; the darker the color, the stronger the acid-producing capacity, and the higher the risk of caries.
[0004] Regarding the aforementioned technologies, there are risks associated with culturing oral microbiota, such as long culture times and culture medium contamination, which can lead to unsatisfactory experimental results. Furthermore, visual observation can introduce human interference when analyzing and judging experimental results, resulting in inconsistent outcomes and inaccurate assessments of caries risk. Therefore, improvements are needed. Summary of the Invention
[0005] To improve the accuracy of dental caries detection, this application provides a method for risk assessment of dental caries detection in children.
[0006] This application provides a method for risk assessment of dental caries detection in children, employing the following technical solution: A method for assessing the risk of dental caries in children, comprising: The teeth are image acquired to obtain initial image data. The initial image data is then used for tooth recognition. Based on the tooth recognition results, the initial image data is divided into intervals to obtain tooth interval image data. Based on the image data of the tooth region, teeth were randomly selected to obtain the first control tooth image; Color judgment is performed on the first control tooth image of each tooth interval to determine the reference hue of the teeth in the initial image data, and the reference hue is compared with the caries hue in the built-in caries sample image to determine whether the reference hue and the caries hue are the same. If the reference color tone is different from the caries color tone, the initial image data is judged based on the reference color tone to determine whether there is a caries problem; If the baseline color tone is the same as the caries color tone, then based on the positional symmetry, the tooth interval image data are grouped into teeth, and the tooth images of two teeth with symmetrical positions are recorded as the second control tooth images. Compare two symmetrically positioned teeth images in the second control tooth image to determine the image difference value; Discrete analysis is performed on the image difference values to determine the discreteness of the image difference values. Based on the discreteness of the image difference values, two symmetrical tooth images with discreteness exceeding the built-in discrete threshold are marked as problem tooth images. Differential markers were made on the second control tooth image to obtain the differential marker region; differential markers were also made on the problem tooth image to obtain the marker region to be judged; the marker region to be judged was compared with the differential marker region to filter out the non-standard marker region; the non-standard marker region was analyzed to determine whether there was caries in the problem tooth image.
[0007] Preferably, color extraction is performed on the first control tooth images of different tooth intervals to obtain the first color dataset on the corresponding teeth; The colors in the first color dataset on different teeth are compared with each other to determine whether the colors extracted from each tooth can match the colors of other teeth. If a color extracted from one tooth cannot be matched on another tooth, the types of colors in the first color dataset of the teeth are judged to obtain the color types, and the base tone of the initial image data is determined based on the color types. If the color extracted from each tooth matches the color of other teeth, then the color combination of each tooth is determined based on the distribution of color on each tooth. The color combinations on each tooth are then matched with each other to determine whether the color combination on each tooth can match other teeth. If the color combination on each tooth can match the color combination of other teeth, then the color combination and its color type are determined to be the base color of the patient's teeth. If the color combination on a patient's teeth does not match the color combination of other teeth, statistical analysis is performed on the color combination and the types of colors to obtain the baseline color tone of the patient's teeth.
[0008] Preferably, the colors in the reference hue are matched with the colors in the caries hue. If the colors in the reference hue fail to match the colors in the caries hue, then the reference hue and the caries hue are determined to be different. If the color in the reference hue matches the color in the caries hue, then obtain the distribution area of the successfully matched color on the tooth and compare this distribution area with the distribution area of the corresponding color in the caries hue. If the distribution area of the successfully matched color in the base hue is the same as the distribution area of the corresponding color in the caries hue, then the base hue and the caries hue are determined to be the same; otherwise, the base hue and the caries hue are determined to be different.
[0009] Preferably, when the reference hue is different from the caries hue, the color data of each tooth in the initial image data is matched with the caries hue to determine whether there is a caries problem in the initial image; If the color data of the teeth in the initial image data matches the color tone of the decayed teeth, then it is determined that there is a caries problem in the initial image; When it is determined that there is a caries problem in the initial image data, the image data of the teeth that match the color tone of the caries are obtained and recorded as the tooth data to be analyzed; The area of caries is obtained by reading the coverage area of the caries plaque based on the color of the caries in the data of the tooth to be analyzed; Based on the built-in plaque corrosion formula, the area of the plaque is calculated to determine the corrosion depth of the plaque under the corresponding conditions. The average thickness of the enamel layer in the corresponding tooth region is determined based on the data of the tooth to be analyzed, and the corrosion depth is compared with the average thickness of the enamel layer to determine the severity of caries in the tooth corresponding to the data of the tooth to be analyzed, and a caries risk report is obtained.
[0010] Preferably, if the color data of the teeth in the initial image data fails to match the color tone of the decayed tooth, then the color data of the teeth in the initial image data is matched with the reference color tone. If the color data of the teeth in the initial image data matches the reference color tone, it is determined that there is no tooth decay problem in the initial image; If the color data of the teeth in the initial image data fails to match the reference color tone, the image of the tooth that fails to match is marked as the second tooth data to be analyzed. Obtain the color data of the occlusal surface of the second tooth to be analyzed, and judge the distribution of the color data to determine whether the color data on the occlusal surface of the second tooth to be analyzed has a clear dividing line. If the distribution of color data on the occlusal surface of the second tooth data to be analyzed is concentrated, then it is determined that the color data on the occlusal surface of the second tooth data to be analyzed has a clear dividing line. When it is determined that the color data on the occlusal surface of the second tooth to be analyzed has a clear dividing line, the color data of the occlusal surface of the second tooth to be analyzed is obtained, and the color data is fused with the caries color to obtain the second color data to be judged. The color data of the occlusal surface of the second tooth to be analyzed is matched with the color data of the second tooth to be matched. If the color data of the occlusal surface of the second tooth to be analyzed is successfully matched with the color data of the second tooth to be matched, it is determined that the tooth corresponding to the second tooth to be analyzed has a caries problem.
