AI-assisted digital oral diagnosis system

Abstract

The invention discloses an AI-assisted digital oral diagnosis system, and belongs to the technical field of oral medical treatment. The system is composed of a data acquisition module, a preprocessing module, an AI image analysis module and a diagnosis result output module. The data acquisition module can acquire an oral X-ray film, a CT scanning image and an oral endoscope image, and can acquire a high-resolution oral CBCT image for root canal therapy. The data preprocessing module processes a root canal image through image enhancement, de-noising and gray level transformation, especially by adopting a specific algorithm. The AI image analysis module is internally provided with a deep learning model trained by a large amount of labeled data, can identify various oral disease features, and can accurately identify root canal conditions for root canal therapy. And the diagnosis result output module presents the result in an intuitive manner and generates a report. The oral disease diagnosis accuracy and efficiency are improved, the work intensity of doctors is relieved, standardized diagnosis is provided, and the oral disease diagnosis method has remarkable clinical application value.

Description

Technical Field

[0001] The present invention relates to the technical field of AI-assisted digital oral diagnosis systems, and in particular to an AI-assisted digital oral diagnosis system. Background Art

[0002] Oral diseases, such as tooth decay, periodontal disease, and inflammation of the oral mucosa, are widespread. Traditionally, the diagnosis of oral diseases relies primarily on the physician's clinical experience and visual observation, as well as routine imaging tests such as X-rays. This diagnostic approach has limitations, such as the tendency to miss early, subtle lesions, and the significant impact of individual physician experience on diagnostic accuracy. With the advancement of digital technology, digital imaging, such as oral X-rays and CT scans, is increasingly used in oral diagnosis. However, the efficient and accurate extraction of valuable information from this massive amount of imaging data to accurately identify, locate, and diagnose oral diseases remains a pressing challenge. The rapid development of AI technology has provided new ideas and methods for addressing this challenge.

[0003] This case is proposed to solve or improve the shortcomings or deficiencies of the existing technology. Summary of the Invention

[0004] The purpose of this invention is to provide an AI-assisted digital oral diagnosis system that uses AI image recognition technology to accurately identify, locate and diagnose various oral diseases such as lesions, tooth decay, periodontal disease, inflammation, etc. in the oral cavity, thereby improving the efficiency and accuracy of oral disease diagnosis.

[0005] The present invention is achieved by taking the following technical solutions:

[0006] This system includes a data acquisition module, a data preprocessing module, an AI image analysis module, and a diagnosis result output module.

[0007] The data acquisition module is used to collect digital oral imaging data from patients, including oral X-rays, CT scans, and oral endoscope images. Equipped with specialized oral imaging equipment, this module ensures high-quality image data. In root canal treatment scenarios, it prioritizes high-resolution oral CBCT images, clearly demonstrating the three-dimensional structure of the root canal.

[0008] The data preprocessing module preprocesses the collected image data, including image enhancement, denoising, and grayscale conversion, to improve image clarity and quality for subsequent AI analysis. For root canal images, a specific edge enhancement algorithm is used to highlight the outline of the root canal.

[0009] The AI image analysis module is the core module of this system, which has built-in multiple AI image recognition algorithm models based on deep learning. These models have been trained with a large amount of labeled oral disease image data and can accurately identify various disease characteristics such as tooth decay, periodontal disease, oral mucosal lesions, inflammation, etc. in the images. Among them, the model for root canal treatment identification has been trained with a large number of CBCT images with different root canal morphologies and lesion conditions. It can accurately identify the number, direction, and curvature of root canals, as well as the presence of blockages, inflammation, and other lesions in the root canals. For example, the model can accurately determine whether the root canal is calcified, determine the location and range of calcification, and provide key information for the formulation of subsequent treatment plans.

[0010] The diagnostic output module presents the AI image analysis module's analysis results to the doctor in an intuitive and easy-to-understand manner, including information such as disease type, lesion location, and severity, and generates a detailed diagnostic report. For root canal treatment, the report clearly identifies various root canal parameters, such as length and diameter, as well as the specific location and severity of the lesion. It also provides treatment options for similar cases to assist doctors in making clinical decisions.

[0011] First, the data acquisition module is used to obtain the patient's oral digital imaging data, especially for root canal treatment needs, to obtain comprehensive and clear CBCT images.

[0012] Then, the data preprocessing module preprocesses the image data and uses enhancement and denoising algorithms suitable for root canal images to improve data quality.

[0013] Next, the preprocessed image data is input into the AI image analysis module, and the root canal treatment recognition model analyzes the image to identify various characteristics of the root canal and possible lesions.

[0014] Finally, the diagnosis result output module generates a diagnosis report specifically for root canal treatment based on the output results of the AI image analysis module and provides it to the doctor.

