Child oral cavity detection method and system based on image recognition

Abstract

The invention provides a child oral cavity detection method and system based on image recognition, and relates to the field of medical detection. The child oral cavity detection system based on image recognition comprises an image acquisition module used for shooting parts of teeth and gingiva in the oral cavity of a child; the image preprocessing unit is used for denoising the acquired image and enhancing preprocessing operation so as to improve the accuracy of subsequent image recognition; the image recognition module is used for analyzing the preprocessed image by adopting a deep learning algorithm; the data analysis and diagnosis unit is used for carrying out comprehensive analysis according to the image recognition result in combination with oral health historical data of the children and giving diagnosis suggestions; the user interaction interface allows doctors or parents to check detection results, historical records and diagnosis suggestions and provides system setting and operation interfaces. Through the automatic image acquisition and recognition process, the time required by manual examination of a doctor is greatly shortened, and the overall detection efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of medical detection technology, and in particular to a method and system for detecting children's oral cavity based on image recognition. Background Art

[0002] In the modern medical field, oral health testing is crucial to the overall health of children. Traditional oral health examinations usually rely on doctors' visual inspections and empirical judgments. This method is not only time-consuming, but also highly subjective and easily affected by the doctor's experience level. With the development of technology, especially the advancement of image recognition and artificial intelligence technology, new possibilities have been provided for oral health testing.

[0003] At present, children's oral health problems are receiving increasing attention, including common problems such as tooth decay and gingivitis. If these problems are not discovered and treated in time, they may affect the long-term health of children. Existing oral health detection systems often lack specialized treatment for children's oral conditions and rely on expensive medical equipment and professional operators in many places, which limits their application in resource-limited environments. In addition, these systems rarely have real-time feedback and early warning mechanisms, and cannot provide test results to doctors and parents in a timely manner. Therefore, technicians in this field provide a children's oral detection method and system based on image recognition to solve the problems raised in the above background technology. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] In view of the deficiencies in the prior art, the present invention provides a method and system for detecting children's oral cavity based on image recognition, which solves the problems of lack of specialized processing of children's oral cavity conditions and lack of real-time feedback and early warning mechanisms in the prior system.

[0006] (II) Technical solution

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: a children's oral cavity detection system based on image recognition, comprising:

[0008] An image acquisition module is used to capture images of the inside of a child's oral cavity, including teeth and gums;

[0009] An image preprocessing unit performs denoising and enhancement preprocessing operations on the collected images to improve the accuracy of subsequent image recognition;

[0010] An image recognition module that uses deep learning algorithms to analyze preprocessed images to identify and mark abnormal areas in the mouth, including tooth decay and gingivitis;

[0011] The data analysis and diagnosis unit conducts comprehensive analysis based on the image recognition results and combines the child’s oral health history data to provide diagnostic recommendations;

[0012] The user interface allows doctors or parents to view test results, history records, and diagnostic recommendations, and provides a system setup and operation interface.

[0013] Preferably, the image acquisition module comprises the following steps:

[0014] S1. Equipment preparation: Choose a high-definition camera or a dedicated intraoral camera suitable for photographing the inside of a child's oral cavity. Make sure the camera resolution is high enough to capture the details inside the oral cavity.

[0015] S2. Environment setting, adjust the light source, ensure that the optical fiber is uniform and sufficient, avoid shadows or overexposure, use sterile covers to ensure hygiene and safety;

[0016] S3. Shooting guidance: let the child take an appropriate posture, keep the head stable, and ensure that the child opens the mouth appropriately to avoid discomfort caused by excessive mouth opening;

[0017] S4. Image capture: Use a camera or endoscope to take images of the inside of the oral cavity and ensure that the images are clear and free of blur or distortion caused by movement;

[0018] S5. Image storage, stores the captured images in a high-resolution format for subsequent processing, and ensures that the naming rules of image files are clear for easy management and retrieval.