[0011] Preferably, the tooth area that matches the second hue data to be matched is recorded as the caries area, and the thickness of the caries area is calculated to obtain the corrosion thickness data; The corrosion thickness data is compared with the average thickness of the glaze layer to obtain the first judgment data; The color ratio of the second hue data to be matched is judged to determine the proportional relationship between the caries hue in the second hue data to be matched and the color data of the occlusal surface of the second tooth data to be analyzed, and the thickness data between the caries cavity and the occlusal surface is evaluated based on the proportional relationship. Based on the location information of the caries patch area, the contact surface of the teeth adjacent to the tooth corresponding to the second tooth to be analyzed is judged, and the position height of the contact surface is determined. The difference between the height and thickness data of the contact surface is calculated to obtain the difference data. Based on this difference data, the contact surface height, the difference data and the thickness data are summed to obtain the corrosion depth. The corrosion depth is compared with the average thickness of the glaze layer to obtain the second judgment data; Based on the first and second judgment data, the severity of caries in the second tooth to be analyzed is determined, and a caries risk report for the corresponding tooth is obtained.
[0012] Preferably, based on the second control tooth image, the image data of each tooth is compared with the unconventional marked area to determine whether the image data of other areas of the tooth are the same as the unconventional marked area; If the image data of other areas of a tooth differs from that of the non-standard marked areas, then the tooth is marked as having a caries problem.
[0013] In summary, this application includes at least one of the following beneficial technical effects: 1. By performing tooth identification and region division on the initial image data, the universality of the selected first control tooth image is ensured. Random selection of teeth from the tooth region image data further guarantees the objectivity of the first control tooth image, making the subsequent determination of the baseline color tone based on the first control tooth image more accurate. The tooth decay detection method is determined by comparing the baseline color tone with the decayed tooth color tone in the caries sample. Different caries detection methods are then used to ensure the accuracy of caries detection. When the baseline color tone and the caries color tone are different, color comparison is used to quickly determine whether the patient has caries. When the baseline color tone and the caries color tone are the same, symmetrical selection of tooth region image data is performed to control variables and reduce interference. Comparison of two teeth at symmetrical positions determines the difference value of the tooth images in each tooth region. Discrete analysis of this difference value identifies a group of teeth with abnormalities. This group of teeth is then compared with other teeth to quickly identify the abnormal area. Analysis of this abnormal area ensures the accuracy of the caries detection results and improves the accuracy of caries detection. 2. Color extraction and analysis are performed on the first control tooth images to ensure the universality and impartiality of the extracted colors. By comparing the first color datasets on different teeth, the matching results of colors on different teeth are determined. Based on the matching results, the baseline hue of the patient's teeth is determined, ensuring the effectiveness and accuracy of subsequent caries identification. When it is determined that the colors of the teeth in the first control tooth images can all match, the distribution of colors on each tooth is further evaluated to determine whether the color distribution of each tooth is the same. This ensures that the final selected baseline hue better represents the overall condition of the patient's teeth, making subsequent caries identification more accurate. When it is determined that the colors of the teeth in the first control tooth images cannot all match, the frequency of each color is statistically analyzed to select the most frequent color as the baseline hue for the patient's teeth, ensuring the effectiveness of subsequent caries identification and improving the accuracy of caries identification. 3. By directly matching the colors in the baseline hue with the colors in the caries hue, it is determined whether the baseline hue is the same as the caries hue. This determines whether caries can be detected by color judgment of teeth. If the baseline hue is different from the caries hue, it indicates that the system can directly detect the patient's teeth using the caries hue. If the baseline hue is the same as the caries hue, the color distribution under the same color is further judged to further determine whether caries problems can be directly judged by color. This makes the caries detection results more accurate in subsequent caries detection processes and improves the accuracy of caries detection. Attached Figure Description
[0014] Figure 1 This is a flowchart illustrating the steps of the risk assessment method for detecting dental caries in children in this embodiment. Detailed Implementation
[0015] The following is in conjunction with the appendix Figure 1 This application will be described in further detail.
[0016] This application discloses a method for risk assessment of dental caries detection in children.
[0017] Example: Figure 1 As shown, the present invention provides a method for risk assessment of childhood dental caries detection, comprising: S1. Acquire images of teeth to obtain initial image data, perform tooth recognition on the initial image data, and divide the initial image data into intervals based on the tooth recognition results to obtain tooth interval image data. S2, Based on the image data of the dental intervals, teeth are randomly selected to obtain the first control tooth image; wherein, the tooth image of one tooth can be selected in each dental interval to jointly construct the first control tooth image.
[0018] S3, perform color judgment on the first control tooth image of each tooth interval, determine the reference color tone of the tooth in the initial image data, and compare the reference color tone with the caries color tone in the built-in caries sample image to determine whether the reference color tone and the caries color tone are the same; S4. If the reference hue is different from the caries hue, then the initial image data is judged based on the reference hue to determine whether there is a caries problem. S5. If the reference color tone is the same as the caries color tone, then according to the positional symmetry, the tooth interval image data are grouped into teeth, and the tooth images of two teeth with symmetrical positions are recorded as the second control tooth images. S6, compare two symmetrical tooth images in the second control tooth image to determine the image difference value; S7. Perform discrete analysis on the image difference values to determine the discreteness of the image difference values, and mark two symmetrical tooth images with discreteness exceeding the built-in discrete threshold as problem tooth images based on the discreteness of the image difference values. The built-in discrete threshold can be obtained by judging the normal distribution of multiple image difference values and selecting the image difference values after normal distribution, thereby obtaining a discrete threshold that matches the actual situation.