[0015] The advantages and positive effects of the present invention are:

[0016] 1. Improved diagnostic accuracy: By using AI image algorithms to learn and analyze large amounts of imaging data, it can identify early-stage subtle lesions that are difficult for the human eye to detect, reducing missed diagnoses and misdiagnoses, and improving the accuracy of oral disease diagnosis. During root canal treatment, it can accurately detect subtle root canal lesions and complex root canal anatomy, avoiding missed treatments.

[0017] 2. Improved diagnostic efficiency: Compared to traditional manual diagnostic methods, this system can rapidly process and analyze oral imaging data, significantly shortening diagnostic time and improving medical efficiency. In root canal imaging analysis, root canal parameter measurement and lesion assessment can be completed quickly, saving doctors time in reading images.

[0018] 3. Reduced workload for doctors: AI technology takes on some of the image analysis work, reducing the workload for doctors and allowing them to focus on more important tasks such as developing patient treatment plans. Doctors no longer need to spend a lot of time manually measuring root canal data and identifying subtle lesions.

[0019] 4. Providing standardized diagnoses: AI models, based on unified algorithms and trained on extensive data, can provide relatively standardized diagnostic results, reducing diagnostic inconsistencies caused by individual physician experience. For root canal treatments, different physicians can reach a more consistent diagnosis based on standardized AI-generated reports. DETAILED DESCRIPTION

[0020] If there are terms related to directional indications or positional relationships in the embodiments of this application (such as up, down, left, right, front, back, inside, outside, top, bottom, center, vertical, horizontal, longitudinal, transverse, length, width, counterclockwise, clockwise, axial, radial, circumferential, etc.), such terms are only used to explain the relative positional relationship, movement, etc. between the components in a specific posture; if the specific posture changes, the directional indication or positional relationship will also change accordingly. In addition, the terms "first" and "second" in the embodiments of this application are only used for descriptive convenience and should not be understood as indicating or implying relative importance.

[0021] Data collection

[0022] High-precision dental X-ray machines are used to capture oral X-rays of patients, ensuring clear images of structures such as teeth and periodontal tissues. An oral CT scanner is also used to obtain three-dimensional CT images of the patient's oral cavity, providing a more comprehensive view of the internal oral structure. Furthermore, an oral endoscope is used to capture images of the oral mucosa and other areas. To collect data related to root canal treatments, advanced CBCT equipment is used, with appropriate scanning parameters, such as high-resolution mode, to ensure a clear representation of the root canal system and obtain complete root canal imaging data from all angles.

[0023] Data preprocessing

[0024] Image enhancement algorithms, such as histogram equalization, are used to enhance image contrast and make lesions more visible. Denoising algorithms are used to remove noise interference from images and improve image quality. For grayscale images, grayscale transformation is performed as needed to adjust the image's grayscale distribution. For root canal images, a specialized image sharpening algorithm is used to enhance the clarity of root canal boundaries. A wavelet-based denoising method is also used to remove noise generated during the scanning process, ensuring the accuracy of subsequent AI analysis.

[0025] AI model training

[0026] A large amount of imaging data containing various oral diseases is collected and annotated by professional dentists. The annotations include information such as disease type, lesion location, and severity. The annotated data is divided into training, validation, and test sets. The training set is used to train the AI image recognition model, and the model parameters are continuously adjusted to improve the accuracy and generalization ability of the model. The training process is monitored through the validation set to prevent model overfitting. Finally, the trained model is evaluated using the test set to ensure that the model performance meets clinical diagnostic requirements. In the training of the root canal treatment recognition model, root canal imaging data of different age groups, different tooth positions, and different lesion severity are collected. Senior dentists carefully annotate information such as the number of root canals, morphology, lesion location, and nature. Using a convolutional neural network (CNN) architecture, the training set is repeatedly trained by continuously adjusting parameters such as the number of network layers and the number of nodes. The validation set is used to evaluate the model performance in real time to prevent overfitting, and ultimately a root canal treatment recognition model with excellent performance is obtained.

[0027] Diagnostic process

[0028] The collected and pre-processed oral imaging data is input into the trained AI image analysis module, and the model analyzes the images and outputs the disease diagnosis results. The diagnosis result output module presents the results to the doctor in a visual manner, such as marking the location of the lesion on the image, displaying information such as the name and severity of the disease, and generating a detailed diagnosis report. The doctor conducts further diagnosis and treatment of the patient based on the diagnosis report and his own clinical experience. In the root canal treatment scenario, the doctor inputs the patient's CBCT imaging data into the system, and the root canal treatment recognition model quickly analyzes the image, and displays the root canal structure in a three-dimensional visual form in the diagnosis report, marking the normal and diseased areas with different colors, and listing in detail the parameters such as the root canal length, diameter, and curvature angle, as well as the specific description and degree classification of the lesion. The doctor formulates an accurate root canal treatment plan based on the content of the report and the actual situation of the patient.