[0019] Preferably, the image preprocessing unit comprises the following steps:

[0020] S1. Image cropping, cropping the inner part of the oral cavity from the original image;

[0021] S2. Grayscale conversion, converting color images into grayscale images to reduce computational complexity. The calculation formula is as follows:

[0022] Gray=0.299×R+0.587×G+0.114×B

[0023] Among them, R, G, and B represent the pixel values ​​of the red, green, and blue color channels respectively;

[0024] S3. De-noising, applying a filter to remove noise in the image, where the calculation formula of Gaussian filtering is:

[0025]

[0026] Where σ is the standard deviation of the Gaussian kernel;

[0027] S4. Contrast enhancement, using histogram equalization to enhance image contrast, the formula is:

[0028]

[0029] Among them, r k is the gray level of the original image, s k is the gray level after equalization, n j is the gray level r j The number of occurrences, n is the total number of pixels in the image;

[0030] S5. Edge sharpening: Use the Sobel operator to enhance the image edge. The calculation formula of Sobel is:

[0031]

[0032] Where A is the image matrix, G x and G y are the gradients in the horizontal and vertical directions respectively;

[0033] S6. Image scaling: scaling the image to a uniform size for subsequent processing, and bilinear interpolation may be used for scaling;

[0034] S7. Standardization: standardize the image pixel values ​​to a specific range, such as 0 to 1 or -1 to 1. The standardization formula is:

[0035]

[0036] Where I is the original pixel value, I min and I max are the minimum and maximum pixel values ​​in the image, respectively.

[0037] Preferably, the image recognition module comprises the following steps:

[0038] S1. Feature extraction, using convolutional neural network CNN to automatically extract image features, the calculation formula is:

[0039] (I*K)(i,j)=∑ m ∑ n I(m,n)K(im,jn)

[0040] Where (i, j) is the position of the convolution kernel center on the image;

[0041] S2. Model training, a large number of oral image datasets with annotations are used to train the deep learning model. During the training process, the model adjusts internal parameters through an optimization algorithm to minimize the difference between the prediction and the actual annotation. The calculation formula of the gradient descent is:

[0042]

[0043] Among them, θ is the model parameter, α is the learning rate, is the gradient of the loss function with respect to the parameters;

[0044] S3. Classifier design, by adopting cross entropy loss to identify different health conditions in the oral cavity, the calculation formula is:

[0045]

[0046] where y i is the true label, p i is the probability predicted by the model;

[0047] S4. Abnormal differentiation and identification: Use the trained model to analyze the newly collected oral images and identify abnormal areas. The model output usually includes the classification results and corresponding confidence scores of each area.

[0048] S5. Result marking and visualization, mark the identified abnormal areas on the image and distinguish them with different colors or marks.

[0049] Preferably, the data analysis and diagnosis unit comprises the following steps:

[0050] S1. Data integration: integrating the abnormal area information identified by the image recognition module with the child’s oral health history data;

[0051] S2. Risk assessment: assess the risk level of children's oral health based on the type, size and location of the abnormal area, and use the weighted summation formula to calculate the risk score:

[0052]

[0053] where ω i is the weight of the ith risk factor, s i is the score of the factor;

[0054] S3. Generate diagnostic recommendations, combining expert knowledge base and existing medical guidelines to generate personalized diagnostic recommendations and treatment plans;

[0055] S4. Report output, which organizes the analysis results and diagnostic recommendations into a report, including images, text descriptions, and recommended next actions.