[0019] S8. Differential markers are added to the second control tooth image to obtain the differential marker region; differential markers are added to the problem tooth image to obtain the marker region to be judged. The marker region to be judged is compared with the differential marker region to filter out non-standard marker regions. The non-standard marker regions are analyzed to determine whether the problem tooth image contains caries. Specifically: Based on the second control tooth image, the image data of each tooth is compared with the non-standard marker region to determine whether the image data of other areas of the tooth is the same as the non-standard marker region; if the image data of other areas of the tooth is different from the non-standard marker region, the tooth is marked as having a caries problem.
[0020] In this embodiment, tooth identification and interval division are performed on the initial image data to ensure the universality of the selected first control tooth image. Random selection of teeth from the tooth interval image data further ensures the objectivity of the first control tooth image, making the subsequent determination of the baseline color tone based on the first control tooth image more accurate. The tooth decay detection method is determined by comparing the baseline color tone with the decayed tooth color tone in the decayed tooth sample. Different tooth decay detection methods are then used to ensure the accuracy of tooth decay detection. When the baseline color tone and the decayed tooth color tone are different, color comparison is used to quickly determine whether the patient has tooth decay. When the baseline color tone and the decayed tooth color tone are the same, the tooth interval image data is symmetrically selected to control variables and reduce interference. Simultaneously, by comparing two teeth at symmetrical positions, the difference value of the tooth images in each tooth interval is determined. Discrete analysis of this difference value identifies a group of teeth with abnormalities. This group of teeth is then compared with other teeth to quickly identify the abnormal area. Analysis of this abnormal area ensures the accuracy of the tooth decay detection results and improves the accuracy of tooth decay detection.
[0021] For example, when detecting and assessing dental caries in pediatric patients, it is first necessary to acquire images of the patient's teeth to obtain the most original initial image data, which includes the patient's normal teeth and decayed teeth.
[0022] Because different teeth present different forms, such as incisors, canines, and molars, the initial image data of a complete tooth is divided into multiple tooth region image data by judging the shape and position of the teeth. For example, tooth images divided into incisor region, canine region, and molar region.
[0023] Before assessing dental caries, it is necessary to first determine the patient's basic dental condition, and then select different assessment methods based on the situation.
[0024] Because the appearance of teeth varies in different dental regions, it is necessary to select teeth according to these regions to determine the general condition of teeth in each region. This is done by randomly selecting teeth from the image data of each dental region to ensure the reliability of the results. For example, in the incisor region, a randomly selected incisor is used as a control image.
[0025] After selecting reference tooth images for each dental interval, the tooth colors in the reference tooth images for each interval are compared to determine the basic color of the child patient's teeth. Since individual constitutions differ, resulting in variations in tooth color, image data from each dental interval is randomly selected to calibrate the basic color within the system, ensuring the accuracy of subsequent caries diagnosis.
[0026] After determining the patient's oral dental condition, the system compares the baseline color of the patient's teeth with the color of caries samples stored in the system to determine whether the patient's dental caries can be directly identified by color. For example, if the baseline color of the patient's teeth is white or pale yellow, and the caries sample color is black, then a black-and-white matching process is performed on the patient's teeth. The tooth with black spots indicates the presence of caries, thus identifying that tooth as having caries. Conversely, if the patient's teeth, for some reason, have a baseline color of black or burnt yellow, and the caries sample color is also black, then a direct color comparison cannot be used to determine caries in this patient, and other methods of assessment are required.
[0027] When color comparison alone is insufficient to diagnose tooth decay, the formation principle of tooth decay can be utilized: tooth decay occurs when cariogenic bacteria adhere to the tooth surface and decompose food debris, producing large amounts of acidic substances. These acids erode the enamel layer, causing plaque and cavities, ultimately leading to tooth decay. Since tooth decay develops from a healthy tooth through erosion, this erosion characteristic can be used to assess the tooth's morphology. A healthy tooth has an enamel layer, making its surface smooth. When acidic substances damage the enamel layer, this smooth surface is lost, allowing for the identification of tooth decay based on this morphological characteristic.
[0028] Firstly, because the morphology of teeth differs between different dentition areas, it is necessary to use teeth within the same dentition area for comparison to improve the accuracy of caries identification. This is achieved by utilizing the symmetry of teeth in the oral cavity and the concentration of acidic substances produced by cariogenic bacteria, by symmetrically selecting teeth within the same dentition area. For example, if the teeth on the left and right sides are symmetrical, then the teeth in symmetrical positions on both sides are grouped together. For instance, the left teeth are numbered Left 1, Left 2, Left 3, Left 4, and the right teeth are numbered Right 1, Right 2, Right 3, Right 4. When selecting Left 1, the corresponding Right 1 must also be selected, and Left 1 and Right 1 are compared as a group to determine the differences between them.
[0029] Assuming that tooth 1 on the left has caries, the enamel on tooth 1 is damaged, while tooth 1 on the right does not have caries, so the enamel on tooth 1 is not damaged. Therefore, when comparing tooth 1 and tooth 1, there will inevitably be differences in their morphology. At the same time, since there are certain image differences between different teeth, in order to ensure that the image differences between tooth 1 and tooth 1 are caused by cariogenic bacteria, it is necessary to perform discrete analysis on the multiple image difference values collected, so as to identify teeth that exceed the conventional tooth differences. That is, through discrete screening, it is possible to screen out tooth 1 and tooth 1 that have caries.