[0029] It should be emphasized that the embodiments described in the present invention are illustrative rather than restrictive. Therefore, the present invention is not limited to the embodiments described in the specific embodiments. Any other embodiments derived by those skilled in the art based on the technical solutions of the present invention also fall within the scope of protection of the present invention.

Claims

1. An AI-assisted digital oral diagnosis system, characterized in that: include: The data acquisition module is used to collect digital oral imaging data of patients, including oral X-rays, CT scans, and oral endoscope images. For root canal treatment scenarios, it can obtain high-resolution oral CBCT images to clearly present the three-dimensional structure of the root canal; The data preprocessing module performs image enhancement, denoising, and grayscale conversion on the collected image data, and uses a specific edge enhancement algorithm to highlight the root canal contour for root canal images; The AI image analysis module has built-in multiple AI image recognition algorithm models based on deep learning. Trained with a large amount of labeled oral disease image data, it can identify disease characteristics such as tooth decay, periodontal disease, oral mucosal lesions, and inflammation. For root canal treatment, it has a model that can accurately identify the number, direction, curvature of root canals, as well as root canal blockage, inflammation, and other lesions. The diagnosis result output module presents the results of the AI image analysis module to the doctor in an intuitive manner, including disease type, lesion location, and severity information, and generates a detailed diagnosis report. It marks the root canal parameters and lesion conditions in the root canal treatment report and provides a reference for treatment plans for similar cases.

2. The AI-assisted digital oral diagnosis system according to claim 1, characterized in that: The oral imaging acquisition equipment in the data acquisition module is a high-precision oral X-ray machine, an oral CT scanner and an oral endoscope. For the acquisition of root canal treatment-related data, the selected CBCT equipment can be set to a high-resolution scanning mode to obtain complete root canal imaging data.

3. The AI-assisted digital oral diagnosis system according to claim 1, characterized in that: The data preprocessing module adopts an image sharpening algorithm for root canal images to enhance the clarity of root canal boundaries, and a denoising method based on wavelet transform is used to remove scanning noise to ensure the accuracy of AI analysis.

4. The AI-assisted digital oral diagnosis system according to claim 1, characterized in that: The model for root canal treatment identification in the AI image analysis module adopts a convolutional neural network (CNN) architecture. By collecting root canal imaging data of different age groups, different tooth positions, and different lesion severity, and having senior dentists annotate information such as the number, morphology, lesion location and nature of the root canals, the model is repeatedly trained with the training set and evaluated in real time with the validation set to prevent overfitting, thereby obtaining a model with excellent performance.

5. The AI-assisted digital oral diagnosis system according to claim 1, characterized in that: The diagnosis result output module displays the root canal structure in a three-dimensional visual form for the root canal treatment diagnosis report, marks the normal and diseased areas with different colors, and lists in detail the parameters such as the root canal length, diameter, bending angle, and the specific description and severity classification of the disease.

6. A digital oral disease diagnosis method based on AI image algorithm, using the AI-assisted digital oral diagnosis system according to any one of claims 1 to 5, characterized in that: The following steps are involved: Step 1: Acquire digital oral imaging data of the patient through the data acquisition module in claim 6, especially obtain comprehensive and clear CBCT images for root canal treatment needs; Step 2: Using the data preprocessing module in claim 1 to perform enhancement and denoising algorithms suitable for root canal images on the image data to improve data quality; Step 3: Input the pre-processed image data into the AI image analysis module in claim 1, and analyze the image by the root canal treatment recognition model to identify various root canal features and possible lesions; Step 4: Through the diagnosis result output module in claim 1, according to the output results of the AI image analysis module, a diagnosis report specifically for root canal treatment is generated and provided to the doctor.

7. The digital oral disease diagnosis method based on AI image algorithm according to claim 6, characterized in that: In the data acquisition step, high-precision oral X-ray machines, oral CT scanners, and oral endoscopes are used to collect image data. Advanced CBCT equipment is selected for root canal treatment and appropriate scanning parameters are set to obtain root canal system images.

8. The digital oral disease diagnosis method based on AI image algorithm according to claim 7, characterized in that: In the AI model training step, a large amount of imaging data containing various oral diseases is collected. Professional dentists annotate the disease type, lesion location, severity and other information, and divide the data into training set, validation set and test set. The training set is used to train the AI image recognition model, and the validation set is used to monitor the training process to prevent overfitting. The test set is used to evaluate the model performance. A convolutional neural network architecture is used to train and optimize the root canal treatment recognition model.