[0056] Preferably, the user interaction interface comprises the following steps:

[0057] S1. Login and authentication: users log in to the system using their username and password;

[0058] S2. Main interface display,After logging in, the user first enters the main interface, which provides a system overview, including the latest test results, warning status, and system notifications;

[0059] S3. Image viewing and analysis: users can view the oral images processed by the image recognition module;

[0060] S4. Diagnosis result display: the system automatically displays the abnormal area identified by the image recognition module and provides corresponding diagnosis information;

[0061] S5. Detailed report and recommendations: Users can view detailed test reports, including detailed descriptions of abnormal areas, possible health risks, and recommended treatment options;

[0062] S6. Historical data query, users can query children’s oral health history records and previous test reports;

[0063] S7. System settings and preferences: Users can access system settings to adjust personal preferences, such as notification settings and language selection;

[0064] S8. Help and support, providing help documents, FAQs or options to contact technical support;

[0065] S9. Feedback and Suggestions. Users can provide feedback or suggestions on the system to help improve system functions and services.

[0066] 8. A detection method of a child oral cavity detection system based on image recognition, comprising the following steps:

[0067] S1. The condition of the teeth and gums inside the child's mouth is measured by using an image acquisition module;

[0068] S2. Preprocess the acquired images to improve the accuracy of subsequent image recognition;

[0069] S3. Analyze the preprocessed images using a deep learning algorithm and identify abnormal areas in the oral cavity;

[0070] S4. Based on the results of image recognition and combined with the child’s oral health history data, a comprehensive analysis is conducted to provide diagnostic recommendations;

[0071] S5. Show the diagnosis results to the doctor and parents.

[0072] (III) Beneficial effects

[0073] The present invention provides a method and system for detecting children's oral cavity based on image recognition.

[0074] Beneficial effects:

[0075] 1. In the present invention, by automating the image acquisition and recognition process, the time required for manual examination by doctors is greatly reduced, and the overall detection efficiency is improved. At the same time, advanced image recognition technologies such as deep learning are used to more accurately identify abnormal areas in the oral cavity and reduce human errors.

[0076] 2. In the present invention, the detailed analysis reports and historical data comparisons provided by the system provide data support for doctors, helping them make more accurate diagnosis and treatment decisions. Automated testing reduces dependence on professional doctors and helps reduce medical costs, especially in resource-limited environments.

[0077] 3. In the present invention, by reducing invasive examinations on children, the discomfort of children is reduced and the overall patient experience is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 It is a schematic diagram of the overall process of the present invention;

[0079] Figure 2 It is a flow chart of the image acquisition module in the present invention;

[0080] Figure 3 It is a schematic diagram of the process flow of the image preprocessing unit in the present invention;

[0081] Figure 4 It is a flowchart of the image recognition module in the present invention;

[0082] Figure 5 is a flow chart of the data analysis and diagnosis unit in the present invention;

[0083] Figure 6 It is a schematic diagram of the flow chart used for the interactive interface in the present invention. DETAILED DESCRIPTION

[0084] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0085] Embodiment 1:

[0086] like Figure 1-6 As shown, an embodiment of the present invention provides a child oral cavity detection system based on image recognition, comprising:

[0087] An image acquisition module is used to capture images of the inside of a child's oral cavity, including teeth and gums;

[0088] An image preprocessing unit performs denoising and enhancement preprocessing operations on the collected images to improve the accuracy of subsequent image recognition;

[0089] An image recognition module that uses deep learning algorithms to analyze preprocessed images to identify and mark abnormal areas in the mouth, including tooth decay and gingivitis;

[0090] The data analysis and diagnosis unit conducts comprehensive analysis based on the image recognition results and combines the child’s oral health history data to provide diagnostic recommendations;

[0091] The user interface allows doctors or parents to view test results, history records, and diagnostic recommendations, and provides a system setup and operation interface.

[0092] The image acquisition module includes the following steps:

[0093] S1. Equipment preparation: Choose a high-definition camera or a dedicated intraoral camera suitable for photographing the inside of a child's oral cavity. Make sure the camera resolution is high enough to capture the details inside the oral cavity.