[0030] Further analysis is needed to determine the presence of cavities between the left and right teeth. By comparing the distinguishing areas between the left and right teeth with the distinguishing areas between other teeth that are considered normal, the areas that are likely to have cavities can be identified. Then, by combining the location information of the cavities, the left and right teeth can be analyzed separately to determine the likelihood of cavities in the left and right teeth, thereby improving the accuracy of cavity diagnosis.
[0031] For example, the distinguishing regions between left and right teeth 1 include region 1, region 2, and region 3. Regions 1 and 2 are common distinguishing regions between other normal teeth. Therefore, region 3 can be determined as a distinguishing region caused by cariogenic bacteria. The next step is to determine which tooth's distinguishing region 3 is caused by cariogenic bacteria. Since the distinguishing region caused by cariogenic bacteria differs from the region of a normal tooth, region 3 can be compared with other regions of the left tooth and the right tooth to determine the tooth with the caries. If distinguishing region 3 differs from other regions of the left tooth, then distinguishing region 3 of the left tooth is determined to be caused by cariogenic bacteria.
[0032] In step S3, the color of the first control tooth image in each tooth interval is determined to establish the baseline hue of the teeth in the initial image data. This includes the following steps: S311, color extraction is performed on the first control tooth images of different tooth intervals to obtain the first color dataset on the corresponding teeth; S312, compare the colors in the first color dataset on different teeth to determine whether the colors extracted from each tooth can match the colors of other teeth; S313, if a color extracted from one tooth cannot be matched on other teeth, the types of colors in the first color dataset of the teeth are determined to obtain the color types, and the baseline hue of the initial image data is determined based on the color types. By filtering multiple color types, the most frequently occurring colors are selected as the baseline hues for the patient. Alternatively, the system's color recognition precision can be adjusted to blur the recognition of multiple color categories, thereby ensuring the accuracy of related judgments based on the baseline hue.
[0033] S314, If the color extracted from each tooth matches the color of other teeth, then determine the color combination of each tooth based on the distribution of color on each tooth, match the color combinations on each tooth with each other, and determine whether the color combination on each tooth can match other teeth except itself. S315. If the color combination on each tooth can match the color combination of other teeth, then the color combination and its color type are determined to be the base color of the patient's teeth. S316. If the color combination on a patient's teeth does not match the color combination of other teeth, statistical analysis is performed on the color combinations and their color types to obtain the baseline color tone of the patient's teeth. By statistically analyzing the color combinations, the most frequently occurring color combination is taken as the baseline color tone of the patient's teeth.
[0034] In this embodiment, color extraction and analysis are performed on the first control tooth image to ensure the universality and impartiality of the extracted colors. By comparing the first color datasets on different teeth, the matching results of colors on different teeth are determined, and then the baseline hue of the patient's teeth is determined based on the matching results, ensuring the effectiveness and accuracy of subsequent caries identification. When it is determined that the colors of the teeth in the first control tooth image can all match each other, the distribution of colors on each tooth is further judged to determine whether the color situation of each tooth is the same, so that the final selected baseline hue can better represent the overall condition of the patient's teeth, making the subsequent caries identification more accurate. When it is determined that the colors of the teeth in the first control tooth image cannot all match each other, the frequency of each color is statistically analyzed to select the color with the highest frequency as the baseline hue of the patient's teeth, ensuring the effective implementation of subsequent caries identification and improving the accuracy of caries identification.
[0035] For example, when performing a caries detection test on a patient's teeth, it is first necessary to assess the current condition of the patient's teeth. This involves extracting the color tone of the patient's teeth to verify the baseline color tone used in the system for caries detection. This ensures the accuracy of subsequent color analysis.
[0036] Because the first control tooth image was randomly selected from teeth in different dentition intervals, the color extraction results were consistent with the overall color of all the patient's teeth, thus ensuring the fairness of the color extraction.
[0037] Because different teeth in a mouth have varying degrees of health, their colors also differ; a single tooth may exhibit multiple colors, or multiple teeth may display the same color. To determine the condition of a patient's teeth in such cases, the colors of different teeth are matched to determine if the color on each tooth represents a common occurrence. If an exception exists, the most frequent color is selected as the baseline color for the patient's teeth through statistical analysis. For example, if the first control image contains five teeth, and these five teeth are white, light yellow, white, white, light yellow, and pink, then pink is considered a special case. Statistical analysis of these five teeth reveals white and light yellow as the baseline colors for the patient's teeth. In subsequent assessments, white and light yellow can be used as the base colors to match the colors of decayed teeth (e.g., black), thus distinguishing between normal teeth and decayed teeth.
[0038] When the color extracted from each tooth can be matched on other teeth, it is necessary to determine the types of colors on a single tooth. If each tooth has only one color, then the distribution of colors on the teeth is unique, and therefore this color can be directly used as the base color of the patient's teeth.
[0039] If multiple colors are present on each tooth simultaneously, the distribution of these colors will vary, and different color distributions may represent different situations. For example, if the tooth edges are dark and the center is light, it may be due to an external cause rather than tooth decay. Conversely, if the edges are light and the center is dark, it may be due to cavities causing the dark center, indicating tooth decay. Therefore, it is necessary to assess the distribution of color on the teeth. When the color distribution on all teeth is the same, this distribution can be used as the baseline color tone for the patient's teeth. If there are variations in the color distribution, it is necessary to analyze and statistically determine the color distribution that best reflects the majority of tooth conditions.
[0040] In step S3, the reference hue is compared with the caries hue in the built-in caries sample image to determine whether the reference hue and the caries hue are the same, including the following steps: S321, Match the color in the base hue with the color in the caries hue. If the color in the base hue fails to match the color in the caries hue, then the base hue and the caries hue are determined to be different. S322, If the color in the reference hue matches the color in the caries hue, then obtain the distribution area of the successfully matched color in the reference hue on the tooth, and compare the distribution area with the distribution area of the corresponding color in the caries hue. S323, if the distribution area of the successfully matched color in the reference hue is the same as the distribution area of the corresponding color in the caries hue, then the reference hue and the caries hue are determined to be the same; otherwise, the reference hue and the caries hue are determined to be different.
[0041] In this embodiment, when matching the reference hue with the caries hue, the system directly matches the colors in the reference hue with the colors in the caries hue to determine whether the reference hue is the same as the caries hue. This determines whether caries can be detected by color judgment of teeth. If the reference hue and the caries hue are different, it indicates that the system can directly detect the patient's teeth using the caries hue. If the reference hue and the caries hue are the same, the system further determines whether caries can be directly detected by color judgment by re-judging the color distribution under the same color. This makes the caries detection results more accurate in subsequent caries detection processes and improves the accuracy of caries detection.
[0042] For example, when using a reference hue to identify dental caries, the reference hue is compared with the hue of the dental caries sample to determine whether the dental caries can be identified by the reference hue.
[0043] By matching a baseline color tone with the color tone of the decayed teeth, if the baseline color tone and the decayed teeth color tone are different, it means that the patient can directly use the color of the decayed teeth to match their teeth, thereby determining whether the patient has a caries problem. For example, if the baseline color tone is white and the decayed teeth color tone is black, then the baseline color tone and the decayed teeth color tone are different. Therefore, by matching the black parts on the patient's teeth, if there are black parts on the patient's teeth, it can be determined that the patient has caries; otherwise, it can be determined that the patient does not have caries.
[0044] When the base color tone matches the color tone of the decayed tooth, it indicates that the patient cannot directly identify the decayed tooth by color, and further judgment is needed to ensure accuracy. For example, if the base color tone is black and the decayed tooth color tone is also black, then the base color tone and the decayed tooth color tone are the same, and therefore it is impossible to match the black color on the patient's teeth to determine the patient's decay problem. In this case, there are two possibilities: one is that the patient's base color tone is only one color. If it is only one color, and it is black, then it is impossible to determine the decayed tooth by color. Another scenario involves a patient whose baseline tooth tone contains multiple colors, one of which matches the tone of the decayed tooth. In this case, the diagnosis can be made by analyzing the combination of these multiple colors, including not only the combinations between colors themselves but also the combinations between the areas where the colors are located. For example, black might be on the outer edge of the tooth, while white is in the center, with a gradual transition between the two. In contrast, the tooth with the decayed tone might have white on the outer edge and black in the center, with a clear boundary between the two. Therefore, by comparing the color combinations in these two scenarios, a diagnosis of tooth decay can be made.
[0045] In step S4, if the reference hue differs from the caries hue, the initial image data is evaluated based on the reference hue to determine whether caries exists. This includes the following steps: S411, when the reference hue is different from the caries hue, the color data of each tooth in the initial image data is matched with the caries hue to determine whether there is a caries problem in the initial image; S412, If the color data of the teeth in the initial image data matches the color tone of the caries, then it is determined that there is a caries problem in the initial image. S413, When it is determined that there is a caries problem in the initial image data, acquire the image data of the teeth that match the color tone of the caries, and record it as the tooth data to be analyzed; S414, based on the color of the caries, read the coverage area of the caries patch in the data of the tooth to be analyzed, and obtain the caries patch area; S415, based on the built-in plaque corrosion formula, calculates the plaque area and determines the corrosion depth of the plaque under the corresponding conditions; wherein, the plaque corrosion formula is the relationship between plaque size and its depth over time obtained through experiments.
[0046] S416. Based on the data of the tooth to be analyzed, determine the average thickness of the enamel layer in the corresponding tooth interval, and compare the corrosion depth with the average thickness of the enamel layer to determine the severity of caries in the tooth corresponding to the data of the tooth to be analyzed, and obtain a caries risk report.
[0047] In this embodiment, by matching the color data of teeth in the initial image data with the hue of the decayed teeth, it is possible to quickly determine whether a patient has dental caries. Once it is determined that the patient has dental caries, the area of the caries patch is determined by performing pixel statistics on the hue of the caries patch. Then, by combining the caries patch corrosion formula, the corrosion depth corresponding to the caries patch area is determined. Based on the corrosion depth, the location of the caries is determined. Finally, based on the definition of the severity of dental caries, the severity of the patient's dental caries is determined, thus improving the accuracy of dental caries diagnosis.
[0048] For example, when the base color tone differs from the caries color tone, it indicates that the presence of caries can be accurately determined by analyzing the color of the patient's teeth. For instance, if the base color tone of the patient's teeth is white and the caries color tone is black, then by comparing the color of all teeth in the patient's initial image data with the black color of the caries color tone, if any teeth in the initial image are black, the patient is determined to have caries. Conversely, if no teeth are black, the patient is determined not to have caries.