[0094] S2. Environment setting, adjust the light source, ensure that the optical fiber is uniform and sufficient, avoid shadows or overexposure, use sterile covers to ensure hygiene and safety;

[0095] S3. Shooting guidance: let the child take an appropriate posture, keep the head stable, and ensure that the child opens the mouth appropriately to avoid discomfort caused by excessive mouth opening;

[0096] S4. Image capture: Use a camera or endoscope to take images of the inside of the oral cavity and ensure that the images are clear and free of blur or distortion caused by movement;

[0097] S5. Image storage, stores the captured images in a high-resolution format for subsequent processing, and ensures that the naming rules of image files are clear for easy management and retrieval.

[0098] The image preprocessing unit includes the following steps:

[0099] S1. Image cropping, cropping the inner part of the oral cavity from the original image;

[0100] S2. Grayscale conversion, converting color images into grayscale images to reduce computational complexity. The calculation formula is as follows:

[0101] Gray=0.299×R+0.587×G+0.114×B

[0102] Among them, R, G, and B represent the pixel values ​​of the red, green, and blue color channels respectively;

[0103] S3. De-noising, applying a filter to remove noise in the image, where the calculation formula of Gaussian filtering is:

[0104]

[0105] Where σ is the standard deviation of the Gaussian kernel;

[0106] S4. Contrast enhancement, using histogram equalization to enhance image contrast, the formula is:

[0107]

[0108] Among them, r k is the gray level of the original image, s k is the gray level after equalization, n j is the gray level r j The number of occurrences, n is the total number of pixels in the image;

[0109] S5. Edge sharpening: Use the Sobel operator to enhance the image edge. The calculation formula of Sobel is:

[0110]

[0111] Where A is the image matrix, G x and G y are the gradients in the horizontal and vertical directions respectively;

[0112] S6. Image scaling: scaling the image to a uniform size for subsequent processing, and bilinear interpolation may be used for scaling;

[0113] S7. Standardization: standardize the image pixel values ​​to a specific range, such as 0 to 1 or -1 to 1. The standardization formula is:

[0114]

[0115] Where I is the original pixel value, I min and I max are the minimum and maximum pixel values ​​in the image, respectively.

[0116] The image recognition module includes the following steps:

[0117] S1. Feature extraction, using convolutional neural network CNN to automatically extract image features, the calculation formula is:

[0118]

[0119] Where (i, j) is the position of the convolution kernel center on the image;

[0120] S2. Model training, a large number of oral image datasets with annotations are used to train the deep learning model. During the training process, the model adjusts internal parameters through an optimization algorithm to minimize the difference between the prediction and the actual annotation. The calculation formula of the gradient descent is:

[0121]

[0122] Among them, θ is the model parameter, α is the learning rate, is the gradient of the loss function with respect to the parameters;

[0123] S3. Classifier design, by adopting cross entropy loss to identify different health conditions in the oral cavity, the calculation formula is:

[0124]

[0125] where y i is the true label, p i is the probability predicted by the model;

[0126] S4. Abnormal differentiation and identification: Use the trained model to analyze the newly collected oral images and identify abnormal areas. The model output usually includes the classification results and corresponding confidence scores of each area.

[0127] S5. Result marking and visualization, mark the identified abnormal areas on the image and distinguish them with different colors or marks.

[0128] The data analysis and diagnosis unit includes the following steps:

[0129] S1. Data integration: integrating the abnormal area information identified by the image recognition module with the child’s oral health history data;

[0130] S2. Risk assessment: assess the risk level of children's oral health based on the type, size and location of the abnormal area, and use the weighted summation formula to calculate the risk score:

[0131]

[0132] where ω i is the weight of the ith risk factor, s i is the score of the factor;

[0133] S3. Generate diagnostic recommendations, combining expert knowledge base and existing medical guidelines to generate personalized diagnostic recommendations and treatment plans;

[0134] S4. Report output, which organizes the analysis results and diagnostic recommendations into a report, including images, text descriptions, and recommended next actions.