[0049] Once a patient is diagnosed with dental caries, the severity of the caries can be determined based on images of the affected teeth. The severity of caries can be categorized as superficial caries (causes confined to the enamel layer), moderate caries (causes reaching the superficial dentin layer), and deep caries (causes reaching the dentin formation layer).
[0050] Because the more severe the caries, the larger the area covered by the caries plaque, the area of the plaque can be calculated. The depth of decay caused by cariogenic bacteria can then be deduced from the area of the plaque, and the location of the caries can be determined based on the depth of decay. For example, if the calculated depth of decay is 1 mm, and the enamel layer of the tooth containing the caries plaque is 3 mm, then the caries is determined to be in the enamel layer, and the severity of the patient's caries can be determined as superficial caries. This improves the accuracy of caries diagnosis.
[0051] In step S4, if the reference hue differs from the caries hue, the initial image data is evaluated based on the reference hue to determine whether caries exists. This includes the following steps: S421, If the color data of the teeth in the initial image data fails to match the color tone of the decayed tooth, then match the color data of the teeth in the initial image data with the reference color tone. S422, If the color data of the teeth in the initial image data matches the reference color tone successfully, it is determined that there is no caries problem in the initial image; S423, If the color data of the teeth in the initial image data fails to match the reference color tone, then the image of the tooth that failed to match is marked as the second tooth data to be analyzed. S424, Obtain the color data of the occlusal surface of the second tooth data to be analyzed, and judge the distribution of the color data to determine whether the color data on the occlusal surface of the second tooth data to be analyzed has a clear dividing line. S425, if the distribution of color data on the occlusal surface of the second tooth data to be analyzed is concentrated, then it is determined that the color data on the occlusal surface of the second tooth data to be analyzed has a clear dividing line. S426, when it is determined that the color data on the occlusal surface of the second tooth data to be analyzed has a clear dividing line, the color data of the occlusal surface of the second tooth data to be analyzed is obtained, and the color data is fused with the caries color to obtain the second color data to be judged. S427, Match the color data of the occlusal surface of the second tooth data to be analyzed with the second hue data to be matched. If the color data of the occlusal surface of the second tooth data to be analyzed is successfully matched with the second hue data to be matched, it is determined that the tooth corresponding to the second tooth data to be analyzed has a caries problem.
[0052] In this embodiment, after the initial image data summarizing the tooth color data fails to match the caries hue, the tooth color data in the initial image data is matched with the reference hue to determine whether the patient's teeth have any color abnormalities. When a successful match with the reference hue is determined, it indicates that the patient does not have caries. When a match with the reference hue fails, the boundary line on the occlusal surface of the second tooth data to be analyzed is judged to determine whether the patient has the risk of proximal caries. When a clear boundary line is determined on the occlusal surface of the second tooth data to be analyzed, it indicates that the patient has the risk of proximal caries. To ensure the accuracy of the judgment, after determining that there is a clear boundary line on the occlusal surface of the second tooth data to be analyzed, the color data of the second tooth data itself is hue-fused with the caries hue to simulate the color of the tooth after proximal caries decay. The simulated tooth color is then compared with the actual tooth color to further determine whether the patient has caries, making the judgment of caries more accurate and improving the accuracy of caries detection.
[0053] For example, when matching the color data of teeth in the initial image data using caries hue, there are two scenarios when the color data of teeth in the initial image data fails to match the caries hue: one is that the patient does not have caries, and the other is that the patient has proximal caries, meaning the caries is on the contact surface between teeth, making it impossible to directly observe the caries in the tooth image acquired during image acquisition. In this case, a baseline hue matching is needed. When a patient has proximal caries, the caries appears intact from the occlusal surface, but its interior is corroded, resulting in a thinner enamel layer on the occlusal surface, making the color on the occlusal surface more transparent. Furthermore, due to the inherent color of the caries and the natural color of the tooth, the superimposed color differs from the color of normal teeth, causing the color data on the occlusal surface to fail to match the baseline hue, thus identifying teeth with a high probability of caries. Similarly, because cariogenic bacteria gradually erode teeth, the enamel layer that has been eroded from the sides will have a clear boundary on the occlusal surface. The eroded portion of the occlusal surface may appear darker, while the uneroded portion will appear lighter. Therefore, the color distribution on the occlusal surface of the second tooth being analyzed can be used to determine the presence of caries. If a clear boundary exists, it indicates a possible caries condition.
[0054] Meanwhile, by hue fusion of the color data with the caries hue, the color that the occlusal surface should present after decay in the case of proximal caries can be simulated and compared with the real color to further determine whether the second tooth data to be analyzed has the possibility of caries, thus ensuring the accuracy of caries judgment.
[0055] In step S427, the color data of the occlusal surface of the second tooth to be analyzed is matched with the second hue data to be matched. If the color data of the occlusal surface of the second tooth to be analyzed successfully matches the second hue data, it is determined that the tooth corresponding to the second tooth to be analyzed has a caries problem, including the following steps: S4271, the tooth area that matches the second hue data to be matched is recorded as the caries area, and the thickness of the caries area is calculated to obtain the corrosion thickness data; S4272, compare the corrosion thickness data with the average thickness of the glaze layer to obtain the first judgment data; S4273, perform color ratio judgment on the second hue data to be matched, determine the ratio relationship between the caries hue in the second hue data to be matched and the color data of the occlusal surface of the second tooth data to be analyzed, and evaluate the thickness data between the caries cavity and the occlusal surface based on the ratio relationship. S4274, Based on the location information of the caries area, the contact surface of the tooth adjacent to the tooth corresponding to the second tooth data to be analyzed is judged, and the position height of the contact surface is determined; S4275, calculate the difference between the height and thickness data of the contact surface position to obtain the difference data, and based on this difference data, calculate the sum of the contact surface position height, the difference data and the thickness data to obtain the corrosion depth; S4276, compare the corrosion depth with the average thickness of the glaze layer to obtain the second judgment data; S4277, based on the first judgment data and the second judgment data, determine the caries severity of the second tooth to be analyzed, and obtain the caries risk report for the corresponding tooth.
[0056] In this embodiment, the caries area of the tooth is removed by using the second color data to be matched, and the thickness of tooth decay is determined based on the caries area. At the same time, the distance between the cavity and the occlusal surface of the tooth is determined based on the second color data to be matched. Then, the depth of tooth decay caused by cariogenic bacteria is determined by combining the possible location of the caries area. After determining the depth and thickness of decay, the severity of tooth decay of the patient's tooth is determined by judging the depth and thickness of decay, thereby improving the accuracy of the judgment of the severity of tooth decay.
[0057] For example, when a patient is diagnosed with proximal caries, to determine its severity, the lateral surface corresponding to the caries is first identified by matching the tooth area with the simulated color (second hue data to be matched). Simultaneously, the thickness of the decayed plaque can be determined based on the boundary line. This thickness of decay is compared with the lateral thickness of the corresponding tooth enamel layer to determine the depth of the caries. For instance, if the enamel layer thickness on the tooth lateral surface is 3mm, and the decayed thickness is also 3mm, it indicates that the caries has reached at least the moderate caries stage.
[0058] Simultaneously, the positional relationship between the cavity and the occlusal surface of the tooth is determined based on the color of the caries patch area. The closer the caries patch area is to the occlusal surface, the closer its color is to the tooth's hue. Therefore, the distance between the corresponding caries patch area and the occlusal surface (thickness data between the cavity and the occlusal surface) is determined based on the current color. Since the caries patch is not observed, its possible location can be determined by the contact surfaces of two adjacent teeth. Assuming the caries erosion proceeds in a circular pattern centered on the contact surface, when it reaches the thickness between the cavity and the occlusal surface, it indicates that the cavity has eroded at least the difference between the height and thickness data of the contact surface. Therefore, the radius of this circle is determined, and the depth of the cavity erosion is determined based on the radius and thickness data. For example, if the distance between the cavity and the occlusal surface is 'a' (thickness data), then the distance from the occlusal surface to the center of the cavity circle is 'a+r', where 'r' is the radius of the cavity. Therefore, the position of the occlusal surface across the center to the other edge is 'a+2r'. If we assume that the distance from the center of the cavity to the occlusal surface is h (the height of the base surface), then we can get h = a + r. Since h and a are known, we can get r, and then we can get a + 2r (the depth of corrosion).
[0059] After determining the depth and thickness of the erosion, the severity of the proximal caries can be assessed. For example, the erosion thickness may indicate moderate caries, but the erosion depth may not.
[0060] Compared to existing methods for assessing the risk of dental caries in children, this invention improves the accuracy of dental caries detection.
[0061] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.
Claims
1. A method for assessing the risk of dental caries in children, characterized in that, include: The teeth are image acquired to obtain initial image data. The initial image data is then used for tooth recognition. Based on the tooth recognition results, the initial image data is divided into intervals to obtain tooth interval image data. Based on the image data of the tooth region, teeth were randomly selected to obtain the first control tooth image; Color judgment is performed on the first control tooth image of each tooth interval to determine the reference hue of the teeth in the initial image data, and the reference hue is compared with the caries hue in the built-in caries sample image to determine whether the reference hue and the caries hue are the same. If the reference color tone is different from the caries color tone, the initial image data is judged based on the reference color tone to determine whether there is a caries problem; If the baseline color tone is the same as the caries color tone, then based on the positional symmetry, the tooth interval image data are grouped into teeth, and the tooth images of two teeth with symmetrical positions are recorded as the second control tooth images. Compare two symmetrically positioned teeth images in the second control tooth image to determine the image difference value; Discrete analysis is performed on the image difference values to determine the discreteness of the image difference values. Based on the discreteness of the image difference values, two symmetrical tooth images with discreteness exceeding the built-in discrete threshold are marked as problem tooth images. Differential points were marked on the second control tooth image to obtain the differential marking region; Distinguishing points are marked on the image of the problematic tooth to obtain the region to be judged. The region to be judged is compared with the distinguishing marked region to filter out the non-standard marked region. The non-standard marked region is analyzed to determine whether there is caries in the image of the problematic tooth.
2. The method for risk assessment of childhood dental caries according to claim 1, characterized in that: The step of performing color judgment on the first reference tooth images of each tooth interval to determine the baseline hue of the teeth in the initial image data includes: Color extraction was performed on the first control tooth images of different tooth regions to obtain the first color dataset on the corresponding teeth; The colors in the first color dataset on different teeth are compared with each other to determine whether the colors extracted from each tooth can match the colors of other teeth. If a color extracted from one tooth cannot be matched on another tooth, the types of colors in the first color dataset of the teeth are judged to obtain the color types, and the base tone of the initial image data is determined based on the color types. If the color extracted from each tooth matches the color of other teeth, then the color combination of each tooth is determined based on the distribution of color on each tooth. The color combinations on each tooth are then matched with each other to determine whether the color combination on each tooth can match other teeth. If the color combination on each tooth can match the color combination of other teeth, then the color combination and its color type are determined to be the base color of the patient's teeth. If the color combination on a patient's teeth does not match the color combination of other teeth, statistical analysis is performed on the color combination and the types of colors to obtain the baseline color tone of the patient's teeth.