[0135] The user interface includes the following steps:

[0136] S1. Login and authentication: users log in to the system using their username and password;

[0137] S2. Main interface display,After logging in, the user first enters the main interface, which provides a system overview, including the latest test results, warning status, and system notifications;

[0138] S3. Image viewing and analysis: users can view the oral images processed by the image recognition module;

[0139] S4. Diagnosis result display: the system automatically displays the abnormal area identified by the image recognition module and provides corresponding diagnosis information;

[0140] S5. Detailed report and recommendations: Users can view detailed test reports, including detailed descriptions of abnormal areas, possible health risks, and recommended treatment options;

[0141] S6. Historical data query, users can query children’s oral health history records and previous test reports;

[0142] S7. System settings and preferences: Users can access system settings to adjust personal preferences, such as notification settings and language selection;

[0143] S8. Help and support, providing help documents, FAQs or options to contact technical support;

[0144] S9. Feedback and Suggestions. Users can provide feedback or suggestions on the system to help improve system functions and services.

[0145] A detection method of a child's oral cavity detection system based on image recognition comprises the following steps:

[0146] S1. The condition of the teeth and gums inside the child's mouth is measured by using an image acquisition module;

[0147] S2. Preprocess the acquired images to improve the accuracy of subsequent image recognition;

[0148] S3. Analyze the preprocessed images using a deep learning algorithm and identify abnormal areas in the oral cavity;

[0149] S4. Based on the results of image recognition and combined with the child’s oral health history data, a comprehensive analysis is conducted to provide diagnostic recommendations;

[0150] S5. Show the diagnosis results to the doctor and parents.

[0151] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A children's oral cavity detection system based on image recognition, characterized in that: include: An image acquisition module is used to capture images of the inside of a child's oral cavity, including teeth and gums; An image preprocessing unit performs denoising and enhancement preprocessing operations on the collected images to improve the accuracy of subsequent image recognition; An image recognition module that uses deep learning algorithms to analyze preprocessed images to identify and mark abnormal areas in the mouth, including tooth decay and gingivitis; The data analysis and diagnosis unit conducts comprehensive analysis based on the image recognition results and combines the child’s oral health history data to provide diagnostic recommendations; The user interface allows doctors or parents to view test results, history records, and diagnostic recommendations, and provides a system setup and operation interface.

2. The method for detecting children's oral cavity based on image recognition according to claim 1, characterized in that: The image acquisition module comprises the following steps: S1. Equipment preparation: Choose a high-definition camera or a dedicated intraoral camera suitable for photographing the inside of a child's oral cavity. Make sure the camera resolution is high enough to capture the details inside the oral cavity. S2. Environment setting, adjust the light source, ensure that the optical fiber is uniform and sufficient, avoid shadows or overexposure, use sterile covers to ensure hygiene and safety; S3. Shooting guidance: let the child take an appropriate posture, keep the head stable, and ensure that the child opens the mouth appropriately to avoid discomfort caused by excessive mouth opening; S4. Image capture: Use a camera or endoscope to take images of the inside of the oral cavity and ensure that the images are clear and free of blur or distortion caused by movement; S5. Image storage, stores the captured images in a high-resolution format for subsequent processing, and ensures that the naming rules of image files are clear for easy management and retrieval.

3. The method for detecting children's oral cavity based on image recognition according to claim 1, characterized in that: The image preprocessing unit comprises the following steps: S1. Image cropping, cropping the inner part of the oral cavity from the original image; S2. Grayscale conversion: convert the color image into a grayscale image to reduce the computational complexity. The calculation formula is as follows: Gray=0.299×R+0.587×G+0.114×B Among them, R, G, and B represent the pixel values ​​of the red, green, and blue color channels respectively; S3. De-noising, applying a filter to remove noise in the image, where the calculation formula of Gaussian filtering is: Where σ is the standard deviation of the Gaussian kernel; S4. Contrast enhancement, using histogram equalization to enhance image contrast, the formula is: Among them, r k is the gray level of the original image, s k is the gray level after equalization, n j is the gray level r j The number of occurrences, n is the total number of pixels in the image; S5. Edge sharpening: Use the Sobel operator to enhance the image edge. The calculation formula of Sobel is: Where A is the image matrix, G x and G y are the gradients in the horizontal and vertical directions respectively; S6. Image scaling: scaling the image to a uniform size for subsequent processing, and bilinear interpolation may be used for scaling; S7. Standardization: standardize the image pixel values ​​to a specific range, such as 0 to 1 or -1 to 1. The standardization formula is: Where I is the original pixel value, I min and I max are the minimum and maximum pixel values ​​in the image, respectively.