3. The method for risk assessment of childhood dental caries according to claim 2, characterized in that: The step of comparing the reference hue with the caries hue in the built-in caries sample image to determine whether the reference hue and the caries hue are the same includes: Match the colors in the base hue with the colors in the caries hue. If the colors in the base hue fail to match the colors in the caries hue, then the base hue and the caries hue are determined to be different. If the color in the reference hue matches the color in the caries hue, then obtain the distribution area of the successfully matched color on the tooth and compare this distribution area with the distribution area of the corresponding color in the caries hue. If the distribution area of the successfully matched color in the base hue is the same as the distribution area of the corresponding color in the caries hue, then the base hue and the caries hue are determined to be the same; otherwise, the base hue and the caries hue are determined to be different.
4. The method for risk assessment of childhood dental caries according to claim 1, characterized in that: If the reference hue differs from the caries hue, the initial image data is evaluated based on the reference hue to determine whether caries exists, including: When the reference hue differs from the caries hue, the color data of each tooth in the initial image data is matched with the caries hue to determine whether there is a caries problem in the initial image. If the color data of the teeth in the initial image data matches the color tone of the decayed teeth, then it is determined that there is a caries problem in the initial image; When it is determined that there is a caries problem in the initial image data, the image data of the teeth that match the color tone of the caries are obtained and recorded as the tooth data to be analyzed; The area of caries is obtained by reading the coverage area of the dental data to be analyzed based on the color of the caries. Based on the built-in plaque corrosion formula, the area of the plaque is calculated to determine the corrosion depth of the plaque under the corresponding conditions. The average thickness of the enamel layer in the corresponding tooth region is determined based on the data of the tooth to be analyzed, and the corrosion depth is compared with the average thickness of the enamel layer to determine the severity of caries in the tooth corresponding to the data of the tooth to be analyzed, and a caries risk report is obtained.
5. The method for risk assessment of childhood dental caries according to claim 4, characterized in that: If the reference hue differs from the caries hue, then based on the reference hue, the initial image data is judged to determine whether a caries problem exists. This also includes: If the color data of the teeth in the initial image data fails to match the color tone of the decayed tooth, then the color data of the teeth in the initial image data will be matched with the reference color tone. If the color data of the teeth in the initial image data matches the reference color tone, it is determined that there is no tooth decay problem in the initial image; If the color data of the teeth in the initial image data fails to match the reference color tone, the image of the tooth that fails to match is marked as the second tooth data to be analyzed. Obtain the color data of the occlusal surface of the second tooth to be analyzed, and judge the distribution of the color data to determine whether the color data on the occlusal surface of the second tooth to be analyzed has a clear dividing line. If the color data on the occlusal surface of the second tooth data to be analyzed is concentrated, then it is determined that the color data on the occlusal surface of the second tooth data to be analyzed has a clear dividing line. When it is determined that the color data on the occlusal surface of the second tooth to be analyzed has a clear dividing line, the color data of the occlusal surface of the second tooth to be analyzed is obtained, and the color data is fused with the caries color to obtain the second color data to be judged. The color data of the occlusal surface of the second tooth to be analyzed is matched with the color data of the second tooth to be matched. If the color data of the occlusal surface of the second tooth to be analyzed is successfully matched with the color data of the second tooth to be matched, it is determined that the tooth corresponding to the second tooth to be analyzed has a caries problem.
6. The method for risk assessment of childhood dental caries according to claim 5, characterized in that: The step of matching the color data of the occlusal surface of the second tooth data to be analyzed with the second hue data to be matched, and if the color data of the occlusal surface of the second tooth data to be analyzed successfully matches the second hue data to be matched, then it is determined that the tooth corresponding to the second tooth data to be analyzed has a caries problem, including: The tooth area that matches the second hue data to be matched is recorded as the caries area, and the thickness of the caries area is calculated to obtain the corrosion thickness data; The corrosion thickness data is compared with the average thickness of the glaze layer to obtain the first judgment data; The color ratio of the second hue data to be matched is judged to determine the proportional relationship between the caries hue in the second hue data to be matched and the color data of the occlusal surface of the second tooth data to be analyzed, and the thickness data between the caries cavity and the occlusal surface is evaluated based on the proportional relationship. Based on the location information of the caries patch area, the contact surface of the teeth adjacent to the tooth corresponding to the second tooth to be analyzed is judged, and the position height of the contact surface is determined. The difference between the height and thickness data of the contact surface is calculated to obtain the difference data. Based on this difference data, the contact surface height, the difference data and the thickness data are summed to obtain the corrosion depth. The corrosion depth is compared with the average thickness of the glaze layer to obtain the second judgment data; Based on the first and second judgment data, the severity of caries in the second tooth to be analyzed is determined, and a caries risk report for the corresponding tooth is obtained.
7. The method for risk assessment of childhood dental caries according to claim 1, characterized in that: The second control tooth image is marked with distinguishing points to obtain the distinguishing marked area; Distinguishing points are marked on the image of the problematic tooth to obtain the region to be judged. This region is then compared with the distinguishing marked regions to filter out non-standard marked regions. These non-standard marked regions are analyzed to determine whether caries is present in the image of the problematic tooth, including: Based on the second control tooth image, the image data of each tooth is compared with the unconventional marked area to determine whether the image data of other areas of the tooth are the same as the unconventional marked area; If the image data of other areas of a tooth differs from that of the non-standard marked areas, then the tooth is marked as having a caries problem.