4. The method for detecting children's oral cavity based on image recognition according to claim 1, characterized in that: The image recognition module comprises the following steps: S1. Feature extraction, using convolutional neural network CNN to automatically extract image features, the calculation formula is: (I*K)(i,j)=Π m Π n I(m,n)K(i-m,j-n) Where (i, j) is the position of the convolution kernel center on the image; S2. Model training, a large number of oral image datasets with annotations are used to train the deep learning model. During the training process, the model adjusts internal parameters through an optimization algorithm to minimize the difference between the prediction and the actual annotation. The calculation formula of the gradient descent is: Among them, θ is the model parameter, α is the learning rate, is the gradient of the loss function with respect to the parameters; S3. Classifier design, by adopting cross entropy loss to identify different health conditions in the oral cavity, the calculation formula is: where y i is the true label, p i is the probability predicted by the model; S4. Abnormal differentiation and identification: Use the trained model to analyze the newly collected oral images and identify abnormal areas. The model output usually includes the classification results and corresponding confidence scores of each area. S5. Result marking and visualization, mark the identified abnormal areas on the image and distinguish them with different colors or marks.

5. The method for detecting children's oral cavity based on image recognition according to claim 1, characterized in that: The data analysis and diagnosis unit comprises the following steps: S1. Data integration: integrating the abnormal area information identified by the image recognition module with the child’s oral health history data; S2. Risk assessment: assess the risk level of children's oral health based on the type, size and location of the abnormal area, and use the weighted summation formula to calculate the risk score: where ω i is the weight of the ith risk factor, s i is the score of the factor; S3. Generate diagnostic recommendations, combining expert knowledge base and existing medical guidelines to generate personalized diagnostic recommendations and treatment plans; S4. Report output, which organizes the analysis results and diagnostic recommendations into a report, including images, text descriptions, and recommended next actions.

6. The method for detecting children's oral cavity based on image recognition according to claim 1, characterized in that: The user interaction interface comprises the following steps: S1. Login and authentication: users log in to the system using their username and password; S2. Main interface display,After logging in, the user first enters the main interface, which provides a system overview, including the latest test results, warning status, and system notifications; S3. Image viewing and analysis: users can view the oral images processed by the image recognition module; S4. Diagnosis result display: the system automatically displays the abnormal area identified by the image recognition module and provides corresponding diagnosis information; S5. Detailed report and recommendations: Users can view detailed test reports, including detailed descriptions of abnormal areas, possible health risks, and recommended treatment options; S6. Historical data query, users can query children’s oral health history records and previous test reports; S7. System settings and preferences: Users can access system settings to adjust personal preferences, such as notification settings and language selection; S8. Help and support, providing help documents, FAQs or options to contact technical support; S9. Feedback and Suggestions. Users can provide feedback or suggestions on the system to help improve system functions and services.

7. A detection method of a child oral cavity detection system based on image recognition, characterized in that: The following steps are involved: S1. The condition of the teeth and gums inside the child's mouth is measured by using an image acquisition module; S2. Preprocess the acquired images to improve the accuracy of subsequent image recognition; S3. Analyze the preprocessed images using a deep learning algorithm and identify abnormal areas in the oral cavity; S4. Based on the results of image recognition and combined with the child’s oral health history data, a comprehensive analysis is conducted to provide diagnostic recommendations; S5. Show the diagnosis results to the doctor and parents.

Images